Adequacy of the VA Persian Gulf Registry and Uniform Case Assessment Protocol

4 - COMMITTEE ON THE EVALUATION OF THE DEPARTMENT OF VETERANS AFFAIRS UNIFORM CASE ASSESSMENT PROTOCOL

5 - FINDINGS AND RECOMMENDATIONS

6 - CONCLUSION

REFERENCES AND SELECTED BIBLIOGRAPHY

Appendix A - Health Consequences of Service During the Persian Gulf War: Initial Findings and Recommendations for Immediate Action

Appendix B - Health Consequences of Service During the Persian Gulf War: Recommendations for Research and Information Systems

Appendix C - Evaluation of the U.S. Department of Defense Persian Gulf Comprehensive Clinical Evaluation Program: Overall Assessment and Recommendations

Appendix - D Adequacy of the Comprehensive Clinical Evaluation Program: Nerve Agents

Appendix E - Adequacy of the Comprehensive Clinical Evaluation Program: A Focused Assessment

Appendix F - VA Phase I Protocol

Appendix G - VA Phase II Protocol

Appendix H - Persian Gulf Registry Code Sheet

Appendix I - VA Persian Gulf Bi-Monthly Report

Appendix J - Examples of Persian Gulf Registry Follow-Up Letters

Appendix K - VA Monitoring Instrument for Appropriate Medical Care

Appendix L - Annual Persian Gulf Conference Agenda

Appendix M - Site Visit Questions

Appendix N - Summary of Responses to Requests for Information

Appendix O - Questions Related to Traumatic Exposure

Appendix P - Patient Satisfaction

NOTICE

NOTICE: The project that is the subject of this report was approved by the Governing Board of the National Research Council, whose members are drawn from the councils of the National Academy of Sciences, the National Academy of Engineering, and the Institute of Medicine. The members of the committee responsible for the report were chosen for their special competences and with regard for appropriate balance.

The Institute of Medicine was chartered in 1970 by the National Academy of Sciences to enlist distinguished members of appropriate professions in the examination of policy matters pertaining to the health of the public. In this, the Institute acts under both the Academy's 1863 congressional charter responsibility to be an adviser to the federal government and its own initiative in identifying issues of medical care, research, and education. Dr. Kenneth I. Shine is the president of the Institute of Medicine.

This study was supported by the Department of Veterans Affairs under Contract Number V101(93)P-1580. The views presented are those of the Institute of Medicine Committee on the Evalution of the Department of Veterans Affairs Uniform Case Assessment Protocol and are not necessarily those of the funding organization.

International Standard Book No. 0-309-06046-X.

Additional copies of this report are available for sale from:

National Academy Press 2101 Constitution Avenue, N.W. Box 285 Washington, DC 20055

Call (800) 624-6242 or (202) 334-3313 (in the Washington metropolitan area), or visit the NAP's on-line bookstore at http://www.nap.edu.

For more information about the Institute of Medicine, visit the IOM home page at http://www2.nas.edu/iom.

Copyright 1998 by the National Academy of Sciences. All rights reserved.

Printed in the United States.

The serpent has been a symbol of long life, healing, and knowledge among almost all cultures and religions since the beginning of recorded history. The serpent adopted as a logotype by the Institute of Medicine is a relief carving from ancient Greece, now held by the Staatliche Museen in Berlin.

COMMITTEE ON THE EVALUATION OF THE
DEPARTMENT OF VETERANS AFFAIRS
UNIFORM CASE ASSESSMENT PROTOCOL

Arthur K. Asbury, M.D.* (Chair), Van Meter Professor of Neurology, Hospital of the University of Pennsylvania, Philadelphia
Dan G. Blazer, II, M.D., Ph.D., M.P.H.,* Dean of Medical Education and J.P. Gibbons Professor of Psychiatry and Community and Family Medicine, Office of the Dean, Duke University Medical Center, Durham, North Carolina
Dedra Stefanie Buchwald, M.D., Associate Professor, Department of Medicine, University of Washington, Harborview Medical Center, Seattle, Washington
Jerrold J. Ellner, M.D., Executive Vice Chair, Department of Medicine, Case Western Reserve University, Cleveland, Ohio
William E. Golden, M.D., Director, Division of General Internal Medicine, University of Arkansas for Medical Sciences, Little Rock, Arkansas
Bonnie Lepper Green, Ph.D., Professor of Psychiatry, Department of Psychiatry, Georgetown University Medical School, Washington, D.C.
Isabel V. Hoverman, M.D., Austin Internal Medicine Associates, L.L.P., Austin, Texas and Clinical Assistant Professor of Medicine, University of Texas Medical Branch, Galveston, Texas
Karl T. Kelsey, M.D., M.O.H., Associate Professor of Occupational Medicine, Harvard School of Public Health, Boston, Massachusetts
David L. Olive, M.D., Department of Obstetrics and Gynecology, Yale University School of Medicine, New Haven, Connecticut
Bernard M. Rosof, M.D., F.A.C.P., Senior Vice President, Medical Affairs, Huntington Hospital, Huntington, New York

Board on Health Promotion and Disease Prevention Liaison
Elena O. Nightingale,* Scholar-in-Residence, Institute of Medicine and Board on Children, Youth and Families, Washington, D.C.

Board on Neuroscience and Behavioral Health Liaison
William E. Bunney, Jr.,* Distinguished Professor and Della Martin Chair of Psychiatry, University of California, Irvine, California

* Member, Institute of Medicine

Staff
Lyla M. Hernandez, Study Director
Donna D. Thompson, Division Assistant
Sanjay S. Baliga, Research Associate
Donna M. Livingston, Project Assistant
Kathleen R. Stratton, Director, Division of Health Promotion and Disease Prevention
Constance M. Pechura, Director, Division of Neuroscience and Behavioral Health

Preface

This report is the third in a series of efforts by the Institute of Medicine to review the protocols used by the Department of Veterans Affairs (VA) and the Department of Defense (DoD) to diagnose the health complaints of Persian Gulf veterans. The first IOM review began in September 1994 and focused on the DoD implementation of the protocol which is referred to as the Comprehensive Clinical Evaluation Program (CCEP). This diagnostic protocol had been jointly developed and simultaneously implemented in mid-1994 by the VA and the DoD. The early initiation of this review provided the opportunity to analyze information on health complaints and programs collected from the very beginning of the implementation process. In January 1996, the committee issued a final report assessing the adequacy of the clinical procedures described in the protocol, and providing detailed recommendations to DoD for refining the original clinical approach.

The second IOM review initiated in July 1996 again focused on the DoD protocol, this time regarding the diagnosis of difficult-to-diagnose and ill-defined conditions, stress and psychiatric disorders, and health complaints that might be related to low-level exposure to nerve agents. Detailed information available from the existing data system was essential for conducting this evaluation on specific diagnoses in these areas. The final report was published in January 1998.

This third review differs from previous efforts in two key respects. First, this committee was charged with a much broader task. In conducting this review of the diagnosis of health problems of Persian Gulf veterans, we were asked to examine the VA health care delivery system, a system which differs markedly from that of DoD in both size and patient population. This analysis of the system included evaluating (1) the adequacy of the protocol as a diagnostic tool for the broad range of medical assessment needs of Persian Gulf veterans, not for specific clinical diagnoses, (2) how well the program was implemented including the process for patients referrals, (3) VA outreach activities, and (4) VA provider education.

The timing of this review is a second critical difference. Because the committee began deliberations in February 1997, almost two and a half years after the initial review of the DoD protocol, we had the opportunity to draw upon and benefit from both the IOM evaluations and additional information not heretofore available. For example, information was available reflecting nearly three years of VA experience in implementing the protocol, we could solicit new information from VA facilities regarding the strengths and challenges of the protocol, and we could draw upon recent research literature on health problems of Persian Gulf veterans and on advances made in developing clinical practice guidelines and pathways aimed at improving efforts to diagnose health complaints. The committee was, therefore, in an excellent position to evaluate the adequacy of the system used by the VA to identify and diagnose the health complaints of Persian Gulf veterans.

Thus, the recommendations of this report are different from those of previous reports, not only because of the much broader scope of the committee's charge, but also because of recent advances in clinical practice guidelines and quality evaluation approaches, and because of new information available to the committee about the strengths and challenges of the ongoing system. The earlier IOM studies provided important recommendations for implementing and improving the DoD and VA protocol. But any serious evaluation of an ongoing diagnostic screening program will lead, ultimately, to new and improved generations of protocols and systems. The timing and resources available to this committee allowed us to develop recommendations toward that end.

Arthur K. Asbury, M.D.
Chair, Committee on the Evaluation of the Department of Veterans Affairs Uniform Case Assessment Protocol

Acknowledgments

The committee wishes to express its appreciation to the many individuals who contributed in various ways to the completion of this project. Those who made time in busy schedules to meet with us during our VA facility site visits include Mack Anderson, Kimberly Arlinghaus, Donald Barnes, William Binkley, Tom Bird, Carol Bodenheimer, Thomas Bowen, Andrea Cohen, Tesfai Gabre-Kidan, Arnold Gorin, Larry Hawkins, Steve Hunt, Raye Hurwitz, Susan Killin, Shirley Laday-Smith, Sum Lee, Linda Lewis, Jeff Lindeman, Leonard Marcella, Miles McFall, Patricia Ordorica, Jaime Ortiz-Toto, Warrenson Payne, Tammy Porter, Ron Ratliff, Alfredo Rohaidy, Arthur Rosenblatt, Richard Schrot, B.J. Searles, Phillip Shenefelt, Richard Silver, Charles Smith, Glenn Smith, Philip While, John Wicher, Ruel Wiley, and Timothy Williams.

We are also grateful to the following individuals with whom the committee met or from whom information was received: Mark A. Brown, Joseph S. Cassells, Timothy Finnegan, Patricia Jones, Stuart Fleishman, Frances M. Murphy, George Poindexter, Joan P. Porter, Matthew Puglisi, and Joseph Violante. The committee also wishes to extend its thanks to the VA health care facilities and the veterans service organizations who responded to our request for information. We hope we have included all those who contributed to this project. Any omission is inadvertent.

This report has been reviewed by individuals chosen for their diverse perspectives and technical expertise, in accordance with procedures approved by the NRC's Report Review Committee. The purpose of this independent review is to provide candid and critical comments that will assist the authors and the IOM in making the published report as sound as possible and to ensure that the report meets institutional standards for objectivity, evidence, and responsiveness to the study charge. The content of the review comments and draft manuscript remain confidential to protect the integrity of the deliberative process. We wish to thank the following individuals for their participation in the review of this report: Mark R. Cullen, M.D., Yale University School of Medicine; Bernard D. Goldstein, M.D., UMDNJ-Robert Wood Johnson Medical School; John E. Helzer, M.D., University of Vermont College of Medicine; Richard T. Johnson, M.D., Johns Hopkins University School of Medicine; Joseph P. Newhouse, Ph.D., Harvard University; Herbert S. Rigberg, M.D., Health Services Advisory Group, Inc.; and M. Donald Whorton, M.D., M.P.H., M. Donald Whorton, M.D., Inc.

While the individuals listed above have provided many constructive comments and suggestions, responsibility for the final content of this report rests solely with the authoring committee and the IOM.

