Document 2767

OMEGA, Vol. 61(4) 269-271, 2010
THE DUAL PROCESS MODEL OF COPING WITH
BEREAVEMENT: A DECADE LATER
VIRGINIA E. RICHARDSON, PH.D., Guest Editor
The Ohio State University, Columbus
PREFACE
Almost exactly 1 decade ago, Dr. Shirley O’Bryant, professor emeritus in human
ecology, asked me to oversee her data and students focusing on older bereaved
men. Dr. O’Bryant had received funding from the AARP Andrus Foundation to
interview 200 older widowers residing in the Central Ohio area who were in their
second year of bereavement. Based on these data, Dr. Shantha Balaswamy and
I published the first evaluation of Margaret Stroebe and Henk Schut’s newly
proposed Dual Process Model of Bereavement (DPM) in the article, “Coping
with Bereavement among Elderly Widowers,” published in Omega: Journal of
Death and Dying in 2001. We had discovered Stroebe and Schut’s paper, “The
Dual Process Model of Coping with Bereavement, Rationale and Description,”
published in Death Studies in 1999, and we were intrigued with their challenge
to “the grief work hypothesis,” which was the most widely accepted viewpoint on
loss and bereavement in popular and scientific journals. The grief work hypothesis
claimed that bereaved persons must focus on their feelings of loss or they
will experience psychosomatic and other maladaptive symptoms and will never
recover from their loss. Bereaved persons must confront their painful feelings
and “work through them” in order to avoid developing disordered grief reactions,
according to proponents of this perspective.
Stroebe and Schut (1999) identified several problems with the grief work
hypothesis, including lack of solid supporting evidence, inadequate clarity of
the concept and processes, and inaccurate operationalization of concepts. By
269
Ó 2010, Baywood Publishing Co., Inc.
doi: 10.2190/OM.61.4.a
http://baywood.com
270 / RICHARDSON
integrating precepts from chronic stress theory (Lazarus & Folkman, 1984) and
Horowitz’s Stress Response Syndrome (Horowitz, 1986), Stroebe and Schut
proposed the DPM as an alternative framework with three central constructs:
loss-oriented coping, restoration-oriented coping, and oscillation. Dr. Balaswamy
and I decided to test this approach using Dr. O’Bryant’s sample of older widowers.
As reported in Omega: Journal of Death and Dying in 2001, we found that loss
variables, such as death circumstances, influenced negative feelings and were
especially critical during the early stages of bereavement, while restorationoriented factors, such as dating, were significantly associated with positive affect
and were more relevant later. Although our findings supported the DPM, we
recognized several limitations in our research: our data were cross-sectional, the
operationalization of constructs was questionable, and our sample was limited
to a self-selected group of older bereaved men.
When the National Institute of Aging awarded Dr. Deborah Carr and her
colleagues funds to organize a workshop on the Changing Lives of Older Couples
(CLOC) data at the University of Michigan during the spring of 2002, I jumped
at the opportunity to test the DPM using a longitudinal research design comprised of older widows and widowers. My analyses of the CLOC data, which were
published in the Journal of Gerontological Social Work in 2007, corroborated
our previous findings reported in Omega: Journal of Death and Dying in 2001.
Meanwhile others were also testing the DPM. Dale Lund, Michael Caserta,
Rebecca Utz, and Brian de Vries received funding from the National Institute
of Aging to compare a DPM-intervention with a comparison group receiving
traditional support (Lund, Caserta, de Vries, & Wright, 2004). Kate Bennett and
her colleagues differentiated copers from noncopers (Bennett, Hughes, & Smith,
2005), and M. Katherine Shear, with support from the National Institute of Mental
Health, used a randomized control trial to evaluate an intervention for complicated
grief using DPM principles (Shear, Frank, Houck, & Reynolds, 2005).
We organized a symposium on the DPM for the annual meeting of the Gerontological Society of America in 2008 and presented our most recent research.
The papers from the symposium comprise this special issue of Omega. We are
grateful to Ken Doka, editor of Omega: Journal of Death and Dying, for providing
us with this opportunity to disseminate our findings on this important model.
We launch the issue with a special contribution from Stroebe and Schut, who
present the introductory article entitled, “The Dual Process Model of Coping
with Bereavement: A Decade On.” This is followed by three empirical studies:
Lund, Caserta, Utz, & de Vries’ article, “Experiences and Early Coping of
Bereaved Spouses/Partners in an Intervention Based on the Dual Process Model
(DPM)”; Bennett, Gibbons, & Mackenzie-Smith’s article, “Loss and Restoration
in Later Life: An Examination of the Dual Process Model of Coping with
Bereavement; and my article, “Length of Caregiving and Well-Being among
Older Widowers: Implications for the Dual Process Model of Bereavement.”
These are followed by Shear’s article, “Exploring the Role of Experiential
DUAL PROCESS MODEL OF COPING WITH BEREAVEMENT /
271
Avoidance from the Perspective of Attachment Theory and the Dual Process
Model,” and Carr’s concluding article, “ New Perspectives on the Dual Process
Model (DPM): What Have We Learned? What Questions Remain?” Carr’s
article is based on her comments as a reactor during the symposium. As we
continue our research on the DPM, we look forward to receiving feedback from
clinicians and academicians and from older bereaved persons who are most
directly affected by our ideas.
REFERENCES
Bennett, K. L., Hughes, G. M., & Smith, P. T. (2005). Psychological response to later life
widowhood: Coping and the effects of gender. Omega: Journal of Death and Dying,
51, 33-52.
Horowitz, M. (1986). Stress response syndromes. Northvale, NJ: Aronson.
Lazarus, R., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.
Lund, D. A., & Caserta, M. S., de Vries, B., & Wright, S. (2004). Restoration during
bereavement. Generations Review, 14, 9-15.
Richardson, V. E., & Balaswamy, S. (2001). Coping with bereavement among elderly
widowers. Omega: Journal of Death and Dying, 43, 129-144.
Richardson, V. E. (2007). A dual process model of grief counseling: Findings from the
changing lives of older couples (CLOC) study. Journal of Gerontological Social Work,
48, 311-329.
Shear, K., Frank, E., Houck, P., & Reynolds, C. (2005). Treatment of complicated grief:
A randomized controlled trial. Journal of the American Medical Association, 293,
2601-2608.
Stroebe, M. S., & Schut, H. (1999). The Dual Process Model of coping with bereavement:
Rationale and description. Death Studies, 23, 197-224.
Direct reprint requests to:
Virginia E. Richardson
The Ohio State University
College of Social Work
1947 College Road
Columbus, OH 43210
e-mail: [email protected]
DEATH, VALUE AND MEANING SERIES
John D. Morgan, Series Editor
Editors: Gerry R. Cox, Robert A. Bendiksen, and Robert G. Stevenson
Spirituality is clearly related to identity. It is a search to
discover our personal make-up: what we are all about, our
innate identity. Letting go of who we are, and what we have
acquired for ourselves, has become such an enterprise that, in
the face of death, deriving meaning out of what is left becomes
an essential project. It is a venture into terra incognita. Yet, this
apparent void may prove to be the beginning of a very creative
process. When confronted by death, the believer finds that
religion can offer a profound sense of direction.
The editors of Making Sense of Death: Spiritual, Pastoral, and Personal Aspects of Death,
Dying and Bereavement provide stimulating discussions as they ponder the meaning of
life and death. This anthology explores the process of meaning-making in the face of
death and the roles of religion and spirituality at times of loss; the profound and
devastating experience of loss in the death of a spouse or a child; a psychological
model of spirituality; the dimensions of spirituality; humor in client-caregiver
relationships; the worldview of modernity in contrast to postmodern assumptions; the
Buddhist perspective of death, dying, and pastoral care; meaning-making in the virtual
reality of cyberspace; individualism and death; the historical context of Native
Americans, the concept of disenfranchised grief, and its detailed application to the
Native American experience; a qualitative survey on the impact of the shooting deaths
of students in Colorado; a team approach with physicians, nursing, social services,
and pastoral care; a study of health care professionals, comparing clergy with other
health professionals; marginality in spiritual and pastoral care for the dying; a
qualitative research study of registered nurses in the northeast United States; and loss
and growth in the seasons of life.
6" × 9", 260 Pages, Cloth, ISBN 0-89503-249-X
$57.95 + $7.00 p/h in U.S. (please inquire for postage rates outside of U.S.)
BAYWOOD PUBLISHING COMPANY, INC.
26 Austin Avenue, P.O. Box 337, Amityville, NY 11701
phone (631) 691-1270 • fax (631) 691-1770 • toll-free orderline (800) 638-7819
e-mail [email protected] • website http://baywood.com
OMEGA, Vol. 61(4) 273-289, 2010
THE DUAL PROCESS MODEL OF COPING
WITH BEREAVEMENT: A DECADE ON*
MARGARET STROEBE
HENK SCHUT
Utrecht University, The Netherlands
ABSTRACT
The Dual Process Model of Coping with Bereavement (DPM; Stroebe &
Schut, 1999) is described in this article. The rationale is given as to why
this model was deemed necessary and how it was designed to overcome
limitations of earlier models of adaptive coping with loss. Although building
on earlier theoretical formulations, it contrasts with other models along a
number of dimensions which are outlined. In addition to describing the basic
parameters of the DPM, theoretical and empirical developments that have
taken place since the original publication of the model are summarized.
Guidelines for future research are given focusing on principles that should
be followed to put the model to stringent empirical test.
It is gratifying to know that precisely a decade after the publication of our
Dual Process Model of Coping with Bereavement (DPM; Stroebe & Schut, 1999),
interest in this model has grown to the extent that a Special Issue in Omega:
Journal of Death and Dying is deemed appropriate. Within the past decade various
research teams have taken up the challenge both to apply and to test the model,
as represented in the following articles in this Special Issue. A scientific model
*Parts of this article have been adapted and updated (with permission) from a more detailed
review of the DPM by Stroebe and Schut in Grief Matters: The Australian Journal of Grief
and Bereavement, 2008, 11, 1-4.
273
Ó 2010, Baywood Publishing Co., Inc.
doi: 10.2190/OM.61.4.b
http://baywood.com
274 / STROEBE AND SCHUT
of this kind is no use unless it can be implemented, and being firm empiricists
ourselves, we have always considered it essential to put the model to the test.
In our view, this should be undertaken first and foremost by research teams
independent of our own, and this, fortunately, is what has occurred. The articles
in this volume represent an excellent variety of different types of investigation
of the model’s parameters.
To understand the significance of these contributions, some background information about the DPM is necessary. Thus, in this article, we first describe the
rationale for the DPM and summarize its main parameters. We compare and
contrast it with other models along a number of dimensions. Throughout, we
indicate where further developments and empirical examinations have taken place
during the past decade. We end with suggestions for future research directions,
focusing on principles that should be followed to put the model to stringent
empirical test.
RATIONALE FOR DEVELOPMENT OF THE DPM
Our purpose in developing the DPM was to provide a model that would
better describe coping and predict good versus poor adaptation to this stressful
life event, and by doing so, to better understand individual differences in the
ways that people come to terms with bereavement. It is a model, then, of coping
with loss, not a generic model aimed at explaining the broad range of phenomena
and manifestations associated with bereavement. Coping refers to processes,
strategies, or styles of managing (reducing, mastering, tolerating) the situation
in which bereavement places the individual. Coping is assumed to impact on
adaptation to bereavement. If coping is effective, then not only the suffering,
but also the mental and physical ill health difficulties that are associated with
bereavement (Stroebe, Schut, & Stroebe, 2007) should be reduced (in time,
usually after quite some struggle and turmoil). To understand its impact on
outcome, coping must be considered a separate entity from the consequences of
bereavement: the former is a process, the latter an outcome variable. Thus (and
we return to this later) it becomes essential to differentiate coping (process)
from consequences (outcomes) in our empirical investigations. Overall, then,
in constructing the DPM, the aim was to postulate regularities in coping processes
that are predictive of (non)adaptive outcomes.
At the time when the DPM was developed, there were a number of models
available that addressed how people go about coming to terms with bereavement
(for a review, see Stroebe, in press). Each of these provided guidelines to understand what needs to be done for successful outcome to occur. Among these, the
most influential and appealing coping models in the bereavement area during
the latter part of the 20th century were the Phase Model (e.g., Bowlby, 1980,
see also Parkes, 1972/1996), which was fundamental to attachment theory, and
the Task Model (Worden, 1982, 1991, 2002, 2009), which became prominent
DUAL PROCESS MODEL: A DECADE ON /
275
in the planning of counseling and therapy programs for bereaved people in
need of help. Working through grief (known as “grief work,” following Freud,
1917/1957) was a fundamental notion underlying the development of both the
phases and tasks that are integral to these models (see Figure 1, left hand and
middle columns). Grief work is understood to refer to the cognitive process of
confronting the reality of a loss through death, of going over events that occurred
before and at the time of death, and of focusing on memories and working toward
detachment from (or relocating) the deceased (Stroebe, 1992). Following this,
it was understood that one has to confront the experience of bereavement in
order to come to terms with loss and avoid detrimental health consequences. It
was in large part due to consideration of these grief work models—with respect
to their major strengths but also their perceived limitations—that led us to
develop the DPM. As such, and as will become evident, the DPM built on and
extended these earlier conceptualizations.
Despite the useful guidelines that these prior models evidently offered, we
had major concerns about the adequacy of their central construct of grief work in
explaining adaptive ways of coping with bereavement. First, there are alternative
ways of coming to terms with bereavement. As examinations of certain nonWestern cultural patterns of grieving show, the types of confrontation involved in
grief work are not universal, nor is non-confrontation systematically linked with
mal-adaptation. Second, the process itself as described in the Phase Model (far
less so in the Task Model) seems somewhat passive (as though the person is being
put through, rather than actively dealing with), neglecting the effortful struggle
that is so much part of grieving. Third, there is no acknowledgment of the need
for “dosage” of grief. It is arduous and exhausting to grieve, respite at times is
recuperative. Fourth, the benefits of denial have not been taken into account
(cf. Bonanno, 2001). Fifth, the grief work notion focuses on the loss of the loved
person him- or herself, neglecting the possibility that there may be other sources of
stress that arise indirectly following a bereavement (e.g., concerns with finances,
legal matters, or upbringing of children as a single parent). Furthermore, our own
research had failed to show evidence that persons who were doing grief work
adjusted better (W. Stroebe, Schut, & Stroebe, 2005). Finally, different types
of “working through” appeared to help different subgroups (Schut, Stroebe,
de Keijser, & van den Bout, 1997), suggesting the need for a more nuanced
approach to understanding effective coping. Our conclusion was that the grief
work model needed revision to define when and for whom—and in what way—
working through is efficacious. 1 This reasoning led to the DPM.
1 Bonanno and Kaltman’s (1999) integrative perspective on bereavement also replaced
what they viewed as a too-narrow focus on the grief work hypothesis (highlighting a lack of
empirical evidence). For a critical appraisal of their approach in comparison with the DPM,
see Archer (2001).
Figure 1. The Dual Process Model of Coping with Bereavement (Stroebe & Schut, 1999).
276 / STROEBE AND SCHUT
DUAL PROCESS MODEL: A DECADE ON /
277
DESCRIPTION OF THE DPM:
MAIN PARAMETERS
The DPM can be understood as a taxonomy to describe ways that people come
to terms with the loss of a loved one (for detailed accounts, see Stroebe & Schut,
1999, 2001). The structure of the DPM parameters owes much to Cognitive Stress
Theory (see Folkman, 2001; Lazarus & Folkman, 1984), particularly insofar
as it defines a number of key components related to coping. These are: stressors
(the nature of the events leading to stress, i.e., the cause, reason for the coping
process to be set in motion); appraisal processes (assessment of threat); coping
processes (ways of dealing with threat); and outcome variables (e.g., mental and
physical health indices). A fundamental contrast with earlier models is that the
DPM defines two categories of stressors associated with bereavement, namely,
those that are loss- versus restoration-oriented. Loss-orientation refers to the
bereaved person’s concentration on, appraising and processing of some aspect
of the loss experience itself and as such, incorporates grief work. It involves a
painful dwelling on, even searching for the lost person, a phenomenon that lies
at the heart of grieving. Restoration-orientation refers to the focus on secondary
stressors that are also consequences of bereavement, reflecting a struggle to
reorient oneself in a changed world without the deceased person. Rethinking and
replanning one’s life in the face of bereavement (a part of restoration orientation)
can also be regarded an essential component of grieving (cf. Parkes’s psychosocial transition theory, e.g., 2006). Caserta and Lund (2007) were able to demonstrate that attention was paid to both types of stressors among a sample of widowed
persons, and to indicate that these were related to bereavement outcomes, as
did Wijngaards, Stroebe, Stroebe, Schut, van den Heijden, et al. (2008).
It is important to note that loss- and restoration-oriented coping are not equivalent to the Cognitive Stress Theory concepts of emotion- and problem-focused
coping (cf. Billings & Moos, 1981, 1984; Folkman, 2001), although at first sight,
one might think that emotion-focused coping seems more loss-oriented, problemfocused coping more restoration-oriented. Emotion-focused coping is directed at
managing the emotion that results from stress, problem-focused coping is directed
at managing and changing the problem causing the distress (Folkman, 2001;
Lazarus & Folkman, 1984). Indeed, some aspects to do with loss orientation may
be better dealt with in an emotion-focused manner (e.g., unchangeable things,
such as relating to the fact that the deceased cannot be brought back), but other
loss-related experiences can also be dealt with in a problem-focused manner
(e.g., to keep the deceased close, one can plant and nurture a tree in his/her
memory). Likewise, both emotion- and problem-focused strategies can be
employed in coping with restoration stressors. For example, consider the need to
repair the financial situation following loss of a spouse’s income: Either one
can deal with this in an emotion-focused way by worrying and feeling anxious
but doing nothing about it, or one can take steps to solve the problem by earning
278 / STROEBE AND SCHUT
money oneself. It becomes evident that many loss- and restoration-stressors can
be dealt with either in an emotion- or a problem-focused manner (and in fact,
both types are typically used: sometimes the former, sometimes the latter will be
more deemed appropriate when dealing with loss- or restoration-stressors).
Both orientations are sources of stress and can be associated with outcomes
such as distress and anxiety. Both are also involved in the coping process, for
example, they are attended to (confronted versus avoided) in varying degrees
(according to individual and cultural variations). The process of attending to or
avoiding these two types of stressor is dynamic and fluctuating, and it also changes
over time. Therefore, the DPM specifies a dynamic coping process, namely, a
regulatory process labeled oscillation, which distinguishes it from the earlier
bereavement models (and also from the more generic cognitive stress theory).
The principle underlying oscillation is that at times the bereaved will confront
aspects of loss, at other times avoid them, and the same applies to the tasks of
restoration. Sometimes, too, there will be “time out,” when the person is not
grieving. Coping with bereavement according to the DPM is thus a complex
regulatory process of confrontation and avoidance. An important postulation
of the model is that oscillation between the two types of stressors is necessary
for adaptive coping. The structural components described above are depicted
in Figure 1. Table 1 compares the DPM with the previous models, illustrating
restoration stressors that need to be dealt with in addition to those postulated in
the Task Model.
At this point in the development of the model, the primary strategy of coping
with the loss and restoration stressors was understood to relate to emotion
Table 1. Comparison of Models
Phase Model
(Bowlby, 1980)
Task Model
(Worden, 1991)
DPM
(Stroebe & Schut, 1999)
Shock
Accept reality of
loss
Accept reality of loss . . . and
accept reality of changed world.
Yearning/protest
Experience pain
of grief
Experience pain of grief . . . and
take time off from pain of grief.
Despair
Adjust to life
without deceased
Adjust to life without deceased . . .
and master the changed
(subjective) environment.
Restitution
Relocate deceased
emotionally and
move on
Relocate deceased emotionally
and move on . . . and develop
new roles, identities, relationships.
DUAL PROCESS MODEL: A DECADE ON /
279
regulation, or more precisely, to confrontation versus avoidance (a major coping
dimension in Cognitive Stress Theory). We extended the original DPM model to
include further analysis of types of cognitive processing in a subsequent paper
(Stroebe & Schut, 2001), as depicted in Figure 2. Oscillation between positive
and negative affect/(re)appraisal is understood to be an integral part of the
coping process, and to be a component of both loss- and restoration-oriented
coping. Persistent negative effect enhances grief, yet working through grief, which
includes rumination, has been identified as important in coming to terms with loss.
On the other hand, positive reappraisals sustain the coping effort. Yet if positive
states are maintained relentlessly, grieving is neglected. We drew on the work
of Folkman (2001) on positive meaning states, and of Nolen-Hoeksema (2001) on
negative appraisals to introduce cognitive pathways into the model.
We continue to believe that processes of confrontation-avoidance are central
mechanisms in adjustment to bereavement. One line of our current research is
directed toward gaining finer-grained understanding of the types of loss- and
restoration-oriented cognitions which are associated with normal versus complicated forms of grieving, focusing on rumination as an avoidance process (see
Stroebe, Boelen, van den Hout, Stroebe, Salemink, & van den Bout, 2007).
Relatedly, Boelen and van den Bout (in press) have used their CognitiveBehavioural Model of Complicated Grief (Boelen, van den Hout, & van den Bout,
2006) to examine assumptions about the role of two types of avoidance in
complicated grief, ones which they see as comparable with avoidance of lossand restoration-orientations in our own conceptualization. They described and
empirically-tested the role of “anxious avoidance” (avoidance of confrontation
with the reality of the loss) and “depressive avoidance” (avoiding engagement in
activities that could foster adjustment), dimensions that are clearly compatible
with the DPM constructs. Anxious and depressive avoidance emerged as distinct
factors and accounted for unique parts of explained variance in grief symptomatology. In our view, their results provide indirect support for our proposition
that the two types of stressor are distinct and relevant to adjustment: difficulty in
dealing with them is associated with poor outcome. Boelen and van den Bout
(in press) provided a fine-grained analysis of pathways between process and
outcome variables, to which we return later.
THE DPM IN COMPARISON WITH
OTHER MODELS
Already above, it has become evident that—rather unusually in bereavement research—the DPM draws heavily on pre-existing generic as well as
bereavement-specific theories for derivation of its parameters. We consider it a
strength that the DPM integrates major theoretical perspectives such as attachment
and cognitive stress theories. But does the DPM add substantially to previous
models, notably, the Phase and Task Models? Worden (2009), in the 4th edition
Figure 2. Appraisal processes in the Dual Process Model (cf. Stroebe & Schut, 2001).
280 / STROEBE AND SCHUT
DUAL PROCESS MODEL: A DECADE ON /
281
of his monograph, actually argued that there is little difference between his
Task Model and the DPM, noting that the tasks are “almost identical” with our
stressors. This is more the case for his revised, post-DPM Task Model. Worden’s
tasks have been reformulated over the years since the DPM was first published,
and they are now more explicitly inclusive of restoration stressors, and more in
line with our additions in Table 1. Worden’s (2009) Task 2 is now: “To process
the pain of grief”; Task 3: “To adjust to a world without the deceased”; Task 4:
“To find an enduring connection with the deceased in the midst of embarking
on a new life” (p. 50). Even though, as evident in these new tasks, there is now
more similarity with the DPM, we consider our explicit distinction of the two
categories of loss- and restoration-stressors to be unique, to reflect the reality
that bereaved people experience, and to be useful for both clinical application
and research investigation.
There are a number of other distinguishing features between the models.
Worden does not distinguish stressors from coping processes, as we do, or make
predictions about (mal)adjustment in relationship specifically to these different
component parts. Furthermore, Worden argued that there is similarity in that,
according to his model, people go back and forth between his tasks as needed,
like our principle of oscillation. Consider oscillation: Although shifting back
and forth between the tasks is implicit in his model, merely stating this fails
to postulate emotion regulation (confrontation and avoidance) as a fundamental
parameter of coping, as we do in our model: The principle of oscillation captures
and highlights the necessity for attention to the different categories of stressors.
Finally, contrary to what Worden claims, our model posits flexibility (it is
fundamental to oscillation), certainly not the “fixed patterns” that he mentions
in considering our model, and it caters for individual and subgroup differences
(see below).
There are a number of additional respects in which we consider the DPM to
add to other formulations, the major ones are briefly described next (for more
details, see Stroebe & Schut, 1999, 2008). First, the DPM provides a framework
for understanding forms of complicated grief, such as chronic, or absent, delayed,
inhibited grief (cf. Lindemann, 1944; Parkes & Weiss, 1983) in a way that was
not nearly so differentiated or explicit in the previous models, with chronic
grievers focusing on loss, absent grievers on restoration-oriented activities, while
those who suffer a complicated form of traumatic bereavement might be expected
to have trouble alternating smoothly between loss- and restoration-orientation,
manifesting extreme symptoms of intrusion and avoidance (for details, see
Stroebe & Schut, 2008). It is important to note that in both loss-oriented
(e.g., chronic) and restoration-oriented (e.g., absent) types of complicated
grief, reactions are extreme, with extensive focus on the one orientation and
avoidance of the other. Such patterns are associated with an absence of the type
of confrontation-avoidance (oscillation) that we have described as characteristic
of “normal” coping with bereavement. In general, there are substantial individual
282 / STROEBE AND SCHUT
differences in the extent to which (normally) bereaved persons focus on lossor restoration-orientation: only in extreme cases of confrontation of the one,
and avoidance of the other are complications in grieving and poor adaptation
likely to occur.
In recent publications, the relationship between complicated grief and patterns
of attachment has been explored within the context of the DPM (e.g., Mikulincer
& Shaver, 2008; Parkes, 2006; M. Stroebe, Schut, & W. Stroebe, 2005). In
fact, one of the most important developments of the DPM has emerged from
exploration of its links with attachment theory constructs (e.g., Mikulincer &
Shaver, 2008; Parkes, 2006; Stroebe, Stroebe, & Schut, 2005; Stroebe, Schut,
& Stroebe, 2005a), and some empirical evidence has accumulated in support
of these connections in the meantime (see Stroebe & Schut, 2008). We have also
explored how attachment style differences influence patterns of disclosure in
coping with bereavement (see Stroebe, Schut, & Stroebe, 2005b).
Turning next to (sub)group differences: As described in Stroebe and Schut
(1999), the DPM also accommodated male and female differences in ways of
grieving better than the previous models described above and, more recently,
we have explored gender differences in coping with bereavement in relationship
to health outcomes, using the DPM framework (see Stroebe, Stroebe, & Schut,
2001). Women appear to be more loss-oriented following bereavement, feeling
and expressing their distress at their loss; men more restoration-oriented, actively
engaging with the problems and practical issues associated with loss (Wijngaards
et al., 2008; cf. Parkes, 2006). Again it is important to note that focusing on lossorientation among women and restoration-orientation among men may generally
work well, unless one or other orientation is adhered to in the extreme (indicated
by an absence of oscillation to the other orientation).
Contrary to subsequent criticism that the DPM is an intra-personal model (like
the preceding models), in our 1999 paper we already described interpersonal
coping processes that the model has the potential to incorporate, acknowledging
that one person’s way of grieving impacts on that of another. The gender differences described above provide an illustration: if, say, a bereaved father is
more restoration-oriented, a mother more loss-oriented, attributions may be made
in terms of differences in extremity of grief, for example, a mother might assume
“he is grieving less than I am” rather than what may actually be the case, that
“he is grieving differently.” Making the former attribution could negatively
impact the couple’s adjustment to bereavement. More recent evidence regarding
the influence of such interpersonal coping processes from a DPM perspective
has been found: Wijngaards et al. (2008) used the DPM framework to examine
the relationship between a bereaved parent’s own and their partner’s way of
coping in relationship to their adjustment to the death of their child. Interpersonal
factors were indeed found to play a part in coping and adjustment. For example,
one of the main findings was that, for fathers, having a wife who was high in
restoration-oriented coping was related to positive adjustment.
DUAL PROCESS MODEL: A DECADE ON /
283
We mentioned earlier that cultural differences in ways of working through
grief were a major reason for the need to revise earlier conceptions. Cultures vary
according to the norms/belief systems which govern manifestations and expressions of grief. These can be understood according to loss- versus restorationoriented coping. A clear example has been provided by the anthropologist Wikan
(e.g., 1988). The Muslim community on the island of Bali would be described
as restoration-oriented, showing little or no overt sign of grief and outwardly
continuing daily life as though nothing untoward had happened. By contrast,
the Muslim community in Egypt expresses their grief openly, gathering together
to reminisce and share anguish over their loss. Other vivid examples of such
cultural differences that are compatible with our DPM formulation can be found
in Rosenblatt (2008).
Although we also mentioned changes in patterns of coping over time, in
our 1999 article we did not elaborate much on this aspect. It is important to note
that, like the Phase and Task Models, changes are expected across the duration
of bereavement according to the DPM. There will gradually (and unevenly)
be less attention to loss-oriented and more to restoration-oriented tasks. For
example, early in bereavement there is generally comparatively little attention
to forming a new identity and far more to going over the events to do with the
death, while over time a gradual reversal in attention to these different aspects
is likely to take place. Furthermore, as time goes on, the total amount of time
spent on coping with loss and restoration tasks will diminish. Some evidence
for these patterns is now available (e.g., Caserta & Lund, 2007; Richardson
& Balaswamy, 2001; Stroebe & Schut, 2008). Of course, both the Phase and
Task Models incorporate changes over time too, but without specifying these so
explicitly in relationship to different types of stressors.
We originally formulated the DPM to address coping after partner loss, since
this had been the focus of our empirical research before that time. By contrast,
the other models were not limited to any particular type of loss (of a child, parent,
spouse, etc.), and we have come to realize that this is probably also the case for
application of the DPM. Indeed, more recently we have explored application of
the model to partners coping with the death of their child (e.g., Wijngaards et al.,
2008) and to bereavement specifically among the elderly (Hansson &. Stroebe,
2007). Other teams of researchers have addressed additional types of bereavement, for example, Stokes, Pennington, Monroe, Papadatou, and Relf (1999)
have applied the DPM to children and other family members. We have also
suggested its application to the phenomenon of homesickness, which can be
regarded as a “mini-grief” experience (Stroebe, van Vliet, Hewstone, & Willis,
2002). We have sometimes been asked about the applicability of the DPM to
other stressful life events, such as divorce or dealing with the chronic illness of
a loved one. There are certain parallels across such different types of events in
loss and restoration domains, but it remains for researchers and clinicians to
explore the usefulness of the DPM to them empirically.
