Arthur H. Crisp, M.D.'
St. George's Hospital Medical School, London, England
ABSTRACT-Emphasis is placed on the opportunities and importance, at this
time, of continuing with clinically oriented psychosomatic research. For instance,
the multidimensional studies of sleep and of depression are beginning t o throw
new light on psychosomatic processes, and so too are the studies of life events
and illness. Short-term prospective clinical investigations provide an attractive
framework for such work and they can be complementary to long-term survey
studies investigating relationships between constitutional characteristics including
personality, and social conditions and disease. There are some useful new tools
for measuring psychological characteristics.
If you are looking for something be very careful or you will be sure to find it.
The notion that disease arises from the complex interplay of multiple factors
over the years, existing and arising within the environment and the individual's
make-up, appeals mostly to those of a divergent turn of mind and perhaps also
those most prepared to tolerate a degree of uncertainty. Such characteristics are
probably not of a kmd that incline a person to systematic research with its often
reductionist ethos and necessary discipline. And yet there are limited-ways in
which the approach can ever advance: either by some universally evident major
preventive or therapeutic breakthrough-the discovery of the Vitamin C or
penicillin of psychosomatic medicine, an unlikely event-or by the relentless
replicable pursuit and identification of the multiple contributory, occasional,
variable and often hidden factors involved.
Meanwhile the clinician, whatever his approach, cannot escape from doing
research. His working life is spent accumulating and sifting information from
case to case and accordingly both shaping his approach to the classification and
'Professor of Psychiatry, University of London at St. George's Hospital Medical School,
Tooting, London,S.W. 17, U.K.
0 1976,Baywood Publishing Co., Inc.
doi: 10.2190/KUPK-3P7G-70U4-TY4G
/ ArthurH.Crisp
manipulation of the diseases he is encountering while at the same time inevitably
changing in his understanding of himself. It is merely a matter of how well he
does these things. At the end of the day, confronted by the enormous complexity of the mind-body apparatus and by a realization at last of why he
became a physician, he will find that someone else has discovered a necessary
condition under which the disease develops and a consequent capacity to
manipulate it mechanistically irrespective of the other human factors bearing
upon it. He is ultimately revealed in his true colors as an anti-mechanist. Fortunately others are close by to help, and today the contributions of biochemists,
physiologists and psychologists are of evident importance both in the animal
and the human experimental fields, as the preceding chapters reveal. It is the
clinician, however, who remains privileged in having unique research opportunities through his special access to the site of human disease. As Inman [ l ]
remarked, “The material for investigation is vast, and to hand daily in every
consulting room in the land. If the notions are far-fetched, they share that with
the oedipus complex and the vestigial cysts in the neck derived from the gill
slits of the embryo.”
In the event, research in this field has to explore the area of relationship and
interaction between the unique qualities of the individual and the more universal
human experiential and biological mechanisms. The differences between individuals, experiential and biological, are both inherited and acquired, and the
nature/nurture debate concerning such capacities as intelligence, personality and
the wider constitution remains inconclusive.
Several papers in this volume have illuminated the growing field of psychophysiology where real progress has been made in this respect. In recent years the
stereotyped autonomic responsivity found by the Laceys [ 2 ] to characterize
many individuals from early life has also been found to be still modifiable in
adult life by operant conditioning procedures [3]. Already, while the degree to
which this is experimentally possible in humans is unclear, therapists have leapt
in where others fear to tread and, with complicated, empirically derived and
little understood ‘‘learning’’ procedures such as hypnosis, systematic desensitization and transcendental meditation, have wrought significant effects on
physical processes and disease, e.g., the mantoux response [4], asthma [S] and
hypertension [6].
