NIAMS Psychosocial and Behavioral Therapies for Musculoskeletal and Rheumatic Disease Outcomes Roundtable Summary

Friday, November 19, 2010

NIH Campus, Building 31, Conference Room 6C10
Bethesda, Maryland

Gayle E. Lester, Ph.D.
James Witter, M.D., Ph.D.
Francis J. Keefe, Ph.D.


The NIAMS organizes four to five roundtables on an annual basis as a part of the Instituteís long-range planning and priority-setting processes. Over the long-term, these discussions help shape the Instituteís thinking about areas of importance in our basic, translational, and clinical portfolios.

The purpose of the November 19, 2010, roundtable was to better understand how research in psychosocial and behavioral medicine could enhance our understanding of the interplay between psychological and behavioral factors and patientsí experience of pain, as applied to the clinical care of musculoskeletal and rheumatic diseases and musculoskeletal disorders and injuries. Psychosocial factors influence how patients experience chronic pain and their health-related quality of life. Differences in psychosocial factors may help explain, for many musculoskeletal and rheumatic conditions, why a patientís overall experience may not directly correlate with clinically observable, physical signs of disease. For example, some osteoarthritis (OA) patients who report having only minor pain and disability have radiographic evidence of severe joint deterioration, whereas others with minor structural joint damage have incapacitating pain. Furthermore, research has documented that cognitive (e.g., beliefs, expectations, coping strategies), affective (e.g., depression, anxiety, stress, heightened concern about illness, anger), behavioral (e.g., avoidance), social (e.g., advantages that patients derive because of their illness), and cultural factors influence how patients who have chronic pain associated with a musculoskeletal or rheumatic disease or injury respond to interventions.

Prior to the roundtable, participants were asked to canvass their respective communities and provide input on the six questions listed below. Discussion expanded on the feedback compiled by participants.

  • How can research help clinicians identify patients suffering from chronic, disease-related pain who are likely to benefit from a psychosocial or behavioral therapy (before, in conjunction with, or after a physical intervention)?
  • What types of research will inform the development of psychosocial screening tools that will allow clinicians to determine which patients with chronic, disease-related pain are likely to respond to a physical intervention (e.g., drugs, surgery)?
  • What research would allow health care providers and their patients to customize a combination of psychosocial, behavioral, and medical strategies for reducing pain and improving function for people who have a musculoskeletal or rheumatic disease?
  • How can the NIAMS facilitate greater interactions and discussions between psychosocial and behavioral researchers, rheumatologists, and orthopaedic surgeons and other musculoskeletal disease clinical investigators with regard to the use of these therapies?
  • Are there other obstacles or barriers to greater implementation of psychosocial and behavioral therapies when treating people who have chronic pain due to rheumatic or musculoskeletal disease?
  • What are the training needs and opportunities for the psychosocial and behavioral research workforce in musculoskeletal and rheumatic conditions?

Personalized Medicine

Several participants suggested that all people who have chronic, painful musculoskeletal or rheumatic disorders would benefit from some form of early psychosocial or behavioral intervention to help them manage their pain or lessen disease-related disability. Therefore, much of the discussion focused on how to determine who would benefit from information that could be administered in a doctorís office or during physical therapy, and who might need more intensive care. For example, some patients immediately adjust their behavior in response to medical advice, while others need considerable help before changing.

Some participants recommended that researchers should compare combinations of therapies for various diseases to determine which particular coping skills should be taught to people with a specific musculoskeletal or rheumatic disease. Others hypothesized that the success of any psychosocial and behavioral intervention was its ability to change patientsí perspectives from passive, reactive, and helpless to being active, resourceful, and empowered; they emphasized that patients need a variety of tools that they can apply as their situations warrant. Some raised concerns about how best to assess the quality of psychosocial therapy protocols, both when reviewing or conducting research and referring patients.

Once psychological or behavioral counseling sessions are implemented, questions exist as to optimal intervals between these sessions for people who are suffering from musculoskeletal or rheumatic diseases. Furthermore, little quantitative data exist regarding when (i.e., how soon after diagnosis) a patient is most likely to benefit from psychosocial treatments. A person who is newly diagnosed as having a rheumatic disease or who recently experienced a traumatic injury may be too overwhelmed to be receptive to cognitive behavioral interventions. Alternatively, waiting to inform a patient about the value of psychological and behavioral therapies may diminish their roles as valid, effective treatments. Moreover, allowing too much time to elapse places patients at risk of developing additional problems due to living with chronic pain.