284 / STROEBE AND SCHUT
GUIDELINES FOR TESTING THE DPM
Although the DPM looks quite straightforward as presented in Figure 1, it
is difficult to test its parameters and/or empirically examine the relationship of
the postulated coping processes to bereavement outcomes. The guidelines are
summarized in Table 2.
Differentiating Stressors, Coping, and Outcome
Given that we have postulated two different categories of stressors, loss- and
restoration-oriented, a useful direction for research is simply first to show that
bereaved people actually have to deal with aspects that fall within these two
categories: here, the focus is not on coping strategies or on the outcomes of dealing
with them, but on the (range of) experiences that have to be dealt with, per se, the
stressors. Thus, in doing such research, it is important to keep clear that the loss
or restoration stressors are not equivalent to “coping with” or “being restored.”
An example hopefully makes this distinction clearer. Consider one restoration
stressor, namely, the problem of changed identity from wife to widow: the stressor
Table 2. Recommended Guidelines for Testing the DPM: Summary
•Separate stressors from process from outcome variables (e.g., make sure no
symptoms are included among coping items).
•Best test of relationship between DPM coping and outcome involves
maladjusted versus adjusted bereaved persons.
•Observe scientific design principles: (e.g., control groups—e.g.,
non-intervention control groups in intervention studies; longitudinal
investigation: before, after, and follow-up).
•Extend beyond questionnaire measures (mobile phones, mobile internet,
PDA’s for monitoring; diaries; intervention principles, etc.).
•Specify (in definitions and operationalizations) the precise parameter under
investigation (the two types of stressors; coping processes; oscillation process;
coping in relationship to outcome).
•Integrate other theoretical perspectives to refine DPM predictions.
•Keep in mind that:
•Normal reactions can vary greatly between individuals and groups with
respect to preferred focus on loss- versus restoration tasks: for some
loss-orientation will dominate, for others, restoration-orientation.
•Only in cases of extreme, unrelenting, exclusive adherence to (focus on)
one or the other type of stressor, or in cases of disturbance of the oscillation
process itself, will maladaptation occur.
•Loss- and restoration-orientation are not equivalent to emotion- and
problem-focused coping.
DUAL PROCESS MODEL: A DECADE ON /
285
should be formulated in words such as “I have trouble finding a place in life
without my spouse.” Coping with this stressor would be tapped with items such
as “I avoid going on dates with potential new partners,” while “I have a new
identity or role in life” indicates outcome.
Assessing Oscillation
It is not easy to investigate the process of oscillation since it is a dynamic
process of confrontation and avoidance that can change not only from moment to
moment, but also in relationship to the duration of bereavement. Methods other
than questionnaire investigations are thus advisable. Questionnaire items asking
about shifting attention from loss- to restoration-oriented stressors have sometimes been used, but rather these fail to capture the dynamics of oscillation.
Some suggestions would be to use:
• Cell phones, mobile internet, personal digital assistants (PDA’s); diaries, and
time sampling (these may indeed usefully include—but not necessarily be
limited to—questionnaires).
• Laboratory techniques to induce shifting (e.g., by presenting stimuli to do
with the loss, and then to do with restoration).
Examining DPM Parameters in Relationship to Outcome
It is important to note that there are individual differences in the amount of
attention paid to LO and RO stressors within the normal range of reactions to
bereavement: some will have a tendency/preference to focus more on their loss,
others more on restoration tasks; some will spend much time, others little time
on either or both of these dimensions. Within a moderate range, that is, one that
does not exclude attention to either or both types of stressor, the prediction will
be that persons adapt to loss in time. Only in extreme cases, such as focusing
unremittingly and exclusively on loss, will there be poor outcome. It follows
from this that the best test of the DPM will be to compare samples of persons
experiencing poor outcomes such as complicated grief with those who are
undergoing a normal grief process. The DPM postulates a number of patterns
that will be predictive of poor outcome, and which could be used to formulate
hypotheses:
1. extreme attention to (i.e., coping with) loss orientation, avoidance of
restoration
2. extreme attention to (i.e., coping with) restoration orientation, avoidance of
loss
3. high scores on 1 and 2 and no “time off”?
4. disturbance of oscillation (disturbed intrusion-avoidance)
5. high scores on the number of stressors (could be both LO and RO)
286 / STROEBE AND SCHUT
The DPM coping-adaptation research paradigm described here can be conducted on an interpersonal level. For example, the Wijngaards et al. (2008)
investigation cited earlier investigated differences in adaptation among couples
as a function of the amount of loss- versus restoration-oriented coping of each
member of the couple. Clearly, additional hypotheses need to be developed to
make couple-level predictions as in this latter project.
It is also important to remember that outcomes can include a variety of consequences besides normal and complicated grief. It is important to look beyond
manifestations of grief and even other symptoms (depression, anxiety, physical
health symptoms, etc.), to think in terms of broader aspects relating to personal
functioning, such as outcome acceptance (i.e. …), sense of control, self-efficacy,
relationship/marital satisfaction, attachment, and emotional equilibrium.
Research Design Features for Testing the DPM
We have already mentioned a number of features for the design of empirical
tests of the DPM. In addition, it is important to stress that longitudinal, preferably
prospective designs should be used in order to establish what is cause and what
is effect. When it comes to testing the effects of intervention using the DPM,
it goes without saying that randomized controlled trials are necessary. This
brings us to our next point.
Intervention Principles: A Test of the DPM
Following DPM principles, if the bereaved person is suffering from complications in their grieving process, intervening to change his or her pattern of
confronting versus avoiding loss- and restoration-stressors should lead to better
adjustment. This is precisely what Shear, Frank, Houck, and Reynolds (2005)
did. They used the DPM as a guideline for designing one intervention program
(labeled Complicated Grief Treatment, CGT) and evaluated the efficacy of this
program against an established one (Interpersonal Psychotherapy, IPT). The
therapist described the DPM to clients with complicated grief and emphasized
the need to focus on restoration as well as loss tasks, which were both addressed
in the therapy sessions. The DPM-type CGT intervention was more effective
(even) than IPT. This suggests that the processes identified in the DPM may
indeed be central in coming to terms with bereavement. Again, the authors
emphasize the need for follow-up investigation, and indeed, it is too early to
conclude that the DPM base was the or even a success factor. Furthermore, in the
absence of a non-bereaved control, one cannot be sure that either the CGT or
IPT interventions were really effective for treating CG. Nevertheless, this study
is an excellent start and good example of how DPM parameters can be included
and tested in intervention.
Testing DPM (see Table 2) using intervention should focus on high risk groups
or those with complicated grief (see Schut et al., 1997; Schut & Stroebe, 2005).
DUAL PROCESS MODEL: A DECADE ON /
287
CONCLUDING REMARKS
There is enormous scope for further research on the DPM, and, in our view, the
best way forward is to follow the sorts of guidelines that we have outlined above.
Both theoretical and empirical contributions are needed, the former to refine and
extend the DPM framework and further identify specific underlying cognitive
processes, the latter to test the model’s parameters and their power in predicting
good versus poor adjustment to bereavement. The input of different research
teams—such as those illustrated in the following articles in this Special Issue—
will be invaluable in this endeavor.
REFERENCES
Archer, J. (2001). Broad and narrow perspectives in grief theory: Comment on Bonanno
and Kaltman (1999). Psychological Bulletin, 127, 554-560.
Billings, A., & Moos, R. (1981). The role of coping resources and social resources in
attenuating the stress of life events. Journal of Behavioral Medicine, 4, 139-157.
Billings, A., & Moos, R. (1984). Coping, stress, and social resources among adults with
unipolar depression. Journal of Personality and Social Psychology, 46, 877-891.
Boelen, P. A., & van den Bout, J. (in press). Anxious and depressive avoidance and
symptoms of prolonged grief, depression, and Post Traumatic Stress-Disorder.
Psychologica Belgica.
Boelen, P. A., van den Hout, M. A. & van den Bout, J. (2006). A cognitive-behavioral
conceptualization of complicated grief. Clinical Psychology: Science and Practice,
13, 109-128.
Bonanno, G. (2001). Grief and emotion: A social-functional perspective. In M. S. Stroebe,
R. O. Hansson, W. Stroebe, & H. Schut (Eds.), Handbook of bereavement research:
Consequences, coping and care (pp. 493-515). Washington, DC: American Psychological Association.
Bonanno, G., & Kaltman, S. (1999). Toward an integrative perspective on bereavement.
Psychological Bulletin, 125, 760-776.
Bowlby, J. (1980). Attachment and loss; Vol. 3, Loss: Sadness and depression. London:
Hogarth.
Caserta, M., & Lund, D. (2007). Toward the development of an inventory of daily widowed
life (IDWL): Guided by the Dual Process Model of coping with bereavement. Death
Studies, 31, 505-535.
Folkman, S. (2001). Revised coping theory and the process of bereavement. In M. S.
Stroebe, R. O. Hansson, W. Stroebe, & H. Schut (Eds.), Handbook of bereavement
research: Consequences, coping and care (pp. 563-584). Washington, DC: American
Psychological Association.
Freud, S. (1917/1957). Mourning and melancholia. In J. Strachey (Ed. & Trans.), The
standard edition of the complete works of Sigmund Freud (Vol. 14, pp. 152-170).
London: Hogarth Press.
Hansson, R. O., & Stroebe, M. S. (2007). Bereavement in later life: Coping, adaptation,
and developmental influences. Washington, DC: American Psychological Association.
Lazarus, R., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.
288 / STROEBE AND SCHUT
Lindemann, E. (1944). Symptomatology and management of acute grief. American
Journal of Psychiatry, 101, 141-148.
Mikulincer, M., & Shaver, P. (2008). An attachment perspective on bereavement. In
M. S. Stroebe, R. O. Hansson, H. Schut, & W. Stroebe (Eds.), Handbook of bereavement research and practice: Advances in theory and intervention (pp. 87-112).
Washington, DC: American Psychological Association.
Nolen-Hoeksema, S. (2001). Ruminative coping and adjustment to bereavement. In
M. Stroebe, R. O. Hansson, W. Stroebe, & H. Schut (Eds.), Handbook of bereavement
research: Consequences, coping and care (pp. 545-562). Washington, DC: American
Psychological Association.
Parkes, C. M. (1972/1996). Bereavement: Studies of grief in adult life. London: Routledge.
Parkes, C. M. (2006). Love and loss: The roots of grief and its complications. London:
Routledge.
Parkes, C. M., & Weiss, R. (1983). Recovery from bereavement. New York: Basic Books.
Richardson, V., & Balaswamy, S. (2001). Coping with bereavement among elderly
widowers. Omega, 43, 129-144.
Rosenblatt, P. (2008). Grief across cultures: A review and research agenda. In M. S.
Stroebe, R. O. Hansson, H. Schut, & W. Stroebe (Eds.), Handbook of bereavement
research and practice: Advances in theory and intervention (pp. 207-222).
Washington, DC: American Psychological Association.
Schut, H. A. W., Stroebe, M. S., de Keijser, J., & van den Bout, J. (1997). Intervention for
the bereaved: Gender differences in the efficacy of grief counselling. British Journal
of Clinical Psychology, 36, 63-72.
Shear, K., Frank, E., Houck, P., & Reynolds, C. (2005). Treatment of complicated grief:
A randomized controlled trial. Journal of the American Medical Association, 293,
2601-2608.
Stokes, J., Pennington, J., Monroe, B., Papadatou, D., & Relf, M. (1999). Developing
services for bereaved children: A discussion of the theoretical and practical issues
involved. Mortality, 4, 291-307.
Stroebe, M. (in press). Coping with bereavement. In S. Folkman (Ed.), Handbook of
stress, coping and health. Oxford: Oxford University Press.
Stroebe, M. S. (1992). Coping with bereavement: A review of the grief work hypothesis.
Omega: Journal of Death and Dying, 26, 19-42.
Stroebe, M. S., Boelen, P., van der Hout, M., Stroebe, W., Salemink, E., & van den Bout, J.
(2007). Ruminative coping as avoidance: A reinterpretation of its function in adjustment to bereavement. European Archives of Psychiatry and Clinical Neuroscience,
257, 462-472.
Stroebe, M. S., & Schut, H. (1999). The Dual Process Model of coping with bereavement:
Rationale and description. Death Studies, 23, 197-224.
Stroebe, M. S., & Schut, H. (2001). Meaning making in the Dual Process Model. In
R. Neimeyer (Ed.), Meaning reconstruction and the experience of loss (pp. 55-73).
Washington, DC: American Psychological Association.
Stroebe, M., & Schut, H. (2008). The Dual Process Model of Coping with Bereavement:
Overview and update. Grief Matters: The Australian Journal of Grief and Bereavement, 11, 1-4.
Stroebe, M., Schut, H., & Stroebe, W. (2005a). Attachment in coping with bereavement:
A theoretical integration. Review of General Psychology, 9, 48-66.
DUAL PROCESS MODEL: A DECADE ON /
289
Stroebe, W., Schut, H., & Stroebe, M. S. (2005b). Grief work, disclosure and counseling:
Do they help the bereaved? Clinical Psychology Review, 25, 395-414.
Stroebe, M., Schut, H., & Stroebe, W. (2007). Health consequences of bereavement:
A review. The Lancet, 370, 1960-1973.
Stroebe, M. S., Stroebe, W., & Schut, H. (2001). Gender differences in adjustment to
bereavement: An empirical and theoretical review. Review of General Psychology,
5, 62-83.
Stroebe, M., Stroebe, W., & Schut, H. (2005). Who benefits from disclosure? Exploration of attachment style differences in the effects of expressing emotions. Clinical
Psychology Review, 26, 66-85.
Stroebe, M., van Vliet, T., Hewstone, M., & Willis, H. (2002). Homesickness among
students of two cultures: Antecedents and consequences. British Journal of
Psychology, 93, 147-168.
Wijngaards, L., Stroebe, M. S., Stroebe, W., Schut, H., van den Bout, J., van der Heijden,
P., et al. (2008). Parents grieving the loss of their child: Interdependence in coping.
British Journal of Clinical Psychology, 47, 31-42.
Wikan, U. (1988). Bereavement and loss in two Muslim communities: Egypt and
Bali compared. Social Science and Medicine, 27, 451-460.
Worden, J. (1982/1991/2002/2009). Grief counselling and grief therapy: A handbook for
the mental health practitioner (4th ed.). New York: Springer.
Direct reprint requests to:
Margaret Stroebe
Department of Clinical & Health Psychology
Utrecht University
Box 80140
3508 TC Utrecht, The Netherlands
e-mail: [email protected]
OMEGA, Vol. 61(4) 291-313, 2010
EXPERIENCES AND EARLY COPING OF BEREAVED
SPOUSES/PARTNERS IN AN INTERVENTION
BASED ON THE DUAL PROCESS MODEL (DPM)*
DALE LUND, PH.D.
California State University San Bernardino
MICHAEL CASERTA, PH.D.
REBECCA UTZ, PH.D.
University of Utah, Salt Lake City
BRIAN DE VRIES, PH.D.
San Francisco State University
ABSTRACT
This study was designed to test the effectiveness of the Dual Process Model
(DPM) of coping with bereavement. The sample consisted of 298 recently
widowed women (61%) and men age 50+ who participated in 14 weekly
intervention sessions and also completed before (O1) and after (O2) selfadministered questionnaires. While the study also includes two additional
follow-up assessments (O3 and O4) that cover up to 14-16 months bereaved,
this article examines only O1 and O2 assessments. Based on random assignment, 128 persons attended traditional grief groups that focused on
loss-orientation (LO) in the model and 170 persons participated in groups
receiving both the LO and restoration-orientation (RO) coping (learning daily
life skills). As expected, participants in DPM groups showed slightly higher
*Funded by a grant from the National Institute on Aging (R01 AG023090). The content
is solely the responsibility of the authors and does not necessarily represent the official views
of the National Institute on Aging or the National Institutes of Health.
291
Ó 2010, Baywood Publishing Co., Inc.
doi: 10.2190/OM.61.4.c
http://baywood.com
292 / LUND ET AL.
use of RO coping initially, but compared with LO group participants they
improved at similar levels and reported similar high degrees of satisfaction
with their participation (i.e., having their needs met and 98-100% indicating
they were glad they participated. Even though DPM participants had six
fewer LO sessions, they showed similar levels of LO improvement. Qualitative data indicate that the RO component of the DPM might be more
effective if it is tailored and delivered individually.
In this article we present and describe the preliminary results of our intervention
study based on the Dual Process Model (DPM) of coping with bereavement
(Stroebe & Schut, 1999). Our study compared the experiences of recently
bereaved men and women age 50+ who received both of the DPM features
(i.e., loss-orientation [LO] and restoration-orientation [RO]) with others in a
comparison group who received only the LO features in their intervention. There
are three primary purposes of this article. First, we describe and compare the
experiences and evaluations of those recently bereaved spouses/partners who
participated in the two group conditions described above. Second, we examine the
extent to which the LO and RO coping processes were enhanced by participating
in each of the two study conditions, DPM treatment versus the comparison group.
Third, and based on the findings of the above, we identify the most valuable and
promising features of the DPM intervention for future research and practice.
While other articles in this special issue deal with different populations of
bereaved persons, we focused exclusively with recently widowed men and women
age 50+ regardless of the level of early coping difficulty. In other words, we did
not select only those who were experiencing complicated or traumatic grief.
Rather, we wanted to test the DPM with a broad representation of bereaved
persons in mid and later life. Furthermore, although the larger study was designed
to test the effectiveness of both study conditions in terms of various bereavement
adjustment outcomes over time, these data will be analyzed in future manuscripts.
Therefore, this report is focused on the early coping processes targeted by
the intervention. The DPM describes loss-orientation and restoration-orientation
as coping responses to two primary types of stressors that bereaved persons
experience with each type of stress requiring specific types of coping behaviors
and strategies. These processes, in turn, influence bereavement adjustment
outcomes related to well-being.
BACKGROUND ON SPOUSE/PARTNER
BEREAVEMENT
The consequences of spousal/partner loss in later life have been well documented (Bennett, Hughes, & Smith, 2005; Bisconti, Bergeman, & Boker, 2004;
Carr, Nesse, & Wortman, 2006; Hansson & Stroebe, 2007; Lee & Carr, 2007;
EXPERIENCES IN INTERVENTION BASED ON THE DPM /
293
Lund, 1989; Lund & Caserta, 2002; Stroebe, Stroebe, & Hansson, 1993).
Although the long-term bereavement process is experienced with considerable
variability, some common elements have included profound sadness, pining,
depression, altered identity, negative health outcomes, loneliness, and the
withdrawal of support networks. Additionally, there is evidence of considerable
stress associated with the role changes that accompany widowhood, particularly
those due to disruptions in life patterns and daily routine, taking on new unfamiliar
tasks, and changes in social activities and relationships (Anderson & Dimond,
1995; Carr, 2004; Moss, Moss, & Hansson, 2001; Utz, Carr, Nesse, & Wortman,
2002; Utz, Reidy, Carr, Nesse, & Wortman, 2004). Persons overwhelmed or
preoccupied with grief often neglect their own nutrition, fail to exercise regularly,
discontinue physical and social activities that they previously did as a married
couple, and become more accident prone from paying less attention to their
personal safety (Johnson, 2002; Quandt, McDonald, Arcury, Bell, & Vitolins,
2000; Schone & Weinick 1998; Shahar, Schultz, Shahar, & Wing, 2001). The loss
of a spouse or partner can be disruptive to existing health care practices, as well
as interfere with the adoption of new healthy behaviors (Chen, Gill, & Prigerson,
2005; Pienta & Franks, 2006; Powers & Wampold, 1994; Rosenbloom &
Whittington, 1993; Williams, 2004).
Spouse/partner bereavement can adversely impact the performance of daily
living tasks that are essential for health and independent functioning. For
example, meal planning and preparation, household maintenance, managing
finances, as well as other tasks often go unattended by the surviving spouse if these
tasks were primarily the responsibility of his or her deceased partner. Those who
fail to acquire new skills to accomplish these tasks are at increased risk for
long-term mental and physical health problems following widowhood (Carr,
House, Kessler, Nesse, Sonnega, & Wortman, 2000; Lund, Caserta, Dimond,
& Shaffer, 1989b; Stroebe & Schut, 1999; Utz, 2006; Wells & Kendig, 1997).
While spouse or partner loss is often associated with a variety of disruptive
and negative outcomes, research and theory also has focused on successful
adaptation, resiliency, and personal growth among the bereaved (Boerner,
Wortman, & Bonanno, 2005; Bonanno, 2004; Calhoun & Tedeschi, 2006;
Caserta, Lund, Utz, & de Vries, 2009; Dutton & Zisook, 2005; Lund, Utz, Caserta,
& de Vries, 2008; Montpetit, Bergeman, Bisconti, & Rausch, 2006; Ong &
Bergeman, 2004; Ong, Bergeman, & Bisconti, 2004; O’Rourke, 2004; Wilcox,
Evenson, Aragaki, Wassertheil-Smoller, Mouton, & Loevinger, 2003; Znoj,
2006). Several conceptual models have emerged to describe these disruptive
processes and outcomes—most notably Worden’s (2002) “tasks of grief” and
other more general stress and coping models (Lazarus & Folkman, 1984). These
approaches, however, pay little attention to other concurrent restorative processes
(i.e., in addition to the LO processes) and potential positive consequences like
opportunities for personal growth through learning, having new experiences and
helping others (Doka & Martin, 2001; Lund, 1999; Lund et al., 2008).
294 / LUND ET AL.
DUAL PROCESS MODEL OF COPING
WITH BEREAVEMENT
The Dual Process Model (DPM) of coping with bereavement has emerged
as a response to the limitations associated with earlier conceptual frameworks.
While the first article in this special issue of Omega (Stroebe & Schut,
2010) presents an overview of and the latest developments regarding the DPM
(Stroebe & Schut, 1999), we provide a description of the key elements of the
model, with particular attention to how it guided the development of our intervention and influenced the selection of our process and outcome measures.
The DPM identifies two concurrent types of stressors and coping processes.
First, loss-orientation (LO) refers to the coping processes (including grief
work) directly focused on the stress attributed to the loss itself, encompassing
many of the grief-related feelings and behaviors that tend to dominate early
but can re-emerge later and sporadically in the bereavement process. Second,
restoration-orientation (RO) refers to those processes the bereaved use to cope
with the secondary stressors that accompany new roles, identities, and challenges
related to the new status as a widow, widower, or bereaved partner. These often
include the need to master new tasks, make important decisions, meet new
role expectations, and take greater self-care initiative. If RO progresses effectively, self-efficacy beliefs emerge and help facilitate greater confidence, independence, and autonomy needed to manage their daily lives (Caserta, 2003; Lund
& Caserta, 2002; Utz, 2006). Another desired outcome is a sense of personal
growth as a result of becoming more independent and learning new skills
(Calhoun & Tedeschi, 2006; Caserta et al., 2009; Dutton & Zisook, 2005; Schaefer
& Moos, 2001).
Another feature of the DPM is the need to take brief periods of respite from
grieving itself, whether to address these new tasks or demands or to keep busy
with diversionary or meaningful activities to help restore a sense of balance
and well-being. There is evidence that engaging in physical activity, hobbies,
volunteer projects, and other leisure activities, as well as socializing and being
involved with others provides an important source of restoration and respite for
the bereaved. As well, our previous research found a strong association between
perceived competencies in tasks of daily living and more favorable adjustments
to psycho-emotional aspects of grief (Lund, Caserta, & Dimond, 1989a) The RO
coping process is amenable to intervention by focusing on self-efficacy, skills
related to new unfamiliar tasks of daily living, self-care, and opportunities to
engage in activities that provide brief periods of respite from grief.
What distinguishes the DPM from the other more global stress and coping
frameworks is the recognition that the bereaved will oscillate between the two
coping processes (RO, LO) throughout the course of bereavement as demands
arise in their daily lives, even on a moment to moment basis (Caserta & Lund,
2007; Richardson & Balaswamy, 2001; Stroebe & Schut, 1999). In previous
EXPERIENCES IN INTERVENTION BASED ON THE DPM /
295
research, we identified and described several potentially important dimensions of
“oscillation” (Caserta & Lund, 2007), including the degree of balance, frequency,
awareness, control, and intent of oscillation.
Although there have been exceptions (Caserta, Lund, & Obray, 2004; Caserta,
Lund, & Rice, 1999), most of the previous and current bereavement interventions focus primarily on emotional impacts of the loss itself (LO) and have rarely
addressed the concurrent RO issues that the bereaved confront in their daily
lives. Our intervention, based on the DPM provides coping assistance directed at
both the LO and RO components and encourages the oscillation between them
by alternating attention to them in group sessions/meetings.
THE DPM-BASED INTERVENTION
We refer to this study as the “Living After Loss” (LAL) project; the name
was chosen as a way to identify the program to the research participants and
community professionals without sounding too academic or theoretical. We
designed the LAL study specifically to compare a DPM-based intervention (with
both the LO and RO components) with the traditional support group format as the
comparison group (which provides attention primarily to LO coping processes).
We hypothesized that those in the DPM condition would have more positive
and broadly based coping processes and outcomes than those who were in the
LO only comparison condition. Our present analyses are focused on the DPM
coping processes and not on the adjustment outcomes as those data are not yet
available. Both the DPM and comparison conditions consisted of 14 weekly
sessions, each about 90 minutes in length. Those in the DPM group had seven of
14 sessions devoted to RO issues, whereas all 14 sessions within the comparison
group were solely loss-oriented in focus. The RO-focused sessions of the DPM
condition were alternated with the LO-focused sessions to simulate the oscillation
between the LO and RO coping processes. For those in the DPM groups, they
had fewer LO sessions but the content for these sessions was the same as those
in the LO only comparison groups. Table 1 provides a detailed description of
the content and theoretical principles addressed in each of the 14 weekly sessions
of the DPM intervention. Additional details about the intervention content and
format are available upon request to the first author and/or in another publication
(Lund, Caserta, de Vries, & Wright, 2004).\
RESEARCH DESIGN AND SAMPLING
PROCEDURES
Eligibility for the LAL project was restricted to widowed persons age 50
and older, whose spouse or partner had died within the previous 2-6 months,
who were English speaking, and who were cognitively and physically able to
complete questionnaires and participate in a 14-week group meeting. Participants
296 / LUND ET AL.
Table 1. Description of the 14-Week Dual Process Model
(DPM) Intervention
Session
1
LO
Content
• Introduction and overview of group and dual
Link to model
• Grief work
processes of grief
• Grief: What is it, typical reactions, how the groups
might help
• Journaling and writing one’s thoughts and feelings
• Circumstances surrounding spouse’s death
• Specific concerns of participants
2, 3
LO
• Physical sensations, cognitions, behaviors
• How grief affects daily functioning
• How to express grief-related feelings
• Grief work
• Intrusion of grief
4
LO
• New responsibilities: One’s unique situation and
• Grief work
• Intrusion of grief
• Avoidance of
common experiences
• New roles: Welcomed, anger, frustration, feeling
overwhelmed, etc.
• Homework: What is one new responsibility you
want to do better?
5
RO
Goal Setting and Personal Priorities
• How to take more control of the situation
• Goal setting as a way to learn new skills and
6
LO
restoration
• Attending to life
changes
behaviors
• Sharing ideas for success
• How to “take a break” from grief; attending to
one’s personal needs
• Homework: Goal Setting—Do something nice for
yourself.
• Doing new things
• Distraction from
• How to put own needs first
• Dealing with loneliness
• Coping with critical time periods (birthdays,
• Intrusion of grief
• Avoidance of
grief
restoration
anniversaries, holidays)
7
RO
Self-Care and Health Care
• Meeting one’s own health needs: To what extent
was spouse a “partner” in health-related needs?
• How to utilize the health care system and service
network effectively (immunizations, screenings,
medication management, communicating with
providers, local services and programs available).
8
LO
• What do participants miss and not miss about
spouse?
• Unfinished business
• Attending to life
changes
• Grief work
• Reframing ties with
spouse
EXPERIENCES IN INTERVENTION BASED ON THE DPM /
297
Table 1. (Cont’d.)
Session
9
RO
Content
Finances and Legal Issues
• Wills and trusts
• Filing legal and important documents
• Understanding statements
• Developing and managing a household budget
• Making decisions about property issues
• Recognizing and avoiding scams
10
LO
• Putting self first: Dealing with requests from others
• Things that have interfered with the need to grieve
• Things that have interfered with the need to be
Link to model
• New roles, identities
• Doing new things
• Attending to life
changes
• Intrusion of grief
• Avoidance of
restoration
independent
11
RO
Household and Vehicle Responsibilities
• Breaking household duties into small, manageable
steps
• Following regular maintenance schedules for
home and car
• Identifying and remedying household hazards
• Who to contact for help, and How to avoid being
defrauded
12
RO
14
RO
LORO
changes
• Doing new things
• Distraction from
grief
Nutrition for One
• Shopping for one and understanding labels
• Preparing one- or two-step meals
• Freezing portions for later use
• Sharing recipes
• Finding nutritional assistance in the community
13
• Attending to life
• Attending to life
changes
• Doing new things
• Distraction from
grief
Remaining Socially Connected
• Functioning “comfortably” as a single person
• Inexpensive and accessible entertainment or
• Attending to life
leisure activities
• Engaging in non-threatening socialization
experiences
• Finding volunteer opportunities
• New roles, identities
• New relationships
• Distraction from
changes
grief
• Rediscovering sources of joy, fulfillment, and growth; • All of the above
renewing old interests and exploring new ones
• “A time to mourn—A time to dance”; taking the
time to grieve and the time away from grieving
• Having realistic expectations
• Where to learn more
LO = Loss-Oriented
RO = Restoration-Oriented
elements
298 / LUND ET AL.
also had to reside in one of the two study areas (San Francisco or Salt Lake City)
for the duration of the group intervention. Potential participants were identified
from death certificate data filed with local and federal health agencies. Names
and addresses of the surviving spouses or partners, as well as information on the
decedent, were included on the death records.
Potential participants received an invitation letter roughly 3 months prior to
the commencement of the groups. Then, approximately 5-7 days after the letters
were mailed, a trained research assistant contacted them by telephone. The
purpose of the call was to verify that they received the letter, that they met
the study criteria, to address any questions about the study, and to solicit their
preliminary agreement to participate. Those who were ineligible or not interested
in participating received a list of available community-based resources related to
bereavement. If all the eligibility criteria were met and the potential participant
agreed, the RA made an appointment to conduct a home visit, in which informed
consent was obtained and the baseline (O1) questionnaire was left for the participant to complete. Participants were randomly assigned in equivalent proportions to one of the two group interventions (DPM or Comparison).