Executive Summary

On August 2, 1990, Iraq invaded the independent nation of Kuwait and the Persian Gulf War began. Within 5 days the United States began to deploy troops, the number of which would ultimately reach almost 700,000, in Operation Desert Shield. Intense air attacks against Iraqi forces began on January 16, 1991 (Operation Desert Storm), and a ground attack was launched on February 24, 1991. Within 4 days Iraqi resistance crumbled. Following the fighting the number of U.S. troops in the area began to decline, and by June 1991 fewer then 50,000 U.S. troops remained.

Most troops returned home and resumed their normal activities. However, a number of those who had been deployed to the Persian Gulf began to report health problems that they believed were connected to their deployment. These problems included the symptoms of fatigue, memory loss, severe headaches, muscle and joint pain, and rashes.

In 1992 the Department of Veterans Affairs (VA) developed a Persian Gulf Registry to assist in addressing questions about the health concerns of Persian Gulf veterans. Queries about exposures, particularly those associated with oil well fires, were included as part of the history taking. With continuing concern about the potential health consequences of service in the Persian Gulf, the Department of Defense (DoD) and VA met in 1994, revised the clinical program, and implemented this revised approach to diagnose veterans' health complaints, called the Comprehensive Clinical Evaluation Program (CCEP) by DoD and the Persian Gulf Registry and Uniform Case Assessment Protocol (UCAP) by VA.

The Persian Gulf Registry is a basic initial examination aimed at diagnosing veterans' health problems. If a veteran's complaint cannot be diagnosed through the Registry examination, he or she is referred to UCAP for specialty consultation and testing.

The President's Advisory Committee, the General Accounting Office, and the Office of Technology Assessment have evaluated these programs and have made recommendations on the basis of their reviews. In addition, the Institute of Medicine (IOM) has conducted assessments of and made recommendations regarding CCEP.

In September 1996 VA asked IOM to evaluate the adequacy of the UCAP and its implementation with specific emphasis on (1) the protocol, (2) its implementation and administration, (3) outreach efforts to inform veterans of available services, and (4) education of providers. A committee of experts met; heard presentations from many groups and individuals including VA, the President's Advisory Committee, the General Accounting Office, the American Legion, and the Disabled American Veterans; conducted site visits of VA health care facilities; and solicited written testimony from VA health care facilities and veterans service organizations.

A great deal of time and a great deal of effort were expended by VA in developing and implementing a diagnostic program for Persian Gulf veterans that could be conducted in all VA facilities, from small rural primary care facilities to large urban tertiary complexes. This effort was begun immediately upon the cessation of hostilities and attempted to build on lessons learned from past program efforts, for example, those directed toward Vietnam veterans' health concerns. The initial program was implemented in 1992 and the revised program was implemented in 1994, veterans were examined, and information about their symptoms and conditions was collected.

The information that has emerged from the diagnostic program, from research studies, and from the veterans themselves has helped indicate where changes and improvements in the Registry and UCAP can be made. Change is part of a natural evolutionary process in developing good screening instruments for diagnosis. This is not to imply that the first efforts were inappropriate but, rather, that time leads to new knowledge, which leads to the ability to improve.

Such is the case with the VA Persian Gulf protocol. Over time, information has been obtained that can be used to help identify areas where change in the protocol and its implementation will be of benefit. This report is intended to assist VA in that effort to improve. As an operational system, the Persian Gulf Registry and UCAP have provided the opportunity for observation, evaluation, and feedback aimed at improvement. That is what the committee has done-observed, evaluated, and reported.

The committee's first recommendations address the process for diagnosing Persian Gulf veterans' health complaints. These are followed by recommenda-tions regarding implementation and administration and the quality of services rendered. Next come recommendations related to outreach efforts and provider education. Broad recommendations appear in boldface type, with the subentries indicating recommendations following from these broader recommendations.

The diagnostic and referral process specified in VA Manual M-10 is laid out as a two-stage protocol. The protocol specifies that if, after an initial history and physical with minimal laboratory testing (Registry; Phase I), a diagnosis is not made, the veteran is referred to the UCAP for specialist consultation and additional testing (Phase II). The division of the diagnostic process into two phases is, however, an artificial designation that does not accurately reflect the way in which medicine is traditionally practiced.

The committee found that the diagnostic process followed in some facilities does not adhere to the written protocol, rather it is more clinically based. For example, it was found that Registry (Phase I) evaluations are supplemented in some facilities by selected consultations and tests from the UCAP (Phase II). Although this approach may be clinically more appropriate and should be encouraged, it is not the process specified by the current protocol. Such deviation introduces the problem of inconsistency in evaluations across facilities as well as variation in data recording and reporting. This failure, in turn, works against achieving one of the purposes for which the system was developed, that is, to identify previously unrecognized major diagnostic entities that could provide an explanation for the symptoms commonly reported in Persian Gulf veterans with unexplained illnesses (VA Manual M-10; see Chapter 3).

The committee believes that the goal of implementing a uniform approach to the diagnosis of Persian Gulf veterans' health problems is admirable and should be encouraged. To accomplish that goal, the committee makes several recommendations.

1. The committee recommends that the diagnostic pathway, illustrated in Figure 1, for the evaluation and referral of Persian Gulf veterans' health problems be adopted and followed by providers in each VA facility.

FIGURE 1. Pathway for diagnosing health problems of Persian Gulf veterans in the VA system.

The major differences between the current written protocol and the pathway recommended by the committee are (1) the primary care provider is encouraged to order additional tests and consultations beyond those specified in Phase I for a patient, based on symptoms and complaints, without the requirement of initiating a Phase II evaluation, (2) patients should be referred to a designated referral center only when the resources necessary to provide appropriate evaluation of presenting complaints are unavailable at the originating facility, and (3) there must be a defined approach to be used for patients who remain undiagnosed or whose major symptoms have not been accounted for (for example, periodic reevaluation, treatment, or referral to a referral center).

Adoption of the described pathway necessitates changes to specific elements of the Registry and UCAP protocols.

1.a. Use of the pathway eliminates the need to designate phases of evaluation; therefore, the distinction between Phase I and Phase II (with all accompanying specifications for specialty examination and referral) should be eliminated.
1.a.(1) The Persian Gulf Registry Code Sheet needs to be redesigned to reflect the elimination of Phase I and Phase II from the protocol.

1.a.(2) The redesign should accommodate the need to aggregate data from the original data collection system with that of the redesigned system.

A minority of patients with persistent symptoms will not receive a definitive diagnosis. Some of these patients could have disease processes that cannot be diagnosed at present because of limitations in scientific understanding and diagnostic testing. They may not benefit from further evaluation now but may receive benefit from reassessment at a later date. This undiagnosed patient cohort, some of whom are designated as having an "unexplained illness," will contain a diversity of individuals who will require monitoring and periodic reassessment.

1.b. VA should plan for and include periodic reevaluations of the clinical needs of these undiagnosed patients.

The pathway specifies an initial evaluation by a primary care provider for both the veteran presenting with complaints and those with no complaints. In traditional medical practice, the comprehensive clinical evaluation of a patient presenting to any physician includes a complete history, physical examination, and laboratory tests. This should be no different for Persian Gulf veterans. The committee believes, however, that VA should consider using an expanded set of tests for the initial laboratory evaluation.

2. The committee recommends that both patients presenting with and those presenting without complaints should receive an initial evaluation which includes (1) a comprehensive history and physical as defined in the American Medical Association publication Physicians' Current Procedural Terminology (1998), (2) a very specific set of questions related to the Gulf War setting, and (3) a standardized laboratory evaluation.

2.a. A national panel of experts should be convened to (1) review the current set of Gulf War-related questions contained on the Persian Gulf Registry Code Sheet to determine whether additions or deletions are needed, (2) identify the set of standardized laboratory tests to be used in the initial evaluation, and (3) conduct periodic reevaluations of the usefulness of each element in the initial evaluation.

Specific Gulf War-related questions to which veterans are asked to respond include the exposure questions contained on the Persian Gulf Registry Code Sheet (see Appendix H). Given the importance placed by veterans and clinicians on the potential contribution of exposures to health complaints of Persian Gulf veterans, the committee decided to examine the exposure questions in detail.

At some facilities the questionnaire portion of this Code Sheet is given to the veteran to complete as a self-report form, whereas at other facilities this is completed during an interview with the provider. The committee believes that many of these questions are appropriate if administered by an interviewer but are not optimal as a self-report. The information collected with this question-naire is not intended to be used for research purposes. Rather, the purpose of the questionnaire is to provide information to the clinician that might be used to assist in the diagnosis of health problems. It is important, therefore, that the patient understand what is being asked of him or her so as to provide the clinician with accurate information.

In addition, the questions related to traumatic experiences may miss important experiences that can affect physical and mental health and about which the physician should know when conducting the patient's evaluation.

2.b. The section on traumatic experiences on the Persian Gulf Registry Code Sheet (Question 19) should be expanded by the addition of (1) specific questions inquiring about experiences not presently assessed that have been reported by Persian Gulf veterans, and (2) an open-ended question(s) that allow(s) the veteran to report idiosyncratic or particularly distressing experiences that may play a role in the veteran's current health status.

2.c. The questionnaire should be administered in an interview format. If the information on environmental exposure, immunizations, and exposure to traumatic situations cannot be collected in an interview format, all yes and don't know responses should be reviewed with the patient in a face-to-face evaluation.

When the Registry and UCAP protocols were developed, they were designed to be very broad and to capture as much information as possible about potential health concerns that could affect Persian Gulf veterans. The information collected through that system as well as the information obtained from other providers and through scientific studies allows efforts to be more appropriately focused on identifying and diagnosing Persian Gulf veterans' health problems. One mechanism that can aid these efforts is the development of clinical practice guidelines which are statements developed for the purpose of assisting the provider and the patient in making decisions about appropriate health care.

3. The committee recommends that VA, to the extent possible, use an evidence-based approach to develop and continuously reevaluate clinical practice guidelines for the most common presenting symptoms and the difficult-to-diagnose, ill-defined, or medically unexplained conditions of Persian Gulf veterans.

Because the Persian Gulf War was the first engagement in which women formed such a large proportion of deployed troops (7%) and because potential exposure of this group of women to stressors, reproductive system toxicants, and other health hazards may produce disorders distinct from those seen in prior conflicts, the committee believes that VA has a unique opportunity to examine the health of women deployed under such conditions. Therefore, there should be increased examination of and attention directed toward women's health issues. The current Registry and UCAP do investigate infertility or subfertility. However, evaluation of miscarriages, stillbirths, and congenital malformations and the evaluation of genitourinary or other hormonally related diseases are limited.

3.a. Clinical practice guidelines for the evaluation and management of women's health issues should be developed.

A major principle in the development and use of clinical practice guidelines is that there must be a mechanism that encourages feedback on the adequacy of the guidelines and their ease of implementation.

3.b. VA should develop a formal mechanism that enables practitioners to provide feedback on the practice guidelines and the diagnostic process used in the VA clinical program for Persian Gulf veterans.

The committee focused its examination of the implementation and administration of the Registry and UCAP on four elements that it believes are of prime importance to the adequate functioning of the program. These four areas are (1) referral for specialty consultation both within and across facilities, (2) quality of services provided, (3) patient satisfaction, and (4) data collection and reporting.