Indeed, adequate techniques for biochemical and endocrinological as well as
physiological research in the psychosomatic field are now at hand and are being
used in this field in many laboratories (e.g., Levi [7]), as previous papers also
testify. It is in the area of psychological measurement at the experiential,
affective and conceptual level that we have the main difficulties. There are in
fact many people who claim that individual subjective experience is not
accessible to measurement and yet also that it is the only aspect of life that is
Psychosomatic Research Today
/ 161
valid. The average practicing physician’s initial reaction seems to be to discredit
this. He traditionally operates according to other principles, nor does he find
many of his patients, representing 90 per cent of the total population annually,
holding such views when they are ill. Although he makes quite evident use of
external and general frames of reference when successfully treating them for
many diseases and disorders that clearly contain an experiential component,
nevertheless at the end of the day he may be aware that it was his patient’s
statement that he experiences, for example, his anger as having been most
tangible and meaningful, more so than any related behavioral manifestations
which are likely to have displayed greater diversity and non-specificity. This
then remains a present-day dilemma for psychosomatic research: the apparent
incompatibility in many respects of the existential position and the scientific
To attempt a comprehensive and effective overview of present trends and
methodological approaches in the face of such a problem and in the wake of
so many definitive papers in this present volume would be both presumptuous
and impossible. These papers cover a wide spectrum of approaches ranging from
the detailed study of individuals to the surveying of large populations. The
uncritical study of what were in effect highly selected populations of clinic
patients fortunately has at last been superseded. Such populations can, of
course, yield priceless information from “within group” studies, but only once
the limits of the study are recognized, especially those imposed by selection
factors, including the neurotic ones, involved. The papers also serve to identify
the psychatric illnesses as psychosomatic processes. Indeed, these may yet
provide a principal means of exploring cerebral chemical processes crucial to
the organization of the mind-body apparatus and its disorders. The main
psychotropic drugs, which have often proved disappointing as means of ultimately controlling psychotic and neurotic disorders, appear to have provided
us with excellent tools for manipulation of cerebral activity in the experimental
laboratory and clinical situations. For instance, in the wake of research into the
barbiturate- and amphetamine-like substances, the widespread central effects of
the phenothiazines on arousal and appetitive mechanisms and on mood and
motor behavior can now be seen to be opposed in certain respects to those of
the dopamine group of drugs currently used in the management of extrapyramidal syndromes. The monoamine oxidase inhibitor drugs, often clinically
disappointing in their effect on disorders of mood, can yet be seen to have
profound effects on the electro-physiological activity of the human brain, for
instance, in their effect on paradoxical sleep.
The current interest in the electrical and hormonal correlates of sleep and the
related discovery of a seemingly pervasive and necessary basic rest/activity cycle
162 / Arthur H.Crisp
throughout the twenty-four hours and its manipulation by such factors as diet
and tricyclic drugs, are potentially fruitful areas in the exploration of experiential and biological interrelations. At the present moment there is a veritable
explosion in the area of sleep research (e.g., Williams [8], Hartmann [9],
Oswald [ l o ] ) which was previously in danger of quietly atrophying after the
initial enthusiams of two decades ago.
Meanwhile, in the field of depressive psychosis, aspects of its psychobiological
correlates are beginning to be dissected out in a number of laboratories. The role
of thyroid metabolism [ l l ] , the status of disturbances of steroid activity [12,
131, the contribution of nutritional changes to the sleep disturbances in the
second half of the night [14], and the status of tryptophan metabolism in the
disorder [ 151 are some of the growing points in this area. They may ultimately
throw some refreshing light upon the central role presumed by some of depression in relation to the development and course of other diseases. Psychosomatically oriented physicians have often commented on the apparent reciprocal
relationship between a number of core psychosomatic diseases and depressive
psychosis. Patients with, for instance, such seemingly diverse diseases and disorders as asthma, obesity, hypertension, peptic ulcer or complaints of pain in
the absence of structural change, have been said to be less capable than others
of experiencing or displaying depression. If they come into close contact with
neurotic patients, if their behavior is manipulated by suggestion or their conflicts
clarified by psychotherapy, if their hypertension is treated with drugs, then the
presenting disorder may remit but sometimes intractable depression supervenes.
Although the dynamics of this process are unclear, psychotherapists might
propose that effective denial mechanisms have been dismantled. Yet few apart
from Hackett [16] have attempted to define and then measure the latter in
behavioral or experiential terms.