All participants, regardless of their assigned study condition, completed selfadministered questionnaires largely consisting of background information and
a variety of measures commonly used in bereavement research: Baseline (O1)
data were collected prior to the intervention at 2-6 months of bereavement.
Post-test (O2) occurred at the conclusion of the 14-week intervention period (at
5 to 9 months of bereavement) with the O3 assessment following 3 months
later. The final data point (O4) was 6 months following O3, equivalent to 14-18
months after the loss. Participants received $25 per questionnaire, for a total of
$100 if they completed all four waves of data collection. The average participant
completed the O1 baseline questionnaire approximately 4 months (15.7 weeks)
after the spouse’s death, with some completing it as early as 5 weeks post-loss and
some as late as 24 weeks post-loss. The group intervention commenced approximately 4.7 months post-loss, with a range of 2 to 7.5 months bereaved.
SAMPLE
A total of 328 widowed persons completed O1 questionnaires, but the analytic
sample for this report includes the 298 participants who completed both the
O1 (pre-test) and O2 (first post-intervention assessment). The most common
reasons for attrition between O1 and O2 included: “I decided that the group
was not for me,” “My grief is too strong and I am unable to participate in group,”
or the participant decided that they “no longer wanted to participate in LAL”
without specifying a reason. One participant died after he completed the baseline
questionnaire, but before the groups started. Two participants experienced significant health declines just prior to the start of the group intervention, and a few
others had to withdraw in order to take care of unexpected family emergencies.
EXPERIENCES IN INTERVENTION BASED ON THE DPM /
299
Finally, the group leaders discovered that a few of the enrolled participants were
not suitable for the group intervention due to cognitive impairment or significant
hearing loss; they were asked to withdraw from the study.
Our analytic sample includes 61% women (n = 183) and 39% men (n = 115).
The average age of our sample was 69.5 years (SD = 10.6), with a range of 50 to
93 years. Participants had been married or partnered for an average of 40.1 years
(SD = 17.0). Our sample was quite educated: only 14% of the sample had a high
school education or less; 41% had some college; and 45% had graduated from
a college. The majority were Caucasian (87%), with 6% Asian, 4% African
American, 2% Latino, and 1% other. Also, 10 participants (or 3% of the sample)
had been in a same-sex union. A little more than half (58%) said they expected
the spouse’s death. Regarding financial status, 69% reported that they were
“comfortable,” 17% said “more than adequate,” and 14% said that it was “not very
good.” A little more than half of the participants (n = 170 or 57%) were assigned
to the DPM condition, while 128 (43%) were assigned to the LO condition. The
number of weekly intervention sessions attended was the same for those in the
LO and DPM conditions, with an average of 11.03 sessions (SD = 3.3).
MEASURES
DPM coping processes were measured by the Inventory of Daily Widowed
Life (IDWL; Caserta & Lund, 2007), consisting of 22 Likert-format items that
inquire into how much time during the past week the respondents spent on
loss-oriented activities (e.g., “Thinking about how much I miss my spouse;”
“Feeling a bond with my spouse”) and restoration-oriented activities (e.g., “Finding ways to keep busy or occupied;” “Took some time away from grieving for
my spouse”). Eleven loss-orientation items and 11 restoration-orientation items
were identified largely from Stroebe and Schut’s (1999) description of the types
of instances that would fall into each category. Construct validity has been
established for both subscales (Caserta & Lund, 2007). LO and RO subscale
scores were calculated by summing the identified items (on scales from 1 “rarely
or not at all” to 4 “almost always,” each subscale ranging from 11 to 44), with
lower scores indicating less LO or RO coping and higher scores indicating more
LO or RO coping. Both subscales range from 11 to 44 and have high internal
consistency (LO subscale, alpha = .91; RO subscale, alpha = .73).
Oscillation balance or the degree to which the participants engaged in equal
amounts of both processes was calculated by subtracting their total LO score
from their total RO score (RO minus LO). Hence, the balance score can range
from –33 (exclusively loss-orientation focus) to +33 (exclusively focused on
restoration-orientation). A score equal to 0 indicates perfect balance between
the two processes.
In addition to examining the extent to which the participants were engaging
in loss- and restoration-oriented coping processes, we also were interested in
300 / LUND ET AL.
knowing to what extent they were consciously aware of where they were focusing
their attention. Consequently, the participants were requested to respond to two
additional items that asked, “During the past week, to what extent have you
focused your attention on issues related to grief, emotions, and feelings regarding
your loss?” to measure if they were aware of their engagement in loss-orientation,
and to what extent they focused on “new responsibilities, activities, and time
away from grieving” as a way to measure awareness of their own restorationoriented coping. Both items were measured on a scale from 1-5 with 1 indicating
very little and 5 a great deal.
Participants’ Assessment of Group Sessions
At the O2 measurement, the participants completed a series of items that
inquired into how well they believe they learned or understood each of the topics
covered in the group meetings (ranging from 1 [not at all well] to 5 [very well])
and to what extent they were applying what they learned in their daily lives (1 =
not at all; 5 = almost always). Each of the items corresponded to the content
depicted in Table 1. There were 11 items related to loss-oriented aspects of coping
with grief and loss that were completed by participants in both study conditions.
Those in the DPM treatment group completed 11 additional items that represented the content addressed in the seven RO sessions. The participants also
completed a checklist to indicate if they sought additional information outside
the group meetings on any of the topics that were covered in their weekly
sessions as well as an open-ended item where they were given the opportunity to
identify any topics they wish were addressed but were not. We also compared the
participants in both groups on items related to their motivation to attend group
meetings (ranging from 1 = very little to 5 = very much), the extent to which
their needs were met by their participation (ranging from 1 = not at all to 5 = very
well), and whether or not they were glad they had participated (yes or no).
RESULTS
Participants’ Assessment of Their Group
Experiences and Content Exposure
Our first interest was in identifying how well participants learned or understood and were applying in their daily lives those intervention topics related to
grief (LO sessions). We also compared the responses of participants in both
study conditions. There were no statistically significant differences between DPM
and comparison participants on their assessments of grief-related topics. The
top five grief topics (based on 1-5 ratings) for “how much was learned” for both
the DPM and LO participants were in the same rank order and with mean scores
ranging from 4.0 to 4.5. They were:
EXPERIENCES IN INTERVENTION BASED ON THE DPM /
1.
2.
3.
4.
5.
301
Experience of grief is unique to each person;
Understanding the need to grieve;
How grief affects everyday functioning;
Understanding what grief is; and
Expectations about how and when to grieve.
With respect to how much they were applying what they learned in their daily
lives, participants in both study conditions had the same top five answers (based
on 1-5 ratings) and nearly in the same order with the highest ones being “Experiences of grief is unique,” “Understanding the need to grieve,” “Using humor as a
way to cope with loss,” “How grief affects everyday functioning,” and “Dealing
with challenges assuming new responsibilities.” The mean scores on these items
ranged from 3.8 (new responsibilities) to 4.2 (grief is unique).
Furthermore, no statistically significant differences were found between those
in the two study conditions regarding their motivation to attend meetings or the
percentages reporting they were glad they participated. Most participants had
very favorable ratings of the program. Regarding motivation for attending, both
groups had mean scores of 4.2 (1-5 scales). Ninety-nine percent of those in the
comparison condition and 95% in the DPM condition said they were glad they
participated. In short, there were far more similarities between those in the DPM
and comparison group condition than there were differences and most participants, regardless of group condition, were very favorable about what they had
learned and were applying, their motivation levels to attend meetings, and being
glad they had participated.
For those in the DPM condition, the top five restoration-oriented skills that
they learned and applied in their daily lives were the same. The highest rated
RO topic was “Taking responsibility for health needs” followed by “Maintaining
clean and safe home,” “Keeping home and auto in good repair,” “Avoiding scams
and fraud,” and “Managing finances and budget.” Participants gave average
ratings ranging from 4.1 to 4.2 on all five of these topics. Two other skill
topics with high scores for “what was learned” included “Nutrition for one”
and “Insurance and legal matters.” There were eight RO skill topics for which
notable proportions of DPM participants indicated they sought additional information beyond what they learned in the group sessions. These were “Insurance
and legal matters” (35%), “Taking responsibility for health needs” (29%), “Setting
personal goals” (29%), “Managing finances and budget” (27%), “Trying
new things” (27%), “Nutrition for one” (24%), “Leisure, social and volunteer
activities” (24%), and “Keeping home and auto in good repair” (23%). One
interpretation of these data is that the RO sessions stimulated the interest of
many participants in these topics as well as many persons recognized the need
for them to learn more than what could be covered during the sessions.
The participants also were asked, “Are there any topics you wish were covered
but were not?” Very few comparison group participants indicated whether there
302 / LUND ET AL.
was anything about which they would have liked to learn more but those in
DPM groups gave many suggestions, including learning about new relationships,
sprinkling systems, hobbies, driving, computers, cooking, books, making television recordings, working cameras, art, and spiritual issues. These responses
might suggest that having participated in RO types of group sessions during the
intervention stimulated their thinking about other topics and skills of interest
which would be a positive outcome according to the Dual Process Model.
Participants’ Engagement in LO and RO
Coping Process
Our next interest was to examine the extent to which the DPM intervention
was enhancing LO and RO coping processes by comparing the participants in the
DPM condition with those in the comparison group. Recall that we expected those
in both study groups to be similar to each other regarding LO coping from O1
(prior to the intervention) to O2 (assessed within 2-4 weeks after the intervention)
because all participants received intervention content devoted to LO coping.
However, we expected those in the DPM condition to show greater gains in RO
coping because the comparison group focused on loss-oriented issues without
any deliberate attempt to foster restoration-orientation. Figure 1 shows the mean
scores for both study groups at O1 and O2 using the IDWL subscales described
earlier. The mean scores for both groups show a statistically significant decline
in LO coping from O1 to O2 (p < .001) which is consistent with the Dual Process
Model in that LO coping gradually subsides while RO coping tends to increase
over time. The decline in LO coping, however, was independent of study condition as indicated by the failure of the group by time interaction to attain statistical significance. This could be interpreted as a positive outcome of the DPM
intervention in that similar decrements were noted regarding LO coping even
though DPM participants received overall less attention in their intervention to
LO coping because they had fewer sessions devoted to those issues.
Figure 2 shows the mean scores for the two study conditions regarding O1 to
O2 changes in RO coping. Those who were in the DPM condition were expected
to show greater gains in RO coping than the comparison group condition because
they had half of their intervention sessions devoted to restoration-oriented issues.
The data show that while those in the DPM condition did show greater gains in
RO coping, those in the comparison group showed some gains as well. Therefore,
both groups changed over time (p < .05) but there was no statistically significant
group by time effect.
The overall decline in the use of LO coping and increase use of RO coping
among the participants is further reflected in the change in oscillation balance
scores from O1 to O2, illustrated in Figure 3. Participants in both groups were
fairly balanced between both processes at baseline but moved to a greater focus on
restoration-oriented coping by the time the weekly sessions ended (p < .001).
EXPERIENCES IN INTERVENTION BASED ON THE DPM /
303
Figure 1. Pre- and post-intervention measures of loss-oriented coping
by study group. Loss-oriented coping is measured with an 11-item scale
(range 11 to 44). Lower values indicate less loss-oriented coping.
Although mean balance scores at O2 were further away from 0 than they were
at the baseline assessment, the score values were of a magnitude that indicates
that loss-oriented coping was still occurring at O2 for participants in both study
conditions. At the same time, a trend appeared to be emerging (although not
statistically significant) in which those in the DPM condition were beginning
to invest more effort into RO (versus LO) coping, than the comparison group.
Figure 4 shows that those in both group conditions had similar LO coping
awareness scores at O1, and consistent with the model were consciously aware
that they were devoting less attention to loss-oriented issues by the O2 assessment.
The significant group by time interaction (p < .05) however, indicates that this
awareness was greater among those in the DPM treatment group. Alternatively
and somewhat unexpectedly, we found that participants in both conditions were
showing greater awareness of RO coping from O1 to O2. Although there is a
statistically significant overall increase in RO awareness (p < .001), both groups
changed in a parallel fashion (see Figure 5). It may be possible that even those
in the comparison condition were learning new skills and engaging in RO coping
without having intervention sessions specifically devoted to these issues. Indeed,
304 / LUND ET AL.
Figure 2. Pre- and post-intervention measures of restoration-oriented
coping by study group. Restoration-oriented coping is measured
with an 11-item scale (range 11 to 44). Lower values indicate
less restoration-oriented coping.
there were informal suggestions from participants, and the research assistants who
attended sessions reported that some participants encouraged each other to seek
assistance with restoration-oriented activities.
While we expected to find stronger and more positive effects for the DPM
intervention, even with using only the O1 and O2 data, the open-ended questions
in the self-administered questionnaires did offer some further explanations and
insights. We wanted to obtain information regarding any criticisms that the
DPM participants might have had so we asked them if they had any suggestions
for how we could improve on the intervention. The following quotes provide
some potential reasons for why we find only modest support in favor of the
DPM condition. “Some sessions of little or no value—could just give us written
material,” “Guest speakers not among my needs or interests,” “Did not enjoy
guest speakers, maybe use only half of the meeting time,” “Guest speakers
weren’t helpful or interesting,” “Some presentations/speakers did not have
enough time,” “Need a doctor as a speaker,” “Some speakers communicated as
though we were totally inexperienced—wrong assumptions,” “Least effective
EXPERIENCES IN INTERVENTION BASED ON THE DPM /
305
Figure 3. Pre- and post-intervention measures of oscillation balance
by study group. Oscillation balance is measured by subtracting LO from RO
with responses ranging from –33 to +33, zero equals perfect balance.
were guest speakers,” “Speakers did not have enough time—rushed and couldn’t
ask questions,” “Instructors were poor use of time,” “Needed better tips on
house cleaning,” “Needed more info on low cost medical insurance,” “More
help with financial matters, but everyone’s situation is different,” “Needed
more guest speakers,” “Less speakers,” “Ask group for other specific topics,”
“More knowledgeable speakers on autos and home services.” In short, these
comments suggest that providing RO coping in group situations may not be
the ideal way to provide what is most needed. Some participants saw the RO
sessions as disruptions to the attention to loss-oriented issues from the previous
week, others were not satisfied because their individual RO needs were not
met in terms of the topics selected, the levels of knowledge, and/or the limited
time available.
CONCLUSION AND DISCUSSION
There are three primary conclusions of this investigation. First, regarding
nearly all measures of motivation and satisfaction with the intervention processes,
306 / LUND ET AL.
Figure 4. Pre- and post-intervention measures of awareness of
loss-oriented coping by study group. Awareness is measured with
a single-item question (In the past week, to what extent have you
focused your attention on issues related to grief, emotions, and
feelings regarding your loss?), with responses ranging from\
1 very little to 5 a great deal.
individuals in both conditions were highly favorable about their participation.
These participants were quite pleased with having their needs met and with what
they were learning and applying to their daily lives, and were feeling glad they had
participated. Although this provides satisfaction for us in the design and implementation of the intervention, it may also signal the natural benefits of group
interventions and the related opportunities to speak and hear from others about
the loss of a spouse or life partner.
Second, although participant satisfaction was high, we identified only modest
positive support for the DPM intervention versus the traditional bereavement
support format represented by the comparison group condition. Those in both
study conditions were engaging in similar LO coping even though the DPM
group had fewer sessions devoted to it. However, although engagement in LO
coping was similar for both groups, those in the DPM condition more consciously
perceived themselves as engaging in less LO coping by the time the weekly
sessions ended compared to those in the comparison group. Also, it appears that
EXPERIENCES IN INTERVENTION BASED ON THE DPM /
307
Figure 5. Pre- and post-intervention measures of awareness of
restoration-oriented coping by study group. Awareness is measured
with a single-item question (In the past week, to what extent have
you focused your attention on issues related to new responsibilities,
activities, and time away from grieving?), with responses ranging
from 1 very little to 5 a great deal.
the comparison group participants were engaging in some RO coping even
though they were not receiving an intervention targeted toward those processes. It is possible they could have learned new skills, for instance, from other
sources like friends, family members, or elsewhere in the community, perhaps
out of necessity (Lund et al., 1989; Utz, 2006). In a few instances as well, our
group leaders occasionally reported to us that restoration-oriented issues
were unintentionally raised at times by comparison group participants in the
course of the weekly sessions. This potentially initiated some group discussion
related to any issue that was raised as well as attention to restoration-oriented
activity on the part of some outside the context of the group. Consequently,
people in both group conditions were improving with respect to LO and
RO coping from O1 to O2. This could suggest that nearly all bereaved persons
might eventually deal with restoration-oriented stressors and engage in some
restoration-oriented coping.
The above conclusion is particularly interesting in light of how the sessions
were sequenced within the DPM treatment group. Focusing the early sessions on
308 / LUND ET AL.
loss-orientation and the latter ones on restoration-orientation was intended to
approximate the intent of the dual process model (Stroebe & Schut, 1999).
We now recognize, however, that this a potentially artificial approximation
as it appears that RO does tend to occur to some extent early on. Furthermore, our recent work with the IDWL suggested that in some instances an
“unbalanced” oscillation favoring greater restoration-orientation, independent
of how long one was bereaved, was associated with better outcomes. A
greater emphasis on RO was not equivalent with avoidance in that grief,
depression and loneliness levels tended to be lower when this occurred (Caserta
& Lund, 2007). Consequently, we argue that how much emphasis on both LO
and RO processes is optimal throughout the bereavement trajectory still needs
to be investigated and developed further. While we hope to begin to address
this as we examine the long-term impact of the DPM treatment on outcomes,
we believe that future intervention designs, as discussed below, should consider
tailoring the sequencing and content of bereavement interventions to be more
responsive to the unique situations and needs of each participant. This brings
us to our third primary conclusion.
While additional data and analyses are certainly needed in order to provide a
more comprehensive test of the DPM intervention in terms of bereavement
adjustment outcomes over time, it is clear at this point that at least some of the
participants might benefit more from an individually targeted and delivered
RO coping intervention. This comment is particularly supported by the openended data revealing the range and type of topics DPM participants identified
as wanting to be covered.
There is sufficient evidence from the qualitative reports included in this
article as well as from comments made from our group leaders during our project
meetings when the intervention was being delivered that there were problems
and limitations due to providing an RO intervention in group settings. We have
learned that, for some participants, they felt that the RO sessions interrupted the
continuity or flow from one week’s meeting to the next. For example, when one
meeting ended with a discussion of someone’s specific problem, the group was
unable to continue that discussion the next week because an RO expert was
scheduled to talk about an already identified RO coping topic. Also, it is clear that
for some participants, the RO group sessions did not always address their most
pressing needs, or specific sessions were not targeted at their specific level of
knowledge or skill, or the sessions were too brief. Even preferences for speaker
styles were mentioned as problematic for some participants. Our decisions about
which topics to include were based on previous studies (Caserta et al., 2004;
Lund et al., 1989), but, our present intervention involves different participants
with differences in specific needs.
While we remain confident that the DPM as a conceptual framework holds
considerable promise for future research, education, and practice, we need additional data to refine and modify the ways in which we make use of the model.
EXPERIENCES IN INTERVENTION BASED ON THE DPM /
309
It may be likely that the optimal way to deliver the RO feature is to do so
with a targeted, tailored, and individually delivered format so that each
bereaved person identifies the RO skills that are most needed, selects the
format in which they learn the new skills, and in a way that fits their schedule
and desired outcome with the least interference. Other bereavement researchers,
educators, and clinicians have suggested more individualized interventions
(Breen & O’Connor, 2007; Hooyman & Kramer, 2006; Hughes, 1995;
Richardson, 2007; Utz, 2006; Worden, 2002). We realize that grief support
groups offer valuable opportunities to bereaved persons who like that type
of format but it should be possible to provide RO interventions simultaneously
to these groups and do so with more individualized formats and with more
flexible sequencing.
With the addition of O3 and O4 data we will not only gain an improved
understanding of how the intervention may have enhanced LO and RO coping
processes, we will be able to assess the extent to which the expected changes in
bereavement adjustment outcomes were achieved. Also, oscillation between the
two coping processes is obviously the least well developed feature of the DPM
but has considerable promise in facilitating more positive adjustment outcomes.
We suggest that those who are prepared to, more aware of, and able to control their
oscillation will be more adaptable to meet the changing needs and demands of
the complex and long-term nature of bereavement. Future interventions should
be targeted toward enhancing oscillation as a coping strategy. We also remain
excited about using theoretically-based interventions to help those who are in
the greatest need and have the most desire to receive assistance, knowing that
not everyone will require sophisticated and expert interventions. We have learned
that many bereaved persons are quite resilient and find ways to manage many
difficult life transitions (Arbuckle & de Vries, 1995; Caserta et al., 2009; Carr,
2004; Richardson, 2007; Schaefer & Moos, 2001).
ACKNOWLEDGMENTS
The authors wish to acknowledge the helpful contributions of three external
consultants; Drs. Bert Hayslip, University of North Texas; Judith Hays, Duke
University; and Marilyn Skaff, University of California, San Francisco.
REFERENCES
Anderson, K. L., & Dimond, M. F. (1995). The experience of bereavement in older
adults. Journal of Advanced Nursing, 22, 308-315.
Arbuckle, N. W., & de Vries, B. (1995). The long-term effects of later life spousal and
parental bereavement on personal functioning. The Gerontologist, 35(5), 637-647.
310 / LUND ET AL.
Bennett, K. M., Hughes, G. M., & Smith, P. T. (2005). Psychological response to later
life widowhood: Coping and the effects of gender. Omega: Journal of Death & Dying,
51(1), 33-52.
Bisconti, T. L., Bergeman, C. S., & Boker, S. M. (2004). Emotional well-being in recently
bereaved widows: A dynamical systems approach. Journals of Gerontology Series B:
Psychological Sciences & Social Sciences, 59B(4), P158-P167.
Boerner, K., Wortman, C. B., & Bonanno, G. A. (2005). Resilient or at risk? A 4-year
study of older adults who initially showed high or low distress following conjugal
loss. Journals of Gerontology Series B: Psychological Sciences & Social Sciences,
60B(2), P67-73.
Bonanno, G. A. (2004). Loss, trauma, and human resilience: Have we underestimated the
human capacity to thrive after extremely aversive events? American Psychologist,
59(1), 20-28.
Breen, L. J., & O’Connor, M. (2007). The fundamental paradox in the grief literature:
A critical reflection. Omega: Journal of Death and Dying, 55(3), 199-218.
Calhoun, L. G., & Tedeschi, R. G. (2006). The foundations of posttraumatic growth:
An expanded framework. In L. G. Calhoun & R. G. Tedeschi (Eds.), Handbook
of posttraumatic growth: Research and practice. Mahwah, NJ: Lawrence Erlbaum
Associates.
Carr, D. (2004). Gender, preloss marital dependence, and older adults’ adjustment to
widowhood. Journal of Marriage and the Family, 66, 220-235.
Carr, D., House, J. S., Kessler, R. C., Nesse, R., Sonnega, J., & Wortman, C. B.
(2000). Marital quality and psychological adjustment to widowhood among older
adults: A longitudinal analysis. Journal of Gerontology: Social Sciences, 55B(4),
S197-S207.
Carr, D., Nesse, R. M., & Wortman, C. B. (Eds.). (2006). Spousal bereavement in late life.
New York: Springer.
Caserta, M. S. (2003). Widowers. In R. Kastenbaum (Ed.), Macmillan encyclopedia
of death and dying (pp. 933-938). New York: Macmillan Reference USA.
Caserta, M. S., & Lund, D. A. (2007). Toward the development of an Inventory of
Daily Widowed Life (IDWL): Guided by the dual process model of coping with
bereavement. Death Studies, 31(6), 505-534.
Caserta, M. S., Lund, D. A., & Obray, S. J. (2004). Promoting self-care and daily living
skills among older widows and widowers: Evidence from the Pathfinders demonstration project. Omega: Journal of Death & Dying, 49, 217-236.
Caserta, M. S., Lund, D. A., & Rice, S. J. (1999). Pathfinders: A self-care and health
education program for older widows and widowers. The Gerontologist, 39,
615-620.
Caserta, M. S., Lund, D. A., Utz, R., & deVries, B. (2009). Stress-related growth among
the recently bereaved, Aging & Mental Health, 13(3), 463-467.
Chen, J. H., Gill, T. M., & Prigerson, H. G. (2005). Health behaviors associated with
better quality of life for older bereaved persons. Journal of Palliative Medicine, 8(1),
96-106.
Doka, K. A., & Martin, T. (2001). Take it like a man: Masculine response to loss. In
Men coping with grief (pp. 37-48). Amityville, NY: Baywood.
Dutton, Y. C., & Zisook, S. (2005). Adaptation to bereavement. Death Studies, 29(10),
877-903.
EXPERIENCES IN INTERVENTION BASED ON THE DPM /
311
Hansson, R. O., & Stroebe, M. S. (2007). Old age and widowhood: Issues of personal
control and independence. Washington, DC: American Psychological Association.
Hooyman, N. R., & Kramer, B. J. (2006). Living through loss: Interventions across the
lifespan. New York: Columbia University Press.
Hughes, M. (1995). Bereavement and support: Healing in a group environment.
Washington, DC: Taylor & Francis.
Johnson, C. S. (2002). Nutritional considerations for bereavement and coping with grief.
Journal of Nutrition, Health and Aging, 6, 171-176.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: Springer.
Lee, M. A., & Carr, D. (2007). Does the context of spousal loss affect the physical
functioning of older widowed persons? Research on Aging, 29(5), 457-487.
Lund, D. A. (1989). Older bereaved spouses: Research with practical applications.
Washington, DC: Taylor-Francis/Hemisphere Press.
Lund, D. A. (1999). Giving and receiving help during later life spousal bereavement.
In Living with grief at work, at school, at worship (pp. 203-212). Levittown, PA:
Brunner/Mazel.
Lund, D., Utz, R., Caserta, M., & de Vries, B. (2008). Humor, laughter & happiness in
the daily lives of recently bereaved spouses. Omega: Journal of Death & Dying, 58(2),
87-105.
Lund, D. A., Caserta, M., de Vries, B., & Wright, S. (2004). Restoration during bereavement. Generations Review, 14, 9-15.
Lund, D. A., & Caserta, M. S. (1998). Future directions in adult bereavement research.
Omega: Journal of Death and Dying, 36, 287-303.
Lund, D. A., & Caserta, M. S. (2002). Facing life alone: The loss of a significant other
in later life. In K. Doka (Ed.), Living with grief: Loss in later life. Levittown, PA:
Brunner/Mazel.
Lund, D. A., Caserta, M. S., Dimond, M. F., & Shaffer, S. K. (1989). Competencies,
tasks of daily living and adjustments to spousal bereavement in later life. In Older
bereaved spouses: Research with practical applications (pp. 135-156). Washington,
DC: Taylor-Francis/Hemisphere Press.
Montpetit, M. A., Bergeman, C. S., Bisconti, T. L., & Rausch, J. R. (2006). Adaptive
change in self-concept and well-being during conjugal loss in later life. International
Journal of Aging & Human Development, 63(3), 217-239.
Moss, M. S., Moss, S. Z., & Hansson, R. O. (2001). Bereavement and old age. In Handbook
of bereavement research: Consequences, coping, and care (pp. 241-260). Washington,
DC: American Psychological Association.
O’Rourke, N. (2004). Psychological resilience and the well-being of widowed women.
Ageing International, 29(3), 267-280.
Ong, A. D., & Bergeman, C. S. (2004). Resilience and adaptation to stress in later
life: Empirical perspectives and conceptual implications. Ageing International, 29(3),
219-246.
Ong, A. D., Bergeman, C. S., & Bisconti, T. L. (2004). The role of daily positive emotions
during conjugal bereavement. Journal of Gerontology: Psychological Sciences, 59B,
P168- P176.
Pienta, A. M., & Franks, M. M. (2006). A closer look at health and widowhood: Do
health behaviors change after the loss of a spouse? In D. Carr, R. M. Neese, & C. B.
Wortman (Eds.), Spousal bereavement in late life (pp. 117-142). New York: Springer.
312 / LUND ET AL.
Powers, L. E., & Wampold, B. E. (1994). Cognitive-behavioral factors in adjustment to
adult bereavement. Death Studies, 18, 1-24.
Quandt, S. A., McDonald, J., Arcury, T. A., Bell, R. A., & Vitolins, M. Z. (2000).
Nutritional self-management of elderly widows in rural communities. Gerontologist,
40, 86-96.
Richardson, V. E. (2007). A dual process model of grief counseling: Findings from the
changing lives of older couples (CLOC) study. Journal of Gerontological Social Work,
48, 311-329.
Richardson, V. E., & Balaswamy, S. (2001). Coping with bereavement among elderly
widowers. Omega, 43, 129-144.
Rosenbloom, C. A., & Whittington, F. J. (1993). The effects of bereavement on eating
behaviors and nutrient intakes in elderly widowed persons. Journal of Gerontology:
Social Sciences, 48, S223-S229.
Schaefer, J. A., & Moos, R. H. (2001). Bereavement experiences and personal growth.
In Handbook of bereavement research: Consequences, coping and care (pp. 145-167).
Washington, DC: American Psychological Association.
Schone, B. S., & Weinick, R. M. (1998). Health-related behaviors and the benefits of
marriage for elderly persons. Gerontologist, 38, 618-627.
Shahar, D. R., Schultz, R., Shahar, A., & Wing, R. (2001). The effect of widowhood on
weight change, dietary intake, and eating behavior in the elderly population. Journal
of Aging and Health, 13, 186-199.
Stroebe, M. S., & Schut, H. (2010). Update on DPM. Omega, 61(4), 275-291.
Stroebe, M. S., & Schut, H. (1999). The dual process model of coping with bereavement:
Rationale and description. Death Studies, 23, 197-224.
Stroebe, M. S., Stroebe, W., & Hansson, R. O. (1993). Handbook of bereavement: Theory,
research, and intervention. New York: Cambridge University Press.
Stroebe, M., Stroebe, W., & Schut, H. (2003). Bereavement research: Methodological
issues and ethical concerns. Palliative Medicine, 17, 235-240.
Utz, R. (2006). Economic and practical adjustments to late life spousal loss. In D. Carr,
R. M. Nesse, & C. B. Wortman (Eds.), Spousal bereavement in late life (pp. 167-192).
New York: Springer.
