compendium research 2014–2015 InnovatIons In HealtHcare DelIvery

Innovations in Healthcare Delivery
A National Science Foundation Industry-University Cooperative Research Center
Table of Contents
The Center for Health Organization Transformation (CHOT) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
The National Science Foundation I/UCRC Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
Becoming an Industry Member . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Industry Testimonials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Selected Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Project 1Characterizing and Reducing Avoidable Outside Utilization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Project 2
Identifying Emergency Department Efficiency Frontiers and the Factors Associated with
Their Efficiency Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Project 3
Predictive Models for System Utilization, Capacity, and Flow Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . 11
Project 4Shared Commons Game Theory Models to Improve Antibiotic Stewardship . . . . . . . . . . . . . . . . . . . . . . . . . 12
Project 5
Understanding the Dual Effect of Hospital Safety Culture on Patients and Care Providers;
Optimizing Hospital Safety Culture and Reducing Safety Events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Project 6
Bundle Science Statistical Models and Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
Project 7Economics and Potential Financial Model of the Perioperative Surgical Home:
Developing a Framework for PSH Design and Action . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Project 8Healthcare Improvement Spread Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Project 9Healthcare System Redesign: Advancing Delivery Quality and Effectiveness . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Project 10 A Data Mining Methodology for Patient Adherence to Home-Based Therapies . . . . . . . . . . . . . . . . . . . . . . . 18
Project 11Applying the Studer Group Evidence Based Leadership Principles to Improve
Physician Engagement and Performance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
Project 12Can ‘Visiting Specialists’ Coverage Agreements Return a Positive ROI
for Sponsoring Institutions? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Project 13Chronic Disease Management: Clinical, Community, and Patient-Centered Approaches . . . . . . . . . . . . . . . . 21
Project 14Analysis of Practice Variance: Outcome and Evidence-Driven Clinical Practice Re-Design . . . . . . . . . . . . . . . . 22
Project 15Hospital Acquired Conditions—Systematic Analysis and Adaptive Approach . . . . . . . . . . . . . . . . . . . . . . . . . 23
Project 16
Quantifying the Impact of Pay-for-Performance Financial Incentives to Reduce
Healthcare-Associated Infections . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Project 17Reinventing the Pediatric Primary Care Model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Project 18
Using Lean Six Sigma to Reduce Hospital Acquired Conditions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
Project 19A Combined Human-Factors and Quality Improvement Approach to Assess Health Information
Technology Usability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Project 20Automated Languate Translation for Improving Care Management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
Project 21
Designing Health Information Technologies to Help Patient Care Teams
Identify and Manage Information Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
Project 22
Gamification for Self-Monitoring of Patients for Enhanced Wellness Outcomes . . . . . . . . . . . . . . . . . . . . . . . 30
The Center for Health Organization
Transformation (CHOT)
The mission of the Center for Health Organization Transformation (CHOT) is to advance the knowledge and
practice of transformational strategies in evidence-based management and clinical practice.
CHOT conducts cooperative research among universities, health systems, and other health-related industries.
The Center relies on multi-disciplinary approaches to advance and link systems design and organizational
technologies in innovation research.
CHOT’s current university partners
CHOT’s current industry members
American Society of Anesthesiologists
Partners HealthCare
Children’s Healthcare of Atlanta
Seattle Children’s Hospital
East Texas Medical Center Regional Healthcare System
Grady Health System
Studer Group, LLC
Texas Children’s Hospital
Meadows Regional Medical Center
University of Texas System
Morehouse School of Medicine
Northside Anesthesiology Consultants, LLC
The National Science Foundation
I/UCRC Model
As a National Science Foundation industry-university cooperative research center (I/UCRC), CHOT
follows a model of an industry-academic partnership that has benefited industry-focused research
across more than 50 disciplines. CHOT creates a safe, mutually beneficial, cooperative environment
where innovative, leading healthcare industry members can come together in collaboration to:
Support important transformation initiatives addressing health organization management
and services
Examine the implementation of transformational strategies
Partner with healthcare management researchers to improve such initiatives and strategies
Participate in research in a cost-effective manner
Play a critical role in shaping the education of future healthcare leaders as managers, engineers,
and health professionals
Becoming an Industry Member
Most observers agree that the old ways of delivering healthcare services are no longer adequate, so
stakeholders are increasingly exploring innovative approaches. Industry membership allows partners to be
on the forefront of leading that innovation. Our research model relies on the knowledge and experience
of healthcare leaders to guide academic research. This cooperative model ensures that the research is both
meaningful and applicable to the healthcare industry and provides immediate decision support for CHOT
NSF contributes $55,000 to each University Partner to cover the administrative costs of CHOT
Each Industry Member contributes $50,000 per year to its respective University Partner
Ninety percent of industry member funding goes directly to research activities
The CHOT Industry advisory board (IAB) defines the research agenda for CHOT researchers on a yearly
basis using a project voting mechanism developed by the IAB
Benefits of Joining CHOT
Provides an objective, third-party, analytical perspective to healthcare organizations who seek to optimize
the impact of innovations at a fraction of the cost of consulting
Exclusive setting for information sharing and networking with other innovative healthcare systems and
industry members across the United States
Using your own organization’s data and a wide range of experienced, skilled researchers and graduate
students to evaluate innovation design and implementation
Partnerships among leading health systems and major universities in driving effective innovation in
Industry Testimonials
“It has been a pleasure working with the
team from CHOT. The projects have helped
our organization think differently and
helped to develop strategies. The research
team has always been very responsive and
helped to educate our leadership team on
the use of data.”
Kay Tittle, President
Texas Children’s Pediatrics
“The outstanding universities who are
the members of CHOT are producing
groundbreaking and outstanding research on
some of the toughest opportunities facing
healthcare today and in the future. In addition,
they are helping to prepare very talented
students, who are our future employees in the
US healthcare system. I am especially impressed
with the caliber of the students.”