Although much of the discussion focused on strategies to reduce disability from chronic pain, opportunities exist to study the psychosocial issues that predispose someone with acute, protective pain due to a musculoskeletal injury to transition to a chronically painful state, and to develop strategies to prevent the transition. The orthopaedic community also would benefit from screening tools that predict which chronic pain patients are less likely to benefit from surgery. Participants suggested that the most successful predictive strategies would likely incorporate a combination of psychological and biological screening tools. Because many patients who are considering surgery are amenable to non-surgical interventions that could improve their outcomes or shorten their recoveries, pre-operative interventions that improve the likelihood of a successful procedure could have a positive effect on the health care system.

In addition to influencing the development and progression of chronic musculoskeletal and rheumatic diseases, genetics affect a personís perception of and response to pain due to these conditions. Studies to elucidate these pathways and interactions would need to distinguish between genetic markers of pain sensitivity and pain severity. Furthermore, participants cautioned that some pain phenotypes commonly used in research (e.g., heat or cold sensitivity) do not correlate with disease-associated pain (e.g., pain caused by inflammation or ischemia) and disability.

Research that moves beyond genetic markers and into the field of epigenomics also holds promise for developing new screening tools for people who have pain related to musculoskeletal or rheumatic conditions and may be at risk of developing chronic pain. However, genetic and epigenetic data ultimately will need to be combined with information about whether people who have musculoskeletal or rheumatic conditions respond to particular interventions. Functional imaging studies that demonstrate changes in brain activity in response to interventions could be useful for such research.

Clinical Study Design

Data from a large number of well-characterized musculoskeletal or rheumatic disease patients are essential for the types of studies mentioned above. Participants expressed a desire for researchers who are investigating psychosocial interventions to collect a core set of subjective and quantitative measures that will allow them, once the original project is complete, to pool their data with that from other studies to enable investigators to carry out meta analyses for increased power and mine the data for other purposes.

NIAMS encourages investigators to propose research into musculoskeletal or rheumatic diseases that they can conduct in conjunction with an existing clinical study ( Information about ongoing clinical research funded by the NIH and others is available at

Investigators designing new studies may wish to consider designs other than the randomized controlled trial. Adaptive trial design (a strategy that uses accumulating data to decide how to modify the study as it continues), choice design (where subjects can be randomized to treatment, control, or choice of treatment or control), and randomized withdrawal design (where everyone receives the intervention when joining the study, then are randomized) were mentioned as possible alternatives.

Other discussion focused on research needs related to strategies for measuring physiologic changes associated with psychosocial interventions, beyond or in addition to existing self-reported outcomes such as those collected by the NIH-funded Patient-Reported Outcomes Measurement Information System (PROMIS;

Translating Therapies into Practice

In the 1970s, when physicians viewed pain intensity as corresponding directly to disease severity or joint damage, any pain that a patient experienced in the absence of identifiable pathology was assumed to be "psychological" or "not real." Although the health care and research communities are discarding that model as members recognize that numerous psychosocial factors directly alter a personís physiology, some continue to view psychosocial therapies as secondary to medications or surgery.

Participants hypothesized that the lack of integration of psychological and medical treatments inhibited patientsí and providersí from being receptive to the use of psychosocial and behavioral interventions. Data about the extent to which people who have musculoskeletal or rheumatic diseases receive psychosocial and behavioral interventions and the barriers to their use are essential before researchers can study strategies to increase their application. Some discussants were concerned that rheumatologists and orthopaedic surgeons do not believe psychosocial and behavioral therapies are effective, while others suggested that they recognize the methodsí benefits, but lack tools for referring patients for appropriate care.

Barriers to psychosocial and behavioral therapies are likely to vary with intervention type, patient background, and provider specialty. The orthopaedic community, for example, embraces physical therapy as a behavioral approach that can restore a personís physical function following injury. Doctors often recommend obese patients lose weight before joint replacement surgery. Referring orthopaedic surgeons may have a stronger sense of which patients should receive those interventions than they do regarding which patients should explore psychosocial therapies. Development of a simple assessment that providers could integrate into each health visit (similar to blood pressure monitoring) would provide clinicians with useful information, and such a tool could help to overcome patient resistance to psychosocial therapies. Because physical therapists and nurse practitioners play major roles in coordinating health care, the development and validation of interventions that these allied health professionals could deliver also might make the interventions more accessible to patients.

Musculoskeletal and rheumatic researchers could learn from colleagues in other fields, such as cancer, that have integrated psychosocial interventions into routine care. The musculoskeletal and rheumatic disease communities also could adapt the long-term health care strategies that are used to treat people who have diabetes. Successful diabetes control, like chronic pain management, requires a combination of medication use and behavior change; health care providers regularly reinforce the notion that disease management requires active involvement on the part of patients and their families.