Referrals

The committee found that problems often exist with referrals for specialty consultations. Within facilities, consultant practices are often booked weeks in advance resulting in long delays for specialty services. The referral specialist is frequently unaware that the referred patient is a Persian Gulf veteran, and the specialist may have little experience with the special needs and concerns of this group of patients. In addition, veterans undergoing this tertiary level of evaluation frequently have unrealistic expectations about the process and the outcome of their visit. One approach to addressing this problem is the use of clinical pathways which are clinical management tools that organize, sequence, and specify the timing of the major patient care activities and interventions of the entire interdisciplinary team for a particular diagnosis, procedure, or process.

4. The committee recommends that the process and procedures for referral be modified.

4.a. In those facilities where specialist consultations are provided, certain individuals within each specialty should be designated as the one(s) who will provide the consultative services to Persian Gulf veterans.

4.b. Clinical pathways should be developed to specify the events and processes involved in referrals for specialty consultation.

4.c. In the case of an inpatient evaluation, a site-specific clinical pathway should be used to facilitate the timely and efficient evaluation of patients.

4.d. The diagnostic pathway should specify that a patient be referred to another facility for evaluation only when the resources necessary to provide appropriate evaluation of the patient's presenting complaints are unavailable at the originating facility.

4.e.VA should develop a transfer protocol that specifies procedures for initial contact and scheduling as well as the materials and processes necessary for a transfer, for example, a full copy of the veteran's record to date including all laboratory tests and consultations, the differential diagnosis, and a procedure for the transfer of records from the tertiary institution to the originating provider upon completion of the diagnostic workup.

Quality

There is a great deal of interest in and concern about the quality of care that Persian Gulf veterans are receiving in VA facilities. Although the VA has developed procedures for what it terms the Quality Management/Assessment Monitor (see Appendix K), this information is not adequate to evaluate the quality of care provided to Persian Gulf veterans. The committee believes that, overall, the clinicians involved in the VA Persian Gulf Registry and UCAP examinations are practicing medicine according to acceptable standards but there does not appear to be, across facilities, a systematic approach to documenting the quality of care provided or to identifying areas where improvement is needed.

Traditional quality improvement programs examined the structure within which care is provided, the process for providing care, or the outcomes of care in an attempt to identify the outliers or "bad apples." More recent approaches focus on performance improvement and are aimed at involving practitioners in the use of nonpunitive efforts that result in more effective changes and improvements to the system than was the case with approaches aimed at identifying practitioners with deficiencies.

The development and use of clinical practice guidelines, as recommended earlier, can be an important tool in a program for continuous quality improvement. Additionally, the participation of a multidisciplinary group of providers in the development of such guidelines increases the likelihood that needed changes will be more readily accepted.

5. The committee recommends that VA should establish an evaluation feedback mechanism that includes the elements of a performance improvement system.

Patient Satisfaction

The VA has implemented a well-developed and structured approach for assessing general patient satisfaction with VA care. However, there is no system in place specifically addressing the substantial number of issues and concerns specific and relevant to Persian Gulf veterans, or the special Persian Gulf diagnostic program.

6. The committee recommends that VA design and implement a brief yet comprehensive questionnaire to survey patient satisfaction with the special program for Persian Gulf veterans.

Data Collection and Reporting

An adequate and accurate data collection and reporting system is crucial to the understanding of Persian Gulf veterans' health problems. Such a system is needed to understand the extent of services provided, to contribute to assessments of the quality of services, to examine overall patterns in health status, and to contribute ideas for potential research studies that could be conducted on the health problems of Persian Gulf veterans. These data are not appropriate, however, for the conduct of scientific research.

Data collection forms need to be standardized and easily computerized. Those completing the forms should be doing so in a consistent manner from facility to facility. The committee found variation across sites in the diagnostic categories that practitioners used to identify their patients' health problems; for example, some practitioners use Chronic Fatigue Syndrome as a diagnosis and some do not. Additionally, the methods used to determine primary versus secondary diagnoses also appear to vary. Such variation has implications for the consistency and accuracy of data collection and reporting.

7. The committee recommends that VA facilitate the consistency of data reporting in the following ways.

7.a. There should be agreement nationally, within VA, on the definition and use of specific diagnostic categories.

7.b. Clear decision rules for determining and recording the primary diagnosis should be developed.

The committee also found that there was no opportunity for updating the database information gathered for each patient, even though a patient's condition can change in ways that are important for the analysis of information collected.

8. The committee recommends that there be established a mechanism by which individual patient information can be updated and incorporated in the database in a systematic fashion.

The committee commends VA for the extensive effort it has put forth to inform Persian Gulf veterans of the services available to them. In only two areas has the committee identified needs.

9. The committee recommends that VA develop informational pamphlets for veterans. These pamphlets could be placed in facility treatment areas and could address common concerns such as the purpose and process of the VA Persian Gulf program, health effects of low-level exposure to chemical warfare agents, research activities related to Persian Gulf veterans and their results to date, and so forth.

When first entering a VA health care facility to receive services, all veterans complete an intake form that requests information about their service in the military. This presents an opportunity for the identification of Persian Gulf veterans who may not yet have participated in the special VA program for Persian Gulf veterans but who may wish to do so.

10. The committee recommends that VA consider redesigning intake forms so that the veteran is asked to identify whether or not she or he was deployed to the Persian Gulf War (or any other specific engagement).

VA has designed a number of high-quality programs to educate its designated Persian Gulf providers. These programs would be more effective if they reached a broader audience.

11. The committee recommends that primary care providers, in addition to the Registry practitioners, as well as the specialists who see Persian Gulf veterans, be provided the opportunity and encouraged to participate in the educational programs.

11.a. The audience for whom existing educational activities are developed related to providing health care for Persian Gulf veterans should be expanded to include other providers involved in the evaluation process, for example, designated specialty consultants.

11.b. VA should consider the following options for educating its providers: periodic team conferences (perhaps quarterly) to be held with all designated providers (including specialists) to discuss activities and findings and to provide updates on Persian Gulf issues and concerns, and the development of site-specific clinical pathways by designated specialists and Registry providers.

Although reproductive issues have been addressed in VA educational efforts, other women's health issues have been less thoroughly explored.

11.c. Future educational efforts should place greater emphasis on women's health concerns. 11.d. VA should provide resources to establish a repository for accumulated knowledge of, expertise in, and experience with the health issues and problems of Persian Gulf veterans. Specialists who possess such expertise should be identified and available for consultation by telephone, e-mail or telemedicine connections with local providers in all VA facilities.

TABLE 1. Summary of Recommendations

Topic	Recommendation
Diagnostic Process	1. A national diagnostic pathway for evaluation of Persian Gulf veterans' health problems should be adopted. 1.a. The distinction between Phase I and Phase II should be eliminated. 1.a(1). The Persian Gulf Registry Code Sheet needs to be redesigned to reflect the elimination of Phase I and Phase II. 1.a.(2) Provision should be made for the aggregation of data from both the original and the revised systems. 1.b. There should be a plan for periodic reevaluation of patients without a diagnosis. 2. All patients entering the special VA program for Persian Gulf veterans should receive an expanded initial evaluation. 2.a. A national panel of experts should be convened to determine the specific questions and tests to be included in this expanded evaluation. 2.b. The section on exposures should be expanded to include additional questions about traumatic experiences. 2.c. The questionnaire should be administered in an interview format, but at a minimum, all yes and don't know responses should be reviewed with the patient. 3. Clinical practice guidelines should be developed for the most common presenting symptoms and the difficult-to-diagnose, ill-defined, or medically unexplained conditions. 3.a. Clinical practice guidelines for the investigation of women's health issues should be developed. 3.b. A formal mechanism that enables practitioners to provide feedback on the practice guidelines and their implementation should be developed.
Referral	4. The committee recommends that the process and procedures for referral be modified. 4.a. Within facilities providing specialty consultation, certain individuals within each specialty should be designated to provide consultative services to Persian Gulf veterans. 4.b. Clinical pathways should be developed to specify the events and processes involved in referrals for specialty consultation. 4.c. If an inpatient evaluation is conducted, a site-specific clinical pathway should be used to facilitate timely and efficient evaluation. 4.d. Patients should be referred to another facility for evaluation only when the necessary resources are unavailable at the originating facility. 4.e. A transfer protocol should be developed to facilitate referral to another facility.
Quality	5. VA should establish an evaluation feedback mechanism that includes the elements of a performance improvement system.
Patient Satisfaction	6. VA should design and implement a brief yet comprehensive questionnaire to survey patient satisfaction with the special program for Persian Gulf veterans.
Data	7. VA should facilitate the consistency of data reporting in the following ways: a. There should be agreement nationally, within VA, on the definition and use of specific diagnostic categories. b. Decision rules for determining and recording the primary and the secondary diagnoses should be developed 8. A mechanism to allow updating of individual patient information should be developed.
Outreach	9. VA should develop for veterans informational pamphlets that address their concerns. 10. Intake forms should be redesigned to more easily identify the war or conflict in which veterans served.
Education	11. Primary care providers, in addition to the Registry practitioners, as well as the specialists who see Persian Gulf veterans, should be provided the opportunity and encouraged to participate in the educational programs. 11.a. The audience for whom existing Persian Gulf educational activities are developed should be expanded to include other providers involved in the evaluation process. 11.b. VA should consider as educational activities team conferences that include specialists and should consider development of site-specific clinical pathways. 11.c. Future efforts should place greater emphasis on women's health concerns. 11.d. Resources should be provided to establish a repository for accumulated knowledge, exper-tise, and experience in Persian Gulf health issues and problems.

Change is inevitable and as such, it is important to plan for that change on the basis of new information and techniques that have emerged from past experiences. The committee believes that the changes recommended in this report build on the strengths and lessons learned through research, the implementation of the Registry and UCAP, and advances made in the field of clinical practice evaluation. It is with the intent to assist VA with better serving Persian Gulf veterans as well as facilitating the practice of VA practitioners that these recommendations are offered.

The committee believes that the recommendations contained in this report will clarify areas of confusion and engage VA practitioners in efforts to design practice guidelines and pathways that will result in the rendering of better, more timely diagnostic services to Persian Gulf veterans. The committee urges VA to make the implementation of these recommendations a priority.

1

Introduction

On August 2, 1990, Iraq invaded the independent nation of Kuwait. In rapid response the United Nations passed Resolution 678, which directed that "all necessary means" be used to get Iraq out of Kuwait. Within 5 days the United States had begun to deploy troops to the Persian Gulf in Operation Desert Shield. Intense air attacks against the Iraqi forces began on January 16, 1991 (Operation Desert Storm). A ground attack was launched on February 24, and within 4 days Iraqi resistance crumbled. Almost 700,000 U.S. troops participated in the Persian Gulf War. Following the fighting the number of U.S. troops in the area began to decline rapidly. By June 1991 fewer than 50,000 U.S. troops remained.

The demographic characteristics of the Persian Gulf troops differed from those involved in previous military engagements. Overall, they were older, a large proportion (about 17%) were from National Guard and Reserve units, and almost 7% of the total forces were women.

U.S. casualties during the Persian Gulf War were low. There were 148 combat-related deaths, with an additional 145 deaths due to disease or accidents. Despite the low numbers of U.S. fatalities and injuries, deployed personnel were exposed to a number of stressors. The rapid mobilization for military service led to the sudden disruption of the lives of large numbers of people. The involvement of large numbers of reservists and National Guard personnel created particular concern because in addition to their rapid mobilization and deployment, they would be returning directly to civilian life at the conclusion of the war.