Steven B. Wagman, Vice President
Enterprise Solutions Implementation Chair,
Healthcare USA Diversity & Inclusion Council,
Siemens Healthcare
“CHOT is a valuable resource to the American
Society of Anesthesiologists (ASA) and
the physician committee responsible for
developing and disseminating a base of
knowledge around the perioperative surgical
home (PSH). There is a real and effective
synergy between the CHOT researchers
and ASA’s in-house research staff. The
CHOT research and reports were the
perfect combination of academic rigor and
objective qualitative information that was
instantly relevant and useful to enhance the
productivity of ASA’s concept development
Thomas Miller, PhD, MBA,
Director of Health Policy Research,
American Society of Anesthesiologists
Selected Publications
Dissemination strategies related to CHOT’s research activities include monthly webinars, presentations
at professional and academic research conferences, and publications in peer-reviewed journals. CHOT
researchers acknowledge the importance of dissemination as a way to influence the thought leadership of
the healthcare delivery domain. This is a list of selected and most recent papers published in leading peerreviewed journals.
Bennett-Millburn, A., Griffin, P., Hewitt, M., Savelsbergh, M. “The Value of Remote Monitoring Systems for
Treatment of Chronic Disease.” To appear in IIE Transactions on Healthcare Systems Engineering, 2014.
Gregory, S.T., Tan, D., Tilrico, M., Edwardson, N., Gamm, L. “Bedside Shift Reports: A Systematic Literature
Review.” To appear in Journal of Nursing Administration, October 2014.
Griffin, P., Yan, I., “Association of Food Environment and Food Retailers with Obesity in US Adults.” To appear
in Health and Place, 2014.
Hagen, M.S., Jopling, J.K., Buchman, T.G., Lee, E.K. (2013). “Advancing Public Health and Medical
Preparedness with Operations Research.” American Medical Informatics Association Annual Symposium
Proceedings: 841-50.
Kang, H., Nembhard, H. B., Rafferty, C., DeFlitch, C. (2014). “Patient Flow in the Emergency Department: A
Classification and Analysis of Admission Process Policies.” To appear in Annals of Emergency Medicine, 2014.
Kash, B.A., Gamm, L.D., and Spaulding, A.C. (2013). “Absorptive Capacity (ACAP) for Transformation in
Healthcare: A Framework for Research.” Change Management: An International Journal 13(1): 1-13.
Kash, B.A., Spaulding A., Johnson, C.E., and Gamm, L.D. (2014). “Success Factors for Strategic Change
Initiatives: A Qualitative Study of Healthcare Administrators’ Perspectives.” Journal of Healthcare
Management 59(1):65-82.
Kash, B.A., Spaulding, A., Johnson, C.E., Gamm, L.D. (2014). “Relevancy of the Resource Based View
in Healthcare Strategic Management: A Comparative Case Study.” To appear in Journal of Strategy and
Management, Volume 7 number 3.
Kraschnewski J.L., Sciamanna, C., Stuckey, H.L., Chuang, C.H., Lehman, E.B., Hwang, K.O., Sherwood,
L.L., Nembhard, H.B. (2013). “A Silent Response to the Obesity Epidemic: Decline in US Physician Weight
Counseling.” Medical Care 51(2):186-92.
Lee, E.K., Atallah, H.Y., Wright, M.D., Post, E.T., Thomas, C.I.V., Wu, D.T., Haley, L.L. “Transforming
emergency department workflow and patient care.” To appear in Interfaces, 2014.
Lee, E.K., Lu, T.W., Tal, S., Jose, J. ”Medical Alert Management: An Automated Decision Support Tool to
Reduce Alert Fatigue.” To appear in American Medical Informatics Association Proceedings, 2014.
Lee, E.K., Pietz, F., Benecke, B., Mason, J., Burel, G. (2013). “Advancing Public Health and Medical
Preparedness with Operations Research.” Interfaces 43(1): 79-98.
Selected Publications
Lee, E.K., Yuan, F., Templeton, A., Yao, R., Kiel, K., Chu, J.C.H. (2013). “Biological Planning for High-Dose
Rate Brachytherapy: Application to Cervical Cancer Treatment.” Interfaces 43(5): 462-47.
Lee, E.K., Yuan, F., Zhou, R.L., Lahlou, S., Post, E., Wright, M., Atallah, H. Haley, L.L. “Modeling and
Optimizing Emergency Department Workflow of Large Urban Public Hospital.” To appear in Interfaces, 2014.
Musdal, H., Shiner, B., Chen, T., Ceyhan, M.E., Watts, B.V., Benneyan, J. (2014). “In-person and Video-based
Post-Traumatic Stress Disorder Treatment for Veterans: A Location-Allocation Model.” Military Medicine
Peck, J.S., Gaehde, S.A., Nightingale, D.J., Gelman, D.Y., Huckins, D.S., Lemons, M.F., Dickson, E.W.,
Benneyan, J.C. (2013). “Generalizability of a Simple Approach for Predicting Hospital Admission From an
Emergency Department.” Academic Emergency Medicine 20(11): 1156-1163.
Schuller, K.A., Kash, B.A., Edwardson, N., Gamm, L.D. (2013). “Enabling and Disabling Factors in
Implementation of Studer Group’s Evidence-Based Leadership Initiative: A Qualitative Case Study.” Journal of
Communication in Healthcare 6(2):90-99.
Watts, B.V., Shiner, B., Ceyhan, M.E., Musdal, H., Sinangil, S., Benneyan, J. (2013). “Health Systems
Engineering as an Improvement Strategy: A Case Example Using Location – Allocation Modelling.” Journal
for Healthcare Quality 35(3): 35-40.
Safety, Access, and Efficiency Cluster
Characterizing and Reducing Avoidable Outside Utilization
Outside referrals or out-of-network “leakage” is
a ubiquitous problem for many health systems,
especially accountable care organizations and
other health systems with risk-sharing insurance
contracts. Leakage occurs when patients within
a health system’s population are referred to or
otherwise receive care outside that system, with both
cost and care continuity implications. Frequently,
for various reasons, an index referral leads to a
chain of additional referrals with unclear patterns
and causality and with poor visibility in billing data
as to how, why, and for whom these referrals are
occurring (e.g. typically just the visit date, specialist,
and original primary care provider are known).