Others [17] meanwhile, have postulated that the underlying state of helplessness and hopelessness is in itself a seed bed for disease, a state arrived at accumulatively and/or in response to immediate life events and which, although not
necessary for the disease to develop, yet is significantly often the trigger to its
activation. For instance, Schmale and Iker [18], in a study meriting attention
because of its design, found such a state, independently identified, to be
significantly associated with the presence, found on subsequent biopsy, of
uterine cervical cancer in asymptomatic women who had entered the experiment
with repeated papanicolaou class 111 changes in their cervical cellular cytology.
It is such complex and conflicting clinical views and findings as these, probably
at least in part rooted in the variable biological and experiential aspects of the
state of depression itself, that the present research into this latter state may yet
help to clarify.
Psychosomatic Research Today
/ 163
The amount of straight clinical psychosomatic research has diminished in
recent years as attempts have focused on problems of measurement and
methodology which have sometimes then led on to other theoretical issues.
Nevertheless adequate tools and designs are available for use [19,20] .
A classic clinical experiment which appears never to have been repeated
either specifically or in terms of further utilization of its design was that
reported by Weiner et al. [21]. They studied healthy army recruits with the
proposition that, on a combination of psychological and physiological grounds,
they could predict which subjects would emerge with peptic ulcers after several
weeks of communal living and intensive drilling. They achieved this and established a model of constitutional predisposition (including personality and
specific physiological vulnerability), combined with individually meaningful life
stress, as providing the substrata of disease onset. Of course they had at their
disposal a gratifyingly homogeneous population and extensive medical and
paramedical resource. Such an approach, apart from the attractiveness of its
design, is likely to serve the purpose of providing straightforward definitive
evidence of a kind still important for convincing the mainstream of clinicians
of both theoretical and clinical importance of psychological factors in such
diseases. More recently the further study and quantification of life events [22241 in general and bereavement in particular [25, 261 has again demonstrated
their role in disease precipitation, thereby complementing Engel’s study of the
pathological subjective experience often attendant on such circumstances.
Longer term prospective studies of a population survey kind have mainly
been undertaken in Scandinavia which is characterized both by detailed statutory registration and documentation of the general population and also by low
social mobility, especially in some areas. Indeed, until recently the only
comparable area in the United Kingdom in this latter respect was Northeast
Scotland, where the epidemiological studies centered on Aberdeen [27] have
been yielding information of a social and physical kind for the last ten years or
more. In the immediate future this part of the country, with the advent of a
major oil industry, is likely to become more socially mobile. Meanwhile a major
survey in Sweden covering the past twenty years has begun to yield data
interesting to the psychosomatic field, linking, for instance, certain personality
types with the ultimate emergence of certain types of cancer [28]. Such a
finding is not surprising if one assumes personality to be one aspect of the
constitution of which others are endocrinological characteristics and immunological propensities, but it also invites a deeper exploration of the relationship
between human behavior, experience and disease vulnerability.
Such studies, seemingly somewhat pedestrian and requiring great patience,
can thus yield vital clues and can prove to be of unexpected heuristic value. The
Arthur H.Crisp
Swedish study has used Sjobring’s personality type measures which have not
been taken up universally. Measures of personality, of coping mechanisms,
neurotic status and of mood abound in the literature in English [29]. There is a
regrettable temptation for everyone to develop his own, thereby substantially
reducing the comparability of reported studies. The search for validity has sometimes masked the fact that adequate reliability is sufficient to allow a tool to
be applied under appropriate conditions in the search for psychological
characteristics and differences. Any such findings, even if their interpretation is
obscure, can then stand as being significant and meriting further exploration.
Equally, tools like the well standardized EPI [30] , which has sometimes been
criticized on the grounds of its limited clinical relevance, can in fact provide a
readily communicable and accurate measure of aspects of behavior, however
much these latter reflect social desirability and other factors. What such tools
offer is a measure which an individual will respond to predictably under standard
circumstances. The Middlesex Hospital Questionnaire [3 11 is another such
measure which seeks to identify six separate categories of psychoneurotic status
and which has been found t o be valid and to produce scale scores from the
general population significantly related to age, sex and social class. With scales
which provide concurrent scores on such states as anxiety, depression, obsessionality, hysteria and somatic complaint, it offers the opportunity of exploring
both the relationship between such scale scores themselves and between them
and concurrently measured physical characteristics such as blood pressure,
fatness, bereavement, and so forth [32, 331. Subsequent rescreening at twoyearly intervals allows the beginnings of a study of sequential relationships
between personality, mood and physical disease. As in Scandinavia, such studies
are much easier in the United Kingdom than in the U.S.A., in the sense that all
patients are registered with a general practitioner, and that the practice office
can also provide a ready agency for the epidemiological research. In the U.K. all
individuals also have a national health insurance number which allows them to
be traced subsequently by bona fide researchers.