Utz, R. L., Carr, D., Nesse, R., & Wortman, C. (2002). The effect of widowhood on older
adults’ social participation: An evaluation of activity, disengagement, and continuity
theories. The Gerontologist, 42(4), 522-533.
Utz, R. L., Reidy, E., Carr, D., Nesse, R., & Wortman, C. B. (2004). The daily consequences of widowhood: The role of gender and intergenerational transfers on
subsequent housework performance. Journal of Family Issues, 25, 683-712.
Wells, Y. D., & Kendig, H. L. (1997). Health and well-being of spouse caregivers and
the widowed. The Gerontologist, 37(5), 666-674.
Wilcox, S., Evenson, K. R., Aragaki, A., Wassertheil-Smoller, S., Mouton, C. P., &
Loevinger, B. L. (2003). The effects of widowhood on physical and mental health,
health behaviors, and health outcomes: The women’s health initiative. Health
Psychology, 22(5), 1-10.
Williams, K. (2004). The transition to widowhood and the social regulation of health:
Consequences for health and health risk behavior. Journals of Gerontology Series B:
Psychological Sciences & Social Sciences, 59B(6), S343-S349.
EXPERIENCES IN INTERVENTION BASED ON THE DPM /
313
Worden, J. W. (2002). Grief counseling and grief therapy: A handbook for the mental
health practitioner (3rd ed.). New York: Springer.
Znoj, H. (2006). Bereavement and posttraumatic growth. In L. G. Calhoun & R. G.
Tedeschi (Eds.), Handbook of posttraumatic growth: Research and practice
(pp. 176-196). Mahwah, NJ: Lawrence Erlbaum Associates.
Direct reprint requests to:
Dr. Dale Lund
Department of Sociology
California State University San Bernardino
5500 University Parkway
San Bernardino, CA 92407
e-mail: [email protected]
OMEGA, Vol. 61(4) 315-332, 2010
LOSS AND RESTORATION IN LATER LIFE:
AN EXAMINATION OF DUAL PROCESS MODEL
OF COPING WITH BEREAVEMENT*
KATE M. BENNETT
KERRY GIBBONS
SUZANNA MACKENZIE-SMITH
University of Liverpool, United Kingdom
ABSTRACT
The Dual Process Model (DPM) of Coping with Bereavement identified two
oscillating coping processes, loss and restoration (Stroebe & Schut, 1999).
The utility of the model is investigated in two studies. In the first, we carried
out secondary analyses on a large-scale qualitative study that we had conducted previously. In the second, we conducted a small-scale study specifically examining the DPM. In the first study we re-examined the interviews
for Loss- (LO) and Restoration-Oriented (RO) Coping and examined
whether these were associated with psychological adjustment. The results
showed that those adjusting well reported the stressors New Roles/Identities/
Relationships and Intrusion of Grief significantly more. Those adjusting less
well reported the stressors Denial/Avoidance of Restoration Changes and
Distraction/Avoidance of Grief significantly more. In the second study, we
asked participants about four RO stressors of the DPM: Attending to Life
*Study 1 was supported by the Economic and Social Research Council (Award No.
L480254034) and is part of the Growing Older Programme of 24 projects studying the quality
of life of older people. However, the findings reported here are entirely the responsibility of
the researcher.
315
Ó 2010, Baywood Publishing Co., Inc.
doi: 10.2190/OM.61.4.d
http://baywood.com
316 / BENNETT, GIBBONS AND MACKENZIE-SMITH
Changes; New Roles/Identities/Relationships; Distraction from Grief; and
New Activities. These data showed that not all participants experienced all
aspects of RO Coping. In particular, participants had diverse views about
the utility of Distraction from Grief as a coping mechanism. The article
concludes by discussing the challenges of testing the DPM empirically.
Over recent years researchers have debated the efficacy of traditional models
of bereavement (Bennett & Bennett, 2000-2001; Footman, 1998; Walter, 1996).
At the same time those working in the field of stress have developed models
to explain how people cope with life events in general (Carver, Scheier, &
Weintraub, 1989; Lazarus & Folkman, 1984; Moos, 1995). Until 1999 these
two areas of research proceeded independently even though bereavement is
a subset of life events. However, in 1999, Stroebe and Schut published their
DPM that aimed to address deficits in both areas. In this issue they describe the
model in detail.
Relatively few studies have examined the DPM empirically (Richardson
& Balaswamy, 2001; Richardson, 2007). Richardson and Balaswamy (2001)
examined the model in a sample of older widowed men during the second year
of their bereavement. They found that both Loss- (LO) and Restoration-Oriented
(RO) stressors were important. The former were more important nearer the time
of bereavement and influenced negative affect. On the other hand, RO stressors
were more relevant later and were often associated with positive affect. In addition, the circumstances of the death were also significant in adjustment, an
aspect not accounted for by the DPM. More recently, Richardson (2007) studied
older widows and widowers over a longer timeframe. She confirmed that both
LO and RO coping were important throughout bereavement. She found a negative
relationship between rumination and affect and argued that it was important to
help widowed people distinguish between constructive and destructive grief
work. Finally, she suggested that widowed people needed to balance out LO and
RO activities; a focus on one over and above the other might be problematic for
recovery. An important question which researchers have so far been unable to
address is whether there is an optimal balance between LO and RO coping, and if
so, does that balance change over time? A second question, that researchers do not
yet know the answer to, is whether there are optimal degrees of LO or RO coping
at different points during the course of bereavement.
Lund, Caserta, de Vries, and Wright (2004) have also considered the DPM
therapeutically. They developed a 14-week intervention tailored around LO and
RO coping. Their proposed sessions included discussions about the physical,
cognitive, and behavioral aspects of grief through to the management of household, motorcar, and nutritional responsibilities. The former discussions focus
on LO, while the latter focus on RO. In 2004, they had been awarded funds to
test this intervention but as yet there has been no published account of the
LOSS AND RESTORATION IN LATER LIFE /
317
empirical results. These are eagerly anticipated and an update is presented in
this issue. In addition, using the DPM as a theoretical basis Caserta and Lund
(2007) developed the Inventory of Daily Widowed Life, which had both LO
and RO subscales. They found that those more recently widowed experienced
greater oscillation, while those bereaved for longer experienced more RO coping
than LO coping.
In 2005, Shear, Frank, Houck, and Reynolds also used the DMP as the basis
for their randomized control trial of treatment for complicated grief. In their
complicated grief treatment not only did they focus on grief but they also
focused on personal life goals, therefore addressing both LO and RO coping.
Their results showed that this approach was more effective than an interpersonal
psychotherapy intervention (see also this issue).
Recently, Stroebe, Folkman, Hansson, and Schut (2006) have proposed a
framework which integrates both the DPM and Cognitive Stress Theory (Lazarus
& Folkman, 1984). Their aim is to enhance the understanding of risk factors
in adaptation to bereavement. They suggest that using this framework may
enable clinicians to focus on those people who are at risk of serious reactions to
bereavement and recognize that for most people bereavement does not require
intervention. They also hope that this approach may be of benefit to theoreticians
in enabling them to test their hypotheses more effectively. As yet this approach
has not been tested empirically.
In two earlier papers we established that it was possible to distinguish between
those participants who demonstrated good psychological adjustment (the term
coping well was used in the original paper) and those who did not (Bennett,
Hughes, & Smith, 2005a, 2005b). In our 2005a paper we found that those who
were less well adjusted had more depressive feelings than those who were more
well adjusted. In the 2005b paper we showed that talking to one’s dead spouse was
associated with good adjustment, while keeping oneself to oneself was associated
with poor adjustment. In addition, we also found that there were three behaviors
that showed an adjustment and gender interaction. Men were more likely to adjust
better if they described themselves as “selfish” and as “upset,” women were more
likely to adjust better if they described themselves as being comfortable “alone.”
Here we present two studies. The first re-examines this earlier data in the light
of the DPM. Codes that reflect the components of LO and RO coping are examined
to see whether they are associated with psychological adjustment. We would
predict, based on earlier work, that some stressors such as Doing New Things and
Breaking Bonds/Ties/Relocation are common to most bereaved people. On the
other hand, we would predict that those engaged in Distraction from Grief
or in Denial/Avoidance of Restoration Changes would be less well adjusted,
and those with New Roles/Identities/Relationships would be more likely to be
well adjusted. In the second study, we focus only on RO stressors to provide
an in-depth understanding of how these are incorporated into the daily lives
of older widows.
318 / BENNETT, GIBBONS AND MACKENZIE-SMITH
STUDY 1
Method
Participants
The participants were 46 widowed men (only 45 of whom were interviewed)
and 46 women. In the analyses presented here, one woman was excluded since
she had been widowed for 60 years. The remaining participants were aged
between 55 and 95 years (mean = 74), living in Merseyside, United Kingdom.
They had been widowed between 3 months to 32 years (mean 8.68 years).
Although, some had been widowed for several years, we decided not to exclude
them because our previous work has suggested that the effects of bereavement
last for a longer time than has previously been found and even those widowed for
a long time may still be making RO adjustments (Bennett, 1997; Davidson, 1999;
Moore & Stratton, 2003). Demographic details are summarized in Table 1 and
include information concerning the excluded woman for consistency with other
papers published from the study.
Recruitment
The research team communicated the aims of the Older Widowed Men and
Women Project to a diverse range of formal and informal groups of older people.
We also made contact with other welfare organizations and agencies, social
services, and sheltered housing schemes, through which links with widowed
people were established. The local ethics committee approved the study and
confidentiality and anonymity were assured. Names have been changed to
preserve anonymity.
Table 1. Demographic Data by Gender of
Study 1 Participants
Mean
Standard deviation
Range
Women (n = 46)
Age
Years married
Years bereaved
73.29
35.75
10.94
8.93
10.70
1.72
57-95
2-63
1-60
Men (n = 46)
Age
Years married
Years bereaved
75.02
39.37
8.18
7.88
12.97
6.72
55-93
5-63
0.25-25
LOSS AND RESTORATION IN LATER LIFE /
319
The Interview
We tape-recorded the semi-structured interviews that lasted from 45 minutes to
1½ hours, conducted in the respondents’ own homes. Respondents gave informed
consent. The interview was not prescriptive; the aim was to learn from the
widowed people what was important to them. The approach was “I am the novice
and you have the experience.” The interview schedule consisted of seven parts.
The first part contained factual questions concerning age, length of marriage,
widowhood, and family relations, followed by four sections inquiring about the
widowed person’s life at various times. The first of these addressed married life
before the death of the spouse, asking questions about hobbies, division of labor,
and marital quality. The second section asked about the time around the death of
the spouse. They were asked to describe what a typical day had been like after the
death, what support they had, and how they had felt. They were then asked what
they did and how they felt 1 year on. They were asked how their lives had changed
by then, what a typical day was like at that stage, whether they were now doing
anything new, and in what ways (if any) their feelings had changed. The fifth
section asked what their lives were like at the present time. Questions related to
what they did with their time, how they felt about their widowhood, how their lives
had changed, and what their emotions were.
Comment on the Method
There are two potential disadvantages with this type of retrospective interview.
First, it assumes that widowed people accurately recall the events surrounding
and following the deaths of their spouses. Second, recall may be subjective and
dependent on the individual experiences of the bereaved, and may indeed differ
from the recollections of bereaved children, for example. However, it is difficult
to obtain contemporaneous qualitative accounts of these experiences, particularly
in relationship to those that occur very close to the death, for both practical and
ethical reasons. We have also found that respondents appear to have detailed
recollections about some events, such as those leading up to the death, and poor or
non-existent recollections of the events immediately after the death—participants
report that these events are often a “blur” (Bennett & Vidal-Hall, 2000). It is
also the case that it is the participant’s subjective experience that is of interest:
there is a great deal of evidence, for example, that subjective rather than objective
measures of health are better predictors of mortality (Benyamini, Blumstein,
Lusky, & Modan, 2003). The lived experience of the bereaved is important in
understanding the DPM, and in particular in the context of its relationship to
wellbeing and adjustment. Another potential disadvantage of the method is the
variety of length of times since bereavement that are used in the study. As
mentioned earlier, other authors have also found that there is great variation in
the impact that time since bereavement has on widowed people’s lives. If we
were to confine our data collection to those bereaved within, for example 2 years
320 / BENNETT, GIBBONS AND MACKENZIE-SMITH
(as many studies do, see Zisook, Paulus, Shuchter, & Judd, 1997), we might
miss important bereavement experiences that occur much later than others have
suggested (Bennett, 1997; Davidson, 1999; Moore & Stratton, 2003).
Analysis
Two members of the original research team coded the interviews using
grounded theory and content analysis methods (Bennett & Vidal-Hall, 2000;
Charmaz, 1995; Glaser & Strauss, 1969; Smith, 1995). Each transcript was first
read through in its entirety to gain an impression of the interview. It was then
re-read line-by-line and coded. This process was reflexive; as new topics emerged
they were looked for in earlier parts of the interview. Examples of coded topics
include: guilt, independence, presentation of husband, quality of marriage, and
death narrative. The transcripts were further examined for broader themes. A
number were common to all the interviews and these included: the domain of
death, the social domain, the emotional domain and the domain of time. Brief
memos were written for each interview (see also Bennett & Vidal-Hall, 2000).
In addition reliability checks were undertaken. The reliability was found to be
satisfactory. A total of 311 codes emerged from analysis of the 91 interviews.
Reliability was assessed and agreement was found to be 80% between the coders.
This article focuses on codes relevant to the DPM (Stroebe & Schut, 1999).
An initial trawl of the codes suggested that 56 might be relevant. Careful reading
of the texts suggested that 44 described aspects of DPM. These codes were then
assigned to the nine individual aspects of LO stressors (Grief Work; Intrusion
of Grief; Breaking Bonds/Ties/Relocation; and Denial/Avoidance of Restoration Changes) and RO stressors (Attending to Life Changes; Doing New Things;
Distraction from Grief; Denial/Avoidance of Grief; and New Roles/Identities/
Relationships). Finally, the frequency with which each of the nine stressors
occurred was calculated.
We assessed psychological adjustment by expert reading of the interviews
and assessment of non-verbal aspects of the interview. For example, coders looked
for reports of medication, contact with primary care, not coping. In addition, the
non-verbal content of the interviews was taken into account. If a participant
mentioned specifically that they were not coping, they were classified as not
adjusting well. They were classified either as psychologically well-adjusted or
not (and in previous work as good coper or poor coper). Characteristically, people
showing good adjustment had developed a life without their spouse, were not
unduly distressed during the interview, were able to discuss the issues surrounding
their bereavement and widowhood in positive as well as negative terms, and
described the events surrounding their bereavement with a degree of distance
(see also Bennett et al., 2005a, 2005b). Two members of the team made the
assessment, again by reading the transcripts independently, and agreement was
found to be 95%. The interviews where there was not agreement were sent to
the third member of the team for final decision.
LOSS AND RESTORATION IN LATER LIFE /
321
We used chi-square analysis to test whether particular stressors were more
common among those with good or poor adjustment.
Results
Table 2 shows the frequency table and c2 for LO and RO coping. Only two
LO stressors differentiated between good and poor adjustment. Significantly
more people with good adjustment reported the stressor Intrusion of Grief than
those with poor adjustment (c2 = 6.58, p £ .001) and the reverse was true for
Denial/Avoidance of Restoration Changes (c2 = 16.68, p £ .001). In addition, it
is worth noting that the majority of respondents reported Breaking Bonds/Ties/
Relocation. On the other hand, relatively few reported Grief Work. There were
two RO stressors that differentiated between good and poor adjustment. More
people with good adjustment reported New Roles/Identities/Relationships (c2 =
5.42, p £ .02) more often and those with poor adjustment reported the stressor
Distraction from Grief more frequently (c2 = 4.05, p £ .04). Many participants
reported the stressors Doing New Things and Avoidance/Denial of Grief.
We were also interested in seeing whether we could use this data to begin
to assess oscillation. As a staring point we began by calculating how many
participants experienced both LO and RO coping. We suggest that those people
who reported only one type of coping were unlikely to be oscillating. Twelve
participants were, therefore, identified as not oscillating. Eleven of those did not
experience LO coping, and only one did not experience RO coping. The remaining
79 did experience both RO and LO coping, therefore, could have experienced
oscillation (87%). We then calculated whether experiencing oscillation was significantly associated with good psychological adjustment but we did not find a
significant association (Fisher’s exact test: p = .67). Our analysis is not as refined
as that conducted by Caserta and Lund (2007), and as they point out it is likely
that oscillation is complex, multi-dimensional, and dynamic. Nevertheless, these
analyses suggest that oscillation, or something resembling it, is common among
widowed people.
STUDY 2
Method
Participants
Thirteen widowed women were interviewed in this small-scale study (one
widow was excluded because she had been widowed for 25 years). These widows
had been widowed from 1.5 to 16 years (mean 7 years) and were aged from 51 to
85 (mean 70 years) (see Table 3). The sample was an opportunity sample recruited
both from the NW of England and the South of England. The University of
Liverpool’s Research Governance Committee approved study and confidentiality
and anonymity were assured. Names have been changed to preserve anonymity.
322 / BENNETT, GIBBONS AND MACKENZIE-SMITH
Table 2. Chi-Square Analysis of Coping and Use of Dual Process
Model Categories from Study 1
Use of Dual Process Model
Categories
Loss-Oriented
Grief Work
Yes
No
Intrusion
Yes
No
Breaking Bonds/Ties
Yes
No
Denial/Avoidance of Restoration
Changes
Yes
No
Restoration-Oriented
Attending to Life Changes
Yes
No
New Activities
Yes
No
Distraction from Grief
Yes
No
Denial/Avoidance of Grief
Yes
No
New Roles
Yes
No
c2
p
Overall % of
participants
using strategy
0.65
.42
23%
6.58
.001
59%
0.08
.78
67%
16.68
0.001
27%
1.01
.32
71%
.93
80%
4.05
0.04
24%
0.43
.84
63%
5.42
0.02
54%
Coping
Yes
No
17
48
9
17
44
21
10
16
43
22
18
8
10
55
15
11
35
7
30
19
52
13
21
5
12
53
10
16
41
24
17
9
40
25
9
17
.007
The Interview
The interviews were conducted by KG and SM-S, who were undergraduate
students, for their final year research project. They were both young women
aged from 20 to 22 at the time of the interviews. They were trained by KMB
who has substantial experience in interviewing widowed women. There was
no evidence that the quality or detail of the interviews was significantly different
from that of Study 1.
LOSS AND RESTORATION IN LATER LIFE /
323
Table 3. Study 2 Demographic Detailsa
Mean
Standard deviation
Range
Age
71.1
9.75
51-85
Years married
34.1
15.0
10-59
6.5
4.3
1.5-1.6
Years bereaved
aExcludes Widow 6 who was widowed for 25 years.
We tape-recorded the semi-structured interviews that lasted from 45 minutes to
1½ hours, conducted in the respondents’ own homes. Respondents gave informed
consent. The interview focused on RO coping. Participants were asked about
practical changes that they had made since their spouse died and how difficult it
had been to make those changes (Attending to Life Changes). They were asked
whether they had taken up any new activities and why they had done so (Doing
New Things). They were asked about the kinds of things they did to keep
their minds off the loss and whether these had helped (Distraction from Grief).
Finally, they were asked whether they had any new roles, relationships, or a
changed sense of identity (New Roles/Identities/Relationships) and what their
impact had been. We did not ask participants about Denial/Avoidance of Grief
because we wished to avoid highly sensitive topics, as undergraduate students
undertook these interviews. In the event, participants did talk spontaneously
about sensitive subjects and about LO coping. When this occurred, participants
were not discouraged from discussing these issues, and the interviewers responded
sensitively. No participants became unduly distressed during the interviews.
Analysis
As with Study 1, the interviews using grounded theory and content analysis
methods (Bennett & Vidal- Hall, 2000; Charmaz, 1995; Glaser & Strauss, 1967;
Smith, 1995). In traditional grounded theory methods of analysis there are
no pre-conceived views about what the data will show. Rather, the data is
read with a view to identifying new areas for theoretical development. However, we were primarily interested in exploring issues raised by the DPM and,
therefore, there were a priori questions that we wanted to address which have
been outlined above.
Each transcript was first read through in its entirety to gain an impression
of the interview. It was then re-read line-by-line and coded. This process was
reflexive; as new topics emerged they were looked for in earlier parts of the
interview. All interviews were read and coded independently by all three authors.
Any discrepancies between coders were discussed and resolved. However, in
324 / BENNETT, GIBBONS AND MACKENZIE-SMITH
general, there were high levels of agreement. The focus, initially, of the coding
was on RO but, since LO themes emerged spontaneously, these were also coded
where they were present.
Results
Attending to Life Changes
The results indicate that one of the most challenging and indeed immediate
changes that had to be made concerned finance. Several of the women had not
managed their finances before and they had to learn how to do this. For example,
Widow 4 closed down all the automatic payments. Her quote illustrates not
only the practical challenges but also the psychological challenges of bereavement, that is, the need to establish control over her new life and to reduce
unnecessary anxiety:
He paid everything on direct debit, and I closed them all down, because
it frightened me to think that this money could go out.
For Widow 14, the attending to life’s changes had already begun prior to her
bereavement with her husband’s illness, but she also illustrates the financial
strains that accompanies widowhood for many women:
That was quite hard because Ken was always in charge of the bills . . . but
Ken was ill I had time to get used to sorting things out.
My finances just hit the floor because when he died, everything died with him.
In other work, with both men and women, pre-bereavement caring and anticipation has also been found to be useful in preparing people for the practical
challenges that bereavement brings (Bennett, 2007). For other women, prior
experience also assisted them in attending to life’s changes:
There was nothing different in that because I always pay the bills.
Thus, from a practical standpoint prior learning and pre-bereavement independence are valuable in attending to life’s changes and contribute to lessening
the stresses of bereavement. The ability to carry out these tasks may also have an
impact in lessening the psychological burden that acquiring new skills often
entails, allowing more psychological resources to be directed where they are
most needed. There is the potential, for soon-to-be-widowed people at least (for
example, those with terminally ill spouses), to receive pre-widowhood training
in the same way that Lund and colleagues proposed post-bereavement (2004).
It is interesting that these widows focused particularly on the financial skills
that they need to acquire. It is likely that this is a cohort effect. Women born later
may have more experience of managing their own finances as they begin to spend
more time living alone, rather than going straight from parental to marital home,
and also spend more time in the labor force. Nevertheless, whenever a spouse dies
LOSS AND RESTORATION IN LATER LIFE /
325
there are likely to be financial matters that need to be resolved and redistributed,
along with other practical life changes.
Doing New Things
In the first study we found that the majority of participants engaged in new
things. This was also the case in this study. Some women took up voluntary work
or education as in the cases of Women 9 and 4. Widow 9 points out that it was
something that she had taken up before her husband died, put on hold while he
was dying and then took up again:
I started this voluntary work when I retired, and I’d just started it funnily
enough, just before he died. But I’d put my life on hold for a year because I
knew he was probably going to die in that time. (Widow 9)
I feel as if I could cope with the studying now. (Widow 4)
For other women, family activities were important new things. In the case of
Widow 8 it was the birth of a grandchild that provided the focus. In the course
of the interview she describes how she transferred her love from her husband
to her granddaughter:
It’s having the granddaughter and stuff that has actually saved me . . . it
seemed like a replacement thing to love.
Finally, there were widows who took up multiple activities. Widow 15 provides a
list of the clubs she has joined and Widow 11 recalls how it was a chosen strategy
(and we shall return to her shortly):
I joined the B* Support group, . . . I’ve joined Fairbrook, . . . a scrabble club.
(Widow 15)
In the initial period I made a point of never saying no . . . a consciously chosen
strategy. (Widow 11)
Widow 11 describes the taking up of new activities as a deliberate strategy and
it is evidence that some behaviors fit into more than one type of DPM coping
behavior, in this case Doing New Things and Distraction from Grief.
Distraction from Grief
When we asked about Distraction from Grief two clear opinions were held
by the widows. Their strength of opinion is interesting for two reasons. First, it
was the only one of the stressors where the women discussed it in strategic terms.
Second, they reflect on a debate that occurs within the bereavement community,
both academic and practice-based. Widows who practice distraction as a means
of coping with bereavement hold the first view, exemplified by Widow 11 above.
Widow 1 who avoided going to sleep at first and later to bed, because at first she
could not face waking up and later not face waking up alone, also illustrates it:
326 / BENNETT, GIBBONS AND MACKENZIE-SMITH
I didn’t go to bed. I couldn’t. . . . And at first I didn’t want to go to sleep.
Other women talk about staying away from home or avoiding rooms or reminders
of their spouses. On the other hand, there were women who firmly believed
that distraction was unhelpful and indeed that found solace from the closeness
with their dead spouse:
I’ve never made any attempt to keep my mind off it cos I don’t think that’s
a particularly good thing to do. (Widow 8)
I used to go into his room and tidy all his things . . . because I felt really
close to him. (Widow 4)
New Roles/Identities/Relationships
Finally, women discussed their new roles, identities, and relationships. For
two of the women there was a clear association between their new role and roles
held by their husbands, what might be seen as identification:
He was chairman of a World War I association and I took that on. (Widow 9)
I felt it was a more positive thing to do to have a kind of living memorial
and pass on something that had been characteristic of them . . . so . . . I set
up a Latin club. (Widow 13)
Many of the women discussed how their characters and outlooks have changed.
Widows 2 and 4 typify these changes, respectively:
I’m like a bird out of a cage. (Widow 2)
I’ve become stronger. (Widow 4)
Finally, we asked about intimate relations. Often the widows do not want the
burden of looking after a man:
I wouldn’t want a relationship, to have to look after a man. (Widow 12)
For others, they see both advantages and disadvantages:
The independence on the one hand that you wish you had someone to help
you with, but independence, on the other hand where you can sit and
watch what you like. (Widow 11)
Among the younger widows there was more interest in finding another partner:
I’m interested but there aren’t any men about are there (laughs), none that
fancy me. (Widow 13, age 51)
Finally, there were two widows who had been widowed, repartnered, and then
widowed again. Widow 10 had been married and widowed twice, and Widow 3
had been widowed once and then had been bereaved of her LAT (living alone
together) partner—of whom she said:
He was such a soul mate for me.
LOSS AND RESTORATION IN LATER LIFE /
327
DISCUSSION
We present data from two studies that looked at the DPM in detail. The first
found that there was an association between some components of the DPM and
psychological adjustment, the second provided qualitative data on the ways in
which widows experienced RO coping.
In the first study we found that four components of the Dual Process Model
differentiated between good and poor adjustment. Two, Intrusion of Grief and
New Roles/Identities/Relationships, were associated with good adjustment and
two, Denial/Avoidance of Restoration Changes and Distraction from Grief, were
associated with poor adjustment. The remainder did not significantly differentiate
adjustment. Many experienced some components such as Doing New Things,
whereas Grief Work was experienced by relatively few.
Turning first to behaviors associated with LO coping, one might not expect
Intrusion of Grief to be used more often by those who are adjusting well,
especially given the negative valence of the word “intrusion.” However, when
we were looking at our existing analysis and codes which fitted with Stroebe
and Schut’s description of “intrusion” we identified some of the following: talking
to the deceased; projecting and identifying with the deceased; and memorials.
These codes appear to be more positive than the word “intrusion” implies, and are
perhaps more associated with the continuation of a bond with the deceased.
In addition, when reading the interviews, it is clear that these experiences give
the bereaved comfort rather than causing distress. Given the association between
the frequency of Intrusion of Grief and good adjustment, we would suggest that
intrusion be relabeled Continuing Bonds.
In contrast, Denial/Avoidance of Restoration Changes is associated with
poor adjustment. The codes which reflect this type of coping behavior are: life
at an end; no change; and kept-self-to-self. In previous work this last code
differentiated, on its own, between good and poor adjustment (Bennett et al.,
2005b). People who report these are clinging onto their previous marital state
and are maintaining an isolation both from their new status and from the social
world, at an emotional level.
These two contrasting findings address a hot topic in bereavement research—
“to continue or relinquish bonds” (Stroebe & Schut, 2005, p. 477; also Schut,
Stroebe, Boelen, & Zijerveld, 2006). In their 2005 paper, Stroebe and Schut
reviewed the evidence for and against the maintenance of the bond. It suggests,
they argue, that certain types of continuing bond may be helpful while others
might be harmful, and similarly it may be the case for relinquishing bonds.
Further, some people may need help to relinquish bonds while others might
not. The current results appear to differentiate between two types of bond. Those
reporting Intrusion of Grief, or as we believe more appropriately, Continuing
Bonds adjust well. These participants do not deny their changed situation nor
are they avoiding making necessary changes. Rather they are using their bonds
328 / BENNETT, GIBBONS AND MACKENZIE-SMITH
with the deceased to make those changes—to seek advice from, and talk things
over with, the deceased and considering what the deceased might have done had
the situation been reversed. Those who adjust less well report Denial/Avoidance
of Restoration Changes. This avoidant situation might reflect what Stroebe
and Schut (2005) describe as clinging to a past attachment, where there is no
acceptance of the necessity for change. It is possible that the Denial/Avoidance
of Restoration Changes and Intrusion of Grief are opposite ends of the same
continuum. Further research is needed to explore these issues.
We found that most people Do New Things but only those who have New
Roles/Identities/Relationships are found to adjust better. In contrast, those who
engage in Distraction from Grief are found to adjust less well. When analyzing
interviews one of the challenges we face is in deciding how to classify a particular
response, especially with respect to these three stressors. The original DPM
paper does not specify the characteristics of these stressors. Following our studies
we believe we can look for two responses in the data, or from the participant that
can enable us to classify a behavior. First, does the response primarily concern
feeling or doing? If the response concerns doing then we could classify that
response as Doing New Things. If, in addition, it also concerns feeling, then we
could classify it as New Roles/Identity/Relationships. This distinction might
explain why although many people do new things, it is only those who take on a
New Role/Identity/Relationship that have better adjustment. Thus, the psychological benefit comes not from doing new things, but from feeling something
new about oneself. Second, what is the motivation for a particular behavior?
For example, a new activity might be undertaken simply for the fun of it, in
which case we might classify that behavior as Doing New Things, on the other
hand a new activity might be undertaken to keep one’s mind of one’s loss, in
which case it might be Distraction from Grief. Although we find that Doing
New Things is not associated with any psychological benefit, Distraction from
Grief is associated with poor adjustment.