This project consists of two objectives, (1) to explore
the utility of a variety of analytic methods to help
understand, characterize, and describe referrals and
leakage patterns, and (2) to help reduce, disrupt, or
prevent leakage. Phase 1 will mostly focus on the first
objective with a variety of methods tested for their
feasibility and utility on pilot data from one or more
ACOs. Potential approaches include data mining,
classifiers, predictive modeling, and social network
analysis. We also will investigate potential approaches
to detect, prevent, or mitigate avoidable out-of-system
referrals, using methods such as network cuts, agentbased simulation, or systems dynamics models.
How this is different
than related research
Most approaches to managing outside utilization
have focused on methods to identify referrals that
were appropriate or not, education within a network,
and contract mechanisms. The current approach
complements this work with data analytic and
operations research methods to better understand,
prevent, and intervene/minimize leakage.
member benefits
• Better understanding of how and why leakage
• Identification of potential sources and patterns of
avoidable leakage
• Approaches to detect, prevent, and mitigate
avoidable out-of-network referrals
Safety, Access, and Efficiency Cluster
Identifying Emergency Department Efficiency Frontiers and
the Factors Associated with Their Efficiency Performance
Emergency department (ED) crowding has been
recognized as a serious concern in hospitals
nationwide. In response, healthcare organizations
have pushed EDs to tackle the issues that result
from crowding and to improve the efficiency of
care. However, there exists no single standardized
metric to assess hospital performance. As a result,
hospitals and external organizations have used
many different performance metrics to assess the
operational efficiency in EDs. Although the metrics
can play a role in representing the efficiency of each
ED in a quantitative way, a simple comparison of the
numbers can lead to inaccurate conclusions because
inputs consumed for the outcomes were not taken
into account. Data Envelopment Analysis (DEA) can
be a useful tool to evaluate the efficiency of each
ED among a set of peer groups and compare their
performance. This study aims to develop a datadriven framework for benchmarking efficient EDs
and determining appropriate stratifications of all
decision making units into peer groups.
How this is different
than related research
Many studies have used time intervals (e.g., door to
doctor, door to bed, and length of stay) to measure
efficiency of EDs. However, a simple comparison of
the numbers can lead to inaccurate conclusions when
the definitions of the metrics are not the same and
when other significant factors affecting the efficiency
are not considered. By using a DEA and statistical
methods, this study will develop a framework to
define appropriate peer groups in which efficiency
of EDs are compared and identify profiles of efficient
EDs in each peer group.
Our industry partnership with Verizon has led to an
understanding that for patients, insurance companies
and hospitals, gamification will transform the
manner in which wellness management is designed
and advanced. IT industries can benefit largely from
the software platforms developed under this project
and a better understanding of the data acquisition,
transfer and management needs.
Safety, Access, and Efficiency Cluster
Predictive Models for System Utilization, Capacity, and
Flow Optimization
This is a phase-1 pilot project to scope and initiate
a subsequent portfolio of work in the general area
of predictive modeling to improve real-time ability
to manage patient flow, system utilization, and
care pathways. The value of predictive information
in healthcare is increasingly appreciated, such as
for patient risk identification but less explored in
other potentially useful logistics contexts. This
project investigates four specific potential general
applications, with the objective of obtaining
preliminary results and viability vetting in order
to influence the focus and direction of subsequent
phase-2 projects. In each case preliminary results
will be generated and evaluated in order to assess
decision making utility and specifics of future
projects. Applications include predicting (1) bed
demand in intensive and critical care units onethrough-seven days in advance on a rolling basis, (2)
system wide patient flow similar to above, (3) long
outlier lengths of stay, such as earlier identification
of long-term acute care (LTAC) patients, and (4)
patients appropriate for palliative care discussions.
Primary modeling approaches are envisioned
to include Monte Carlo simulation, probability
convolutions, logistic regression, and time series
analysis. Each application also will be evaluated
for the usefulness of predictive information to
decision making, via direct analysis and stakeholder
How this is different
than related research
While predictive models per se are used in many
healthcare contexts, most uses tend to focus more on
patient risk, changes, or health status than on system
status and rates of change. Each above application is
of significant interest to a number of health systems
in order to better manage the delivery system and its
most effective utilization. Additionally, our approach
to system flow and bed demand uses a recently
developed new probability model, resulting in a fairly
large and previously intractable convolution, and
a novel generating function approach to its rapid
• Understanding how to use predictive modeling
for bed demand, system utilization, and patient
• Identification of challenges and opportunities
• Improved system utilization, costs, flow, and
Safety, Access, and Efficiency Cluster
Shared Commons Game Theory Models to Improve
Antibiotic Stewardship
Antibiotic resistance remains a growing problem
of broad health and cost concern, with significant
focus on antimicrobial stewardship as one important
intervention. This project develops and uses
game theoretic models of stewardship policies,
participation rates, and intervention design to help
understand resistance spatial-temporal dynamics and
how to best limit resistance locally and regionally.
In behavioral economics, stewardship can be viewed
as a “tragedy of the commons”, Hardin’s analogy
of a shared town pasture for which each individual
herder has incentive to graze their sheep without
concern for the others, thereby reducing the longterm value to everyone. For antibiotic stewardship,
this equates to over-use reducing their effectiveness,
where short-term incentives exist to use antibiotics for
individual care episodes but at the consequence of
reducing long-term effectiveness across a community.
Results will help analyze regional spread and growth
of antibiotic resistance over time as a function of
stewardship participation percentages, distribution,
and compliance, which will be used to help inform
policies and influence awareness, participation, and
cooperation in such programs.