Stressors to which U.S. troops were exposed included oil smoke, diesel and jet fuel, solvents and other petrochemicals, CARC (chemical agent resistant coating) paint, depleted uranium, sand, endemic infections such as leishmaniasis, the extreme heat, and primitive living conditions. In addition, some soldiers were given anthrax and botulinum vaccines and some soldiers ingested Pyridostigmine Bromide pills to protect against chemical warfare agents.

Other stressors affecting the troops were the unfamiliar character of the region, the requirement that U.S. military personnel have virtually no interaction with the indigenous populations, the wait for the fighting to begin, the fear that chemical warfare agents would be used by the Iraqis, and the immense destruction visited on the whole nation of Iraq, including exposure to dead and mutilated Iraqis.

After the war most veterans returned home and resumed their normal activities. Within a relatively short time, however, some active-duty military personnel and veterans began to report various health problems that they believed were connected to their service in the Persian Gulf. Symptoms commonly described included fatigue, memory loss, severe headaches, muscle and joint pain, and rashes (Iowa Persian Gulf Study Group, 1997).

In 1992, the Department of Veterans Affairs (VA) developed and implemented the Persian Gulf Registry. Its original purposes were to ease returning veterans into the VA health care system, to create a registry containing medical and other data on Persian Gulf veterans that would assist in addressing questions about possible future effects of air pollutant exposure and other environmental agents, and to serve as the basis for future medical surveillance. Exposures, particularly those associated with the oil well fires, were included as part of the history taking. As time passed it became apparent that a number of exposure issues and a host of symptoms needed further investigation.

As concern escalated over the health problems of those deployed to the Persian Gulf, the Department of Defense (DoD) also decided to develop and implement a Persian Gulf clinical program. DoD and VA met, used experts to develop clinical protocols, and by 1994 both had implemented similar and parallel clinical evaluation programs. VA's in-depth clinical program is called the Uniform Case Assessment Protocol (UCAP) whereas DoD's is called the Comprehensive Clinical Evaluation Program (CCEP).

By early 1994 over 20,000 veterans had been examined as part of VA's Persian Gulf Registry program. There were concerns, however, about whether those veterans were being appropriately diagnosed and cared for under the VA program. In response to those concerns, in 1994 the Congress passed P.L. 103-446 which stated: "In each year after the implementation of the protocol, the Secretary shall enter into an agreement with the National Academy of Sciences under which agreement appropriate experts shall review the adequacy of the protocol and its implementation by the Department of Veterans Affairs."

In September 1996, VA charged the Institute of Medicine (IOM) with evaluating the adequacy of the UCAP and its implementation. The expert Committee on the Evaluation of the VA Uniform Case Assessment Protocol was convened to review the VA Persian Gulf clinical protocol and data collection, the adequacy of its implementation of the programs, outreach efforts to veterans, and the education of providers.

2

The Institute of Medicine's Persian Gulf Activities

IOM has undertaken several activities focusing on the potential health implications of deployment in the Persian Gulf War and on DoD and VA responses to health concerns. The IOM Medical Follow-up Agency examined the health consequences of service in the Persian Gulf and developed recommendations for research and information systems. The first report of that group (IOM, 1995) recommended that "The VA Persian Gulf Health Registry should be limited and specific to gathering information to determine the types of conditions reported. . . . There should be efforts to implement quality control and standardization of data collected by the registry." The report went on to recommend that improved publicity regarding the availability of the Registry was needed. For a complete list of recommendations, see Appendix A.

The second and final report of the Medical Follow-up Agency (IOM, 1996a) provides 16 recommendations, with accompanying findings, concerning research and information systems needed regarding the health consequences of service during the Persian Gulf War (see Appendix B).

In 1994, DoD asked IOM to assemble a group of medical and public health experts to evaluate the adequacy of the CCEP. That committee concluded that specific changes in the protocol would help increase its diagnostic yield but that, overall, the CCEP was a comprehensive effort to address the clinical needs of those who served in the Persian Gulf War. That report also stated that the CCEP was not appropriate as a research tool but that the results could and should be used to educate Persian Gulf veterans and the physicians caring for them, to improve the medical protocol itself, and to evaluate patient outcomes. (See Appendix C for a complete set of recommendations.)

DoD asked IOM to continue its evaluation of the CCEP giving special attention to three issues: (1) approaches to addressing difficult-to-diagnose individuals and those with ill-defined conditions, (2) the diagnosis and treatment of stress and psychiatric conditions, and (3) the assessment of health problems of those who may have been exposed to low levels of nerve agents. A report addressing the adequacy of the CCEP relative to nerve agents was released in April 1997 (IOM, 1997). It concluded that, overall, the CCEP provides an appropriate screening approach to the diagnosis of neurological diseases and conditions but that certain refinements would enhance the program. (See Appendix D for a complete set of recommendations.)

The issues of medically unexplained conditions and stress and psychiatric disorders were addressed in a separate and final report. That report concluded that information provided over time has led to the ability to focus CCEP evaluation efforts on emerging areas of importance. To that end the committee made several suggestions including the need to emphasize treatment of symptoms, whether or not a diagnosis has been determined; the need to provide increased screening for depression, traumatic exposure, and substance abuse; the importance of conducting an evaluation across facilities to determine consistency in terms of examinations and patterns of referral; and the need for greater communication between DoD and VA, particularly as it relates to the ongoing treatment of patients. A complete set of recommendations appears in Appendix E.

In September 1996, VA asked IOM to convene a panel of experts to fulfill the congressional mandate to review the VA Persian Gulf protocol and its implementation. Specific areas to be addressed included the adequacy of (1) the protocol to address the wide range of medical assessment needs of Persian Gulf veterans, (2) the implementation of the protocol and administration of the program, (3) outreach efforts to inform veterans of available services, and (4) education of providers.

Over the course of the project the committee heard presentations and reviewed written material provided by representatives of VA, the Presidential Advisory Committee, the General Accounting Office, the American Legion, and Disabled American Veterans. In addition, the committee reviewed the VA manual and protocol for the Persian Gulf Registry and UCAP; received reports of the health and research activities being undertaken in Britain and Canada regarding their Persian Gulf War veterans; examined the relevant scientific literature; solicited written input from national veterans organizations and all VA health facilities; conducted site visits to three VA facilities; and received staff updates on testimony, research, quality assurance activities, and other information related to the health of Persian Gulf veterans. This report contains the findings and recommendations of that committee.

3

The Department of Veterans Affairs Persian Gulf Registry and Uniform Case Assessment Protocol

This section of the report describes the special VA program designed to diagnose the health complaints of Persian Gulf veterans. The topics covered include the initial examination (the Persian Gulf Registry), the secondary level of examination (the UCAP), the process for referral, the data collection and reporting system, the designated VA providers and their responsibilities, the performance monitoring system, some descriptive statistics of program participants, VA outreach efforts, and VA provider education activities.

In 1992, VA implemented a program aimed at diagnosing the health problems of Persian Gulf veterans. In 1994, a revised VA clinical program for diagnosing the health problems of Persian Gulf veterans was implemented. This revised program is formally divided into two phases, the Registry examination (Phase I) and the UCAP (Phase II). During the Phase I Registry examination, (see Appendix F) a complete medical history is to be taken and is to include a family history; an occupational history; a social history including tobacco, alcohol, and drug use; a civilian history of possible exposures to toxic agents; a psychosocial history; and finally, a review of systems. The examiner is to record the time of onset of symptoms or condition, their intensity, the degree of physical incapacitation, and details of any treatment received. In addition, each veteran is to be given a complete blood count (CBC), an electrolyte glucose (SMA-6, SMA-12) or equivalent blood chemistries and enzyme studies, and a urinalysis.

The manual specifies that, "Individuals who after completing Phase I or Registry evaluations, have a disability and do not have a clearly defined diagnosis which explains their symptoms, must receive the following supplemental baseline laboratory tests and consultations." This level of examination is called Phase II or UCAP.

According to VA Manual M-10, "The concept behind the protocol was to identify previously unrecognized major diagnostic entities which could provide an explanation for the symptoms commonly reported in Persian Gulf veterans with unexplained illnesses." The protocol states that those referred to UCAP "must receive" the tests listed for the Phase II protocol. For example, all patients are to receive a set of supplemental baseline laboratory tests that include a CBC, Sedimentation Rate Erythrocyte Sedimentation Rate, C-Reactive Protein, Rheumatoid Factor, Anti-Nuclear Antibody, Liver Function, Creatine Phosphokinase, Hepatitis Serology, Human Immunodeficiency, VDRL, B-12 and Folate, Thyroid Function Test, Urinalysis, and Tuberculosis skin test Purified Protein Derivative.

In addition, the patient is to receive certain consultations including (1) dental, but only if the participant's annual screening is not done; (2) infectious disease; (3) psychiatry, but only with physician-administered instruments; and (4) neuropsychological testing, but only as indicated by a psychiatry consult. Additional examinations ordered are symptom-specific, that is, tests are specified based on the symptom(s). For example, if a patient complains of a headache, he or she undergoes magnetic resonance imaging and receives a lumbar puncture, which includes glucose, protein, cell count, VDRL, oligoclonal (IgC), myelin basic protein, opening pressure, and neurology. Other tests are required if patients exhibit other symptoms. (See Appendix G for the complete Phase II protocol.)

If, after completing the UCAP investigation of the veteran's symptoms and conditions, a diagnosis is still elusive, the veteran may be referred to one of four Persian Gulf Referral Centers. The centers located in Washington, D.C., Houston, and Los Angeles were established in 1992, and the one in Birmingham, Alabama was established in 1995. The referral centers offer inpatient stays during which the veteran is observed, multidisciplinary consultations and serial physical examinations are conducted, and lengthy occupational and exposure histories are documented. Each center has available providers with clinical and academic expertise in multispecialty areas including pulmonary and infectious diseases, neurology, immunology, neuropsychology, and toxicology. Once veterans have completed their stay at the referral center, information is to be sent to the veteran's home VA medical center which is then to assume responsibility for providing follow-up care.

A Persian Gulf Registry Code Sheet (see Appendix H) has been developed to collect basic information on all veterans seen through the Persian Gulf Registry, UCAP, or Referral Center evaluations. Part I of the code sheet is completed with the cooperation of the veteran and requests basic demographic information, exposure history, the veteran's evaluation of his own health status, and reproductive information. Part II of the Code Sheet is completed by the examining physician and requests information on symptoms, consultations needed, diagnoses, and the disposition of the patient (e.g., examination completed, hospitalized for future tests, or referred for outpatient care).

If UCAP is undertaken because of significant symptoms but no identifiable diagnosis, the remaining three pages of the Code Sheet are completed. The information requested on these pages includes tests performed, consultations completed, diagnoses made, and whether the physician believes that the veteran has an unexplained illness. A copy of each coded form is submitted to the Austin Automation Center, and the information on the form is entered into the computerized database. The original code sheet is to be filed in the patient's medical record. Copies are not returned to the medical center of origin unless corrections are required. The code sheets are then forwarded to the central VA headquarters in Washington, D.C. The information gathered by this process is to be analyzed for detection of patterns of disease.