The value or benefit of a particular RO coping strategy may also change over
time. For example, an activity that initially was undertaken as a distraction, may
take on new meaning and significance. An excellent illustration comes from
an earlier study of widows that recruited from a Widows Club (Bennett &
Vidal-Hall, 2000). I found women had joined the club to distract themselves
from the loneliness of Sunday afternoons (Distraction from Grief), who then went
on holiday with the club (New Activities) and then became organizers of the
club and developed “true” friendships (New Roles/Identities/Relationships). For
these women, and for the women in the current studies, a single behavior may
have multiple functions in relationship to the DPM, which can depend on the time
course, on motivation, and on other situational factors. It is, therefore, important
not to confine the analysis of the DPM to a narrow time period. This example also
illustrates the value of qualitative research in capturing the multi-dimensional
nature of LO and RO that might be less easily captured by quantitative methods.
LOSS AND RESTORATION IN LATER LIFE /
329
Alongside Doing New Things there are also other behaviors that are undertaken
by most people. Almost everyone begins to break bonds. Most people discuss
the decision to dispose of their spouses’ possessions and this is a task that must
be tackled to some degree. The majority of people discard most things at some
point (sometimes early on, sometimes later) although many keep hold of one or
two significant things. Thus, people are simultaneously continuing and breaking
bonds. It is not a case of one or the other. In addition, the majority of people also
discuss times when they experienced Denial/Avoidance of Grief. People often
talk of not wanting to be in the house, or of the distress they feel at returning to
an empty house, or sleeping in an empty bed. They talk about the numbness and
the lack of memories concerning the days immediately after the death. These
appear to be common, and are part of the normal pattern of grief. Thus, some
periods of avoidance may be necessary in order to survive the pain of bereavement. Finally, although most people have to Attend to Life Changes, it is important to consider what the components of Attending to Life Changes might be.
At first glance the focus is on practical tasks and illustrated by our participants.
The ease to which some of these tasks were attended was influenced by at least
two factors. Those widows who had prior experience, either as a consequence of
an independent personality or lifestyle, or those women who as a consequence
of prior caregiving, were more able to attend to these changes (and this has been
found elsewhere; Bennett, Hughes, & Smith, 2003). In addition, there is some
evidence that there is also a psychological component to Attending to Life
Changes, and this illustrated by Widow 4 who changed the way her bills were
managed to exercise control. It would be interesting to explore the psychological
components of this type stressor, since this may have an impact on psychological adjustment. Thus, as with the other RO stressors, it may be important to
consider the underlying meaning of a particular activity—does it represent simply
a functional change or does it represent a psychological change?
Richardson and Balaswamy (2001) found that LO coping was associated
more often with negative affect and restoration with positive. Although positive
and negative affect cannot be directly mapped onto good and poor adjustment
there may be similarities. If that is the case the current results are different from
theirs. However, these results do reflect another literature, that of the traditional
stress’ literature (Carver et al., 1989; Folkman & Lazarus, 1980). Those behaviors
associated with poor adjustment, Denial/Avoidance of Change and Distraction
from Grief, may be seen as avoidant coping strategies. On the other hand, those
associated with good adjustment, Intrusion of Grief and New Roles/Identities/
Relationships, may be seen as confrontative coping strategies.
There were some interesting attitudes toward new relationships. Many men
in the first study discussed the question of whether they would remarry and
considered the idea positively, whereas the women in that study were unenthusiastic about remarriage or repartnering. However, relatively few men or women
had embarked on new relationships, although this may have been because the
330 / BENNETT, GIBBONS AND MACKENZIE-SMITH
study was explicitly for people who remained widowed. In that study the participants mean age was 74 and all were 55 years of age and older. In the second
study the participants’ mean age was 70 and the youngest was 51. Here there
were mixed views from the women with respect to new relationships. There is
evidence that the attitude toward new relationships is age dependent: younger
widows would like to re-partner; older widows would not. This effect may also
be cohort-dependent. Attitudes toward re-partnering and sexual relationships
changed greatly during the 20th century, and that may also explain the range of
views the widows had about re-partnering.
That both people with good and poor adjustment share some coping experiences
suggests that the model is doing what it is supposed to be doing, describing and
explaining the everyday experiences of bereaved people. At the same time some of
those coping experiences/behaviors could be placed in the integrated risk-factor
model proposed by Stroebe et al. (2005). The reporting of Denial/Avoidance of
Change and Distraction from Grief might be useful in identifying those who are
at risk of complicated grief, while those who report Intrusion of Grief or New
Roles/Identities/Relationships might be selected out of interventions.
ACKNOWLEDGMENTS
Thanks are due to the men and women who participated in these studies and
to Georgina Hughes and Philip Smith.
REFERENCES
Bennett, K. M. (1997). A longitudinal study of wellbeing in widowed women. International
Journal of Geriatric Psychiatry, 12(1), 61-66.
Bennett, K. M. (2007). “No Sissy Stuff”: Towards a theory of masculinity and emotional
expression in older widowed men. Journal of Aging Studies, 21, 347-356.
Bennett, K. M., & Bennett, G. (2000-2001). “And there’s always this great hole inside
that hurts”: An empirical study of bereavement in later life. Omega, 42(3),
237-251.
Bennett, K. M., & Vidal-Hall, S. (2000). Narratives of death: A qualitative study of
widowhood in women in later life. Ageing and Society, 20(4), 413-428.
Bennett, K. M., Hughes, G. M., & Smith, P. T. (2003). “I think a woman can take it”:
Widowed men’s views and experiences of gender differences in bereavement. Ageing
International, 28(4), 408-424.
Bennett, K. M., Hughes, G. M., & Smith, P. T. (2005a). Coping, depressive feelings
and gender differences in late life widowhood. Aging and Mental Health, 9(4),
348-353.
Bennett, K. M. Hughes, G. M., & Smith, P. T. (2005b). The effects of strategy and gender
on coping with widowhood in later life. Omega, 51(1), 33-52.
LOSS AND RESTORATION IN LATER LIFE /
331
Benyamini, Y., Blumstein, T., Lusky, A., & Modan, B. (2003). Gender differences in
the self-rated health-mortality association: Is it poor self-rated health that predicts
mortality or excellent self-rated health that predicts survival? The Gerontologist,
43(3), 396-405.
Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies: A
theoretically based approach. Journal of Personality and Social Psychology, 56(2),
267-283.
Caserta, M. S., & Lund, D. A. (2007). Toward the Development of an Inventory of
Daily Widowed Life (IDWL); guided by the Dual Process Model of Coping with
Bereavement. Death Studies, 31(6), 505-535.
Charmaz, K. (1995). Grounded theory. In J. A. Smith, R. Harré, & L. Van Langenhove
(Eds.), Rethinking methods in psychology (pp. 27-49). London: Sage.
Davidson, K. (1999). Marital perceptions in retrospect: A study of older widows and
widowers. In R. Miller & S. Browning (Eds.), With this ring: Divorce, intimacy
and cohabitation from a multicultural perspective (pp. 127-145). Stamford, CT:
JAI Press.
Folkman, S., & Lazarus, R. S. (1980). An analysis of coping in a middle-aged community
sample. Journal of Health and Social Behavior, 21, 219-239.
Footman, E. B. (1998). The loss adjusters. Mortality, 3, 291-295.
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for
qualitative research. Chicago: Aldine.
Lazarus, R. S., & Folkman, S. (1984). Stress, appraisal, and coping. New York: SpringerVerlag.
Lund, D., Caserta, M., de Vries, B., & Wright, S. (2004). Restoration after bereavement.
Generations Review, 14(4), 9-15.
Moore, A. J., & Stratton, D. C. (2003). Resilient widowers: Older men adjusting to a
new life. New York: Prometheus Books.
Moos, R. H. (1995). Development and applications of new measures of life stressors,
social resources, and coping resourced. European Journal of Psychological Assessment, 11, 1-13.
Richardson, V. E. (2007). A dual process model of grief counseling: Findings from the
Changing Lives of Older Couples (CLOC) study. Journal of Gerontological Social
Work, 48(3/4), 311-329.
Richardson, V. E., & Balaswamy, S. (2001). Coping with bereavement among elderly
widowers. Omega, 43(2), 129-144.
Schut, H. A. W., Stroebe, M. S., Boelen, P. A., & Zijerveld, A. M. (2006). Continuing
relationships with the deceased: Disentangling bonds and grief. Death Studies, 30,
757-766.
Shear, K., Frank, E., Houck, P. R., & Reynolds, C. F. III. (2005). Treatment of complicated
grief: A randomized controlled trial. Journal of the American Medical Association,
293(21), 2601-2608.
Smith, J. A. (1995). Semi-structured interviewing and qualitative analysis. In J. A. Smith,
R. Harré, & L. Van Langenhove (Eds.), Rethinking methods in psychology (pp. 9-26).
London: Sage.
Stroebe, M. S., & Schut, H. (2005). To continue or relinquish bonds: A review of the
consequences for the bereaved. Death Studies, 29(6), 477-494.
332 / BENNETT, GIBBONS AND MACKENZIE-SMITH
Stroebe, M., & Schut, H. (1999). The dual process model of coping with bereavement:
Rationale and description. Death Studies, 23, 197-224.
Stroebe, M. S., Folkman, S., Hansson, R. O., & Schut, H. (2006). The prediction of
bereavement outcome: Development of an integrative risk factor framework. Social
Science & Medicine, 63(9), 2440-2451.
Walter, T. (1996). A new model of grief. Mortality, 1, 7-25.
Zisook, S., Paulus, M., Shuchter, S. R., & Judd, L. L. (1997). The many faces of depression
following spousal bereavement. Journal of Affective Disorders, 45, 85-95.
Direct reprint requests to:
Dr. Kate M. Bennett
School of Psychology
University of Liverpool
Eleanor Rathbone Building
Bedford Street South
Liverpool L69 7ZA UK
e-mail: [email protected]
OMEGA, Vol. 61(4) 333-356, 2010
LENGTH OF CAREGIVING AND WELL-BEING
AMONG OLDER WIDOWERS: IMPLICATIONS FOR
THE DUAL PROCESS MODEL OF BEREAVEMENT
VIRGINIA E. RICHARDSON, PH.D.
The Ohio State University, Columbus
ABSTRACT
The intent of this study was to examine if length of caregiving was associated
with older widowers’ adjustment to bereavement and to identify factors,
based on principles underlying the Dual Process Model of Bereavement,
that might mitigate the potential adverse effects of time spent caring. Twohundred men over the age of 60 and in the second year of bereavement were
identified from death records of older women who had died within a 12-month
period. Interviews lasted about 2 hours and focused on widowers’ experiences
surrounding their wives’ deaths along with questions about social support,
health, retirement, and other demographic information. The Bradburn Affect
Scale was used to measure positive and negative affect. Restoration-oriented
coping, such as starting new relationships and activities were measured.
These variables included extent of family contact, number of friends, having
a confidante, involvement with neighbors, and participation in sports and
clubs. Time since death and demographic variables were used as controls.
Hierarchical linear regression was conducted on positive and negative affect
after which potentially moderating effects were analyzed. Results indicated
that the most important influences on negative affect were time since death,
ethnicity, and participation in clubs while for positive affect the most
significant factors included length of caregiving, number of friends,
and having a confidante. Although no interaction effects were significant,
patterns emerged. Implications for applying the DPM with older bereaved
men are made.
333
Ó 2010, Baywood Publishing Co., Inc.
doi: 10.2190/OM.61.4.e
http://baywood.com
334 / RICHARDSON
Widowhood is the most significant life event for most older persons. Spousal
bereavement is associated with poor health, increased mortality rates, reductions
in income, and problems in self-care that may require learning new skills, such
as meal planning and preparation (Lund, Caserta, de Vries, & Wright, 2004;
Lund, Caserta, & Dimond, 1993; Stroebe & Stroebe, 1993). Recent studies of
bereavement have used prospective longitudinal research designs to examine
bereavement (Carr, 2008). We now know that older widowed persons vary in their
response to bereavement depending on the social context, e.g., place of death
(at home or at the hospital), circumstances of death (sudden or protracted), and
psychological influences, including interpersonal features, such as the quality
of the marital relationship. Caregiving is another important factor that affects
widowed persons’ post-bereavement adjustment.
Most scholars who have studied caregiving have used either the stress-relief
or chronic strain models (Schulz, Boerner, & Hebert, 2008). Those who have
applied the stress-relief model expect that caregivers will feel relief after care
recipients die, especially if the death was protracted and required extensive
medical care. Proponents of the chronic strain model maintain that caregiving
leads to short- and long-term stress reactions, which sometimes continue for
years after bereavement (Kiecolt-Glaser, 1999; Selye, 1993). This is often
referred to as the general adaptation syndrome (GAS). An alternative model
used in this study (and discussed earlier in this issue), is The Dual Process
Model of Bereavement (DPM), which has several advantages over these
other models.
One advantage of the DPM is that it is based on the premise that bereaved
persons oscillate between positive and negative reactions, that is, a dialectic
tension underlies grieving between focusing on the loss, i.e., loss-oriented coping,
and restoration-oriented coping, such as meeting new challenges or starting
new relationships. According to Stroebe and Schut (1999), restoration-oriented
coping requires attending to the secondary stressors that result from bereavement, including participating in different roles, mastering new tasks, engaging in
instrumental activities, or making decisions about living alone or relocating.
Archer (2008, p. 57) describes restoration-coping as “. . . coping with the loss by
turning attention away from it and engaging in new tasks and relationships.”
Although people differ in how they cope with loss, most people go back and forth
(i.e., they oscillate, between loss- and restoration-oriented coping). Instead of
assuming that people grieve in orderly sequences as many stage theories of
bereavement have suggested, Stroebe and Schut emphasize how variously
bereaved persons respond to their loss.
Another advantage of the DPM is its potential relevance for practitioners
assisting bereaved adults. By assessing widowed persons’ coping styles, clinicians
can help bereaved persons balance their coping approaches and identify those
at-risk for complicated grief reactions.
LENGTH OF CAREGIVING AND WELL-BEING /
335
In this study I apply principles from the DPM to determine whether
bereaved caregivers’ engagement in restoration-oriented coping might moderate
the potential adverse effects that some older adults experience during bereavement. In previous investigations, researchers have observed that older widowed
persons who focused exclusively on their loss without attending to new, previously avoided challenges, struggled more during bereavement than those who
laughed, visited friends, or took up hobbies (Keltner & Bonnano, 1997; Lund, Utz,
Caserta & de Vries, 2008-2009; Richardson, 2007). Ong, Bergeman, and Bisconti
(2004) found that older widowed persons who had fun and entertained themselves became less depressed or anxious. These studies underscore the need for
restoration-oriented coping during bereavement and challenge assumptions that
healthy grieving only occurs after concentrating intensely on the loss.
RESEARCH ON THE EFFECTS OF CAREGIVING
POST-BEREAVEMENT
Several studies have shown that most caregivers felt at peace after their loved
one died, especially if the cause was Alzheimer’s disease (AD). Schulz and
colleagues (2001) found that 90% of dementia caregivers felt relieved following
their loved ones’ death although about 25% of these caregivers were depressed a
year later. Zhang, Mitchell, Bambauer, Jones, and Prigerson (2008) observed
that one-half of the caregivers evidenced depression 1.5 years after their care
recipients died from Alzheimer’s disease, and many evidenced long-term stress
reactions, including elevated blood pressure (Grant, Adler, Patterson, Dimsdale,
Ziegler, & Irwin, 2002). Similarly, Kiecolt-Glaser, Dura, Speicher, Trask, and
Glaser (1991) found that caregivers’ immune systems remained compromised
many years after bereavement and that some never returned to previous levels
of functioning.
The caregivers who struggle most during bereavement often had experienced
high levels of caregiver burden, felt overwhelmed with caring, or were overly
invested in the caregiver role (Boerner, Schulz, & Horowitz, 2004; Schulz,
Boerner, Shear, Zhang, & Gitlin, 2006; Schulz et al., 2008). Those who care
extensively for long periods without respite especially are at risk for complicated
grief reactions (Aneshensel, Botticello, & Yamamoto-Mitani, 2004; Richardson,
2007; Richardson & Balaswamy, 2001; Robinson-Whelan, Tada, MacCallum,
McGuire, & Kiecolt-Glaser, 2001; Zhang et al., 2008).
Research findings vary depending on the outcome and control variables
used. Most scholars have focused on negative bereavement outcomes, such as
depression and other mental health symptoms. Few have examined positive
responses, such as humor, positive affect or stress-related growth reactions, postbereavement (Lund et al., 2008-2009; Znoj, 2006). In this investigation I consider
negative and positive outcomes.
336 / RICHARDSON
STUDY OBJECTIVES
I ask two questions in this study:
1. Is there a significant association between length of time spent caregiving
and well-being, operationalized as positive affect (PA) and negative affect
(NA), among older widowers during the second year of bereavement, and,
if so, does the association hold when time since bereavement is considered?
2. Does engagement in restoration activities—specifically involvement in
relationships and pleasurable activities—mitigate potentially adverse postbereavement reactions?
METHOD
Sample
The sample was identified through county obituaries and death records that
appeared in a Midwestern daily newspaper of women approximately 60 years
of age or older. All obituaries for older women who had died within a 12-month
period from the summer of 1992 until the spring of 1993 were considered. County
death records were accessed to identify decedents for whom no obituary had
been published, or to obtain the address, marital status, and ethnic background of
potential respondents. A pool of 357 widowers was initially identified, and 200
of the older widowers were interviewed. Within this pool several men were
ineligible for the study. Specifically, 18 (5%) were too ill to participate; 34 (9%)
died soon after their wives within the first year of bereavement; 12 (3%) were
under age 60; 31 (8.5%) had already relocated outside of the area; and 11 (3%)
could not be located. Of the remaining respondents who were eligible to participate in the study 51 (20%) refused to be interviewed.
The sample of 200 widowers ranged in age from 58 to 91 years old, and the
average age was 75. A description of the sample is presented in Table 1. Almost
three-quarters (73%) of these men lived alone in a place at which they had
resided, on average, for about 20 years. Most widowers were in good health,
averaged about 13 years of education, and brought in between $1400 and $1500
of income each month. Eighty-two percent were Caucasian and 18% were
African-American. These bereaved men averaged 520 days since the death of
their spouses. About 56% (113) reported that it had been between 13 and
18 months since bereavement and 43% (87) were between 19 and 24 months
since bereavement.
The interview was conducted at least 1 year after the death of the spouse
(between 12-22 months) to allow respondents time to recover from the initial
loss and grief (O’Bryant & Hansson, 1995). This also improves response rates
(O’Bryant, 1983, 1987). Among studies that contacted widows or widowers
before 12 months after the spouse’s death, refusal rates have been as high as 65%
LENGTH OF CAREGIVING AND WELL-BEING /
337
Table 1. Description of Sample (N = 200)
Characteristics
Age
Mean (SD)
Mdn
75
74
Ethnicity
% Caucasian
% African-American
82%
18%
Living alone
% Alone
% With others
74%
26%
Education # Years
Monthly income
13
12
$1400
$1400
Driving status
% Yes
% No
90%
10%
Health
1 = Poor
2 = Fair
3 = Good
4 = Excellent
10%
27%
49%
14%
(Lund, 1989). Widowed women, over the age of 60, conducted the interviews
because they had participated in an earlier study and were familiar with the
interview schedule. Widows from African-American backgrounds interviewed
African-American widowers. We also used interviewers with similar characteristics (age, marital status, and ethnic background) to enhance empathy and
rapport. In-depth training was administered until all interviewers correctly and
consistently asked questions to reduce potential interviewer effects. Those
who refused to participate were compared with those who agreed to be interviewed based on available data: days since wife’s death, wife’s age, respondent’s
ethnic background, and the median income in their particular census tract.
No significant differences emerged between the refusers and participants on
these variables.
The interviews lasted about 2 hours and focused on many issues including
information on social support, self-esteem, health, occupational history, retirement, and other demographic factors. They also included questions about
338 / RICHARDSON
experiences surrounding the death of the spouse, psychological well-being,
socialization patterns, and relationships with family, friends, and neighbors.
The 14-page interview schedule included close-ended questions and several wellestablished scales. Although open-ended contingency questions were included,
I focused on the close-ended items in this analysis.
Measures
Bradburn Affect Scale
Affect was measured using the Bradburn Affect Balance Scale (ABS)
(Bradburn, 1969). The ABS asks participants to indicate whether or not they
have experienced each of 10 feelings in the past week. Negative affect (NA)
items included:
1.
2.
3.
4.
5.
During the past week did you feel depressed or unhappy (yes or no).
Did you feel lonely?
That people disliked you?
Bored?
Restless?
Positive affect (PA) items included:
1. During the past week did you feel particularly excited or interested in
something? (yes or no).
2. Pleased about having accomplished something?
3. That things were going your way?
4. Proud because someone complimented you on something you had done?
These questions originally were coded as 1 = yes and 2 = no, and recoded to
0 = no and 1 = yes.
A factor analysis (requested eigenvalues > 1.0) using varimax rotation of these
10 items yielded two major factors, referred to as positive affect (PA) and negative
affect (NA), that corresponded to those that previous researchers have identified.
These two components, with 2.6 and 1.8 eigenvalues for positive and negative
affect, respectively, contributed to 44.5% of the variance (PA explained 26% and
NA explained 18%). The alpha coefficient for PA was .65, and for NA it was .67,
which was consistent with what other investigators have reported for these scales
(Kim & Mueller, 2001; Maitland, Dixon, Hultsch, & Hertzog, 2001; Sikorska,
1999). The Positive Affect (PA) index was comprised of items asking about
positive sentiments while the Negative Affect (NA) scale included items asking
about negative feelings. High scores reflected high levels of Positive Affect
(PA) or Negative Affect (NA), respectively.
LENGTH OF CAREGIVING AND WELL-BEING /
339
Independent Variable
Length of caregiving was based on participants’ responses to the question:
How long did you provide care for your wife? Answers were coded in weeks.
While 49.2% of the men reported that they did not spend any time caring for their
wives, the other half (49.8%) spent, on average, about 132 weeks or 33 months;
the median was 52 weeks. They ranged from no time spent caregiving to 728
weeks or 182 months. The mean for the total sample was 67 weeks.
Restoration Variables
The restoration variables were organized into relationship and activity
measures. The relationship variables included: amount of contact with family,
number of friends, extent of interaction with neighbors, and whether or not the
participant had a confidante. The activity variables focused on joining clubs,
playing sports, and relaxing since bereavement.
The family contact variable (family contact), was created based on questions
regarding the frequency with which they saw or talked with their siblings in
the past year (1 = never, 2 = a few times a year, 3 = every month, 4 = every
other week, 5 = daily). Most participants (90%) had children who lived nearby;
the average number of children they had was two. A variable reflecting the
total frequency of contact with children and siblings was computed into a total
family contact scale, which was normally distributed and evidenced minimal
kurtosis or skewness.
The size of friendship network (#friends) was based on participants’ responses
to the question asking them how many individual friends they spent time with
in the past week prior to the interview. The confidante now variable was coded as
0 = no and 1 = yes based on respondents’ answers to this question: Is there
anyone in particular now that you confide in or talk to about your problems?
A neighbor interaction (neighbor interaction) scale, previously developed by
Lopata (1973), included 11 questions focusing on various activities, including:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
casually talking outside with neighbors;
borrowing tools from them;
dropping by each other’s homes;
visiting by invitation each other’s homes;
going to club meetings together;
going out together for meals or entertainment;
providing each other with transportation;
helping with the mail or newspaper if either was away;
helping with household and yard maintenance;
assisting if someone became ill; or
helping out in emergencies.
340 / RICHARDSON
Respondents used a Likert scale (1= never, 2 = seldom, 3 = occasionally, 4 =
frequently) to respond to these questions. Although a factor analysis of these
items indicated three underlying constructs, the 11 items were added together to
create a total score for amount of interaction with neighbors because the total
scale, which had an alpha coefficient .76, evidenced higher construct validity with
other measures than the separate factors. In addition, the total item scale was
normally distributed and had low scores on kurtosis and skewness.
The following variables represented the activity restoration variables: Have
you joined any clubs since your wife’s death? Do you take time to relax 15-20
minutes daily? Do you play sports, jog, or participate in other physical activities
at least three times weekly? These variables were coded as 0 = no and 1 = yes.
The mean and standard deviation scores for each variable are shown in
Table 2. A histogram and analysis of skewness and kurtosis were conducted
for each variable, and outliers were identified. The variables measuring weeks of
Table 2. Descriptive Statistics of
Study Variables (N = 200)
Variables
Weeks of caregiving
Days since death
Family contact
M
SD
67
135.8
520
97.9
33.0
16.4
Number of friends
1.82
1.25
Neighborhood interaction
8.61
2.69
Confidante now
.69
.464
Join club since wife’s death
.140
.256
Play sports often
.275
.448
Relaxation
.930
.692
Ethnicity
.18
.381
Live alone
.74
.442
Health
2.67
.839
Economic status
2.87
.802
Months retired
127
71.9
Ethnicity
1.18
.381
Health
2.67
.839
Negative affect
2.10
1.30
Positive affect
2.29
1.21
LENGTH OF CAREGIVING AND WELL-BEING /
341
caregiving and number of friends had two participants who reported numbers
that were obvious outliers. In these two instances the outliers were recoded to
the highest value of the variable to improve these variables’ normal distribution.
No data were missing for these variables.
Control Variables
The following variables were included as control measures. Time since death
was coded as the number of days since bereavement (Days since death). In
addition, we controlled for demographic variables, including ethnicity (0 = White
and 1 = Black), and whether they lived alone (0 = no and 1 = yes). The participants’
economic status was coded as: 1 = restricted, 2 = rather short, 3 = comfortable,
4 = fairly well-fixed. Health was measured using a self-rating of health with:
1 = poor, 2 = fair, 3 = good, 4 = excellent. Income and years of education, which
were highly correlated with other variables, were omitted to avoid problems
with multicollinearity.
Analysis Plan
Hierarchical ordinal least squares (OLS) regression analyses were conducted
for negative affect (NA) and positive affect (PA). In step one, days or time since
death was entered into the model, and in step two, length of time caregiving was
added. The relationship restoration variables—family contact, number of friends,
neighbor interaction, and confidante now—were added in step three while the
activity restoration variables—join club, engage in sports, and relax since death—
were included in step four. Finally, the control variables, specifically, ethnic
status, live alone, current health, and economic status were included in the
last step.
Potentially significant interaction effects were analyzed in two ways. First,
variables that were statistically significant and were theoretically important were
analyzed. Computing the product of two significant predictor variables with a
potential moderator variable created variables representing interaction effects.
The interaction effects subsequently were included as an additional step in the
multiple regression analyses. This strategy is supported by Aiken and West
(1991) who emphasize the importance of analyzing associations between predictor and outcome variables and of determining how they work together, i.e.,
whether the association is weakened (moderated) or strengthened (amplified)
under certain conditions. Second, significant interaction effects were probed to
determine the condition(s), i.e., the value(s), in which a potentially moderating
variable, was significantly associated with the outcome variable. MODPROB, a
software program developed by Hayes and Matthes (2009) that can be downloaded from www.comm.ohio-state.edu/ahayes/macros.htm, was used to examine
possible interaction effects. MODPROB is programmed to use the JohnsonNeyman technique, which determines “. . . at what values of q does t equal or
342 / RICHARDSON
exceed the critical t so as to produce a p value for t no greater than a.” These values
define limits of the regions of significance for a focal predictor and a moderator
variable. According to Hayes and Matthes (2009), if no qs are statistically significant then the overall interaction effect is significant across all values of the
whole observation. In addition to the J-N technique, the program produces a
table that shows the estimates of the effects of a focal variable that is specified
for all values on a previously identified moderator variable. The output identifies
“regions of significance” and computes the data needed to visually depict the
associations among variables. Graphs were created to show how the focal and
moderator variables and effects were related.
RESULTS
Descriptive Characteristics
Descriptive statistics of variables used in the subsequent analyses are shown in
Table 2. Causes of death are presented in Table 3; most wives died from either
cancer (39%) or heart disease (32%) although 7% died from a stroke. Table 4
indicates the number of weeks spent caregiving for each disease; the length of
caregiving was highest for those who died from chronic obstructive pulmonary
disease, such as emphysema, respiratory failure, and pulmonary embolism, and
lowest for those who died from cancer.
Association between Length of Caregiving
and Well Being (PA and NA)
Tables 5 and 6 present the results from the regression analyses for Positive
and Negative Affect (PA and NA, respectively). After time since bereavement
Table 3. Cause of Death
Disease
Percent
Heart disease
32%
Cancer
39%
Stroke
7%
COPD
2%
Liver/Kidney
4%
Pneumonia/Influenza/Infections
3%
Diabetes
3%
Alzheimer’s disease
4%
Other
6%
LENGTH OF CAREGIVING AND WELL-BEING /
343
Table 4. Length of Caregiving (Weeks)
Disease
Weeks
Heart disease
57
Cancer
49
Stroke
51
COPD
202
Liver/Kidney
84
Pneumonia/Influenza/Infections
114
Diabetes
169
Alzheimer’s disease
143
Other
80
was included as a control in step one, length of caregiving, which was inversely
related to PA (Beta = –.169, p < .01), was added in step two. These results
indicated that the more time a widower spent caring for his wife, the lower the
level of positive feelings post-bereavement. The Beta coefficient (b) decreased to
.136 when restoration relationship variables, specifically, family contact, number
of friends, neighbor interaction, and confidante now were added in step three.
When activity variables—join club, engage in sports, and relax since death—were
added in step four the regression coefficient for length of caregiving increased
to –.150 and was –.148 when ethnicity, living alone, health, and economic
status were included in the last step. Length of caregiving remained statistically
significant with PA in all five models. Despite substantial time since death the
widowers (during the second year of bereavement) who engaged in many months
of caregiving maintained lower levels of positive affect than those who did no
caregiving or engaged in caregiving for a shorter duration.
In the final model, the friends and confidante variables remained statistically
significant. Those who maintained these relationships evidenced higher levels of
positive affect than those who were less involved with friends and confidantes.
The amount of variance explained, however, was .089.