How this is different
than related research
While antibiotic stewardship has been promoted by
numerous patient safety, epidemiology, and infection
control organizations, to our knowledge little-to-no
work has been conducted to model the impact of such
programs to help understand and inform policy and
• Improved understanding of how stewardship
policies, participation rates, and consistency
impact resistance
• Methodology to identify the most effective
interventions to reduce the extent and spread
of resistance
Safety, Access, and Efficiency Cluster
Understanding the Dual Effect of Hospital Safety Culture
on Patients and Care Providers; Optimizing Hospital Safety
Culture and Reducing Safety Events
The healthcare industry in the United States
continues to report among the highest rates of
workplace injury and illness of all industries. Many
studies examine care provider personal safety
perceptions and have found these perceptions
influence care provider health and wellness. With
respect to patient safety, hospitals continue to
struggle with effective tools and processes to reduce
patient safety events. Retrospective data shows that
many of the facets that promote a safe environment
for care providers are the same facets as those that
promote a safe environment for patient care. This
project will identify and assess the facets of safety
culture that influence both care provider and patients
safety events and determine how safety events may
influence patient satisfaction scores (as measured by
the Hospital Consumer Assessment of Healthcare
Providers and Systems–HCAHPS).
studies that examine both patient safety and care
provider safety in tandem. Further, safety culture
studies have yet to include the influence of poor
safety culture on patient and family satisfaction with
their care experience. The inclusion of HCAHPS
patient satisfaction scores presents the financial
imperative for hospitals to optimize their safety
culture, a relatively unexplored imperative in safety
The results of this research will assist all hospitals in
developing a better understanding of the relationship
between patient and care provider safety and
the effect of safety events on HCAHPS scores. By
identifying the patient/provider commonalities in
safety, these relationships will provide hospitals with
critical areas of focus to improve the hospital’s safety
culture and reduce safety events for both providers
and patients and help improve patient satisfaction.
How this is different
than related research
While there are substantive literature bases in both
employee and patient safety, there is a dearth of
Macro/Policy Cluster
Bundle Science Statistical Models and Analysis
Use of evidence-based bundles has become common
for monitoring evidence base compliance in many
patient safety contexts. Examples include surgical site
infections, ventilator pneumonia, acute myocardial
infarction, total joint replacement, coronary artery
bypass graft surgery, and others. Despite becoming
part of routine improvement projects, the evidence
base on evidence based bundles is limited at best.
Little actually is known about the best way to form
bundles, relationships between care elements,
individual and combined adverse events (AE)
predictive ability and efficacy, or how to use them as
intervention triggers. This project therefore addresses
several important questions and needs, using data
from two or more health systems: (1) Statistical
Process Control (SPC) methods for monitoring
bundle compliance, including self-starting methods
and prototyping an automated surveillance systems,
(2) regression analysis of bundle elements, including
quantification of interaction terms and the so-called
culture-of-safety “bundle effect”, and (3) how to best
use these results as an intervention trigger.
How this is different
than related research
Bundles are used increasingly to measure and
motivate patient safety improvement activities,
typically defined by expert consensus and literature
review. Little-to-no statistical work, however, has been
conducted on the science of creating and validating
bundles themselves, nor using them for prediction
and surveillance. Our prior work, additionally,
developed new control charts for bundles, with
significant detection improvements but also
highlighting several limitations and research needs
(weighting, start-up, parameter estimation).
• Increased understanding of how to create and use
bundles for patient safety quality improvement
• Validated statistical methods for comparing
and monitoring bundle compliance over time,
manually or in an automated surveillance or
triggering system
• Understanding of the relative importance and
interaction terms of different bundle elements
and, more broadly, development of a general
bundle science framework
Macro/Policy Cluster
Economics and Potential Financial Model of the Perioperative
Surgical Home: Developing a Framework for PSH Design
and Action
The “perioperative surgical home” (PSH) is a
relatively new concept that is based, at least in part,
upon the patient-centric characteristics of the
medical home combined with foci on team science,
micro-systems, service line management, carecoordination, and bundled payment. The purpose
of this study is to continue to define the “surgical
home” conceptually and to identify and describe
the economics and detailed financial model of one
selected PSH model in the United States.
How this is different
than related research
Unlike the related concept of patient-centered
medical home that dates back over 50 years, the PSH
is a product of a new environment of care concerned
with improved safety, effectiveness, timeliness, and
efficiency of surgical care. This research is heavily
driven by both theory and practice to more clearly
define the financial model of the PSH and its variants
across the health care industry. Furthermore, it
requires close collaboration with professionals
associated with the selected PSH at all stages of the
The sponsor as well as other associations, hospitals,
and policy makers will benefit from a clear under­
standing of the nature, operational design, and
financial model for leading PSH programs in the U.S.
Specific attention will be given to characteristics of a
viable PSH financial model starting with one specific
surgical specialty.
Macro/Policy Cluster
Healthcare Improvement Spread Models
This is a continuation project of a current phase-1
project in response to our industrial advisory board
request for proposals looking at the slow spread of
improvement in healthcare. The focus is to develop
analytic models of the spread of innovations and
improvement knowledge across healthcare systems
and healthcare quality improvement networks.
Phase-1 consisted of two activities: (1) applying
social network analysis tools to “map” the structure
of several healthcare quality improvement networks
to investigate their interconnectedness relating to
spread, and (2) developing two proof-of-concept
agent-based simulation models of the spread of
improvements across such networks. Phase-2 now will
continue this work and apply results to two specific
applications to: (1) validate our simulation model
using real data from two identified networks, and (2)
develop an optimization framework to maximize the
spread of innovations across the network. We will also
generalize knowledge and explore how to bridge the
gap between our theoretical work and practical use.
How this is different
than related research
While a significant amount of work by others has
focused on project management and the challenge
of improvement implementation (e.g. Mayo Clinic
Model for Diffusion, IHI Framework for Spread, etc.),
less is known about how such projects and innovations
actually spread across quality improvement networks,
resulting in a need to better understand and
accelerate the spread of good ideas across healthcare
improvement communities.
• Better understanding of the evolution and
structure of effective and ineffective quality
improvement networks
• Identification of ideal network structures to
promote effective spread of ideas and innovation
Macro/Policy Cluster
Healthcare System Redesign:
Advancing Delivery Quality and Effectiveness
Individual health systems provide various services
and allocate different resources for patient care.