Such survey studies are subject to the criticism that, in the search for
universal characteristics of a psychological kind, the baby has been thrown out
with the bath water. They do, however, often throw up clues which invite
further study of the matter in depth, and in this way survey and intensive
clinical studies, even including single case studies, often serve to fertilize and
refertilize each other.
In single case studies it is sometimes easier to identify and bring important
contributory variables under control while intensively manipulating one factor
with the prediction that certain changes will then occur elsewhere in the
system [34]. The investigation of the effect of treatment by drugs, of course,
Psychosomatic Research Today
I 165
uses this method but it is also possible to concentrate the technique so as to be
able to apply it usefully to the individual subject [34-371. This kind of
investigation allows the use of time-consuming methods which on the psychological side are well represented by the repertory grid techniques for displaying
and providing a basis for the analysis of individual construct (conceptual)
systems [38]. Construct theory provides a basis for the understanding of and a
means of quantifying the uniqueness of an individual’s conceptual organization.
Developed in the United States [39], it has caught the imagination of European
researchers more than those in America. Application of construct theory itself
requires the elicitation, from the individual under investigation, of “constructs”
in relation to a series of “elements.” It is then possible, for instance, to calculate
correlations between each construct and also for that matter between each
element, to identify their hierarchical relationship to each other and the dimensions or components of their organization [40, 411. Imposition of constructs
by the investigator allows a more general study of subjects [33] but this will
be primarily in terms of the investigator’s own constructs system. The method
is still subject to such overall influences as social desirability set. As a tool for
existential enquiry it defies the normal requirements for validity. In practice
it provides us with a projective technique for quantifying the meanings of
symptoms and changes in them which has not previously been available. The
growing dialogue between clinical scientists from many disciplines is rooted in
such developments.
1 . Inman WS: Emotion, cancer and time: coincidence or determinism. Brtt J Med Psychol
40:222-31, 1967
2. Lacey J, Lacey B: Verification and extension of the principal of autonomic response
stereotopy. Amer J Psychol 7 1 5 0 - 7 3 , 1 9 5 8
3. Miller NE: Learning of visceral and glandular responses. Science 163:43445, 1969
4. Black S, Humphrey J, Niven JSF: Inhibition of mantoux reactions by direct suggestion
under hypnosis. Brit Med J 1:56267, 1963
5. Moore N: Behaviour therapy in bronchial asthma: a controlled study. J. Psychosom Res
9:256-76, 1965
6. Patel CH: Yoga and bio-feedback in the management of hypertension. Lancet 1053-55,
7. Levi L: Stress, distress and psychosocial stimuli. Occup Ment Hlth, 3:2-10, 1973
8. Williams HL: The new biology of sleep. J Psychiat Res 8:445-78, 1971
9. Hartman EL: The Functions of Sleep. New Haven and London, Yale University Press,
10. Oswald I : Sleep, the great restorer. New Scientist 23:170-72, 1970
11. Whybrow P, Parlatore A: Melancholia, a model in madness: a discussion of recent
psychobiological research into depressive illness. Int J Psychiat in Med 4:351-78, 1973