In addition to the Persian Gulf Registry Code Sheet, a bimonthly report is required from all VA facilities (see Appendix I). This report provides statistical information on the total number of initial and follow-up examinations as well as cumulative totals, pending appointments, next appointment date, and number of missed appointments.

Each VA health facility is to designate a Veterans' Registry Physician (VRP) who is responsible for the clinical management of the Persian Gulf veteran and who serves as the primary health care provider until another one has been assigned. As defined in the VA manual, the major responsibilities of the VRP include (1) explaining the purpose of the physical examination to the veteran, (2) determining the focus of the initial examination and care provided for the symptomatic veterans, and (3) conducting and documenting the physical examination. This physician is also responsible for ensuring that a follow-up letter is mailed to each veteran explaining the results of the examination and laboratory tests (see Appendix J for sample follow-up letter).

In addition, each of the facilities is to have an identified Veterans' Registry Coordinator (VRC) who is responsible for the administrative management of the program including scheduling of appointments, monitoring time frame compliance, reviewing records for accuracy and completeness, and collecting data for the reporting process.

VA has developed an instrument (Appendix K) to be used "to assess and monitor the appropriateness of medical care being provided" through the Registry. Each VA medical center is to use the instrument to review at least 10% of its Registry examinations. The form elicits information on the required components of the Registry examination including the presence or absence of a Persian Gulf Registry Code Sheet in the health record; a "thorough history and physical exam such as completion of an SF88"; a breast examination and a gynecological examination for women; laboratory test results including CBC, blood chemistries, urinalysis, and chest X ray; record of follow-up visit; record of follow-up letter containing examination results and recommendations; and whether or not specific examinations were ordered if the veteran complained of persistent diarrhea, memory loss, shortness of breath, or chronic cough. A summary form for recording the monitoring results is also completed by each facility.

Phase I Persian Gulf Registry examinations have been provided in 170 VA health facilities. Between August 1992 and May 1997, 67,989 veterans had received initial evaluations in these facilities. In September 1995 data began to be collected in such a manner that it became possible to identify whether veterans received Phase II examinations. From September 1995 to May 1997 data were collected on 5,970 patients. Of those patients, 695 received a Phase II examination and 422 patients were referred to one of the four Referral Centers for further evaluation.

Of the 5,970 veterans for whom data were collected on the revised Code Sheet, 90.2% were male and 9.8% were female. In addition, 70.6% were in active-duty units during the Persian Gulf War, 12.7% in Reserve units, and 11.1% were in the Guard; the unit status was unknown for 5.6% of those participating in the program. Most frequent complaints are listed in Table 3.1 while Table 3.2 lists the distribution of diagnoses.

TABLE 3.1 Most Frequent Complaints among 5,970 Patients

Complaint	Number	Percent
Loss of memory and other general symptoms	1,687	28.3
Headache	1,552	26.0
Fatigue	1,492	25.0
Muscle, joint pain	1,423	23.8
Skin rash	1,377	23.1
Sleep disturbances	718	12.0
Shortness of breath	679	11.4
Diarrhea and other gastrointestinal symptoms	657	11.0
Choking sensitivity	365	6.1
Chest pain	275	4.6
Abdominal	263	4.4
Other symptoms involving skin and integumentary tissue	244	4.2
Cough	222	3.7

TABLE 3.2 Distribution of Diagnoses among 5,970 Patients Diagnosis Number

Diagnosis	Number	Percent
Musculoskeletal and connective tissue	2,111	35.4
Mental disorders	1,909	32.0
Loss of memory and other general symptoms	1,687	28.3
No medical diagnosis	1,252	20.9
Skin and subcutaneous tissue	1,156	19.4
Respiratory system	1,120	18.8
Nervous system	1,022	17.1
Digestive system	971	16.3
Injury and poisoning	666	11.1
Circulatory system	613	10.3
Infectious diseases	538	9.0
Genitourinary system	333	5.6
Neoplasm	45	0.8

Information obtained from the VA on the 5,970 patients seen through May 1997 indicates that 80.1% of patients presenting with symptoms and examined through the Registry and UCAP received a diagnosis whereas 19.9% of patients with symptoms received no diagnosis. Of those who did not receive a diagnosis, 11.6% were classified as having an unexplained illness. An unexplained illness is defined by VA as "a single or multiple signs or symptoms which do not fulfill the definition of any conventional medical diagnosis but have a significant impact on the veteran's quality of life or functional status." It is possible, therefore, for an individual veteran to have symptoms but no diagnosis and not fulfill the criteria for unexplained illness.

The Environmental Agents Service is the focal point within VA for clinical issues related to Persian Gulf veterans. It is from this office that many outreach efforts are organized and implemented. Extensive efforts have been expended to inform Persian Gulf veterans of the services available through the VA health care system. VA maintains a toll-free number (1-800-PGW-VETS or 1-800-749-8387) for persons with questions about Persian Gulf health issues, research findings, and applying for VA disability compensation. Veterans can request information over this automated helpline 24 hours a day, 7 days a week, or speak with a personal representative between 7:30 a.m. and 8:30 p.m. (central time). Helpline representatives encourage veterans to obtain a Registry examination. A VA Persian Gulf War veterans' illnesses home page is also available on the world wide web (http://www.va.gov/gulf.htm).

The primary vehicle for disseminating updated Persian Gulf War-related information to veterans is the Persian Gulf Review newsletter. This publication is produced three to four times annually and is sent directly to all Registry participants as well as to veterans who have asked to be placed on the mailing list. Approximately 180,000 veterans receive the publication, as do all VA medical centers, VA regional offices, and veterans' centers. To date, 14 issues have been published, the most recent being in September 1997. Other items produced for mass distribution include a brochure entitled "Persian Gulf Veterans' Illnesses: Questions and Answers" and a two-page fact sheet entitled "A Report to Veterans-Department of Veterans Affairs-Persian Gulf Research." A few facilities have developed their own brochures describing their programs for Persian Gulf veterans.

VA also offers an on-line bulletin board with updated information on Persian Gulf issues. On that bulletin board one may access the most recent issues of Persian Gulf Review, the most recent developments in Persian Gulf concerns, and facts about VA benefits and medical care for Persian Gulf War veterans.

Extensive educational materials aimed at VA health care providers have been developed. The Environmental Agents Service periodically mails material related to Persian Gulf veterans' concerns directly to the Registry physicians and Registry coordinators in each VA health care facility.

Medical center staff also receive a substantial amount of information via interactive satellite teleconferences, interactive quarterly national telephone conference calls, and annual educational conferences. The most recent national conference on health consequences of service in the Persian Gulf War was held in early June 1997. This program covered a range of subjects related to Persian Gulf concerns (see Appendix L for a detailed agenda). In addition, a manual filled with reference material was distributed to each participant.

4

Committee on the Evaluation of the Department of Veterans Affairs Uniform Case Assessment Protocol

IOM was charged by VA with evaluating the adequacy of the UCAP, its implementation, VA efforts at outreach to veterans, and education of providers. IOM assembled a committee with expertise in neurology, neurobiology, internal medicine, infectious disease, gastroenterology, psychology, psychiatry, occupational and environmental medicine, obstetrics and gynecology, chronic fatigue syndrome, trauma, and clinical assessment. The full committee met three times, heard numerous presentations, and reviewed published material. In addition, the committee conducted site visits of VA health facilities to obtain firsthand information about the system and its implementation. The committee also solicited written testimony from VA health care providers and veterans service organizations on their views of the adequacy of the protocol and its implementation.

The committee determined that it was necessary to visit a range of VA health care facilities to understand how the providers at the local level are using the Registry and UCAP protocols, and implementing the program for Persian Gulf veterans. It must be emphasized that these visits were in no way intended to be a comprehensive survey of how the system was working. Rather, it was the intention of the committee to collect information that would provide a general picture of how individual VA health care facilities have organized to implement the protocols, and how the program was viewed by front-line providers in terms of its strengths and weaknesses.

The committee chose the sites to be visited primarily on the basis of geographic region and the total number of examinations of Persian Gulf veterans conducted. In addition, the committee decided that one of the facilities should be a referral center. With these general criteria in mind, the committee elected to visit (1) the VA Puget Sound Health Care System in Seattle, Washington; (2) the James A. Haley VA Medical Center in Tampa, Florida; and (3) the Houston Veterans Affairs Medical Center in Houston, Texas (a referral center).

Before the visits the committee developed a question guide for members to use during their visits. This guide was not meant to be an interview protocol but, rather, to serve as a reminder of the topics that committee members wished to cover during their visits to the facilities (see Appendix M for a list of the questions). Meetings were scheduled with (1) designated Persian Gulf Registry providers including physicians, nurse practitioners, and administrative staff; (2) specialists commonly called upon to see Persian Gulf veterans, for example, neurologists, psychiatrists, and dermatologists; (3) primary care providers (physicians, nurse practitioners, and nurses) who were not part of the Persian Gulf provider team but who would likely be involved in providing primary care to Persian Gulf veterans; (4) outreach personnel who were the first point of contact for the veterans and who conduct the local outreach activities; and (5) Persian Gulf veterans who were patients at the facility.

The committee divided into three subgroups to conduct the visits. Each subgroup consisted of at least two committee members plus one IOM staff person. All visits were conducted during April 1997 and each visit took two days. It was readily apparent that the Persian Gulf Registry and UCAP were implemented differently at each facility.

One facility had formed a Persian Gulf Clinic in which all Persian Gulf veterans receive the initial Registry and UCAP examinations as well as their follow-up care from a single physician. This physician integrates all of the medical and laboratory work conducted during both Phase I and Phase II examinations. Only when the provider/patient relationship is well established is primary care provided through a different clinic.

The second facility is organized so that Registry examinations are provided on certain days of the week by one of two designated physicians. The examinations are scheduled by a coordinator who arranges for the routine laboratory and X-ray studies to be carried out before the appointment. If the patient's first contact with the facility is through the Registry, the follow-up care is typically provided by one of the Registry physicians. It is not unusual, however, to assign patients to other primary care clinics for follow-up once the health Registry and UCAP examinations are completed.

In the third facility, all Phase I Registry examinations are conducted by a physician's assistant and are coordinated by the Registry physician. A social worker follows all Registry patients in an attempt to ensure that scheduled specialty consultations are completed and reported. Follow-up care is provided in the primary care clinics by other than the designated Persian Gulf providers. Referral center patients are generally cared for on an inpatient basis, with services coordinated by a designated physician. Follow-up care is provided at the home VA facility.

Letters were mailed to 167 VA health care facilities and to 37 veterans service organizations asking them to provide comments, based on their own experiences, about the adequacy of the VA Persian Gulf Registry and UCAP. Specifically, each was asked to submit written information that described its experience with the VA clinical program as it related to (1) the adequacy of the clinical protocol to address the wide range of medical assessment needs of Persian Gulf veterans, (2) how well the protocol had been implemented and administered by VA, and (3) the adequacy of education for providers and veterans about the program.

Responses were received from 89 VA health care facilities. About one-third of those responding stated that the protocol was adequate, that it had been implemented and administered well, and that education for providers and veterans had been sufficient. The remainder of the responses provided more specific information on both the strengths and the challenges of the program. For a more complete summary of responses, see Appendix N.