Healthcare resources including professional and
staff time are constrained. Patients are ‘sicker’ often
with a combination of chronic diseases. It would
already take 16–18 hours daily to do everything the
guidelines recommend that primary care provide for
their patients. Patient lifestyle patterns are mostly
suboptimal with adherence with pharmacotherapy
is often limited. This study aims to (1) identify
critical variables that impact outcomes (e.g.
control of risk factors and prevention of hospital/
ED admission) and inform allocation of limited
time and resources for greater effect; (2) address
realistically modifiable social determinants of health
that will improve community health; and (3) seek
greater use of treatment evidence (e.g. secondary
EMR usage, “OMICs” data) to advance quality
and effective of care delivery. We aim to increase
quality and timeliness of care, maximize financial
performance, and decrease practice variability across
the organization.
How this is different
than related research
This study attempts to combine social-economic
and demographics demands, hospital resources,
and evidence of treatment (including EMR, Omics,
and other laboratory data) to redesign the delivery
process for quality and effectiveness of healthcare
delivery. While efficiency is often performed via
process improvement, patient risk factors, disease
patterns and treatment characteristics may shed lights
on resource needs and care requirement, and provide
holistic health systems redesign opportunities for
improving care quality and effectiveness.
Improve quality and efficiency of care
Reduce waste
Serve more needed patients
Improve demand-resource alignment
Reduce prolonged LOS (and thus reduce hospital
acquired conditions)
• Improve capability in the event of pandemic or
disaster response.
From the patient standpoint, it offers timeliness
and personalized evidence-based care, and reduces
unnecessary hospital stays and the associated risks
and costs. This work has the potential to reduce
healthcare delivery disparities.
Patient-Centered Care Cluster
A Data Mining Methodology for Patient Adherence to
Home-Based Therapies
Patient non-adherence to physician-prescribed
disease and wellness management protocols is a major
challenge in the healthcare industry and has led to
an increase in hospital visits, health risks and medical
costs. For example, the non-adherence to prescribed
medication results in over 125,000 deaths per year
and a financial burden to the healthcare system
exceeding $100 billion in direct costs. This project
will explore patient adherence for those who adopt a
proposed sensor and visualization system for remote
wellness management and feedback.
How this is different
than related research
Systems such as AutoCITE reveal that remote
patient supervision has tangible impact on patient
health outcomes. The main limitations of existing
techniques are that they are physically invasive, often
requiring patients to wear some digitally connected
device for an extended period of time. Furthermore,
these systems do not provide an integrated healthcare
delivery strategy that connects the sensing results to
the patients and healthcare officials in a seamless,
visually straightforward manner. The proposed
project aims to not only predict patient adherence,
but also provide feedback to both patients and
physicians, which can then help physicians prescribe
alternative solutions if a patient is non-adherent.
Our industry partnership with Verizon has led to an
understanding that for patients, insurance companies
and hospitals, a convenient and automated technique
to monitor treatment progress can lead to large
time and money savings. In particular, industries
can benefit largely from the research into sensor
placement and data management and transfer. This
will be an increasingly important field, as low cost
sensors we use in our homes become more prevalent.
Patient-Centered Care Cluster
Applying the Studer Group Evidence Based Leadership
Principles to Improve Physician Engagement and Performance
Studer Group, an outcomes-based firm dedicated
to improving the patient experience, has been
developing its Evidence-Based Leadership technology
to help address the issues faced by both health
systems, physician practices, and individual physicians
successfully integrate. This project proposes the
evaluation of the Studer Physician Feedback
System (SPFS), (referred to as the intervention),
recently developed to assist health system leaders
and physicians with hospital/physician integration
initiatives. The integration of physicians from
independent practice into health systems has proved
challenging over the past several decades, and has
been met with mixed financial, economic, and
physician engagement results.
How this is different
than related research
Current research regarding hospital and physician
integration tends to focus on distal outcomes (e.g.
financial performance). Further, there is a paucity
of evidence regarding tools and methods for health
system leaders to guide these efforts, and provide
real-time feedback across a variety of dimensions,
including clinical, safety, engagement and
This study offers a unique opportunity to assess
such a comprehensive intervention, and disseminate
important findings to a practitioner-oriented
audience, thus impacting the quality and nature of
care in a variety of settings.
Patient-Centered Care Cluster
Can ‘Visiting Specialists’ Coverage Agreements Return a
Positive ROI for Sponsoring Institutions?
Access to specialty care has been and continues
to be a pressing issue for rural patients. While it
has always been desirable to push care back to
these smaller, underserved markets, volume is not
typically sufficient to support specialty physician
coverage. This problem has been compounded by
a reimbursement system that pays physicians on
volume rather than value basis. Thus, health systems
have begun to spend large amounts of money by
subsidizing employed and independent specialists
to offer clinics in rural areas. Sponsor systems
stand to benefit from offering such subsidies by
potentially reducing unnecessary hospitalizations
and rehospitalizations for patients who would have
otherwise not been seen by a specialist. However,
little if any research has been conducted to date to
determine if this ‘visiting specialist’ model yields
a positive return on investment (ROI) for the
sponsoring health system.
How this is different
than related research
Little research has been conducted to measure the
visiting specialist model’s impact on quality and
costs. None of this research was conducted in light of
new CMS penalties on 30-day readmissions, stricter
community needs assessment requirements, or more
generally on the recent shift towards population
health management.
• Determine if the system’s CHF 30-day re-admission
rate for patients in targeted rural markets is
affected by the visiting specialist model
• Determine if the ROI for the sponsor is system is
positive or negative
• Identify local factors (e.g., physician-population
ratio, frequency of service offerings, mean patient
severity index, distance to system etc.) with the
biggest impact on the sponsor system’s ROI
• Results should also allow researchers to
calculate the value of physician impact on the
community—a key requirement for community
needs assessments
Patient-Centered Care Cluster
Chronic Disease Management: Clinical, Community, and
Patient-Centered Approaches
Sixty-eight percent of Medicare spending goes to
people with five or more chronic diseases. Reports
found that between 44%–57% of older patients take
more than one unnecessary drug. The management
of multiple diseases is complicated and offers
daunting challenges to healthcare providers. More
drugs are prescribed for treatment, which causes
reduced adherence of patient to drug therapy, higher
possibility of drug-drug interactions, more side effects
observed on patients, less effective treatment, and
more frequent changes in drug therapies. This results
in more hospital visits, heavier burden on the use of
health resources and higher medical expenses. The
objective of this study covers both the clinical visits,
and a patient-home-centric approach to optimize the
outcome and sustained health of patients.