12. Davies B, Carroll BJ, Mowbray RM: Depressive Illness: Some Research Studies.
Springfield, Illinois, Charles C. Thomas, 1972
13. Sachar EJ, Halpern S, Rosenfeld RS, et al: Disrupted 24-hour patterns of cortisol
secretion in psychotic depression. Arch Gen Psychiat 28:19-24,1973
14. Crisp AH, Stonehill E: Aspects of the relationship between sleep and nutrition: a study
of 375 psychiatric out-patients. Brit J Psychiat 122:379-94, 1973
/ ArthurH.Crisp
15. Coppen A, Whybrow PC, Noguera R: The comparative antidepressant value of
L-tryptophan and imipramine with and without attempted potentiation by liothyronine. Arch Gen Psychiat 26:23441,1972
16. Hackett TP, Weisman AD: Denial as a factor in patients with heart disease and cancer.
Ann NY Acad Sci 164:802-11,1969
17. Engel GL: The concept of psychosomatic disorder. J Psychosom Res 11:3-9, 1967
18. Schmale AH, Iker HP: The affect of hoplessness and the development of cancer: 1.
Identification of uterine cervical cancer in women with atypical cytology. Psychosom
Med 28:714-21,1966
19. Crisp AH: Some approaches to psychosomatic clinical research. Brit J Med Psychol
20. Crisp AH: Some psychosomatic aspects of neoplasia. Brit J Med Psychol, 43:313-31,
21. Weiner H, Thaler M, Reiser MF, Mirsky IA: Aetiology of duodenal ulcer. Psychosom
Med 19:l-10,1957
22. Rahe RH: Lifechange measurement as a predictor of illness. Proc Roy SOC Med
23. Brown GW, Sklair F, Harris TO, Birley JLT: Life events and psychiatric disorders: Part
1. Some methodological issues. Psychol Med 3:74-87, 1973
24. Paykel ES, Myers JK, Dienelt MN, et al: Life events and depression: a controlled study.
Arch Gen Psychiat 21:75360, 1969
25. Parkes CM, Benjamin B, Fitzgerald RG: Broken hearts: a statistical study of increased
mortality among widowers. Brit Med J 1:74043,1969
26. Rees WD,Lutkins SG: Mortality of bereavement. Brit Med J 4:13-16, 1967
27. Baldwin JA (ed): Aspects of the Epidemiology of Mental Illness: Studies in Record
Linkage. Internat Psychiat Clin Vol8, No 3. Boston, Little, Brown, 1971
28. Hagnell 0: The premorbid personality of persons who develop cancer in a total
population investigated in 1947 and 1957. Ann NY Acad Sci 125:846-55, 1966
29. Aitken RCB, Zealley AK: Measurement of moods. Brit J Hosp Med 4:215-24, 1970
30. Eysenck HJ, Eysenck SBG: Manual of the Eysenck Personality Inventory, London,
University of London Press, 1074
31. Crown S, Crisp AH: A short clinical diagnostic self-rating scale for psychoneurotic
patients. The Middlesex Hospital Questionnaire. Brit J Psychiat 112:917-23, 1966
32. Crisp AH, Priest RG: Psychoneurotic profiles in middle age. Brit J Psychiat 119:385-92,
33. Crisp AH, McGuinness B: The relationship between obesity and psychoneurosis in the
general population. Brit Med J 1975 (in press)
34. Crisp AH: “Transference,” “symptom emergence” and “social repercussion” in
behaviour therapy-a study of 54 treated patients. Brit J Med Psychol 39:179-96, 1966
35. Frank CM, Wilson GT (eds): Annual Review of Behaviour Therapy, Theory and
Practice. Section VIII: Clinical studies and case reports. London, Butterworth, 1973, pp
36. Crisp AH, Stonehill E: Sleep patterns, daytime activity, weight changes and psychiatric
status: a study of three obese patients. J Psychosom Res 14:353-58, 1970
37. Crisp AH, Fransella F : Conceptual changes during recovery from anorexia nervosa. Brit
J Med Psychol 45:395405,1972
38. Bannister D, Mair JMM: The Evaluation of Personal Constructs. London and New York,
Academic Press, 1968
39. Kelly GA: The Psychology of Personal Constructs. Vols I and 11. New York, Norton,
40. Mair JMM, Crisp AH: Estimating psychological organisation, meaning and change in
relation to clinical practice. Brit J Med Psychol 41:15-29, 1968
41. Bannister D, Fransella F: Inquiring Man: The Theory of Personal Constructs.
Harmondsworth, Middlesex, Penguin Books 197 1