High praise was given for the VA Environmental Agents Service educational efforts aimed at providers, including the annual national Persian Gulf Conference, video conferences, teleconferences, and newsletter updates, although many believed that the potential audience was much greater than the number of people who were currently participating.

Respondents reported difficulties regarding communication with and scheduling of specialty consultations and referral center visits, problems in completing the Registry Code Sheet, and reported the need for more information that could be distributed to the Persian Gulf veterans.

Overall, the respondents indicated they believed that the program was working fairly well in their own facilities, that minor adjustments were needed, and that they were committed to meeting the needs of their Persian Gulf veteran populations.

Five veterans' service organizations responded to the request for information. There were comments on a broad range of issues, some of which are beyond the scope of this IOM committee's charge, for example, funding levels for VA health care.

Regarding the Persian Gulf Registry and UCAP, VA was praised for its extensive efforts to inform Persian Gulf veterans of the existence of this program. Respondents also indicated that most Registry providers appear dedi-cated, concerned, and familiar with the program and Persian Gulf issues, in general.

Some expressed concern however, that although the VA protocol appeared appropriate for diagnosing common ailments, it did not appear effective for veterans with difficult-to-diagnose or ill-defined conditions. Although referral centers were established to aid in this area, only a small portion of those with undiagnosed illnesses are referred to such centers. Other respondents believed that there was a failure to pursue testing for possible causes of problems among the Persian Gulf veterans such as exposure to depleted uranium. It was suggested that chronic health problems may have led to poor mental health or depression that has gone unrecognized and not tested for in the VA protocol.

Respondents indicated that a major concern was that once patients receive a diagnosis and are referred for care, there is no assurance that follow-up care is received or is adequate.

5

Findings and Recommendations

VA has a large health care system, and as such, it not only reaps the benefits of but is also subject to the strains and limitations of any large organization. VA facilities run the gamut from the small rural primary care facility to the large urban tertiary care complex. It was for this system that an effort was undertaken to develop and implement a clinical diagnostic program that would identify the health problems of those who served in the Persian Gulf War and that could be implemented in each facility, regardless of its size. That effort was begun immediately upon the cessation of hostilities and drew upon lessons learned from past efforts to respond to the needs of Vietnam veterans.

A great deal of time and effort was expended to develop and implement a program that would provide high-quality diagnostic services to those deployed to the Persian Gulf. However, time brings new information and experiences that serve to indicate improvements that can be made. Change is part of a natural evolutionary process in developing good screening instruments and processes. This is not to imply that the first efforts were inappropriate but, rather, that time leads to new knowledge, which leads to the ability to improve.

Such is the case with the VA Persian Gulf protocol. Over time, information has been obtained which can be used to help identify areas where change in the protocol and its implementation will be of benefit. This report is intended to assist VA in that attempt.

The conclusions and recommendations reported here are not meant to serve as an indictment of the efforts of the VA providers who are working to provide high-quality services to the veteran population. Rather, as an operational system, the Persian Gulf Registry and UCAP provide the opportunity for observation, evaluation, and feedback aimed at improvement. That is what the committee has done-observed, evaluated, and reported.

The following section provides findings and recommendations, organized as follows. First, the committee provides its recommendation on the overall process to be used in a program focused on diagnosing Persian Gulf veterans' health problems. This is followed by specific recommendations regarding the elements of the process, its implementation and administration, and the quality of services rendered. Next come recommendations related to outreach efforts to veterans and provider education. Broad recommendations appear in boldface type, with subentries indicating recommendations following from these broader recommendations.

The diagnostic and referral process specified in VA Manual M-10 is laid out as a two-stage protocol. The protocol specifies that if, after an initial history and physical with minimal laboratory testing (Registry; Phase I) the veteran does not receive a diagnosis, he or she is referred to UCAP (Phase II) for specialist consultation and testing. Division of the diagnostic process into two phases is, however, an artificial division which does not accurately reflect the way in which medicine is traditionally practiced.

The committee found that the diagnostic process followed in some facilities does not adhere to the written protocol but is, instead, more clinically driven. That is, evaluations as carried out in some facilities are often tailored to the symptoms and complaints of the patient, not blind adherence to the written protocol. Phase I evaluations in such facilities are frequently supplemented by selected consultations and tests from Phase II (e.g., the Registry provider does not wait until a Phase II designation to order pulmonary function tests if the patient complains of shortness of breath). It also appears that the entire Phase II protocol is not necessarily implemented in some facilities if a diagnosis that accounts for the primary complaints of the veteran can be arrived at without the full workup.

Although it is encouraging that providers are using their clinical judgment to evaluate veterans who present with symptoms, such practices result in confusion about where the patient is in the diagnostic process when compared to the written protocol mandated by VA. The current differentiation between a Phase I and a Phase II evaluation varies from facility to facility, that is, it depends on the facility in which a patient is examined rather than, as the protocol specifies, on whether the patient has received a diagnosis that explains his or her significant symptoms. Such variation introduces the problem of inconsistency in data recording and reporting across facilities, which, in turn, works against achieving one of the purposes for which the system was developed, that is, to identify previously unrecognized diagnostic entities that could provide an explanation for the symptoms commonly reported in Persian Gulf veterans with unexplained illness.

The committee believes that the desire to implement a uniform approach to the diagnosis of health problems in Persian Gulf veterans is admirable and should be encouraged. In order to accomplish this goal, however, the committee believes that several changes need to be made in the system. First, the system needs to be conceived of as a diagnostic pathway or process along which every patient flows. The committee believes that the following pathway (see Figure 5.1) provides the appropriate framework for clinical diagnosis and referral.

1. The committee recommends that the diagnostic pathway, illustrated in Figure 5.1, for the evaluation and referral of Persian Gulf veterans' health be adopted and followed by providers in each VA facility.

Following this pathway, a Persian Gulf veteran entering the system with no complaints would be given the initial evaluation as specified above, and if nothing was found upon examination and testing, that veteran's evaluation would stop. A veteran presenting with the complaint of diarrhea would be given the initial evaluation as specified above with the addition of such tests as stool examination and endoscopy. If the results of the examination and testing were consistent with a diagnosis that explained the complaints (for example, intestinal parasites), that patient's diagnostic evaluation would stop and she or he would be referred for treatment. If, however, the results were not consistent with the complaint or if something unexplained appeared as a result of the tests or the examination, the patient would continue on the diagnostic pathway for additional evaluation and testing. Eventually, if a diagnosis cannot be determined, the provider must decide whether (1) the symptoms or problems are serious enough to cause disruption in the patient's life and therefore warrant continued evaluation at a special center or (2) symptoms and complaints are not causing disruption in the patient's life and therefore the patient should receive periodic reevaluations to determine if his or her condition changes over time.

The major differences between the current written protocol and the pathway recommended by the committee are (1) the primary care provider is encouraged to order additional tests and consultations beyond those specified in Phase I for a patient, based on symptoms and complaints, without the requirement of initiating a Phase II evaluation, (2) patients should be referred to a designated referral center only when the resources necessary to provide appropriate evaluation of presenting complaints are unavailable at the originating facility, and (3) there must be a defined approach to be used for patients who remain undiagnosed or whose major symptoms have not been accounted for (for example, periodic reevaluation, treatment, or referral to a referral center).

FIGURE 5.1. Pathway for diagnosing health problems of Persian Gulf veterans in the VA system.

Adoption of the described pathway necessitates specific changes to the protocols for conducting the Registry and UCAP examinations. A patient's position in the pathway is dependent upon whether the patient has received a diagnosis that accounts for his or her major symptoms. Although that was the apparent intent of the original protocols, the distinction between the Phase I Registry examination and the Phase II UCAP examination resulted in confusion among providers about which phase was being conducted and where information should be entered on the data recording form. If laboratory tests beyond those specified in the Phase I Registry (i.e., CBC, urinalysis, and blood chemistry tests) were ordered, did that mean that the patient was in Phase II? If a consultation with a neurologist was ordered, did that mean that the patient was in Phase II? Those questions are no longer relevant or important in the new diagnostic pathway.

1.a. Use of the pathway eliminates the need to designate phases of evaluation; therefore the distinction between Phase I and Phase II (with all accompanying specifications for specialty examination and referral) should be eliminated.
1.a.(1) The Persian Gulf Registry Code Sheet needs to be redesigned to reflect the elimination of Phase I and Phase II from the protocol.

1.a.(2) The redesign should accommodate the need to aggregate data from the original data collection system with that of the redesigned system.

A minority of patients with persistent symptoms will not receive a definitive diagnosis. Some of these patients could have disease processes that cannot be diagnosed presently because of limitations in scientific understanding and diagnostic testing. They may not benefit from further evaluation now but may receive benefit from reassessment at a later date. This undiagnosed patient cohort, some of whom are designated as having an "unexplained illness," will contain a diversity of individuals who will require monitoring and periodic reassessment.

1.b. VA should plan for and include periodic reevaluations of these undiagnosed patients' clinical needs.

The pathway specifies an initial evaluation by a primary care provider for the Persian Gulf veteran. In traditional medical practice, the comprehensive clinical evaluation of a patient presenting to any physician includes a complete history, physical examination and laboratory tests appropriate to the presenting complaints or clinical problems. This should be no different for Persian Gulf veterans. In addition, some veterans may not be experiencing difficulties but may wish to participate in the program. An initial evaluation with a basic set of elements should be given to both groups of patients, with additional tests performed as necessary based upon presenting complaints for those veterans with clinical problems or complaints.

The results of that initial evaluation will determine whether the patient proceeds further in the diagnostic process, as reflected in the recommended clinical pathway. The committee believes that VA should consider using an expanded set of tests and additional exposure questions for the initial evaluation.

2. The committee recommends that both patients presenting with and those presenting without complaints should receive an initial evaluation which includes (1) a comprehensive history and physical as defined in the American Medical Association publication Physicians' Current Procedural Terminology (1998), (2) a very specific set of questions related to the Gulf War setting, and (3) a standardized laboratory evaluation.

2.a. A national panel of experts should be convened to (1) review the current set of Gulf War-related questions contained on the Persian Gulf Registry Code Sheet to determine whether additions or deletions are needed, (2) identify the set of standardized laboratory tests to be used in the initial evaluation, and (3) conduct periodic reevaluations of the usefulness of each element in the initial evaluation.

Specific Gulf War-related questions to which veterans are asked to respond include the exposure questions contained on the Persian Gulf Registry Code Sheet (see Appendix H). Given the importance placed by veterans and clinicians on the potential contribution of exposures to health complaints of Persian Gulf veterans, the committee decided to examine the exposure questions in detail.

In some facilities the questionnaire portion of this Code Sheet is given to the veteran to complete as a self-report form, whereas in other facilities this is completed during an interview with the provider. The information collected with this questionnaire is not intended to be used for research purposes. Rather the purpose of the questionnaire is to provide information to the clinician that might be used to assist in the diagnosis of health problems. It is important, therefore, that the patient understand what is being asked of him or her so as to provide the clinician with accurate information.

With regard to Question 18 and its subsections, covering smoking and war-related toxic exposures, the committee believes that these questions are appropriate if administered by an interviewer, but not optimal as a self-report. Since this section is frequently handed to the veteran to be filled out as a self-report form, the extent to which responses are checked by an interviewer is not clear.