How this is different
than related research
First, the project focuses on co-existing multiple
conditions, rather than a single disease. Thus, it is
more challenging, interesting, and clinically relevant.
The project will bring together a multi-team of
providers to identify guidelines of multiple disease
treatment. It will reduce the time pressure of doctors
on unnecessary patient visits, and assists doctors to
manage complex treatments. Chronic disease also
requires pro-active patient participation as well as
fostering a community and culture for healthy living.
Active home and community engagement provides a
supporting environment. Remote sensors can be fun
and offers unique opportunity for health engagement
and communication between providers and patients
for sustained health improvement.
The study attempts to deal with chronic disease
from both clinical as well as home-community
levels. The study will return optimal outcome-driven
treatment for multiple conditions with lower cost and
better control of disease symptoms. The resulting
treatment will also use a minimum amount of drugs,
thus reducing the risk of adverse/side effects and
increasing the efficacy of the treatment (more drugs
mean high risk of non-compliance). This all will
translate to improve the quality of care and quality
of life of patients. Positive and healthy home and
community environments facilitate pro-active patient
health engagement, and promote healthy eating.
Remote sensors offer care continuation (outside
clinic), promote active engagement to sustain broader
health improvement.
Quality and Safety
Analysis of Practice Variance:
Outcome and Evidence-Driven Clinical Practice Re-Design
Numerous studies have shown that surgical
outcomes differ among hospitals. Why do some
sites achieve better outcomes? This is a complex
question with many contributing elements. A large
factor is the variability in patient characteristics
and risk factors. With regard to non-patient factors,
it is likely that outcomes are affected by a host of
factors broadly related to experience, resources, and
experimentation. For example, some centers may
commit greater resources to certain procedures.
Other centers may encourage experimentation,
resulting in adoption of changes in surgical and
medical care that appear promising and divergence
in management practices from those at other
institutions. Practice variance is an important issue to
analyze as a means to optimize care delivery (quality
and efficiency) and to encourage collaborative
learning for broad quality improvement.
How this is different
than related research
Collaboration has the added potential of stimulating
new ideas for investigation or new management
techniques, and increases our ability to conduct
prospective research in a highly specialized
clinical setting. Experimentation and discussion
among colleagues can lead to the rapid adoption
of innovations and avoid the replication of
disadvantageous techniques. Collaboration and site
visits have not yet been applied to pediatric cardiac
surgery. Collaborative learning in pediatric cardiac
surgery requires a multi-institutional approach due
to relatively low volumes. A national structure for
collaborative site visits has never been tried, to our
knowledge, in any field.
• Improve quality and efficiency of care
• Successful dissemination of best practice
• Reduce length-of-stay through early extubation
and improving care coordination and management
• Establishment of important CPG for broad
national dissemination.
Quality and Safety
Hospital Acquired Conditions—
Systematic Analysis and Adaptive Approach
A Hospital Acquired Condition (HAC) is a medical
condition or complication that a patient develops
during a hospital stay, which was not present at
admission. About one in 25 U.S. patients has at
least one infection contracted during the course
of their hospital care, according to a 2104 study
released by the U.S. Centers for Disease Control
and Prevention (CDC), resulting in about 75,000
patients with healthcare-associated infections (HAI)
died during their hospitalizations. Hospitals have
worked to mitigate HAC as unnecessary resources are
tied up, and outcome of patients are compromised.
The progress and urgency have been accelerated
as the Affordable Care Act imposes HAI penalty.
The challenges here are multiple folds, including
suboptimal adherence to current prevention
recommendations; limitations in surveillance
strategies; lack of efficient mechanism for reporting
adverse events; inconsistent metrics of measurement;
and at times, lack of system-wide research. Most
studies are site-specific, e.g., ICU-focus, antibioticsfocus, etc. The interdependencies and multi-faceted
potential personnel and process contribution to
HACs make it difficult to pinpoint sources for early
detection and intervention. Our team has previously
made good SSI advances in open heart surgery
through system advances.
NICU, MRSA, ED, and environmental service, and
multiple stakeholders (care givers and providers,
patients, and facility/cleaning workers). Terminal
cleaning tools and processes will also be observed.
Our study is designed to uncover susceptible areas/
processes/procedures over the entire hospital stay
period where infection/conditions are acquired with
the objective to cultivate a pro-active surveillance
system of awareness of infection-prone situations.
The team will completely immerse in the day-to-day
processes and will map out the multi-faceted interdependencies across processes and systems. Multi-site
comparison will be performed.
• Improve quality of care and treatment outcome
for patients
• Reduce unnecessary length of stay and extra
medical care
• Improve provider and patient compliance
• Improve hospital surveillance
• Improve hospital resource utilization
• Improve providers’ morale and confidence
• Establish a conducive atmosphere for sustainable
process and change transformation where HAC
awareness is integral and second nature to service
How this is different
than related research
This large-scale system-wide study involves multiple
hospitals, units, and services, including OR, ICU,
Quality and Safety
Quantifying the Impact of Pay-for-Performance Financial
Incentives to Reduce Healthcare-Associated Infections
Healthcare-associated infections (HAIs) are
infections that patients contract while receiving
treatment for medical or surgical conditions,
which impose a considerable economic burden
on the U.S. healthcare system. According to the
Centers for Disease Control and Prevention (CDC),
approximately 1 out of every 20 hospitalized patients
contract some form of HAI. Further, the estimated
medical costs of HAIs to U.S. hospitals range
from $30-45 billion. As a result, HAIs have greatly
contributed to the escalating costs of hospital care
as well as both morbidity and mortality. Pay-forperformance (P4P) initiatives are increasingly used
to incentivize providers to improve both care quality
and performance. The system-wide implementation
of P4P models may help drive down HAIs for
participating hospitals, but what are the incentives
for hospitals to participate? In this project, we seek
to quantify the economic benefit of participating
hospitals in Highmark’s P4P financial incentive
program in terms of return-on-investment (ROI). We
aim to evaluate the effects of hospital P4P program
participation on existing levels of care quality and
whether there is a decline in the HAI incidence rates
for these participating hospitals.