The same steps and variables were repeated using negative affect (NA) instead
of positive affect (PA) as the outcome measure, and the results, shown in Table 6,
revealed no significant associations with length of caregiving. What was most
significant in all models was time since bereavement, and the regression coefficient for this variable increased from –.164 to –213 in models one and five,
respectively. The longer the time since bereavement the lower the negative
affect. When the activities since bereavement variables, specifically join clubs,
344 / RICHARDSON
Table 5. Multiple Regression of Caregiving and Positive Affect
(Beta Coefficients) (N = 200)
Models
Model 1
Model 2
Model 3
Model 4
Model 5
Variables
Days since death
.055
Caregiving length
.065
.055
.051
.068
–.169*
–.136*
–.150*
–.148*
.000
–.011
–.045
Family contact
#Friends
.180***
.190***
.175*
Neighbor interaction
.100
.120
.129
Confidante now
.144*
.139*
.151*
.111
.122
Sports
–.059
–.073
Relax
–.022
–.006
Join club since death
.127
Ethnicity
–.033
Live alone
.053
Current health
.048
Economic status
R2
Adjusted R2
.003
.032
.117
.131
.149
–.002
.022
.089
.090
.089
*p £ .05; **p £ .01; ***p £ .001.
participate in sports, and relax often, were added in model four the amount of
variance explained increased by about 4%, and join clubs emerged as a significant
variable. Contrary to expectations, joined clubs was inversely related to NA,
that is, those who joined clubs had higher levels of NA than those who did not
engage in this activity. Finally, significant ethnic differences emerged: Caucasian
widowers evidenced higher negative affect than Black widowers. The adjusted R2
was .124. Interaction effects between time since bereavement and join club
and between time and ethnicity were added in the sixth step, but these were not
statistically significant.
Potential Interaction Effects
Positive Affect Analyses
The MODPROB program was used to test possible interaction effects. In
this program one identifies the outcome variable (PA), a focal variable (length
of caregiving), and a potential moderator variable (friends or confidante). The
LENGTH OF CAREGIVING AND WELL-BEING /
345
Table 6. Multiple Regression of Caregiving and Negative Affect
(Beta Coefficients) (N = 200)
Models
Model 1
Model 2
Model 3
Model 4
Model 5
–.164*
–.171*
–.173**
Variables
Days since death
–.183**
–.213**
–.094
.072
.069
Family contact
–.066
–.094
–.041
#Friends
–.099
.056
–.064
Neighbor interaction
–.105
–.067
–.080
Confidante now
–.050
–.055
–.072
Caregiving length
.114
Join club since death
.212**
.194**
Sports
–.090
–.066
Relax
–.079
–.105
Ethnicity
–.199**
Live alone
.061
Current health
–.099
Economic status
–.078
R2
.027
.040
.079
.134
.182
Adjusted R2
.022
.030
.051
.093
.124
*p £ .05; **p £ .01; ***p £ .001.
program yields a regression summary and conditional effects of the focal predictor
for values of the moderator program, which automatically categorizes continuous
variables into three categories: low (one SD below the mean), moderate (sample
mean), and high (one SD above the mean). Unlike other approaches that analyze
interaction effects, one is not required to dichotomize the predictors to probe
the interactions. Values t and p for coefficients measuring conditional effects
for each category of the potential moderator variable are used to test the null
hypothesis that the conditional effects are equal to 0.
The conditional effects revealed no significant interaction effects with number
of friends although a trend was found for the high (b1 = –.0013, t = –1.79, p < .07)
and moderate levels of number of friends (b2 = –.0011, t = –1.75, p < .08). These
effects are shown graphically in Figure 1. Length of caregiving reduced positive
affect among respondents regardless of the size of their friendship network. Both
length of caregiving and friends had independent effects although those who
had more friends demonstrated less of a reduction in positive feelings than those
who had few friends.
346
/
RICHARDSON
Figure 1. A visual depiction of the interaction between length of
caregiving and number of friends with positive affect.
A visual depiction of the interaction between length of caregiving and having
a confidante is presented in Figure 2. The results of these conditional effects
revealed a significant association for those who reported that they did not have a
confidante now (b1 = –.0017, t = –1.93, p < .05). Those without a confidante
evidenced a steeper decline in PA than those who had a confidante, but regardless
of the duration of caregiving, those with a confidante indicated higher levels of
PA than those without this relationship. When widowers without a confidante
were involved in prolonged caregiving their levels of positive affect declined;
having a confidante partially moderated the potential adverse consequences
related to lengthy caregiving.
LENGTH OF CAREGIVING AND WELL-BEING /
347
Figure 2. A visual depiction of the interaction between length of
caregiving and having a confidante now with positive affect.
Negative Affect Analyses
Although the overall interaction effects for predictors yielded nonsignificant
results, patterns emerged. The focal predictor variable, days or time since bereavement, which was statistically significant in the analysis of NA, and the potential moderator variable, ethnicity in which 0 = Caucasian and 1 = Black, also
significant, yielded significant conditional effects for the Caucasian widowers
348
/
RICHARDSON
(b1 = –.0023, t = –2.47, p < .01) and for those who did not join clubs (b1 = –.0021,
t = –211, p < .04). These data, shown in Figures 3 and 4, raise questions about
the applicability of the assumption that widowhood is the most significant crisis
in late life for all older persons. The results indicating that those who joined a
club had more negative affect also underscore the importance of considering
various relationship types, and perhaps differentiating between the effects of
formal and informal relationships on bereavement.
Figure 3. A visual depiction of the interaction between time since
bereavement and ethnic status with negative affect.
LENGTH OF CAREGIVING AND WELL-BEING /
349
Figure 4. A visual depiction of the interaction between time since
bereavement and joined club since bereavement with negative affect.
DISCUSSION
The primary objective of this study was to determine whether there was an
association between length of caregiving, as reported by older widowers during
their second year of bereavement, and their current well-being, specifically
positive and negative affect (PA and NA, respectively). A second objective was to
assess if older widowers’ engagement in selected restoration-oriented activities,
350 / RICHARDSON
such as socializing, joining new clubs, or playing sports, moderated the potentially
adverse effects related to prolonged caregiving on widowers’ well-being.
The findings from this investigation corroborated results from previous
research that caregiving prior to the death of a spouse adversely influences
older persons’ well-being during bereavement (Schulz et al., 2006). In addition,
involvement in relationships partially moderated the association between length
of caregiving and reduction of positive affect although number of friends
and having a confidante enhanced positive affect independent of the duration
of caregiving.
The results from these analyses also concurred with Boerner and colleagues
(2004) observations that those who were most involved in the caregiver role
struggled more during bereavement than those who were less engaged in this
role. Extensive caring for a spouse frequently precludes social interactions with
others, and informal social participation often declines as a spouse’s illness
worsens (Utz, Carr, Nesse, & Wortman, 2002). Caregivers who withdraw from
others and focus intensely on their caring role increase their risks for poor
bereavement outcomes and complicated grief reactions, as mentioned earlier.
At the same time, most bereaved caregivers eventually rebound. For example,
Utz and colleagues (2002) observed that the bereaved persons who reduced
their social participation during the worst part of their spouses’ illness typically
resumed their interactions with friends and relatives within 6 months after
their spouses died. Although caregivers might resume their social interactions,
they still might demonstrate long-term stress-related reactions especially if they
were responsible for spouses who died from Alzheimer’s disease and illnesses
that require extensive caregiving.
The lack of significant findings between length of caregiving and negative
affect (NA) most likely was due to when these widowers were interviewed.
Most bereaved persons’ negative feelings typically subside by the second
year although, as shown in this research, the circumstances surrounding a
spouse’s death substantially influence how quickly and successfully older
widowed persons adjust to their losses. The outcome variable used in this investigation is another possible explanation for the lack of significant findings with
NA. Although NA is comprised of multiple emotions including depression,
loneliness, restlessness, and boredom, it excludes many other indices that reflect
negative emotions that investigators typically have considered, such as in the
Geriatric Depression Scale or the Center for Epidemiologic Studies Depression
Scale. NA is a broader measure of negative emotions than depression (Watson,
2005). Previous investigators also have suggested that NA is primarily influenced
by adverse events and circumstances surrounding a loss (Richardson &
Balaswamy, 2001; Watson, 2005). In addition, Bennett, Smith, and Hughes
(2005a) found that gender differences influenced responses. These researchers
observed that widows and widowers’ responses differed depending on the instrument used to measure concepts.
LENGTH OF CAREGIVING AND WELL-BEING /
351
The significant association between NA and ethnicity was consistent with
Carr’s (2004a) findings that African-American bereaved persons reported lower
levels of anger and despair compared to Caucasian widows and widowers and
with McCallum and Yarry’s (2008) conclusions that African-American caregivers appraise caregiving as less stressful and less negatively than caregivers
from other ethnic groups. The substantial differences in levels of negative affect
between older Black widowers and older Caucasian widowers underscore the
need for more studies of older bereaved persons from various cultures and
backgrounds.
Time since bereavement was the most salient predictor of NA in this study
and corroborated previous findings using data from the Changing Lives of Older
Couples (CLOC). In an analysis of these longitudinal data, again using the
Bradburn Affect Scale, Richardson (2007) found that over time older widows
and widowers’ negative affect decreased while their positive affect increased.
The finding that joining a club post-bereavement increased older widowers’
negative affect was surprising. Based on assumptions from DPM, one would
have expected the opposite result, but different restoration activities apparently
satisfy different needs. Although I was unable to assess individual differences in
restoration-oriented coping in this research, I concur with other investigators’
conclusions that practitioners should assess which activities work best for each
person. Restoration-oriented coping, apparently, is a broad construct that is
manifested in various ways.
Several issues need to be explored in greater depth. Most importantly, we
lack consensus on how best to operationalize loss- and restoration-oriented
coping. Investigators also need to more closely examine how pre-bereavement
circumstances, such as caregiving, affect post-bereavement responses. This is
especially important given that increasing numbers of older persons are dying
from protracted illnesses, many of which require extensive medical care. More
older bereaved persons will have been involved in prolonged caring than in
previous years. If we know which subgroups are at greatest risk for potentially
complicated grief reactions, practitioners can intervene earlier. For example,
increased respite support and other resources might counteract the social restrictions that many bereaved caregivers engage in that inevitably reduce their positive feelings. Such supports also might decrease the stress-related reactions that
previous researchers have observed among some bereaved caregivers.
Although investigators need to replicate these analyses using longitudinal
data, we know that positive sentiments and outlooks during bereavement enhance
older widowers’ well-being. For example, Lund and colleagues (2008-2009)
observed that laughter and humor increased older bereaved persons’ resilience.
Their findings concur with what Frederickson (2005) and Watson (2005)
have emphasized in their broaden-and-build theory of positive emotions.
Frederickson (2005) asserts that positive emotions augment an individual’s
personal, physical, and social resources and thereby enhance coping and resilience
352 / RICHARDSON
when a person is confronted with a threatening event, such as bereavement
(Stein, Folkman, Trabasso, & Richards, 1997). Proponents of the broaden-andbuild theory also argue that positive affectivity—but not negative affectivity—
increases social behavior, especially with respect to number of friends, frequency
of contact with friends, making new friends, involvement in organizations, and
overall socializing. Positive affectivity is both a cause and an effect of social
behavior. While interacting with friends and confidantes improves bereaved
persons’ well-being their enhanced positive sentiments presumably motivate
them to socialize more.
I have identified a subgroup of older widowers who are at risk for developing
problems during post-bereavement that corroborate previous results that Schulz
et al. (2008) have observed. The older bereaved men who were caregivers for
prolonged periods demonstrated less positive affect than the older bereaved men
who were not caregivers in this investigation. The deprivation of positive feelings
as a result of prolonged caregiving presumably places older widowers at greater
risk for developing complicated grief reactions than their peers who are not
caregivers. Future research is needed to determine whether practitioners can
intervene with older widowed persons to prevent or mitigate potentially adverse
reactions that can arise from prolonged caregiving. However, the findings
observed here and discussed in the other articles suggest that restoration-oriented
coping, especially that which involves socializing with friends, might buffer the
reduction in positive feelings that extensive caregiving causes even during the
second year of bereavement.
A significant limitation of this investigation was that the results applied
only to widowers although, as previously mentioned, Richardson (2007) found
similar findings among widows and widowers using longitudinal data from
CLOC. Given the gender differences that Bennett, Hughes, and Smith
(2005b) and Carr (2004b) have observed, however, we need more studies that
examine within gender effects instead of between groups’ analyses. This research
is one of the few that has examined within-group differences among older
bereaved men.
Another limitation was the cross-sectional research design used in this study
that precluded any conclusions about causality although caretakers’ well-being
obviously cannot influence length of caregiving. A longitudinal research
design following the same bereaved persons over time would shed light on the
dynamics between affect and coping style during bereavement and yield more
information about which type of restoration or loss activity is most beneficial
at various points during bereavement. Finally, my results underscore the
importance of PA and restoration-oriented coping during bereavement, although
we need more information regarding the function of loss-oriented coping
and identifying moderators that have the potential not only to enhance positive
affect but also might reduce older bereaved persons’ negative feelings during
bereavement.
LENGTH OF CAREGIVING AND WELL-BEING /
353
REFERENCES
Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and interpreting
interactions. Thousand Oaks, CA: Sage.
Aneshensel, C. S., Botticello, A. L., & Yamamoto-Mitani, N. (2004). When caregiving
ends: The course of depressive symptoms after bereavement. Journal of Health and
Social Behavior, 45, 422-440.
Archer, J. (2008). Theories of grief: Past, present, and future perspectives. In M. S.
Stroebe, R. O. Hansson, H. Schut, & W. Stroebe (Eds.), Handbook of bereavement
research and practice: Advances in theory and intervention (pp. 45-67). Washington,
DC: American Psychological Association.
Bennett, K. M., Smith, P. T., & Hughes, G. M. (2005a). Coping, depressive feelings and gender differences in late life widowhood. Aging & Mental Health, 9,
348-353.
Bennett, K. M., Hughes, G. M., & Smith, P. T. (2005b). Psychological response to later
life widowhood: Coping and the effects of gender. Omega: Journal of Death and
Dying, 51(1), 33-52.
Boerner, K., Schulz, R., & Horowitz, A. (2004). Positive aspects of caregiving and
adaptation to bereavement. Psychology and Aging, 19, 668-675.
Bradburn, N. M. (1969). The structure of psychological well-being. Chicago, IL: Aldine.
Bradley, E. H., Prigerson, H., Carlson, M. D. A., Cherlin, E., Johnson-Hurzeler, R.,
& Kasl, S. V., et al. (2004). Depression among surviving caregivers: Does
length of hospice enrollment matter? American Journal of Psychiatry, 161,
2257-2262.
Carr, D. (2004a). Black/White differences in psychological adjustment to spousal loss
among older adults. Research on Aging, 26, 591-622.
Carr, D. (2004b). Gender, preloss marital dependence, and older adults’ adjustment to
widowhood. Journal of Marriage and the Family, 66, 220-235.
Carr, D. (2008). Factors that influence late-life bereavement: Considering data from the
Changing lives of older Couples Study. In M. S. Stroebe, R. O. Hansson, H. Schut, &
W. Stroebe (Eds.), Handbook of bereavement research and practice: Advances in
theory and intervention (pp. 417-440). Washington, DC: American Psychological
Association.
Frederickson, B. (2005). Positive emotions. In C. R. Snyder & S. J. Lopez (Eds.),
Handbook of positive psychology (pp. 120-134). New York: Oxford University
Press.
Grant, I., Adler, K. A., Patterson, T. L., Dimsdale, J. E., Ziegler, M. G., & Irwin, M. R.
(2002). Health consequences of Alzheimer’s caregiving transitions: Effects of placement and bereavement. Psychosomatic Medicine, 63, 477-486.
Hayes, A. F., & Matthes, J. (2009). Computational procedures for probing interactions
in OLS and logistic regression: SPSS and SAS implementations. Behavior Research
Methods, 41, 924-936.
Kiecolt-Glaser, J. K., Dura, J. R., Speicher, C. E., Trask, O. J., & Glaser, R. (1991). Spousal
caregivers of dementia victims: Longitudinal changes in immunity and health.
Psychosomatic Medicine, 53, 345-362.
Kiecolt-Glaser, J. K. (1999). Stress, personal relationships, and immune function: Health
implications. Brain, Behavior and Immunity, 13, 61-72.
354 / RICHARDSON
Keltner, D., & Bonanno, G. A. (1997). A study of laughter and dissociation: Distinct
correlates of laughter and smiling during bereavement. Journal of Personality and
Social Psychology, 73, 687-702.
Kim, A. K., & Mueller, D. J. (2001). To balance or not to balance: Confirmatory
factor analysis of the Affect Balance Scale. Journal of Happiness Studies, 2,
289-306.
Lopata, H. Z. (1973). Widowhood in an American city. Cambridge, MA: Schenkman.
Lund, D. (1989). Conclusions about bereavement in later life and implications for interventions and future research. In D. Lund (Ed.), Older bereaved spouses (pp. 3-16).
New York: Hemisphere.
Lund, D. A., Caserta, M. S., & Dimond, M. F. (1993). The course of spousal bereavement
in later life. In M. S. Stroebe, W. Stroebe, & R. O. Hansson (Eds.), Handbook of
bereavement: Theory, research, and intervention (pp. 240-254). London: Cambridge
University Press.
Lund, D., Caserta, M., de Vries, B., & Wright, S. (2004). Restoration after bereavement.
Generations Review, 14, 9-15.
Lund, D. A., Utz, R., Caserta, M. S., & de Vries, B. (2008-2009). Humor, laughter, and
happiness in the daily lives of recently bereaved spouses. Omega: Journal of Death
and Dying, 58, 87-105.
Maitland, S. B., Dixon, R. A, Hultsch, D. F., & Hertzog, C. (2001). Well-being as a moving
target: Measurement and equivalence of the Bradburn Affect Balance Scale. Journal
of Gerontology: Psychological Sciences, 56B, P69-P77.
McCallum, T. J., & Yarry, B. A. (2008, Spring). Applying stress-related growth concepts
to the caregiving process among African Americans. African American Research
Perspectives, 12, 54-70.
O’Bryant, S. (1983). The subjective value of “home” to older homeowners. Journal of
Housing for the Elderly, 1, 29-44.
O’Bryant, S. (1987). Precursors of physical, economic, and psychological well-being in
widowhood. Final report to the AARP Andrus Foundation. Washington, DC.
O’Bryant, S., & Hansson, R. (1995). Widowhood. In R. Bliesner & V. Kilkevitch Bedford
(Eds.), Handbook of aging and the family (pp. 440-458). Westport, CT: Greenwood
Press.
Ong, A. D., Bergeman, C. S., & Bisconti, T. L. (2004). The role of daily positive emotions
during conjugal bereavement. Journal of Gerontology: Psychological Sciences, 59B,
168-176.
Richardson, V. E., & Balaswamy, S. (2001). Coping with bereavement among elderly
widowers. Omega: Journal of Death and Dying, 43, 129-144.
Richardson, V. E. (2007). A dual process model of grief counseling: Findings from the
Changing Lives of Older Couples (CLOC) study. Journal of Gerontological Social
Work, 48, 311-329.
Robinson-Whelan, S., Tada, Y., MacCallum, R. C., McGuire, L., & Kiecolt-Glaser, J. K.
(2001). Long-term caregiving: What happens when it ends? Journal of Abnormal
Psychology, 110, 573-584.
Rosenberg, M. (1979). Conceiving the self. New York: Basic Books.
Ross, C. E., & Sastry, J. (1999). The sense of personal control: Social structural causes
and emotional consequences. In C. S. Aneshenel & J. C. Phelan (Eds.), Handbook of
the sociology of mental health (pp. 369-394). New York: Plenum Press.
LENGTH OF CAREGIVING AND WELL-BEING /
355
Roth, D. L., Mittelman, M. S., Clay, O. J., Madan, A., & Haley, W. E. (2005). Changes
in social support as mediators of the impact of a psychosocial intervention for
spouse caregivers of persons with Alzheimer’s disease. Psychology and Aging, 20,
634-644.
Schulz, R., Boerner, K., & Hebert, R. S. (2008). Caregiving and bereavement. In M. S.
Stroebe, R. O. Hansson, H. Schut, & W. Stroebe (Eds.), Handbook of bereavement research and practice: Advances in theory and intervention (pp. 265-285).
Washington, DC: American Psychological Association.
Schulz, R., Boerner, K., Shear, K., Zhang, S., & Gitlin, L. N. (2006). Predictors of
complicated grief among dementia caregiver: A prospective study of bereavement.
American Journal of Geriatric Psychiatry, 14, 650-658.
Schulz, R., Mendelsohn, A. B., Haley, W. E., Mahoney, D., Allen, R. S., Zhang, S.,
et al. (2004). End of life care and the effects of bereavement among family
caregivers of persons with dementia. New England Journal of Medicine, 349,
1936-1942.
Schulz, R., Beach, S. R., Lind, V., Martire, L., Zdaniuk, B., Hirsch, C., et al. (2001).
Involvement in caregiving and adjustment to death of a spouse: Findings from the
Caregiver Health Effects Study. Journal of the American Medical Association, 285,
3123-3129.
Schulz, R., Newsom, J. T., Fleissner, K., Decamp, A. R., & Nieboer, A. P. (1997).
The effects of bereavement after family caregiving. Aging & Mental Health, 1,
269-282.
Seyle, H. (1993). History of the stress concept. In L. Goldberger & S. Breznitz (Eds.),
Handbook of stress: Theoretical and clinical aspects (pp. 2-17). New York: The Free
Press.
Shaw, B. A. (2005). Anticipated support from neighbors and physical functioning during
later life. Research on Aging, 27, 503-525.
Sikorska, E. (1999). Organizational determinants of resident satisfaction with assisted
living. The Gerontologist, 39, 450-456.
Stein, N. L., Folkman, S., Trabasso, T., & Richards, T. A. (1997). Appraisal and goal
processes as predictors of psychological well-being in bereaved caregivers. Journal
of Personality and Social Psychology, 72, 872-884.
Stroebe, M., & Schut, H. (1999). The dual process model of coping with bereavement:
Rationale and description. Death Studies, 23, 197-224.
Stroebe, M., & Stroebe, W. (1993). The mortality of bereavement: A review. In
M. S. Stroebe, W. Stroebe, & R. O. Hansson (Eds.), Handbook of bereavement:
Theory, research, and intervention (pp. 175-195). London: Cambridge University
Press.
Utz, R. L., Carr, D., Nesse, R., & Wortman, C. (2002). The effect of widowhood on
older adults’ social participation: An evaluation of activity, disengagement, and
continuity theories. The Gerontologist, 42, 522-533.
Watson, D. (2005). Positive affectivity: The disposition to experience pleasurable emotional states. In C. R. Snyder & S. J. Lopez (Eds.), Handbook of positive psychology
(pp. 106-119). New York: Oxford University Press.
Zhang, B., Mitchell, S. L., Bambauer, K., Jones, R., & Prigerson, H. G. (2008). Depressive
symptom trajectories and associated risks among bereaved Alzheimer disease caregivers. American Journal of Geriatric Psychiatry, 16, 145-155.
356 / RICHARDSON
Znoj, H. (2006). Bereavement and posttraumatic growth. In L. G. Calhoun & R. G.
Tedeschi (Eds.), Handbook of posttraumatic growth: Research and practice
(pp. 176-196). Mahwah, NJ: Lawrence Erlbaum Associates.
Direct reprint requests to:
Dr. Virginia Richardson
College of Social Work
The Ohio State University
1947 College Road
Columbus, OH 43210
e-mail: [email protected]
OMEGA, Vol. 61(4) 357-369, 2010
EXPLORING THE ROLE OF EXPERIENTIAL
AVOIDANCE FROM THE PERSPECTIVE
OF ATTACHMENT THEORY AND THE
DUAL PROCESS MODEL*
M. KATHERINE SHEAR, M.D.
Columbia University School of Social Work, New York
ABSTRACT
Avoidance can be adaptive and facilitate the healing process of acute grief
or it can be maladaptive and hinder this same process. Maladaptive cognitive
or behavioral avoidance comprises the central feature of the condition of
complicated grief. This article explores the concept of experiential avoidance
as it applies to bereavement, including when it is adaptive when it is problematic. Adaptive avoidance is framed using an attachment theory perspective and incorporates insights from the dual process model (DPM). An
approach to clinical management of experiential avoidance in the syndrome
of complicated grief is included.
Avoidance is sometimes an adaptive strategy in coping with adversity and sometimes maladaptive. In the case of bereavement, experiential avoidance usually
plays a role in facilitating the healing process. The emotional pain associated with
new information that a loved one has died is so severe that people need time
interspersed with periods of respite in order to be able to fully acknowledge
the unwanted reality. Respite can be achieved using cognitive avoidance, and
*This work was supported by a grant from the National Institute of Mental Health: R01
MH70741.
357
Ó 2010, Baywood Publishing Co., Inc.
doi: 10.2190/OM.61.4.f
http://baywood.com
358 / SHEAR
sometimes by also avoiding contact with triggers of emotion. When avoidance
is used adaptively, it facilitates processing of the painful information as well
as restoration of the capacity for a satisfying ongoing life. As processing and
restoration are achieved, the need for avoidance diminishes and the strategy
must be relinquished. If it is not, or if avoidance is over-used in the wake of
bereavement, the strategy can backfire. Processing difficult information is
impeded rather than facilitated and acute grief is prolonged.
This article explores the boundary of adaptive and maladaptive use of avoidance
during acute grief, considering both behavioral and cognitive strategies. We
define acute grief as the abrupt onset of a bereavement response that is usually
severe and that has a relatively short course, measured in months rather than
years. Acute grief, analogous to acute inflammation, is a painful healing process
that usually resolves over time if there are no complications. Healing is associated
with restoration of functioning and a permanent residue of integrated grief.
The discussion that follows is framed using an attachment theory perspective
and incorporates important insights from the dual process model (DPM). The
hypothesis is that avoidance facilitates the healing process, providing a needed
respite from severe emotional pain. On the other hand, over-use or persistence
of avoidance strategies beyond a certain point can hamper mourning and
complicate acute grief. Once established, complicated grief is often chronic
and unremitting with experiential avoidance at its core. Avoidance is notoriously
difficult to bring to light as it can be overlooked by clinicians. Focused assessment
and intervention are usually needed to identify and ameliorate the problem. This
article concludes with a summary of our approach to clinical management of
bereavement-related avoidance.
Avoidance is often used as a method of evading external situations that are
appraised, accurately or not, as dangerous. Avoidance is adaptive when real
danger is present. However, people with anxiety disorders over-estimate the
probability that a situation is dangerous and misjudge the likely consequences.
For example, people with post-traumatic stress disorder often appraise neutral
situations as dangerous because they have become associated with a life
threatening event. These people fear that the external danger will strike again.
Avoidance is also used as a way of achieving distance from emotions and other
internal experiences. This type of avoidance, called “experiential,” was defined
by Haves, Wilson, Strosahl, Gifford, and Follette (1996) as “the phenomenon
that occurs when a person is unwilling to remain in contact with particular
private experiences (e.g., bodily sensations, emotions, thoughts, memories,
behavioral predispositions) and takes steps to alter the form or frequency of these
events and the contexts that occasion them” (Haves et al., 1996, p. 1154).
Bereaved people utilize avoidance to manage perceived internal rather than
external threats. Haves and colleagues outline a number of ways that experiential
avoidance can be adaptive, yet they contend that experiential avoidance is not
advisable during grief: “When an unchangeable loss occurs, the healthy thing to
ATTACHMENT, LOSS, AND AVOIDANCE /
359
do is to feel what one feels when losses occur” (Haves et al., 1996, p. 1157). Their
belief in the categorical need to experience painful grief, disputed by grief
researchers, is a good example of the confusion engendered when the process
and goals of healing after bereavement are not spelled out. Bereaved people
regularly utilize experiential avoidance in the process of adapting to a painful loss.
Bowlby (1980) is clear that experiential avoidance used judiciously is a natural
feature of acute grief. In fact, Bowlby goes farther and posits that there is an
adaptive form of defensive exclusion (i.e., cognitive avoidance) in the resolution
of acute grief. In the discussion that follows we consider adaptive as well as
maladaptive use of experiential avoidance during acute grief. We first discuss
adaptive avoidance using an attachment theory perspective on bereavement to
provide the context for the discussion (Bowlby, 1980).
ATTACHMENT, LOSS, AND ADAPTIVE
AVOIDANCE
There is good evidence that humans are instinctively oriented to seek, form,
and maintain close relationships and to respond to separation and loss of these
relationships. The biobehavioral attachment system underlying these innate tendencies operates throughout the lifespan. Attachment security contributes to
optimal psychological functioning and fosters a sense of wellbeing. Attachment
relationships impact daily life in countless ways, including learning, mastery and
performance success, overall effective functioning, emotion regulation, psychophysiological reactivity, sleep quality, self-esteem and self-concept, cognitive
functioning, coping skills and problem solving, and general interpersonal functioning (e.g., Carmichael & Reis, 2005; Cassidy, 1994; Feeney & Collins, 2003;
Gillath, Bunge, Shaver, Wendelken, & Mikulincer, 2005; Kim, Carver, Deci, &
Kasser, 2008; Mikulincer, Dolev, & Shaver, 2004; Mikulincer, Florian, Cowan, &
Papa, 2002; Mikulincer, Shaver, & Pereg, 2003; Pereg & Mikulincer, 2004). The
attachment working model is an internal mental representation thought to entail
episodic and semantic memory systems that specifically map each attachment
relationship. Attachment working models are believed to be dynamic both in the
sense of their day-to-day operations and in the sense of being altered as needed by
important changes in the actual relationship to the attachment figure (Collins &
Feeney, 2004; Mikulincer & Shaver, 2003). The working model is thought to be
the mechanism by which the impact of attachment on psychological functioning
operates (Bowlby, 1980; Bretherton, 1999; Collins, 1996; Collins & Feeney,
2004; Gillath, Makulincer, Fitzsimons, Shaver, Schachner, & Bargh, 2006;
Grossman, 1999; Meins, 1999; Mikulincer & Shaver, 2003; Roisman, Collins,
Sroufe, & Egeland, 2005; Shaver & Mikulincer, 2002; Simpson, Winterheld,
Rholes, & Oriña, 2007; Waters & Waters, 2006; Zimmermann, 1999).
The working model makes it possible to experience day-to-day physical
separation from an attachment figure without distress. However, people resist
360 / SHEAR
prolonged separation and react strongly to perceived uncertainty about a loved
one’s safety. The permanent loss of an attachment relationship is highly impactful.
Bowlby posited that when the death of a loved one occurs, the unwanted
and emotionally painful reality of its finality is only gradually accepted and
integrated into the attachment working model. He contended that the response
to bereavement evolves in quality during the process of integration and that
avoidance regularly plays a role in this process (Bowlby, 1980).