The instructions to skip over certain questions on items 18A to 18F (the smoking questions) are also confusing. The problems with questions 18G to 18Z have to do primarily with comprehension. Some of the options are not explained, so the veteran may respond negatively even if he or she has been exposed. For example, CARC is spelled out, but no description is provided. Likewise, exposure to depleted uranium and mustard gas may not be acknowledged if the veteran does not know how to tell if he or she might have been exposed. Although "don't know" is a valid option, the committee assumes that the examining physicians would want to know as much as possible about probable exposure. For this reason, the veteran should be able to discuss these possible exposures.

The primary problem with the questions related to traumatic experiences (items 19A to 19F) is that the questionnaire may miss important experiences or stressors that can affect physical and mental health, and about which the physician should know when doing the patient workup. Research has shown that stressors have been associated with major depression, substance abuse, and a variety of physical health problems, including immune system dysfunction.

With men and women serving together in a difficult situation such as war, a unique series of concerns emerges regarding the incidence of physical or sexual harassment or assault or both. Chapter 1 of VA Manual M-10, Part III identifies some special health needs of women veterans of the Persian Gulf including the long-term consequences of rape, other sexual assault, sexual harassment, exposure to combat during military service, or mistreatment as a prisoner of war. However, there is no specific reference to rape or abuse in the protocol history, and there is no routine evaluation of these activities in the absence of "reproductive health problems."

Veterans have indicated that seeing others dead (including Iraqi soldiers) is a very stressful experience. There may have been other frightening experiences that were relatively uncommon but that would be very upsetting and stressful if one were exposed to them.

There needs to be a way for veterans to report traumatic or stressful experiences to the clinician so that these experiences can be taken into account in evaluating a veteran's health complaints. Even less classically traumatic experiences such as harassment may play a role in health outcomes. An open-ended question(s) would also be useful for nonnormative, yet highly stressful, experiences. VA has a number of research centers with trauma specialists who could assist with the specific wording of such questions. Examples of such questions are given in Appendix O, some of which are taken from Southwick et al. (1997).

2.b. The section on traumatic experiences on the Persian Gulf Registry Code Sheet (Question 19) should be expanded by the addition of (1) specific questions inquiring about experiences not presently assessed that have been reported by Persian Gulf veterans, and (2) an open-ended question(s) that allows the veteran to report idiosyncratic or particularly distressing experiences that may play a role in the veteran's current health status.

2.c. The questionnaire should be administered in an interview format. If the information on environmental exposure, immunizations, and exposure to traumatic situations cannot be collected in an interview format, all yes and don't know responses should be reviewed with the patient in a face-to-face evaluation.

In addition to discussion of the exposure questions, the committee focused on examining the list of the consultations and tests required by the UCAP protocol if a veteran presented with specific symptoms. At the time of its development the VA protocol was an appropriate attempt to collect a wide variety of information that covered all known potential health concerns that could affect Persian Gulf veterans. Much has been learned since it was first implemented. It is now known that certain symptoms (e.g., fatigue, memory loss, severe headaches, muscle and joint pain, and rashes) are commonly reported by these veterans. Now that this additional information is available, areas upon which to focus efforts at identifying and diagnosing health problems can be discerned. One mechanism that can aid in these efforts is the develop-ment of clinical practice guidelines.

Great strides in methods for developing clinical practice guidelines have been made in the past few years. Clinical practice guidelines are defined as systematically developed statements that assist the practitioner and the patient in making decisions about appropriate health care for specific clinical circumstances (Field and Lohr, 1990). Once developed, clinical practice guidelines can be used to assist clinical decision making by patients and practitioners, to educate individuals or groups, to assess and assure the quality of care, and to guide the allocation of resources for health care (Field and Lohr, 1992).

There are two major approaches to the development of practice guidelines. The first approach, the evidence-based approach, emphasizes the significance of the science base for guidelines and the use of quantitative modeling for estimating and comparing outcomes. The other approach emphasizes professional judgment in areas in which the science is weak or nonexistent (Field and Lohr, 1992). It may be that for some of the conditions being seen in the VA Persian Gulf program, a melding of experience and judgment with scientific evidence, where it exists, is the best possible approach.

Efforts are already under way for the development of practice guidelines in the VA health care system. In January 1997, VA distributed to all of its facilities clinical guidelines for major depressive disorder, posttraumatic stress disorder, and addictive disorder (Veterans Health Administration, 1997). These guidelines were developed by and for VA clinicians.

3. The committee recommends that VA, to the extent possible, use an evidence-based approach to develop and continuously reevaluate clinical practice guidelines for the most common presenting symptoms and the difficult-to-diagnose, ill-defined, or medically unexplained conditions of Persian Gulf veterans.

These guidelines need to be specific, comprehensive, and flexible enough to be useful in everyday clinical practice. Multidisciplinary groups of those providing care should be involved in the development process.

The Persian Gulf War differed from previous U.S. military engagements in that 7% of those deployed (about 49,000) were women. Potential exposure by this group of women to stressors, reproductive system toxicants, and other health hazards may produce disorders distinct from those seen in prior conflicts or among the men who served in the Gulf. The committee believes that VA has a unique opportunity to examine the health of women deployed under such circumstances. Therefore, there should be increased examination of and attention directed toward women's health issues. The current Registry and UCAP do evaluate infertility or subfertility among males and females, miscarriages, stillbirths, and congenital malformations.

The evaluation of genitourinary or other hormonally related disease is limited, however. Evaluation of this area in the current system is symptom-driven, allowing for errors of omission to be made in the absence of patient awareness. In addition, as discussed above, the special health needs of Persian Gulf veterans related to physical or sexual harassment and assault should be systematically addressed.

3.a. Clinical practice guidelines for the evaluation and management of women's health issues should be developed.

During the site visits and in the VA responses to requests for input, it was noted that no mechanism for providing feedback on the adequacy of the protocol and its use is available. For example, the current protocol states that every patient who is not diagnosed after Phase I and who presents with headache is to undergo magnetic resonance imaging of the head and receive a lumbar puncture. However, lumbar punctures are rarely ordered. Feedback from providers on the usefulness of the tests recommended for the diagnostic process and on the clinical practice guidelines once they are developed would provide important information on what changes, if any, should be made in the evaluation process.

3.b. VA should develop a formal mechanism that enables practitioners to provide feedback on the practice guidelines and the diagnostic process used in the VA clinical program for Persian Gulf veterans.

The committee focused its examination of the implementation and administration of the Registry and UCAP on four elements that it believes are of prime importance to the adequate functioning of the program. These four areas are (1) referral for specialty consultation, both within and across facilities, (2) quality of services provided, (3) patient satisfaction, and (4) data collection and reporting. The following sections discuss the committee's findings and recommendations for these four areas.

According to information received by the committee, referrals for specialty consultation or to a Referral Center present problems. Within facilities, consultant practices are often booked weeks in advance, resulting in long delays for specialty services. Workup and consultant appointments are often at the convenience of the clinic or specialty service, without consultation with the patient about his or her availability. Appointments may be spread over days or weeks, requiring patients to return frequently to the VA facility. For all veterans this can be a lengthy and time-consuming process. For the employed veteran, this can create additional difficulties related to missed work, missed pay, and poor employee evaluation.

In addition, once specialty appointments are scheduled and the veterans are seen, the referral specialist is frequently unaware that the patient is a Persian Gulf veteran or the specialist has little or no experience with the particular needs and concerns of this group of patients. This may create the perception that the care being provided is less than adequate, whether or not such is the case.

It is also the case that veterans undergoing evaluations at a tertiary level often have unrealistic expectations of the process and the outcome of the visit. They are unaware that many of the same questions and tests that they have already undergone will be repeated and that the process is extremely time-consuming. This often results in frustration and anger, exacerbating an already sensitive situation.

One approach to addressing this problem is the use of clinical pathways which are clinical management tools that specify the various activities involved in a project from start to finish and the amount of time anticipated to complete each of the activities (Hoffman, 1993).

Clinical pathways organize, sequence, and delineate the timing of the major patient care activities and interventions of the entire interdisciplinary team for a particular diagnosis, procedure, or process, defining key processes and events in the day-to-day management of care and identifying expected outcomes. They should be developed locally, because they are specific to the particular setting and the team who uses them (Veterans Health Administration, 1997).

4. The committee recommends that the process and procedures for referral be modified.

4.a. In those facilities where specialist consultations are provided, certain individuals within each specialty should be designated as the one(s) who will provide the consultative services to Persian Gulf veterans.

Designated specialists should receive initial orientation concerning Persian Gulf War-related medical issues in their area of expertise.

4.b. Clinical pathways should be developed to specify the events and processes involved in referrals for specialty consultation.

Facilitation of the workup for Persian Gulf veterans is an important consideration. The evaluation should be done in a manner that is as timely, efficient, and convenient as possible. If travel, distance, time off from work, or other obstacles make completion of the evaluation as an outpatient too onerous for a given veteran, then preplanned, rapid inpatient evaluation may be considered.

4.c. In the case of an inpatient evaluation, a site-specific clinical pathway should be used to facilitate the timely and efficient evaluation of patients.

Referral to another facility presents additional problems, for both the veteran and the practitioners. For the veteran, referral to another center for specialty services often creates problems related to travel, time off from work, and family considerations. For the providers, there is frequently difficulty related to initial contact with the referral facility, unreturned calls, and an inability to obtain copies of reports. There may also be a lack of communication between the originating facility and the referral facility regarding proposed treatment and follow-up. In addition, there appears to be a great deal of inconsistency from facility to facility in terms of when it is deemed appropriate to refer a veteran for this tertiary level of evaluation.

4.d. The diagnostic pathway should specify that a patient be referred to another facility for evaluation only when the resources necessary to provide appropriate evaluation of the patient's presenting complaints are unavailable at the originating facility.

4.e. VA should develop a transfer protocol that specifies procedures for initial contact and scheduling as well as the materials and processes necessary for a transfer, for example, a full copy of the veteran's record to date including all laboratory tests and consultations, the differential diagnosis, and a procedure for the transfer of records from the tertiary institution to the originating provider upon completion of the diagnostic workup.

There is a great deal of interest in learning more about the quality of care that Persian Gulf veterans are receiving in VA facilities. Although the committee believes that, overall, the clinicians involved in the VA Persian Gulf Registry and UCAP examinations are practicing medicine according to acceptable standards, there does not appear to be across facilities a systematic approach to documenting the quality of care provided to Persian Gulf veterans or to identifying areas where improvement is needed.

VA has developed procedures for what it terms the Quality Management/Assessment Monitor (see Appendix K). According to the manual, this form (VA Form 10-9009C-1) is to be used by VA medical centers "to assess and monitor the appropriateness of medical care being provided" through the Persian Gulf program. The form is to be used to abstract information from the charts of at least a 10% sample of all Persian Gulf Registry physical examinations conducted at each facility. Unfortunately, the form only collects such information as whether an examination was done or ordered, whether laboratory results were obtained, and whether follow-up letters are in the record or were mailed to the veteran. This information does not represent an adequate assessment of the quality of care provided to Persian Gulf veterans.

In 1990, IOM defined quality of care as "the degree to which health services for individuals and populations increase the likelihood of desired health outcomes and are consistent with current professional knowledge." (Lohr, 1990).