How this is different
than related research
Previous research on the impact of P4P models have
focused on improved hospital quality, efficiency,
patient care and safety, but a critical gap remains
with measuring the actual ROI associated with
hospital participation in such P4P financial incentive
programs. The objective of our research is to
measure the true economic benefit of these financial
incentives for both Highmark and participating
hospitals, while evaluating the extent to which the
QB program may help reduce HAI incidence rates;
thereby, serving as a motivator for system-wide
Highmark, as a NSF-CHOT partner, has identified
the strategic priority around a better understanding
of financial incentives for HAI. This project
is potentially significant for all NSF-CHOT
hospital partners, and we expect to leverage their
participation in the effort as appropriate.
Quality and Safety
Reinventing the Pediatric Primary Care Model
Pediatric primary care has evolved from a reactive
delivery model to a more coordinated and proactive
model of care over the past 40 years. Today,
advancements in pediatric practice guidelines,
chronic disease prevention, diagnostic and
treatment technologies, and an increasingly engaged
parent population present this field with a unique
opportunity to reinvent itself. Modern pediatric
care networks are now pursuing strategies to engage
patients and parents earlier and more often by using
innovative technologies and approaches. Further,
pediatric primary care networks stand to benefit
from improved integration with obstetrics in order
to create a continuous stream of healthy parents
and children. Finally, pediatric care networks are
becoming increasingly proactive with their high
acuity patients through the use of remote monitoring
technologies and mobile health. In designing these
new care models it is important to make informed
judgments on what is best suited for well-defined
customer segments and existing organization
infrastructure. The purpose of this study is to
identify best practices of innovative pediatric primary
care models (IPCM) and to define operational
and financial details of relevant models for future
How this is different
than related research
As suggested in our initial description, the IPCM is a
product of a new environment of care concerned with
improved access, effectiveness, timeliness, patient/
parent engagement, and efficiency of pediatric care.
IPCM calls for evolving care teams and professional
leadership in the reengineering of work processes
from obstetrics (when the patient first enters the
model) all the way through their transition into
adult-oriented care. Thus, this research is heavily
driven by both theory and practice to more clearly
define IPCMs and their variants across the health
care industry. Results of this research will provide a
model for IPCM and guide CHOT members in IPCM
planning and implementation.
The sponsor as well as other associations,
hospitals, and policy makers will benefit from
a clear understanding of the nature, evolution,
design components and role of the IPCM in the
healthcare industry. Specific attention will be given
to characteristics of the IPCM and its contributions
to improved, patient/parent engagement, and access
to coordinated care models of practice for IPCM
Quality and Safety
Using Lean Six Sigma to Reduce Hospital Acquired Conditions
In fiscal year 2015, CMS will implement the HospitalAcquired Condition (HAC) Reduction Program.
This program mandates that hospitals in the lowest
quartile for hospital-acquired infections (conditions
that patients did not have when they were admitted to
the hospital) or the lowest quartile for medical errors,
will receive a 1% penalty on reimbursement, meaning
they will only be paid 99% of what otherwise would
be paid under inpatient prospective payment system
(IPPS). With the average American hospital earning
approximately 5% margin, a loss of 1% revenue has
the potential to be a significantly negative effect on
the financial viability of some hospitals. Further,
hospital-acquired conditions are largely preventable
and thus programs that serve to reduce HACs are an
important facet of optimal patient care.
How this is different
than related research
Limited research has examined retained surgical
items using process improvement methodologies.
Utilizing a rapid improvement event to test the
emergent themes will serve as a unique validation of
our findings. Finally, by using both clinical (clinical
documentation) and non-clinical (coding process)
workflow processes to examine the data and identify
process breakdown, this project will serve to optimize
the surgical process with respect to preventing
retained surgical items.
The results of this research will assist all hospitals in
better utilization of Lean Six Sigma methodologies
to examine deficient hospitals processes that
result in HACs. Further, by incorporating a rapid
improvement event, the project will offer hospitals an
important “next–step” in utilizing the study results
to improve patient care thereby fostering greater
utilization of this research.
Enabling HIT and Care Coordination Cluster
A Combined Human-Factors and Quality Improvement Approach
to Assess Health Information Technology Usability
Electronic Health Records (EHR) play a major role
in the efficiency of clinical operations. Although
the main objective of EHR is to provide support
on clinical activities, several studies have reported
that usability issues have caused inefficiencies and
dissatisfaction of clinicians. As a result, EHR systems
have suffered from lack of acceptance and adoption.
The American Recovery and Reinvestment Act
(ARRA) of 2009 put the “meaningful use” of EHR
as a central priority for the Centers of Medicare &
Medicaid Services (CMS) with the main objective of
effective use of EHRs to achieve health and efficiency.
As a way to support this priority, a three-phase EHR
incentive program was developed to implement EHRs
in a meaningful way to improve quality and safety
of the U.S. healthcare systems. After 2015, financial
penalties will be imposed on Medicaid eligible
professionals that do not meet all the criteria for
meaningful use.
on efficiency and satisfaction of clinical users. In
addition, most of those studies discuss the EHR
usability problem without providing details on
how they could be addressed and quantified. Our
combined HF-QI framework provides a quantification
of EHR usability at the task level and a more detailed
mapping of usability issues. Therefore, informed
recommendations can be made to improve usability
and as a consequence, improve efficiency in clinical
Identifying and quantifying EHR usability issues
at the task level represent a huge opportunity to
inform EHR interface designers and identify areas
of opportunity for EHR training programs. This
will address the efficiency of clinical operations
and clinician satisfaction. Therefore, it will have a
positive impact not only on people’s health but also in
healthcare costs.