In the early aftermath of a death most bereaved people naturally employ
strategies to avoid the severe pain entailed in fully acknowledging its decisiveness and consequences, nourishing instead a hope, however vague and slender,
that reunion is possible. The hope grows primarily from the fact that the attachment working model does not initially register the meaning of the death. In
this situation a separation response is instinctively activated and generates an
irrational feeling, often unsettling, that the person could reappear. This is associated with yearning and searching for the deceased and preoccupation with
thoughts and memories of this person. Inevitably, though, a belief in reunion is
on a collision course with reality. Bowlby suggests that this means a newly
bereaved person faces a painful dilemma:
So long as he does not believe that his loss is irretrievable, a mourner is
given hope and feels impelled to action; yet that leads to all the anxiety
and pain of frustrated effort. The alternative, that he believes his loss is
permanent, may be more realistic; yet at first it is altogether too painful and
perhaps terrifying to dwell on for long. (Bowlby, 1980, p. 139)
The private experience of the recent death of an attachment figure is extremely
painful and it is natural for a person to be unwilling to remain in contact with
this experience. Bowlby calls attention to the instinctive use of cognitive and
behavioral avoidance strategies that are observed in this situation. He then alerts
us to the limitations of these strategies and the need for a permanent adaptive
resolution of the bereavement dilemma:
It may be merciful, therefore, that a human being is so constructed, that
mental processes and ways of behaving that give respite are a part of
his nature. Yet such respite can only be limited and the task of resolving
the dilemma remains. On how he achieves this turns the outcome of his
mourning—either progress toward a recognition of his changed circumstances, a revision of his representational models, and a redefinition of his
goals in life, or a state of suspended growth in life in which he is held prisoner
by a dilemma he cannot solve. (Bowlby, 1980, p. 139)
He provides specific examples of avoidance behavior:
[the mourner] may then oscillate between treasuring . . . reminders and
throwing them out, between welcoming the opportunity to speak of the
dead and dreading such occasions, between seeking out places where they
have been together and avoiding them. (Bowlby, 1980, p. 92)
ATTACHMENT, LOSS, AND AVOIDANCE /
361
He cites Parkes who listed types of processes of “defensive exclusion,” that are
strategies for cognitive avoidance:
(a) processes that result in a bereaved person feeling numb and unable to
think about what has happened
(b) processes that direct attention and activity away from painful thoughts
and reminders and towards more neutral or pleasant ones
(c) processes that maintain the belief that the loss is not permanent and
reunion is still possible
(d) processes that result in recognition that loss has in fact occurred combined
with the feeling that links with the dead nevertheless persist, manifest often
in a continuing sense of the continuing presence of the lost person. (Bowlby,
1980, p. 140; Parkes, 1970)
Bowlby posits that the bereavement dilemma is best resolved through gradually
integrating the emotionally painful information about the death with positive
thoughts and memories of the deceased person. Although not mentioned by
Haves, the existence of contradictory elements of the private experience is another
example of a situation in which avoidance of one or the other of these elements
might be adaptive until the conflict is resolved. In the case of acute grief there
are contradictory experiences of belief in the possibility of union and appraisal
of the reality of the death. To manage this conflict, people experiencing acute grief
make use of cognitive avoidance strategies to alternately dismiss either the belief
that reunion is possible (bouts) or painful awareness of the death (moratoria.)
Bowlby explains the process as follows:
When an affectional bond is broken there is usually a preliminary registering
of the relevant information combined with the inability to evaluate it to
more than the most cursory extent. . . . Thereafter further evaluation proceeds
in bouts plentifully interspersed with moratoria. During a bout certain of
the implications already received are considered or reconsidered though
others are still avoided; whilst additional information may be sought. . . .
During each moratoria by contrast, some or all of the information regarding
change already received is likely to be excluded and the old models partially
or wholly reinstated. Hence the oscillation of feeling already documented.
(Bowlby, 1980, p. 239)
According to this view the response to bereavement is an evolving process in
which avoidance is a dynamic and adaptive element. To say that avoidance is
maladaptive following loss is to oversimplify the process of adjustment. To decide
if experiential avoidance is adaptive during bereavement, we must consider
when and how it occurs. Ultimately, a bereaved person must relinquish experiential avoidance focused on evading painful thoughts and emotions associated
with acknowledging the reality of the death. Bowlby suggests the best way to
resolve the dilemma is by merging love and loss. The bereaved person links
recognition of the loss with ongoing feelings of love in the form of a continued
sense of connectedness. Bowlby (1980) and Parkes (1970) label this sense of
362 / SHEAR
connectedness in the face of a death as defensive exclusion, a form of experiential
avoidance, and this is considered adaptive.
In summary, an attachment theory view of successful outcome of bereavement
centers on the need to reconcile the dilemma of conflicting inner experiences of
love and loss, and places experiential avoidance at the center of the adaptive
process by which these conflicting realities are ultimately combined. There is both
a sense of ongoing connection to the deceased and awareness of the painful reality
that they are gone. Bowlby suggests that the two sides of the bereavement dilemma
are gradually integrated during a process of oscillation between processing and
excluding private experiences of each. Most people achieve this. People do not
forget loved ones who die, nor do they stop caring about them. Instead they feel
a permanent sense of connection and responsibility to the person who died.
OTHER CONSIDERATIONS RELATED TO THE
PROCESS OF ACUTE GRIEF
Two other biobehavioral motivational systems are linked to the attachment
system and are affected by the loss. Attachment relationships are dyadic and
require a caregiver. In an adult attachment relationship giving and receiving
care are shared functions, so that loss of an attachment figure is also loss of
a caregiving recipient. Caregiving is also thought to have instinctive roots
focused on sensitive and responsive attention to nurturing and protecting others
(Feeney & Collins, 2003; Gillath et al., 2005; Kim & Carver, 2007; Mikulincer &
Goodman, 2006). Caregiving loss instinctively triggers a sense of failure of the
protector role. Anger is triggered if it seems that the death could have been
prevented by others. From the standpoint of caregiving, fully acknowledging
the death evokes a painful sense of failure, with self-blame or anger. Adaptive use
of experiential avoidance pertains to these difficult feelings as well as those
associated with attachment. These feelings may also be the reason for over-use
of avoidance strategies.
Exploration is another instinctive system that is related to attachment. The
exploratory system provides instinctive motivation for learning, mastery, and
performance (Elliott & Reis, 2003). Both acute stress and threatening separation
experiences inhibit the exploratory system. This system could play an important
role in effective restoration of a satisfying life, yet the initial effect of bereavement is to shut it down. Inhibition of exploration impedes restoration through
reducing enthusiasm for, and confidence in doing new activities and taking on
new roles. Reconciling the bereavement dilemma reactivates exploration and
fosters restoration activities.
In resolving the instinctive attachment dilemma as well as responses related
to the caregiving and exploration systems, a bereaved person must find ways to
adjust to changes relevant to her or his unique situation. How does this process
occur? The DPM (Stroebe & Schut, 1999) provides a framework for thinking
ATTACHMENT, LOSS, AND AVOIDANCE /
363
about how coping occurs during the process of adaptation and considers this a
different problem from that of processing bereavement-related information. The
model posits that there are countless stressors contained in any bereavement
episode and that these can be generally grouped as loss-related or restorationrelated. Restoration-focused coping includes attending to life changes, doing new
things, taking on new roles and identities, as well as distraction from, denying/or
avoiding grief. By contrast, the loss focus includes grief work, intrusion of
grief, breaking bonds/ties/ relocation, as well as denial/avoidance of restoration
related changes.
The DPM is consistent with Bowlby’s premise that acknowledgment of the
finality of the loss and its consequences leads the mourner to appropriately revise
the working model (loss) and redefine life plans and goals (restoration). However,
the DPM places a focus on stresses related to these objectives rather than the
processing of new information, and introduces the seminal insight that loss and
restoration-related processes proceed in tandem. Bowlby believed that bereaved
people must “resolve the loss” before they “move on.” The DPM suggests
otherwise. This model posits that people begin to cope with restoration-related
stress even as they work to cope with acceptance of the loss. This is an important
idea because progress in restoration-related activities can facilitate processing
of the loss needed to revise the working model. Coping with restoration-related
stressors makes the finality of the loss less frightening. Addressing issues pertaining to ongoing life in the absence of the deceased loved one opens possibilities
for satisfaction and pleasure in this new situation. A daunting problem of daily
life not only stymies restoration-related coping but also complicates acute grief.
The DPM concept of a partnership between loss and restoration is therefore
very important. However its authors propose that mourners oscillate between
loss and restoration-focused coping. Bowlby’s description of oscillation toward
and away from emotional pain is closer to the bereaved person’s lived experience
than is oscillation between loss and restoration coping. Avoidance of grief is
not necessarily focused on restoration, and restoration-related coping strategies
are not necessarily associated with respite from grief. Sometimes restorationrelated coping entails activities that directly foster coping with the loss. Loss and
restoration might better be illustrated visually as overlapping Venn diagrams
that progress in tandem than as separate activities that are undertaken in an
oscillating sequence. What oscillates is the private experience of thoughts and
emotions. Oscillation progresses through use of experiential avoidance.
A useful strategy for experiential avoidance of emotional pain includes
directing attention and activity away from painful thoughts and reminders and
toward more neutral or pleasant ones. There is now extensive data on the importance of positive emotion to psychological as well as physical health (Fredrickson,
1998, 2001; Fredrickson & Joiner, 2002; Fredrickson & Levenson, 1998;
Fredrickson, Tugade, Waugh, & Larkin, 2003; Tugade & Fredrickson, 2004).
Successful mourning is likely to be facilitated when periods of emotional pain
364 / SHEAR
oscillate with respite from that pain, and preferably with thoughts or activities
associated with positive emotions. There is some empirical support that such
oscillation in emotion might occur (Ong, Bergeman, & Bisconti, 2004). Positive
emotion may be elicited by fond memories or amusing anecdotes that are lossrather than restoration-related. On the other hand, restoration-related coping
processes may be associated with emotionally painful anxiety or self-doubt.
The notion of a partnership between loss and restoration coping during acute
grief makes sense. The proposed oscillating pattern of attention of the two
foci may not be as helpful.
AVOIDANCE THAT COMPLICATES
ACUTE GRIEF
There is good evidence that most bereaved people find a way to adjust to
the loss. However, a subgroup of about 10% experiences complicated grief in
which avoidance often plays a central role. From the discussion above, we might
conclude that experiential avoidance can become a hindrance to adjustment if
avoidance strategies are not used judiciously or effectively or if they are not
adaptively altered as healing progresses. People with complicated grief (CG) are
held prisoner by a dilemma they cannot solve, just as Bowlby (1980) suggests
above. Maladaptive use of experiential avoidance is often an important hindrance
to the resolution process. Avoidance of private experiences of grief, including
thoughts and emotions, through cognitive or behavioral strategies is a core symptom of complicated grief.
Adaptive avoidance is dynamic in the sense that it is active and changing and
fluid in the sense that it is not stable and fixed but rather responsive to the need
for respite in the service of resolution. Maladaptive avoidance is more stable and
less sensitive and responsive. Resolution of the bereavement dilemma requires
cognitive engagement, including consideration of different ways of appraising
new information. Excessive and fixed use of defensive exclusion of this information blocks this process. Processing new unwanted information about the
death also requires some respite. A natural avenue for respite is through irrational
hope for reunion. Defensive exclusion of the belief in reunion can also be
employed excessively and this can lead the bereaved person to be barraged by
the reality to an extent that they cannot engage in processing activities. Coming
to terms with the death is a process that works best if it is grappled with, set
aside, and revisited. Each mourner must find an effective balance between bouts
and moratoria.
Behavioral strategies are less sensitive and less effective than cognitive strategies for experiential avoidance. This is because cognitive strategies are also
private experiences and can be implemented quickly and extensively. Moreover,
there is less ambivalence about their use. Behavioral avoidance frequently
entails situations that are both aversive and desirable. Perhaps for this reason,
ATTACHMENT, LOSS, AND AVOIDANCE /
365
behavioral avoidance is more variable in frequency and extensiveness among
people with CG than is cognitive avoidance.
In the presence of unremitting pain related to the reality of the death, some
people resort to behavioral avoidance as a desperate measure to try to control
experiential stimuli. Examples of this kind of avoidance includes visiting the
final place of rest, going to the place where death occurred, reading the obituary
or reading letters of condolence, looking at photographs, thinking about the
person, talking about the person, dealing with personal belongings, spending time
in certain rooms of home, eating certain foods, listening to favorite music or
watching favorite movies, going to places they went with the person, or going out
with others. Many of these activities were a source of pleasure or satisfaction and
avoidance not only leads to ineffective processing of the death but also impedes
restoration of the capacity for joy and satisfaction in life. Moreover, we found that
grief-related avoidance correlates with intrusive thoughts (r = 0.37) suggesting
that extreme efforts to avoid reminders of the loss may oscillate with intrusive
thoughts and each may fuel the other. Both avoidance and intrusions can interfere
with information processing.
There is evidence that avoidance behavior is related to poor outcomes of bereavement. For example, Bonanno, Papa, Lalande, Nanping, and Noll (2005) found that
avoidance 4 months post loss is related to avoidance at 18 months and to worse
physical health at 18 months. In a clinical population of individuals with CG, more
avoidance scores correlate with CG severity (r = 0.40) and with impairment from
grief (r = .33) (Shear, Monk, Houck, Melhen, Frank, Reynolds, et al., 2007). In
another study, experiential avoidance correlated with CG (r = 0.63) (Boelen &
Reijntjes, 2008). However, avoidance behavior can be difficult to recognize
because people evade thinking and talking about situations that they are avoiding.
SUGGESTIONS FOR CLINICAL WORK WITH AVOIDANCE
BEHAVIOR IN COMPLICATED GRIEF
Complicated grief appears to be a unique syndrome that occurs when a person
bereaved of an attachment figure is unsuccessful in resolving the bereavement
dilemma and is held prisoner by the continued pain and disruption of a prolonged
period of acute grief. Acute grief can become complicated by a range of different
pathways. However, all result in an internal response that is painful and a focus
of experiential avoidance. We designed a targeted complicated grief treatment
(CGT) that addresses experiential avoidance in several ways. Additionally,
the treatment accepts the DPM premise that grief progresses most successfully
when loss and restoration are addressed contemporaneously. Therefore, each
of the 16 sessions of CGT addresses both loss- and restoration-oriented issues.
Our objective is to help individuals caught in a seemingly endless cycle of grief
to re-imagine their ongoing relationship to their deceased loved one and to
re-envision their own lives in the present and future.
366 / SHEAR
We address cognitive avoidance using a technique called imaginal revisiting
similar to prolonged exposure developed for treatment of posttraumatic stress
disorder. Imaginal revisiting is modified in order to foster the processing of the
bereaved person’s private grief experience. We followed Bowlby in designing the
treatment to encourage a pattern of oscillating confrontation with, and respite
from, the loss. The technique entails visualizing the experience of having learned
of the loved ones death with eyes closed, telling the story of that period of time
out loud and in the present tense, while visualizing the scenes, and responding to
the therapist’s periodic queries about emotional intensity (subjective units of
distress.) At the end of about 15 minutes, the person opens her eyes and debriefs
the visualization/story telling experience. During this period the therapist facilitates identification and processing of emotionally painful complicating thoughts
or beliefs. At the end of the debriefing period, the therapist does another visualization exercise in which the person imagines rewinding a tape with the story on it,
and putting it away in a safe place. The therapist then asks the bereaved person to
think about how she might reward herself for this hard work and makes a concrete
plan to do so. The remainder of the session is focused on activities expected to
generate positive feelings.
Goals work is the primary method for generating positive emotions and for
working on restoration. CGT includes a segment of each session that focuses on
considering what the bereaved person would want to do with her life if her grief
were at a manageable level. The person is encouraged to think of long-term goals
that she believes will provide significant gratification. The therapist talks with
the person about how she will know that she is working toward her goal, how
committed she is to her goal, and what she might expect could stand in her way.
The therapist asks if there is anyone else who can help the person achieve her
goal and together they make a specific plan for the upcoming week.
Situational revisiting exercises are used to address experiential avoidance using
behavioral changes. The approach is to identify activities that the patient is not
doing because they do not want to be reminded that their loved one is gone and
to rate these according to their desirability. Ideally the situational revisiting
exercise will entail activities that have been pleasurable or satisfying in the past,
have the potential to be so again, and that the person would like to be doing.
This procedure helps with processing information about the finality of the death
and it also helps restoration of ongoing life by reducing restrictions.
There are other elements of the treatment approach that help to engage the
person with complicated grief and provide a space in which she can feel safe. The
person is provided with a handout that describes a model of complicated grief
similar to the one presented in this article. Other work is done to help with
integrating loss and love in order to facilitate resolution of the bereavement
dilemma. We evaluated this treatment in a pilot study (Shear, Frank, Foa, Cherry,
Reynolds, Vander Bilt et al., 2001) and tested it in a prospective randomized
controlled trial (Shear, Frank, Houck, & Reynolds, 2005).
ATTACHMENT, LOSS, AND AVOIDANCE /
367
CONCLUSION
Avoidance is an important strategy used by bereaved people and is focused
on private experiences rather than external danger. Experiential avoidance is
identified by the DPM as a component of the normal process of coping with
loss and restoration and is also described as a natural aspect of the loss of an
attachment figure. Experiential avoidance can function as an adaptive strategy in
the difficult process of resolving the bereavement dilemma. However, this strategy
must be used judiciously and in a fluid dynamic way in order to facilitate healing.
When over-used or when used in a rigid unchanging way throughout the acute
grief period, avoidance can become an encumbrance to mourning and lead to
the development of complicated grief. When this occurs, the bereaved person
often needs assistance in decreasing cognitive and behavioral avoidance and
facilitating a dual process model of coping.
REFERENCES
Boelen, P. A., & Reijntjes, A. (2008). Measuring experiential avoidance: Reliability
and validity of the Dutch 9-item acceptance and action questionnaire (AAQ). Journal
of Psychopathology and Behavioral Assessment, 30, 241-251.
Bonanno, G. A., Papa, A., Lalande, K., Nanping, Z., & Noll, J. G. (2005). Grief processing
and deliberate grief avoidance: A prospective comparison of bereaved spouses and
parents in the United States and the People’s Republic of China. Journal of Consulting
and Clinical Psychology, 73, 86-98.
Bowlby, J. (1980). Loss, sadness and depression, New York: Basic Books.
Bretherton, I. (1999). Updating the “internal working model” construct: Some reflections.
Attachment & Human Development, 1, 343-357.
Carmichael, C. L., & Reis, H. T. (2005). Attachment, sleep quality, and depressed affect.
Health Psychology, 24, 526-531.
Cassidy, J. (1994). Emotion regulation: Influences of attachment relationships. Monographs of the Society for Research in Child Development, 59, 228-249.
Collins, N. L., & Feeney, B. C. (2004). Working models of attachment shape perceptions
of social support: Evidence from experimental and observational studies. Journal of
Personality and Social Psychology, 87, 363-383.
Collins, N. L. (1996). Working models of attachment: Implications for explanation,
emotion and behavior. Journal of Personality Social Psychology, 71, 810-832.
Elliot, A. J., & Reis, H. T. (2003). Attachment and exploration in adulthood. Journal
of Personality & Social Psychology, 85, 317-331.
Feeney, B. C., & Collins, N. L. (2003). Motivations for caregiving in adult intimate
relationships: Influences on caregiving behavior and relationship functioning. Personality & Social Psychology Bulletin, 29, 950-968.
Feeney, B. C., & Collins, N. L. (2001). Predictors of caregiving in adult intimate relationships: An attachment theoretical perspective. Journal of Personality and Social
Psychology, 80, 972-994.
Fredrickson, B. L. (2001). The role of positive emotions in positive psychology. The
broaden-and-build theory of positive emotions. American Psychology, 56, 218-226.
368 / SHEAR
Fredrickson, B. L. (1998). What good are positive emotions? Review of General
Psychology, 2, 300-319.
Fredrickson, B. L., & Joiner, T. 2002. Positive emotions trigger upward spirals toward
emotional well-being. Psychological Science, 13, 172-175.
Fredrickson, B. L., & Levenson, R. W. (1998). Positive emotions speed recovery from the
cardiovascular sequelae of negative emotions. Cognition and Emotion, 12, 191-220.
Fredrickson, B. L., Tugade, M. M., Waugh, C. E., & Larkin, G. R. (2003). What good
are positive emotions in crises? A prospective study of resilience and emotions
following the terrorist attacks on the United States on September 11th, 2001. Journal
of Personality and Social Psychology, 84, 365-376.
Gillath, O., Bunge, S. A., Shaver, P. R., Wendelken, C., & Mikulincer, M. (2005).
Attachment-style differences in the ability to suppress negative thoughts: Exploring
the neural correlates. Neuroimage, 28, 835-847.
Gillath, O., Mikulincer, M., Fitzsimons, G, M., Shaver, P. R., Schachner, D. A., & Bargh,
J. A. (2006). Automatic activation of attachment-related goals. Personality & Social
Psychology Bulletin, 32, 1375-1388.
Grossmann, K. E. (1999). Old and new internal working models of attachment: The organization of feelings and language. Attachment & Human Development, 1, 253-269.
Haves, S. C., Wilson, K. G., Strosahl, E. V., Gifford, E. V., & Follette, M. (1996). Experiential avoidance and behavioral disorders: A functional dimensional approach to diagnosis and treatment. Journal of Consulting and Clinical Psychology, 64, 1152-1168.
Kim, Y., & Carver, C. S. ( 2007). Frequency and difficulty in caregiving among spouses of individuals with cancer: Effects of adult attachment and gender. Psychoncology, 16, 714-723.
Kim, Y., Carver, C. S., Deci, E. L., & Kasser, T. (2008). Adult attachment and psychological well-being in cancer caregivers: The mediational role of spouses’ motives for
caregiving. Health Psychology, 27, 144-154.
Meins, E. (1999). Sensitivity, security and internal working models: Bridging the transmission gap. Attachment & Human Development, 1, 325-342.
Mikulincer, M., Florian, V., Cowan, P. A., & Papa, C. (2002). Attachment security in
couple relationships: A systemic model and its implications for family dynamics.
Family Process, 41, 405-434.
Mikulincer, M., & Goodman, G. (Eds.). (2006). Dynamics of romantic love: Attachment,
caregiving, and sex. New York: Guilford Press.
Mikulincer, M., Dolev, T., & Shaver, P. R. (2004). Attachment-related strategies during
thought suppression: Ironic rebounds and vulnerable self-representations. Journal
of Personality & Social Psychology, 87, 940-956.
Mikulincer, M., & Shaver, P. (Eds.). (2003). The attachment behavioral system in adulthood: Activation, psychodynamics, and interpersonal processes. San Diego, CA:
Elsevier Academic Press.
Mikulincer, M., Shaver, P. R., & Pereg, D. (2003). Attachment theory and affect regulation: The dynamics, development, and cognitive consequences of attachment-related
strategies. Motivation and Emotion, 27, 77-102.
Ong, A. D., Bergeman, C. S., & Bisconti, T. L. (2004). The role of daily positive emotions
during conjugal bereavement. Journal of Gerontology: Psychological Sciences, 59B,
158-167.
Parkes, C. M. (1970). “Seeking” and “finding” a lost object: evidence from recent studies
of the reaction to bereavement. Social Science Medicine, 4, 187-201.
ATTACHMENT, LOSS, AND AVOIDANCE /
369
Pereg, D., & Mikulincer, M. (2004). Attachment style and the regulation of negative
affect: Exploring individual differences in mood congruency effects on memory and
judgment. Personality & Social Psychology Bulletin, 30, 67-80.
Roisman, G. I., Collins, W. A., Sroufe, L. A., & Egeland, B. (2005). Predictors of
young adults’ representations of and behavior in their current romantic relationship:
Prospective tests of the prototype hypothesis. Attachment and Human Development, 7,
105-121.
Shaver, P. R., & Mikulincer, M.( 2002). Attachment-related psychodynamics. Attachment
& Human Development, 4, 133-161.
Shear, M. K., Frank, E., Foa, E., Cherry, C., Reynolds, C. F. III, Vander Bilt, J., et al.
(2001). Traumatic grief treatment: A pilot study. American Journal of Psychiatry, 158,
1506-1508.
Shear, K., Frank, E., Houck, P. R., & Reynolds, C. F. III. (2005). Treatment of complicated
grief: A randomized controlled trial. The Journal of the American Medical Association, 293, 2601-2608.
Shear, K., Monk, T., Houck, P., Melhem, N., Frank, E., Reynolds, C., et al. (2007).
An attachment-based model of complicated grief including the role of avoidance.
European Archives of Psychiatry and Clinical Neuroscience, 257, 453-461.
Simpson, J. A., Winterheld, H. A., Rholes, W. S., & Oriña, M. (2007). Working models
of attachment and reactions to different forms of caregiving from romantic partners.
Journal of Personality & Social Psychology, 9, 466-477.
Stroebe, M., & Schut, H. (1999). The dual process model of coping with bereavement:
Rationale and description. Death Studies, 23, 197-224.
Tugade, M. M., & Fredrickson, B. L. (2004). Resilient individuals use positive emotions
to bounce back from negative emotional experiences. Journal of Personality and
Social Psychology, 86, 320-333.
Waters, H. S., & Waters, E. (2006). The attachment working models concept: Among
other things, we build script-like representations of secure base experiences. Attachment & Human Development, 8, 185-197
Zimmermann, P. (1999). Structure and functions of internal working models of attachment
and their role for emotion regulation. Attachment & Human Development, 1, 291-306.
Direct reprint requests to:
Katherine Shear, M.D.
Marion E. Kenworthy Professor of Psychiatry
Columbia University School of Social Work
1255 Amsterdam Avenue
New York, NY 10027
e-mail: [email protected]
OMEGA, Vol. 61(4) 371-380, 2010
NEW PERSPECTIVES ON THE DUAL PROCESS
MODEL (DPM): WHAT HAVE WE LEARNED?
WHAT QUESTIONS REMAIN?
DEBORAH CARR
Rutgers University, New Brunswick, New Jersey
In 1999, Stroebe and Schut published their seminal article on the Dual Process
Model (DPM), a conceptual model which changed the direction of bereavement
research. While earlier models of grief focused primarily on psychological adjustment in the wake of a severed emotional attachment, the DPM model places
equal emphasis on practical—even mundane—daily life strains that follow from
bereavement, such as learning new household management skills and establishing
new relationships. In order to cope effectively, bereaved persons must “oscillate”
between loss-oriented (LO) coping and restoration-oriented (RO) coping. The
former refers to coping processes that focus directly on the stress of the loss itself,
including symptoms of grief, loss, and sadness; the latter includes the processes
one uses to cope with the secondary stressors that accompany one’s new status as
a bereaved spouse. Oscillation is essential for optimal psychological adjustment;
bereaved persons must attend to practical as well as emotional matters, and many
may turn to RO activities as respite from negative emotions associated with the
lost attachment.
In the 10 years since Stroebe and Schut’s (1999) publication, many bereavement researchers have conducted empirical evaluations of specific components
of the model. At the 2008 Gerontological Society of America’s annual meeting,
a multidisciplinary panel of researchers spanning the fields of psychology,
psychiatry, social work, and sociology came together to test, refine, and debate
the model. The four papers presented in the symposium are published in this
special issue. These papers represent a range of research methods, including
371
Ó 2010, Baywood Publishing Co., Inc.
doi: 10.2190/OM.61.4.g
http://baywood.com
372 / CARR
in-depth qualitative interviews (Bennett, Gibbons, & Mackenzie-Smith, 2010),
quantitative analysis of survey data (Richardson, 2010), and program evaluation (Lund, Caserta, Utz, & de Vries, 2010), as well as a theoretical essay (Shear,
2010). In this article, I summarize and critique the contributions of each of these
works and highlight important yet unresolved questions about the DPM.
Bennett and colleagues (2010) use data from two qualitative studies to explore
whether the coping tactics adopted by widows and widowers are associated
with their psychological adjustment to loss. Their first analysis uses data from
in-depth interviews with 90 older bereaved spouses. Study participants had
been bereaved anywhere from 3 months to 32 years, with an average of 9 years
since loss. The interview obtained rich information about life prior to loss,
psychological and social adjustment post-loss, and a retrospective assessment of
“what a typical day had been like after the death, what support they had, and how
they felt.” Their second analysis involved in-depth interviews with 13 widowed
women, and focused on RO coping. Of particular interest was obtaining detailed
reports about “practical changes” that were made since the death, and one’s
appraisal of how difficult those changes had been.
The analysis shows that particular clusters of stressors are associated with
positive versus negative adjustment. Specifically, persons who adjust well to loss
had experienced stressors related to New Roles, Identities, and Relationships
(RO) and Intrusion of Grief (LO), while persons who adjust less well experienced
stressors related to Denial/Avoidance of Restoration Changes and Distraction/
Avoidance of Grief. The article also elucidates the specific changes that widowed
women made following loss; the key challenge they reported was managing
personal finances, yet women also reported positive aspects of their newlyacquired independence, including the pursuit of new activities and relationships,
and becoming self-sufficient emotionally and instrumentally.
Bennett and colleague’s (2010) work makes important strides in pinpointing
specific aspects of RO that facilitate adjustment to loss. However, the analysis is
based on retrospective data only, and is thus susceptible to the critiques typically
levied against such designs. Retrospective recall bias is a serious concern given
that some of the study participants were bereaved more than 2 decades earlier.
Moreover, retrospective data collection techniques—especially when focused on
a stressful life event such as loss—often elicit overly rosy reports of adjustment.
Emerging research on post-traumatic growth suggests that individuals tend
to construct narratives that emphasize themes such as personal growth and
improvement. It is psychologically and emotionally protective to believe that
one has “grown” or learned from a traumatic experience; by contrast, it is
distressing to admit that a bad event has no positive consequences (e.g., Aldwin
& Levenson 2004).
Like Bennett and colleagues (2010), Lund and colleagues (2010) focus primarily on the nature and utility of RO coping. Their research team developed
the Living After Loss (LAL) project which evaluates the effectiveness of an
NEW PERSPECTIVES ON THE DPM /
373
innovative DPM-based intervention (compared to a “traditional” support group).
The DPM-based intervention provided bereaved spouses with lessons on how to
cope with both LO and RO challenges, whereas the traditional intervention
focused on LO issues only. Each of the concrete lessons taught mapped directly
onto a component of the DPM model, such as “grief work” or “attending to life
changes.” Further, in an effort to emulate the process of “oscillation,” participants
in the DPM-based group were exposed to LO and RO lessons on alternating
weeks. The authors expected that persons in the DPM-based intervention would
evidence better coping outcomes at the end of the 14-week program. Study
participants were recently bereaved spouses ages 50 and older, residing in the
San Francisco and Salt Lake City regions.