Because successful delivery of information is the lynchpin of all psychosocial and behavioral therapies, the musculoskeletal and rheumatic disease research communities could benefit greatly from information dissemination studies. Participants hypothesized that it would be worthwhile to "re-educate" patients and physicians with regard to pain and disability in a way that empowers patients to manage their pain rather than blaming them for it. In this vein, the NIH Office of Behavioral and Social Sciences Research (OBSSR) supports research on strategies to reduce disease and disability by improving health literacy. The NIH also is interested in leveraging mobile communication devices as health promotion and research tools, and recently partnered with the Foundation for the National Institutes of Health and the mHealth Alliance on the 2010 Mobile Health (mHealth) Summit.

Many issues regarding the integration of psychosocial and behavioral therapies with surgical or pharmacologic interventions to improve musculoskeletal or rheumatic disease outcomes are covered in depth in the Agency for Healthcare Research and Quality (AHRQ) Technical Brief on Multidisciplinary Pain Programs for Chronic Non-Cancer (, which was available for public comment at the time of the roundtable. Participants stated that health care systems would be informed by research demonstrating that psychosocial and behavioral therapies save money (alone or combined with other interventions for musculoskeletal and rheumatic diseases).

Basic Research into Mechanisms

Participants acknowledged that psychological treatments invoke physiological responses, and were curious to learn more about the underlying mechanisms. Certain psychosocial and behavioral therapies might directly alter a personís musculoskeletal physiology if they cause tense muscles to relax or reduce the levels of destructive cytokines that tissues release in response to pain. Elucidating the mechanisms underlying these effects could provide a foundation for development of medications or alternate treatment strategies that could restore function or protect against further damage. Functional imaging studies that demonstrate changes in brain activity could be useful when investigating the combinations of interventions that benefit people who have acute or chronic pain related to a musculoskeletal or rheumatic condition.

Workforce Issues

Recognizing that continuing education is an important part of clinical practice, participants discussed several suggestions for how they or others could disseminate information so health care providers could make evidence-based decisions. Suggestions included grand rounds, educational "webinars," and workshops at professional society meetings. The NIH Pain Consortium and the American College of Rheumatology are attempting to raise health professionalsí awareness of research results that could influence the diagnosis and treatment of people who are in pain.

Meeting Participants

ANG, Dennis C., M.D.
Assistant Professor of Medicine, Division of Rheumatology
Indiana University School of Medicine

CARTER, Robert, M.D.
Deputy Director

GEORGE, Steven Z., Ph.D.
Associate Professor and Assistant Department Chair, Department of Physical Therapy
University of Florida

KATZ, Stephen I., M.D., Ph.D.

KEEFE, Francis J., Ph.D.
Professor, Department of Psychology and Neuroscience
Duke University Medical Center

LESTER, Gayle, Ph.D.
Director, Osteoarthritis Initiative and Diagnostic Imaging of Bones and Joints Program
Division of Musculoskeletal Diseases

LILLER, Tamara K., M.A.
President and Director of Publications
National Fibromyalgia Partnership, Inc.

Fred and Julie Soper Professor and Chair, Department of Health Policy and Management
Johns Hopkins School of Public Health

McGOWAN, Joan, Ph.D.
Director, Division of Musculoskeletal Diseases

MOEN, Laura, Ph.D.
Director, Division of Extramural Research Activities

NAYLOR, Magdalena R., M.D., Ph.D.
Professor, Department of Psychiatry
University of Vermont

NILSEN, Wendy J., Ph.D.
Health Scientist Administrator, Office of Behavioral and Social Sciences Research
National Institutes of Health

PANAGIS, James S., M.D., M.P.H.
Director, Orthopaedics Program
Division of Musculoskeletal Diseases

RIDDLE, Daniel L., Ph.D.
Otto D. Payton Professor, Department of Physical Therapy
Virginia Commonwealth University

RINI, Christine, Ph.D.
Research Assistant Professor, Department of Health Behavior and Health Education
University of North Carolina at Chapel Hill

Director, Division of Skin and Rheumatic Diseases

Professor, Department of General Internal Medicine
University of Texas MD Anderson Cancer Center

TONKINS, William (Phil) Jr., Ph.D.
Director, Rheumatic Diseases Biobehavioral and Biopsychosocial Research Program
Division of Skin and Rheumatic Diseases

TURK, Dennis C., Ph.D.
John and Emma Bonica Professor of Anesthesiology and Pain Research
Department of Anesthesiology and Pain Medicine
University of Washington

WITTER, James, M.D., Ph.D.
Director, Rheumatic Diseases Clinical Program
Division of Skin and Rheumatic Diseases