How this is different
than related research
Although the ARRA claimed for a meaningful use
of certified EHR technologies, only a few studies
have investigated the impact of EHR usability issues
Enabling HIT and Care Coordination Cluster
Automated Language Translation for Improving
Care Management
Language barriers pose problems for communication
and interaction among patients and healthcare
providers. Yet, proper communication is critical
for optimal health management and outcomes. To
improve patient-provider communication for patients
with limited English proficiency (LEP), it is necessary
to interpret spoken language and translate written
clinical documents to the patient’s primary language
of communication. There is mounting evidence
that LEP is a risk factor for reduced healthcare
accessibility, reduced quality of care, decreased
patient satisfaction, poor understanding of provider’s
instructions, increased length of hospital stay and
increased adverse events and misdiagnoses. Thus,
limited patient–provider communication due to the
language barrier is a burden to payers, providers and
the community as a whole.
In this study, we plan to address the translation
services and plan to test computer-assisted translation
and machine translation (MT), utilizing freely
available open source tools such as Google Translate,
along with our advanced computing machine
translation services to translate discharge summaries
to various other languages to improve the accuracy of
translations. We will use a combination of carefully
customized user dictionaries/templates, based on
correct terminology and fine tuning of MT tools, to
increase the accuracy of machine translation.
How this is different
than related research
The overall objective of this project is to study the
language interpreter/translation services workflows
and find opportunities where advanced informatics
solutions could provide a robust solution to the
language barriers. Our system is the first attempt to
automation where the resulting machine translator
will continue to learn and improve through multilevel usage.
Reduced time for language translators to edit the
machine translated summaries, reduced time to
translate documents and improved quality of the
discharge process by providing the documents
in the language the patient understands. It will
also enhance the discharge for patients speaking
languages for which there are no translators. This
allows the hospital to set up a community language
bank. Once successful, these language access tools
could be applied in a variety of settings across the
entire healthcare system where language barriers
pose problems and to materials such as health
education and disease related documents, brochures,
health guides and research briefs.
Enabling HIT and Care Coordination Cluster
Designing Health Information Technologies to Help Patient
Care Teams Identify and Manage Information Problems
Patient-care teams frequently encounter information
problems during their clinical decision making
process. These information problems include
wrong, outdated, conflicting, incomplete, or
missing information. Information problems can
negatively impact the patient-care workflow, lead to
misunderstandings about patient information, and
potentially lead to medical errors. Although these
information problems have existed for some time in
paper records, there is an increasing need to focus
on them in electronic records due to the tremendous
growth in the use of health information technologies
(HIT). Consequently, we will investigate the role
that HIT plays in supporting or hindering patient
care team members’ ability to identify and manage
information problems in an inpatient unit of Hershey
Medical Center (HMC). The goals of the project will
be to (1) identify requirements for HIT features to
better support identification and management of
information problems and (2) develop low-fidelity
prototypes of these features and get feedback from
users on their usability/usefulness.
How this is different
than related research
Current medical informatics research focuses
primarily what causes information problems and
the impact that the information problems have on
the workflow of the hospital staff. However, there is
little research that examines how these information
problems are identified and managed by patient care
teams and the role that HIT plays in this process. The
intellectual merit of this work lies in addressing an
issue in the medical informatics field for which there
is currently little research. The broader impact of this
research lies in its ability to potentially improve the
delivery of care and reduce medical errors.
Our work is relevant to all the industry members of
CHOT. Identifying features that can help reduce
information problems can improve the quality of
healthcare delivery in hospitals, decrease the chances
of medical errors occurring, and lead to the better
design of health information technologies.
Enabling HIT and Care Coordination Cluster
Gamification for Self-Monitoring of Patients for
Enhanced Wellness Outcomes
The objective of this project is to investigate the
fundamental aspects of gaming (both traditional
hardcore gaming and casual mobile gaming) that
make them engaging, rewarding and stimulating
and apply those research findings towards a more
immersive healthcare wellness management solution
that can be adopted by patients. The video game
industry has grown to become a ~$100 billion
industry, with the average age of gamers being 30.
The success of mobile games such as angry birds,
candy crush, etc. has extended the definition of a
“gamer” to include a broad range of individuals of
all ages and demographics. The term “gamification”
is an emerging paradigm that aims to employ game
mechanics and game thinking to change behavior.
The current physician-patient relationship is topdown in nature; a physician provides a patient
with a specific set of instructions that they must
comply with and a patient goes home and is left to
manage their wellness until the next hospital visit.
In the context of healthcare, gamification aims to
transform the patient-physician relationship into
a more collaborative experience, where patients
themselves are motivated to succeed in their wellness
management goals.
How this is different
than related research
The goal of our project is to create the “angry
birds/candy crush” of wellness systems, based on
the gamification paradigm that appeals to a broad
range of individuals (that may not have considered
wellness management systems in the past). This
project will focus on maintaining engagement in the
wellness management apps through a theoretical
understanding of how/why the gaming industry is
often successful in maintaining user engagement for
extended periods of time.
Our industry partnership with Verizon has led to an
understanding that for patients, insurance companies
and hospitals, gamification will transform the
manner in which wellness management is designed
and advanced. IT industries can benefit largely the
software platforms developed under this project and a
better understanding of the data acquisition, transfer
and management needs.
Innovations in Healthcare Delivery
Contact Information
Bita Kash, Ph.D., M.B.A., FACHE
Eva K. Lee, Ph.D.
Center DirectorCo-Director
[email protected]
Jim Benneyan, Ph.D.
Harriet Black Nembhard, Ph.D.
[email protected]
Additional information on CHOT research projects from previous years are available to our members at
This material is based upon work supported by the National Science Foundation under Grant No. IIP-1361509.
Any opinions, findings, and conclusions or recommendations expressed in this material are those of the
author(s) and do not necessarily reflect the views of the National Science Foundation.
32501 714