The analyses yielded several findings that have important implications both
for gerontological practice and for model refinement. First, they (Lund et al.,
2010) found that the DPM group (i.e., both LO and RO treatments) and the
LO-only group did not differ significantly in terms of how much they learned
about grief-related topics. Likewise, those in the DPM group did not show greater
gains in RO coping than those in the comparison group. The authors speculate that
discussion of RO issues might have arisen during the “traditional” LO group, or
that bereaved persons may naturally consult their friends and family outside of the
support group for assistance with RO issues.
Perhaps the most interesting finding was that the needs of study participants
varied tremendously in the RO group. When subjects were asked to name additional topics they would like to learn more about, multiple non-overlapping
suggestions were offered—ranging from cooking to computers. The authors
(Lund et al., 2010) conclude that “at least some of the participants might benefit
more from an individually targeted and delivered RO coping option.” These
findings suggest that while LO coping tactics and symptoms pertaining to
emotional aspects to loss may be universalistic, RO coping tactics and the
consequences thereof for bereaved people’s adjustment may be individualistic.
The latter may depend heavily on one’s education, skill set, availability of social
support, degree of preparation for the death, and both gender and cohort-specific
socialization. This finding has important implications for policy and practice;
while bereavement support groups may be appropriate for treating the nearly
universal, short-term emotional aspects of loss, the practical challenges are much
more idiosyncratic, and may require social workers or bereavement counselors to
have a fairly in-depth knowledge of the bereaved person’s skills, needs, and other
sources of support.
While the LAL project focused heavily on practical and instrumental aspects
of RO, Richardson’s analysis investigated a different aspect of RO coping: social
relations and integration. Richardson (2010) explores the extent to which social
relations and activities post-loss protect against the strains of pre-loss caregiving.
She proposes that individuals who engage in long spells of spousal caregiving
prior to loss may disengage from meaningful social roles and relations and thus
374 / CARR
evidence poorer adjustment post-loss. Specifically, Richardson uses crosssectional survey data from a sample of 200 older widowers to investigate whether
the length of caregiving affects widowers’ levels of positive and negative affect
2 years post-loss, and whether the caregivers’ engagement in RO activities,
operationalized as social contact with friends, families, and neighbors and participation in new activities, moderates the adverse psychological consequences
of prolonged caregiving.
Richardson’s analysis, like the work of Lund and colleagues and Bennett and
colleagues, finds some support for the effectiveness of RO coping. Lengthy
durations of pre-loss of caregiving are associated with poorer psychological
adjustment post-loss, yet this effect is moderated by social support. Having many
friends and having a confidante with whom one can share their private thoughts
each mitigate against the psychological strains associated with prolonged caregiving. These results are encouraging, yet provide only a partial test of the
DPM. Several of the measures of social support, including number of friends
and having a confidante, do not distinguish between relationships established prior
to the loss versus post-loss. Although both measures focus on social contact
occurring in the very recent past, it is not clear whether these relationships
have persisted in the long term, or whether the intensity of the relationships
has changed since bereavement. A core component of the DPM, by contrast, is
a focus on new roles and relations developed post-bereavement. Despite this
limitation, Richardson’s (2010) work makes an important contribution by considering a broad range of social relationships—including those with neighbors,
friends, family members, and activities including clubs and sports.
Whereas Bennett and colleagues, Lund and colleagues, and Richardson focus
primarily on RO aspects of coping, Shear (2010) focuses on the complex interplay
between RO and LO coping. Her essay sets forth a provocative hypothesis:
avoidance, typically conceptualized as a maladaptive coping tactic, may be adaptive—at least in the short term following loss. Her argument draws heavily on
attachment theory and DPM theory. Shear notes that the pain of losing a loved
one is distressing, and individuals may need to temporarily and sporadically
distance themselves from the harmful emotions and cognitions that accompany
the loss. Unlike Stroebe and Schut (1999), however, Shear (2010) believes that
bereaved persons do not necessarily oscillate between LO and RO coping, rather
the two processes are overlapping and one may give rise to the other. For instance,
one who needs to “achieve distance from emotions and other internal experiences”
may rely on experiential avoidance and turn to new activities and relationships
for a reprieve from their grief symptoms.
Shear (2010) cautions that this strategy of experiential avoidance can
“backfire”; however, if avoidance persists “beyond a certain point [it] can hamper
mourning and complicate grief.” Bereaved persons engaging in unhealthy
avoidance may stay away from activities or emotions that could provide them
solace; for instance, bereaved persons may stop listening to the favorite music of
NEW PERSPECTIVES ON THE DPM /
375
their deceased spouse, or may withdraw from married couple friends with whom
they used to socialize. Shear recommends that strategies for treating grief recognize the important role of experiential avoidance. She proposes a cognitive
strategy called “imaginal revisiting” which requires the bereaved persons to recall
the death, talk about the experience, and set goals for the future. This process
allows the bereaved person to identify the potentially pleasant or rewarding
activities that they are avoiding because they fear reminders of the loss.
Taken together, these four articles make important contributions to testing
and refining DPM. They highlight the importance of RO coping for positive
adjustment to loss; underscore the difficulty in conceptualizing and operationalizing oscillation; and show that avoidance may be adaptive, at least in small
doses. They also reveal the importance of heterogeneity: whether, how, and
to what end one copes with loss is conditioned by characteristics of the death,
the late marriage, and the psychological and social resources of the bereaved
spouse. However, these studies also highlight four important issues that require
further theoretical development and empirical analysis: identifying the psychosocial factors that affect both one’s coping tactics and the implications of such
tactics for psychological adjustment to loss (i.e., social selection); the role of
agency and intent in coping with loss; the time course and relative balance of
LO and RO coping; and the consideration of multiple outcomes when studying
the effectiveness of LO and RO coping. I briefly summarize each of these
limitations and suggest strategies for future study.
SOCIAL SELECTION PROCESSES
An overarching question posed by both the DPM and the articles presented
in this issue is: to what extent do LO and RO coping facilitate psychological
adjustment to loss? For instance, Bennett et al. (2010) find that persons who report
stressors related to Denial/Avoidance of Restoration Chances and Distraction/
Avoidance of Grief report poor psychological adjustment to loss, while
Richardson (2010) shows that persons with more friends and a confidante fare
better post-loss, even in the aftermath of long caregiving spells. However, the
analyses do not consider the social, economic, and psychological resources
that may give rise to both stress/coping factors and post-loss distress, and thus
may account for an observed correlation between coping and psychological
adjustment.
The issue of social selection is a critically important concern for researchers
studying adjustment to loss; individuals with the fewest economic resources,
poorest physical and mental health, and weakest social ties prior to loss may
be most likely to face difficulties in managing practical tasks of loss and of
adjusting emotionally to the loss. For instance, a poorly educated, economically
distressed widow may not have the economic resources to engage in new activities
(i.e., RO coping) yet also may be at particular risk of secondary stressors such as
376 / CARR
financial strain, that further compromise her psychological well-being. Likewise,
a widow with relatively high levels of trait neuroticism may select ineffective
LO strategies, such as rumination or denial, which in turn compromise her
psychological adjustment to loss. Prospective multiwave studies which obtain
information on the psychosocial traits and resources of an individual prior to
loss are necessary if researchers hope to distinguish the effects of coping tactics
on adjustment. Longitudinal data sets such as the Changing Lives of Older
Couples (CLOC), Health and Retirement Study (HRS), and Wisconsin Longitudinal Study (WLS) may be useful resources for bereavement researchers
hoping to document both the precursors to and psychological consequences of
particular coping strategies and secondary stressors related to loss.
AGENCY AND INTENT IN COPING WITH LOSS
One of the greatest strengths of the DPM model is that it allows for individuallevel agency and innovation. Unlike classic “stage theories” of grief, which
assume that bereaved persons proceed through a series of stages in lock-step
fashion, and where deviation from this progression is viewed as problematic for
adjustment, the DPM model allows that there are multiple paths to adjustment,
and that individuals will oscillate between LO and RO coping based on one’s own
needs and demands. However, the model does not adequately address whether
individuals are actively and purposively choosing these strategies, or whether
they are passively defaulting to strategies due to lack of options.
Understanding a bereaved person’s intent and motivation is particularly important when assessing the implications of “avoidance” for adjustment to loss.
Shear (2010) proposes a plausible argument: that avoidance can be adaptive,
provided it does not persist too long. However, Bennett finds that avoidance is
associated with poorer adjustment to loss. These conflicting findings raise questions about the role of intent and agency. Is a bereaved person actively avoiding
a thought or activity because it is too painful? Or are they passively avoidant
because they lack the resources to engage in a particular activity? Or, are they
avoiding a task or emotion because they simply have other more desirable or
satisfying options? Importantly, both “push” and “pull” factors may cause a
bereaved older adult to engage in new activities, roles, and relationships. For
example, a widow may seek out new activities because she cannot bear to be in her
home alone (i.e., push factor), or may seek out new activities because she happily
embraces the prospects of developing new skills and friendships (i.e., pull factor).
The extent to which one plays an active versus passive role in selecting a new
activity or relationship, and one’s rationale for why one has chosen a particular
path may condition the psychological consequences of that coping practice. Future
qualitative studies could probe bereaved spouses for the reasons behind their
choices to engage in particular practices; the simple question of “why?” may elicit
valuable insights into the role of agency in the bereavement process.
NEW PERSPECTIVES ON THE DPM /
377
TIME COURSE AND RELATIVE BALANCE OF LO AND
RO AND COPING PROCESSES
Each of the four studies presented here points out an important gap in bereavement research: neither the DPM model nor empirical tests thereof have yet
established when, how much, and to what end one engages in LO versus RO
coping. Of particular interest would be an investigation of whether such processes
can begin prior to loss. Among the 2 million deaths in the United States each year,
nearly three-quarters are to older adults—most of whom suffered from long-term
chronic illnesses that required intensive caregiving (Federal Interagency Forum
on Aging-Related Statistics, 2008). As such, soon-to-be bereaved spouses may
begin to oscillate between RO and LO coping even prior to the loss.
The concept of “anticipatory grief” suggests that individuals often begin to
disengage emotionally from and mourn the loss of their loved one even prior to
the death, as terminal illness and severe cognitive impairment irrevocably alter
the nature of one’s relationship (Rando, 2000). Likewise, Bennett and colleague’s
(2010) interviews reveal that some women began to cut back their activities, such
as volunteering, when their husbands became ill. Yet others may ramp up their
social engagement in preparation for the impending death. For example, research
on divorce reveals that unhappily married women who anticipate a future divorce,
may return to the labor market even prior to the divorce, so that they will be
financially prepared for the transition (Johnson & Skinner, 1986). One might
suspect that married persons may begin to learn new skills, such as cooking,
financial management, or earning a driver’s license prior to the loss of their
spouses in anticipation of the RO stressors they may ultimately encounter. Married
caregivers may re-establish old friendships or re-invest in relationships with
siblings in anticipation of the social and emotional support they will require upon
bereavement. This attention to RO coping prior to loss may also provide respite
from the emotional strain associated with watching a loved one die; thus it is
plausible that oscillation is adaptive both pre- and post-loss.
A further goal of DPM researchers is to ascertain the time points post-loss
when RO versus LO coping is particularly valuable, and whether there are optimal
time points for invoking one set of strategies more frequently than the other, or
for oscillating more or less frequently. Each of the articles presented in this issue
implicitly recognizes the importance of time, although none made this concern
the focus of their study. For instance, while Bennett and colleague’s (2010)
sample included persons who had been bereaved for as many as 32 years, she
did not stratify her analysis by duration since loss. Both Richardson (2010) and
Lund and colleagues (2010) focused exclusively on newly bereaved person (i.e.,
1 to 2 years, and 2 to 6 months post-loss, respectively); this limited focus is
well justified, given that psychological symptoms and practical challenges are
most acute during the early stages of loss. Still, future studies could explore the
extent to which and the effectiveness with which one oscillates between LO and
378 / CARR
RO in the near term (i.e., less than 6 months) versus longer term (6 to 12
months) post-loss. This approach would be particularly useful in evaluating
Shear’s (2010) claim that avoidance is adaptive in the short term only. Studies
that focus explicitly on the time course of symptoms could help to identify the
specific time period during bereavement when avoidance becomes maladaptive
rather than adaptive.
The issue of balance, or how much one focuses on LO versus RO coping, also
remains an unresolved question. Bennett and colleagues (2010) found that a
full 87% of bereaved persons in their sample reported both LO and RO coping,
but they captured the presence of rather than oscillation between the two. Lund
and colleagues (2010) attempted to emulate an oscillation process in their DPM
intervention, but acknowledge that alternating weeks of LO and RO sessions
may not be a realistic approximation of the “real” oscillation process. They
note that “oscillation . . . is the least well-developed feature of the DPM but
has considerable promise in facilitating more positive adjustment outcomes.”
Of particular interest is whether the “optimal” balance varies based on one’s
psychosocial characteristics including one’s gender, physical health, economic
resources, social support systems, and other stressors in one’s life.
Researchers now have the methodological tools to assess both the optimal
balance of LO and RO coping, and to document the point in the bereavement
process when one is invoked more heavily than the others. As Stroebe and
Schut (1999) note in their introductory article, new data collection techniques
including daily diary studies or “beeper” studies that ask bereaved persons to
indicate what they are doing at every moment of the day may provide rich
descriptive information on the coping tactics used, and the extent to which
oscillation occurs. Large-scale time-diary studies such as the American Time
Use Survey (ATUS) and National Survey of Daily Experiences (NSDE) provide
methodological templates for developing diary studies of bereavement. This
type of data could be used to predict psychological outcomes measured at a
subsequent time point. Moreover, as existing studies of bereavement obtain
multiple waves of data across time, data analysts will be able to use sophisticated
analytic tools such as latent growth curves. This analytic approach allows
researchers to track both the initial levels of RO and LO coping post-loss, changes
in these levels over the course of bereavement, and the psychosocial correlates
of such trajectories.
IMPORTANCE OF MULTIPLE OUTCOMES
A further strength of the DPM model is that it was developed to explore a
range of behavioral and psychosocial outcomes. Future studies should continue
to consider a diverse range of outcomes including physical, emotional, and social
well-being; different aspects of adjustment may respond to different aspects of
LO and RO coping. For example, Richardson (2010) found that the duration of
NEW PERSPECTIVES ON THE DPM /
379
caregiving pre-loss predicts positive affect—but not negative affect—in her
sample of 200 recently bereaved widowers. Similarly, positive affect only is
associated with having a confidante and number of friends.
It is important to study multiple outcomes because the time course of specific
symptoms may vary and important consequences may go undetected. Studies
that focus on the consequences of LO and RO coping at only one time point,
such as 6 months post-loss, may fail to detect individual-level differences in
adjustment that occur immediately after loss, as well as those effects that are
lagged and manifest only in the longer-term post-loss. For instance, depressive
symptoms may be most acute during the first 6 months post-loss, whereas indicators of social adjustment, such as interest in dating or forming new relationships,
may emerge only in the longer-term, given cultural norms prohibiting relationship formation “too soon” after loss.
Further, researchers should continue to consider the level of one’s psychological symptoms rather than solely one’s diagnostic category, such as a diagnosis
of complicated grief. Studies that focus solely on discrete categories cannot
provide information on the coping processes of those who barely fail to meet
diagnostic criteria, or who experience severe symptoms at one point in time
post-loss but whose symptoms have subsided by the time of data collection. For
example, Lund and colleagues (2010) focus on a broad range of participants in
the LAL, rather than only those who are coping poorly. This strategy enables
them to focus on a diverse range of coping strategies and outcomes.
In sum, influential theoretical and empirical advances have been made in the
past decade as researchers have tested and refined the Dual Process Model. This
model has influenced the development of innovative interventions to treat
the bereaved (Lund et al., 2010; Shear, 2010) and has promise to inform practice
and theory even more powerfully in the next decade, facilitated in part by the
promising research presented in this issue.
REFERENCES
Aldwin, C., & Levenson, R. (2004). Post-traumatic growth: A developmental perspective.
Psychological Inquiry, 15, 19-21.
Bennett, K. M., Gibbons, K., & Mackenzie-Smith, S. (2010). Loss and restoration in
late life: An examination of Dual Process Model of Coping with Bereavement. Omega,
61(4), 317-334.
Federal Interagency Forum on Aging-Related Statistics. (2008). Older Americans 2008:
Key indicators of well-being. Federal Interagency Forum on Aging-Related Statistics,
Washington, DC: U.S. Government Printing Office.
Johnson, W. H., & Skinner, J. (1986). Labor supply and marital separation. American
Economic Review, 76, 455-469.
Lund, D., Caserta, M., Utz, R., & de Vries, B. (2010). Experiences and early coping
of bereaved spouses/partners in an intervention based on the Dual Process Model
(DPM). Omega, 61(4), 293-315.
380 / CARR
Rando, T. A. (Ed.). (2000). Clinical dimensions of anticipatory mourning: Theory and
practice in working with the dying, their loved ones, and their caregivers. Champaign,
IL: Research Press.
Richardson, V. E. (2010). Length of caregiving and well-being among older widowers:
Implications for the Dual Process Model of Bereavement. Omega, 61(4), 335-358.
Shear, M. K. (2010). Exploring the role of experiential avoidance from the Perspective
of Attachment Theory and the Dual Process Model. Omega, 61(4), 359-371.
Stroebe, M. S., & Schut, H. (1999). The Dual Process Model of Coping with Bereavement: Rationale and description. Death Studies, 23, 197-224.
Direct reprint requests to:
Deborah Carr
Department of Sociology and Institute for Health,
Health Care Policy and Aging Research
Rutgers University
54 Joyce Kilmer Avenue
Piscataway, NJ 08854
e-mail: [email protected]
OMEGA, Vol. 61(4) 381-382, 2010
OMEGA: JOURNAL OF DEATH AND DYING
Index—Contents of Volume 61, 2010
ABAKOUMKIN, GEORGIOS: See Stroebe, Wolfgang, jt. author
BELL, JO: The Role of Perfectionism in Student Suicide: Three Case Studies from
the UK, No. 3, p. 251
BENNETT, KATE M.: Loss and Restoration in Later Life: An Examination of
Dual Process Model of Coping with Bereavement, No. 4, p. 315
BURKE, LAURIE A.: African American Homicide Bereavement: Aspects of
Social Support that Predict Complicated Grief, PTSD, and Depression, No. 1, p. 1
CARR, DEBORAH: New Perspectives on the Dual Process Model (DPM): What
Have We Learned? What Questions Remain? No. 4, p. 371
CASERTA, MICHAEL: Sampling, Recruitment, and Retention in a Bereavement
Intervention Study: Experiences from the Living After Loss Project, No. 3, p. 181
CASERTA, MICHAEL: See Lund, Dale, jt. author
CLUTE, MARY ANN: Bereavement Interventions for Adults with Intellectual
Disabilities: What Works? No. 2, p. 163
COOLEY, ERIC: Reactions to Loss Scale: Assessing Grief in College Students,
No. 1, p. 25
DE VRIES, BRIAN: See Caserta, Michael, jt. author
DE VRIES, BRIAN: See Lund, Dale, jt. author
GIBBONS, KERRY: See Bennett, Kate M., jt. author
HAMID, TENGKU AIZAN: See Momtaz, Yadollah Abolfathi, jt. author
HIJMANS, ELLEN: See Yang, William, jt. author
HOLLAND, JASON M.: An Examination of Stage Theory of Grief among Individuals Bereaved by Natural and Violent Causes: A Meaning-Oriented Contribution, No. 2, p. 103
HOUSEMAN, CLARE A.: See Lundgren, Burden, S., jt. author
IBRAHIM, RAHIMAH: See Momtaz, Yadollah Abolfathi, jt. author
JOHNSON, CELESTE M.: African-American Teen Girls Grieve the Loss of
Friends to Homicide: Meaning-Making and Resilience, No. 2, p. 121
381
Ó 2010, Baywood Publishing Co., Inc.
doi: 10.2190/OM.61.4.h
http://baywood.com
382 / INDEX—CONTENTS OF VOLUME 61, 2010
LIVINGSTON, KATHY: Opportunities for Mourning When Grief is Disenfranchised: Descendants of Nazi Perpetrators in Dialogue with Holocaust Survivors,
No. 3, p. 205
LUND, DALE: Experiences and Early Coping of Bereaved Spouses/Partners in
an Intervention Based on the Dual Process Model (DPM), No. 4, p. 291
LUND, DALE: See Caserta, Michael, jt. author
LUNDGREN, BURDEN S.: Banishing Death: The Disappearance of the Appreciation of Mortality, No. 3, p. 223
MACKENZIE-SMITH, SUZANNA: See Bennett, Kate M., jt. author
MALLON, SHARON: See Bell, Jo, jt. author
MANTHORPE, JILL: See Bell, Jo, jt. author
MCDEVITT-MURPHY, MEGHAN E.: See Burke, Laurie A., jt. author
MOMTAZ, YADOLLAH ABOLFATHI: Mediating Effects of Social and Personal Religiosity on the Psychological Well Being of Widowed Elderly People,
No. 2, p. 145
NEIMEYER, ROBERT A.: See Burke, Laurie A., jt. author
NEIMEYER, ROBERT A.: See Holland, Jason M., jt. author
RICHARDSON, VIRGINIA E.: Length of Caregiving and Well-Being among
Older Widowers: Implications for the Dual Process Model of Bereavement, No. 4,
p. 333
RICHARDSON, VIRGINIA E.: The Dual Process Model of Coping with
Bereavement: A Decade Later, No. 4, p. 269
ROSCOE, LAUREN: See Cooley, Eric, jt. author
SCHUT, HENK: See Stroebe, Margaret, jt. author
SHEAR, M. KATHERINE: Exploring the Role of Experiential Avoidance from
the Perspective of Attachment Theory and the Dual Process Model, No. 4, p. 357
SILVERMAN, SAM: The Death of Socrates: A Holistic Re-examination, No. 1,
p. 71
STANLEY, NICKY: See Bell, Jo, jt. author
STAPS, TON: See Yang, William, jt. author
STROEBE, MARGARET: See Stroebe, Wolfgang, jt. author
STROEBE, MARGARET: The Dual Process Model of Coping with Bereavement: A Decade On, No. 4, p. 273
STROEBE, WOLFGANG: Beyond Depression: Yearning for the Loss of a Loved
One, No. 2, p. 85
TORAY, TAMINA: See Cooley, Eric, jt. author
UTZ, REBECCA: See Caserta, Michael, jt. author
UTZ, REBECCA: See Lund, Dale, jt. author
YAHAYA, NURIZAN: See Momtaz, Yadollah Abolfathi, jt. author
YANG, WILLIAM: Existential Crisis and the Awareness of Dying: The Role of
Meaning and Spirituality, No. 1, p. 53
The goal of Suicide and Homicide-Suicide
Among Police is to fully explore what the
author refers to as “the near epidemic levels
of suicide and homicide-suicide” among law
enforcement officers, and ultimately to offer
some recommendations and best practices
with which to better address the problem.
The book begins by discussing suicide in
some depth, for one has to know suicide,
unequivocally, to understand a suicidal or
homicidal-suicidal officer. Briefly defined,
suicide is the human act of self-inflicted, selfintentioned annihilation, precipitated by a
multidimensional malaise—a general feeling
of being unwell. Suicide is not a disease.
Suicide is not a sin. Suicide is not a crime.
Suicide and homicide-suicide are complex,
multidetermined events—the result of an
interplay of individual, relational, social,
cultural, and environmental factors. This
complexity of causation necessitates a parallel complexity of knowledge. There are
at least two avenues to understanding: the nomothetic (general) approach, which
deals with generalizations using empirical, statistical, and demographic methods
or techniques; and the idiographic (specific) approach, which typically involves the
intense study of individuals. This book explores both.
Attempting to be mindful of the needs of the officer on the street, the mental health
provider, the administrator, the forensic specialist, and the survivors of these needless
tragedies, the belief is that by amalgamating the concerns of a diverse audience, we
can meet the challenge of identifying at-risk individuals and situations and saving
lives among police, their families, their fellow officers, and the community. Rather
than a cognitively constricted approach, there needs to be continuing development of
a multidimensional (or ecological) approach to understanding suicide and homicidesuicide, and suicide-prevention policies and procedures need to be in place to facilitate
effective prediction and control.
Format Information
6" × 9", Cloth, 222 Pages
ISBN 978-0-89503-390-1
$48.95 + postage & handling*
*please inquire for current rates
Baywood Publishing Company, Inc.
26 Austin Ave., PO Box 337, Amityville, NY 11701
phone 631.691.1270 • fax 631.691.1770 • toll-free orderline 800.638.7819
email [email protected] • website http://baywood.com
Neil Thompson, Ph.D.
Death, Value and Meaning Series
Series Editor: Dale A. Lund
T
he workplace is not immune to the problems, pressures, and challenges
presented by experiences of loss and trauma and the grief reactions they
produce. This clearly written, well-crafted book offers important insights and
understanding to help us appreciate the difficulties involved and prepare ourselves
for dealing with such demanding situations when they arise. People’s experiences of
loss and trauma are, of course, not left at the factory gate or the office door. Nor are
loss and traumatic events absent from the workplace itself. Loss, grief, and trauma
are very much a part of life—and that includes working life. Executives, managers,
human resource professionals, and employee assistance staff need to have at least a
basic understanding of how loss, grief, and trauma affect people in the workplace.
This book provides that foundation of understanding and offers guidance on how to
find out more about these vitally important workplace issues.
The text provides a valuable blend of theory and practice that will be of interest
to both students and professionals involved in management, human resources, and
organizational studies as well as those interested in the social scientific study of loss,
grief, and trauma—and, of course, to those involved in the helping professions,
i.e., social work, counseling and psychotherapy, health care, ministry, chaplaincy,
and pastoral studies. It is essential reading for anyone concerned with making the
workplace a more humane and effective environment, or anyone wishing to develop
an understanding of the complexities of loss, grief, and trauma in our lives.
Format Information
6" × 9", 138 Pages, Cloth, ISBN 978-0-89503-342-0
$43.95, plus $7.00 p&h in U.S. (please inquire for postage outside of U.S.)
Baywood Publishing Company, Inc.
26 Austin Ave., PO Box 337, Amityville, NY 11701
phone 631.691.1270 • fax 631.691.1770 • toll-free orderline 800.638.7819
e-mail [email protected] • website http://baywood.com
Complicated Grieving
and Bereavement
DEATH, VALUE AND MEANING SERIES • SERIES EDITOR: JOHN D. MORGAN
UNDERSTANDING AND TREATING PEOPLE EXPERIENCING LOSS
Editors: Gerry R. Cox, Robert A. Bendiksen and Robert G. Stevenson
In praise . . .
While one derives significance through the experience of loss, another may encounter
bereavement with less consequence. Complicated Grieving and Bereavement:
Understanding and Treating People Experiencing Loss examines complicated grief in
special populations, including the mentally ill, POW-MIA survivors, the differentiallyabled, suicide survivors, bereaved children, death at birth, death in schools, and
palliative-care death. Through humor, music, puppeteering, drama, family systems,
spiritual care and support groups, the book presents practical suggestions to those
managing grief in the face of traumatic death.
“I received this book to review soon after the events of September 11. It was an awkward experience to
review a book about loss when the number of lives lost in the terrorist attacks that day was so high. I
wondered how any book could help individuals and communities grieve and cope with such loss.
However, I was impressed by the content of this book and its emphasis on what are probably the very
ideas that are essential in helping people cope. The book is well organized and highly readable. The
authors of Complicated Grieving and Bereavement identify important issues involved in bereavement,
but, more importantly, they emphasize the use of the many strengths our patients can draw on. They
address a variety of of practical issues, many of which seem especially applicable in the aftermath of
the recent terrorist attacks. This book is a valuable addition to the resources of any clinician who will
deal with loss and bereavement.”
—Jeffrey L . Geller, M.D., M.P.H.
Editor, Psychiatric Services, May 2002, Volume 53, Number 5
“While the book covers a wide scope, it’s strengths lie in an exploration of grief in specific populations,
rather than in a systematic review of complicated grief. Of particular interest is the inclusion of
chapters on spiritual abuse and dementia, which alerts practitioners and researchers to issues not
commonly dealt with.”
—Jane Powell
Department of Forensic Medicine Counselling Unit, Sydney, Australia
Grief Matters 9(2), Winter 2006
Format Information: 6" × 9", 334 pages, Cloth, ISBN 0-89503-213-9
$70.95 + 7.00 p/h in U.S. (please inquire for rates outside of U.S.)
BAYWOOD PUBLISHING COMPANY, INC.
26 Austin Ave., PO Box 337 • Amityville, NY 11701
phone (631) 691-1270 • fax (631) 691-1770 • toll-free orderline (800) 638-7819
e-mail [email protected] • website http://baywood.com
Death, Value and Meaning Series
John D. Morgan, Series Editor
ETERNITY AND ME
The Everlasting Things
in Life and Death
Allan Kellehear
The 40 short reflections in this book address the ways in
which we face the prospect of death and loss. The first 20
reflections are designed to be read by (or to) anyone
living with a life-threatening illness; the other 20 are
reflections on living with grief, especially bereavement.
Each reflection is based on a single story drawn from one
of three sources: Dr. Kellehear’s professional experience
with individuals living with dying or loss; his own
experiences and stories from childhood; and the retelling
of some of the great myths and legends about life, love,
and death, selected from around the world—from Ireland
to Japan, from Melanesia to China. The book is written to
be accessible to a wide general audience. It can be read
from beginning to end like a conventional book; each
self-contained piece is also suitable for reading on a bus,
train, or plane journey, or before bed at night. Each piece
can be selected as a stand-alone meditation for use as a
discussion topic in pastoral care, counseling, or sermons.
These reflections are stories about how we can make the
most of life in the shadow of death and loss. They are
designed to instill hope and meaning in the difficult
times that can accompany human mortality.
6" × 9", 194 Pages, Paper, ISBN 0-89503-298-8
$37.95 + 7.00 p/h in U.S. (please inquire for rates outside of U.S.)
BAYWOOD PUBLISHING COMPANY, INC.
26 Austin Ave., PO Box 337, Amityville, NY 11701
phone 631-691-1270 • fax 631-691-1770 • toll-free orderline 800-638-7819
email [email protected] • website http://baywood.com
`