Traditional quality assurance programs examined the structure within which care is provided, the process for providing care, or the outcomes of care in an attempt to identify the outliers or "bad apples." Current approaches focus on performance improvement and are based on a set of principles for implementing change. Their aim is to involve practitioners in the use of nonpunitive tactics for quality assurance that result in more effective changes and improvements to the system than was the case with traditional approaches that were aimed at identifying the practitioners with deficiencies.

The principles for performance improvement models emphasize the relationship between the partners in the health care transaction, the fact that errors are more often the result of defects in the system rather than individual deficiencies, the use of statistical and scientific precepts and techniques, reliance on self-improvement as opposed to external regulation, standardized processes, the provision of feedback to practitioners on their pattern of practice as compared to that of their peers, a visible commitment to quality by top leadership, and a striving for continuous improvement as opposed simply to the achievement of preset goals (Field and Lohr, 1992).

The development and use of clinical practice guidelines, as recommended earlier in this report, can be an important tool in a program for continuous quality improvement. Such guidelines can clearly define what is appropriate care, what are acceptable outcomes, and the contributions of practitioners and patients to those outcomes. These guidelines can be used to help structure the medical review criteria used to collect data on how the system is operating. Additionally, the participation of a multidisciplinary group of providers in the development of clinical practice guidelines increases the likelihood that needed changes will be more readily accepted.

5. The committee recommends that VA should establish an evaluation feedback mechanism that includes the elements of a performance improvement system.

The VA health care system, as is the case with any health care system that diagnoses and manages medical problems, must attend not only to objective outcomes (e.g., morbidity and mortality) but also to more subjective ones such as patient satisfaction. In fact, the measurement of patient satisfaction, as pioneered by the Medical Outcomes Study (Rubin et al., 1993), has become one of the most widely used and important outcomes both in the clinical arena (e.g., in assessing the success of interventions and other aspects of health care delivery) and in the health services research arena. The many questions that have arisen regarding the health consequences of the Persian Gulf War experience, coupled with widespread publicity on a myriad of potential adverse medical outcomes, make patient satisfaction a critical component of any evaluation of the adequacy of the UCAP.

VA has implemented a well-developed and structured approach for assessing general patient satisfaction with the care provided at VA facilities. However, no system specifically addressing the substantial numbers of issues and concerns specific and relevant to Persian Gulf veterans or the special Persian Gulf diagnostic program is in place.

6. The committee recommends that VA design and implement a brief yet comprehensive questionnaire to survey patient satisfaction with the special program for Persian Gulf veterans.

An example of the kind of instrument that might be developed can be found in Appendix P.

An adequate and accurate data collection and reporting system is crucial to the understanding of Persian Gulf veterans' health problems. One explicitly stated purpose of the VA Persian Gulf Registry and UCAP is the creation of a registry that contains medical and other data on Persian Gulf veterans. This registry would assist in identifying previously unrecognized major diagnostic entities that could provide an explanation for the symptoms commonly reported in Persian Gulf veterans with unexplained illnesses (Manual M-10, Part III, Chapter 3, page 3-1). Toward that end, VA has established a database that is meant to be comparable and retrievable across VA health care facilities. These data are not, however, intended to be used for scientific research, nor are they adequate for the conduct of scientific research.

Because VA has only partially automated inpatient record-keeping, current on-line retrieval of data is not possible. However, VA has plans for construction of a computerized database for outpatient records. This should facilitate the review of any data collected through the Persian Gulf program.

Data collection and reporting should, in general, be completed by using standardized and field-tested instruments. The data should be recorded on forms that are easily computerized. Individuals completing the reporting forms should be trained in how to do so to ensure standardization of the quality of data collected across sites. Any data entry should be done in a systematic fashion, with routine error checking involved, and double entry should be used whenever possible. Error rates should be checked and reported. Frequent data quality meetings across VA sites would be advantageous for ensuring standardization.

Finally, the goals of all data collection and data reporting must be explicit and must be the guiding principle behind their development. That is, each data collection system and each system of reporting must be completed with specific goals in mind. This will provide a means for the evaluation of the success and of the strengths and weaknesses of this approach.

The Persian Gulf Registry and UCAP have been established by VA with the intent of implementing a uniform and consistent evaluation across facilities. Data are available; that is, medical records of Gulf War veterans are complete and have been recorded. However, with respect to the application of the medical protocols, standardization and reporting are problematic.

As discussed earlier, current evaluation of the patient tends, in many facilities, to proceed along medically indicated lines rather than being strictly protocol-driven. Thus, it is clear that there is little consistency with respect to designation of phase of evaluation. Such variation hampers evaluation efforts based on these reported data.

Until the national diagnostic pathway has been implemented, Phase I and Phase II eliminated, and the Code Sheet revised to reflect such changes, it is anticipated that this variation will continue.

The committee found that there was variation across sites in the diagnostic categories that practitioners used to categorize patients' health problems. Providers at some facilities used such diagnostic categories as chronic fatigue syndrome and fibromyalgia, while providers at other VA health facilities did not. Such variation in the use of diagnostic categories has implications for the consistency and accuracy of data collection and reporting.

Further, the method used to determine primary versus secondary diagnoses varies. While a primary diagnosis is to be recorded on the code sheet, no instructions are given regarding how to determine which diagnosis is primary. The committee found that some physicians list a medical diagnosis as primary over a psychiatric diagnosis. Other physicians do not follow this custom. Such variation introduces uncertainty about the consistency of the data.

7. The committee recommends that VA facilitate the consistency of data reporting in the following ways.

7.a. There should be agreement nationally, within VA, on the definition and use of specific diagnostic categories.

7.b. Clear decision rules for determining and recording the primary and the secondary diagnoses should be developed.

The committee also found that there was no opportunity for updating the database information gathered for each patient. That is, intake questionnaires and data gathered at the presentation of the patient cannot be updated later in any systematic fashion. This is true for both patients who receive a diagnosis and those who remain symptomatic but undiagnosed. A patient's health status can and does change over time, and the system must have some way of capturing that new information.

8. The committee recommends that there be established a mechanism by which individual patient information can be updated and incorporated in the database in a systematic fashion.

This includes revision of original diagnoses and revision of status including data that are related to life style or demographics.

For Persian Gulf veterans to participate in the special program that VA has designed to address their health concerns, they must be aware that the program exists and know how to register for it. The committee commends VA for the extensive outreach efforts put forth to inform Persian Gulf veterans of the services available to them. As described earlier, the toll-free number (1-800-PGW-VETS or 1-800-749-8387) for those with questions about Persian Gulf health issues, the newsletters, the pamphlets, and the computer bulletin board (http://www.va.gov/gulf.htm) are designed to provide a number of venues through which veterans can obtain information.

However, there does appear to be a lack of reliable, easily understood information for Persian Gulf veterans regarding exposures and their health consequences. This information needs to be readily accessible and available.

9. The committee recommends that VA develop informational pamphlets for veterans. These pamphlets could be placed in facility treat-ment areas and could address common concerns such as the purpose and process of the VA Persian Gulf program, health effects of low-level exposure to chemical warfare agents, research activities related to Persian Gulf veterans and their results-to-date, and so forth.

When a veteran first visits a VA health care facility, she or he completes what is known as an intake form. Such forms in general use throughout the VA system are not designed to easily identify Persian Gulf veterans. Although intake forms request dates of active-duty service, no questions specifically ask whether the veteran served in the Persian Gulf War. Unless veterans so identify themselves, those performing the intake function would have to know about the VA Persian Gulf Registry and specifically ask about service in the Persian Gulf War. It is unlikely that this occurs, and therefore, an opportunity to identify veterans who may wish to participate in the program is lost.

10. The committee recommends that VA consider redesigning intake forms so that the veteran is asked to identify whether or not she or he was deployed to the Persian Gulf War (or any other specific engagement).

One of the charges to the committee was to determine the adequacy of the provider education activities for those who participate in the diagnosis of Persian Gulf veterans' health problems.

As discussed earlier in the report, VA has designed a number of programs to educate its designated Persian Gulf providers including interactive satellite teleconferences, quarterly national telephone conference calls, direct educational mailings, and an annual conference on health consequences of service in the Persian Gulf War.

The committee was impressed with the high quality of these efforts, hopes that those for whom they have been designed avail themselves of the opportunity to participate, and believes that the audience who receives such education needs to be expanded.

Although the designated Persian Gulf providers are given the opportunity to participate, other primary care personnel and specialists do not usually receive the materials nor do they participate in the educational programs. Some specialists appeared to know little about the VA Persian Gulf Registry and UCAP, had little or no orientation to the program, and were unable to identify whether the patients whom they saw were Persian Gulf veterans. Although a number of educational opportunities to learn about the program exist (e.g., the yearly national conference, quarterly newsletters, conference calls, and videotapes), participation in such activities appears to be limited to the designated Registry providers.

11. The committee recommends that primary care providers, in addition to the Registry practitioners, as well as the specialists who see Persian Gulf veterans, be provided the opportunity and encouraged to participate in the educational programs.

11.a. The audience for whom existing educational training activities are developed related to providing health care for Persian Gulf veterans should be expanded to include other providers involved in the evaluation process, for example, designated specialty consultants.

The specialists with whom the committee spoke and those from whom the committee received input often felt frustrated that veterans sometimes appeared more knowledgeable about the latest Persian Gulf veterans' health issues and research results than they were. Many believed that such a lack of information could be interpreted as a lack of caring or a lack of expertise in treatment and could lead to patient dissatisfaction with the services he or she is receiving, even if those services were of high quality.

This lack of knowledge undermines Persian Gulf veterans' confidence in the system, worsening patient satisfaction and heightening public concerns about the adequacy of the VA system for addressing the health issues of Persian Gulf veterans.

The lack of in-depth involvement of the specialists also means that they do not have access to continuing medical education opportunities related to Persian Gulf veterans' health issues and that they may not participate in an evaluation-feedback mechanism to determine the value of various parts of the protocol or system. The specialists are the ones ordering, providing, and evaluating the more sophisticated tests conducted during the evaluation process. They are in an excellent position to determine how appropriate or beneficial these might be for individual patients. Such feedback could be collated to look more closely at the tests that do result in some positive information about the nature of veterans' complaints.

11.b. VA should consider the following options for education of its providers: periodic team conferences (perhaps quarterly) to be held with all designated providers (including specialists) to discuss activities and findings and to provide updates on Persian Gulf issues and concerns; and development of site-specific clinical pathways by designated specialists and Registry providers.

Reproductive issues have been addressed in the educational efforts of VA. Although these concerns seem to have been discussed, aspects of health unique to women have been given a lower profile.

11.c. Future educational efforts should place greater emphasis on women's health concerns.

A tremendous amount of knowledge about the diagnosis and treatment of Persian Gulf veterans' health problems is being amassed in various sites around the country. It behooves VA to identify where and with whom this special expertise exists and to develop mechanisms whereby others can benefit from the lessons that have been learned.

11.d. VA should provide resources to establish a repository for the accumulated knowledge of, expertise in, and experience with the health issues and problems of Persian Gulf veterans. Specialists who possess such expertise should be identified and available for consultation by telephone, e-mail, or telemedicine connections with local providers in all VA facilities.

6

Conclusion

Change is inevitable, and as such, it is important to plan for that change on the basis of new information and techni