ORIGINAL INVESTIGATION
Emerging Credentialing Practices, Malpractice
Liability Policies, and Guidelines Governing
Complementary and Alternative Medical Practices
and Dietary Supplement Recommendations
A Descriptive Study of 19 Integrative Health Care Centers in the United States
Michael H. Cohen, JD; Andrea Hrbek; Roger B. Davis, ScD; Steven C. Schachter, MD; David M. Eisenberg, MDBackground:Little is known about policies governing the integration of complementary and alternative medi-cal (CAM) therapies and providers.
Methods:To document emerging approaches in 19 US
hospitals regarding credentialing, malpractice liability, and pharmacy policies governing integration of CAM therapies and providers into conventional medical set-tings, we surveyed 21 academic medical centers and 13 non–academically affiliated hospitals that are nationally visible and are integrating CAM therapies into conven-tional medical settings. Of the 19 respondents, 11 were tertiary care hospitals, 6 were community hospitals, 1 was a freestanding center associated with a community-based hospital, and 1 was a university-community-based rehabilita-tion hospital.
Results:Institutions had no consistent approach to pro-vider mix and authority within the integrative care team, and minimum requirements for professional liability in-surance, informed consent disclosure, and hiring
sta-tus. Less than a third had a formal (stated) policy con-cerning dietary supplements; those selling supplements in their pharmacy lacked consistent, evidence-based ra-tionales regarding which products and brands to in-clude or exin-clude. Although many hospitals confiscated patient supplements on admission, institutions had in-consistent criteria regarding allowance of home supply.
Conclusions:Hospitals are using heterogeneous
ap-proaches to address licensure, credentialing, scope of prac-tice, malpractice liability, and dietary supplement use in developing models of integrative care. The environ-ment creates significant impedienviron-ments to the delivery of consistent clinical care and multisite evaluations of the safety, efficacy, and cost-effectiveness (or lack thereof ) of CAM therapies (or integrative models) as applied to management of common medical conditions. Consen-sus policies need to be developed.
Arch Intern Med. 2005;165:289-295
H
OSPITALS IN THEUNITEDStates increasingly are ex-ploring clinical use of complementary and al-ternative medical (CAM) therapies such as chiropractic, acupunc-ture, massage therapy, and nutritional and herbal care.1This research documents emerging approaches, across 19 US hos-pitals, regarding credentialing practices, malpractice liability policies, and phar-macy and therapeutic guidelines govern-ing the integration of CAM therapies and providers into conventional health care de-livery systems. Legal rules and regula-tory policies in each of these 3 domains leave an uneven map for clinicians and re-searchers. For example, while every state licenses chiropractors, not all states li-cense practitioners of acupuncture and tra-ditional oriental medicine, massage therapy, and naturopathic medicine.2,3
State licensing laws vary in the scope of practice they allow each licensed CAM pro-vider; for example, only some specifi-cally permit acupuncturists to recom-mend Chinese herbal medicine, while at least 1 state expressly prohibits the prac-tice.3Similarly, at least 3 states allow un-licensed CAM providers to offer health care services, under certain conditions,3while in other states, unlicensed providers risk prosecution for unlawful medical prac-tice.2Such variability makes institutional selection of, and credentialing for, CAM providers highly heterogeneous.
In the liability arena, few judicial opin-ions specifically address clinicians’ use of CAM therapies,2,4although increased liti-gation is likely as physicians are asked to counsel patients regarding use or avoid-ance of CAM therapies or for referrals to CAM providers.5,6Finally, although ques-tioning patients regarding ingestion of di-Author Affiliations: Harvard
Medical School Osher Institute and the Division for Research and Education in
Complementary and Integrative Medical Therapies, Harvard Medical School, Boston, Mass. Financial Disclosure: Mr Cohen and Dr Eisenberg each receive honoraria from speaking engagements; Mr Cohen represents clients in the area of complementary and alternative medicine through his law office and the
etary supplements arguably should be part of history tak-ing,7the lack of agreed-on manufacturing standards and insufficiency of information concerning safety and effi-cacy for dietary supplements (and individual brands of supplements) has led some health care institutions to cat-egorically ban patient use of dietary supplements.8,9 Ide-ally, determining the status of current practices can help clarify whether common approaches are emerging to fa-cilitate reproducible models of integrative health care and generalizable clinical research.
METHODS
Integrative medicine is defined as health care that “combines
mainstream medical therapies and CAM therapies for which there is some high-quality scientific evidence of safety and ef-fectiveness.”10Because the universe of US hospitals offering integrative medicine cannot be accurately determined—and is rapidly changing—this pilot survey identified centers from National Institutes of Health–funded research studies con-cerning CAM therapies, from the Consortium of Academic Health Centers for Integrative Medicine,11from among at-tendees at Harvard Medical School Continuing Medical Edu-cation courses on delivery of CAM therapies, and from an ex-amination of major, operating integrative care centers.12We identified a convenience sample of 21 such academically af-filiated and 13 non–academically afaf-filiated hospitals with in-tegrative care centers. We defined the academic medical center as a health care center that includes an allopathic or osteo-pathic school of medicine, at least one other health profes-sions school or program, and one or more university-owned or -affiliated teaching hospitals.13
We sent invitational letters, followed up by a survey and telephone calls, to medical directors of the integrative care centers in these 34 institutions. Twelve (57%) of the 21 academically affiliated centers and 7 (54%) of the 13 non– academically affiliated centers responded with completed sur-veys from May 1, 2001, through June 1, 2002. Of the 19 total respondents, 11 were tertiary care hospitals, 6 were commu-nity hospitals, 1 was a freestanding center associated with a com-munity-based hospital, and 1 was a university-based rehabili-tation hospital (see acknowledgments). Institutional review
board approval from Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Mass, was obtained.
RESULTS
LICENSURE, CREDENTIALING, AND SCOPE OF PRACTICE
Practitioner Mix
The professional mix of practitioners participating in in-tegrative care teams reflected no discernible pattern, across institutions, geographic regions, or provider groups within a given institution (Tableand Web Appendix 1 [Mind Body Experts and Other Providers]). (All Web appen-dixes referred to in this article can be found on the Har-vard Medical School Osher Institute’s Web site: http: //www.osher.hms.harvard.edu/pu_general_pub.htm).
Practitioner Scope of Practice and Requirement for Physician Referral
Only 40 of the 170 total providers were explicitly re-quired to obtain physician referrals before delivering clini-cal services to patients (Web Appendix 2 [Practitioner Scope of Practice]). Practitioner autonomy was hetero-geneous, with no apparent pattern with regard to aca-demic vs nonacaaca-demic affiliation. Slightly more provid-ers were approved (71) than not approved (63) to recommend dietary supplements (ie, herbs, vitamins, and nonherbal supplements). Six of 15 non-MD acupunc-turists and 9 of 17 nutritionists were so approved. More than twice as many providers (99 of 170) had authority to treat inpatients than did not have such authority (43 of 170). When asked whether centers had any specific limitations on provider authority other than those indi-cated in Web Appendix 2, few centers added written com-ments (Web Appendix 3 [Limitations on Provider Au-thority and Other Risk Management Strategies]). Table. Provider Mix in 19 Centers, by Frequency
Provider Type
Academic Institutions Nonacademic Institutions
Total No. of Providers 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Mind-body 3 1 1 2 1 1 2 1 1 1 1 2 1 1 1 4 1 1 26 Nutritionist 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 17 Massage therapist 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 16 Nurse practitioner 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 16 Acupuncturist (non-MD) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 15 Acupuncturist (MD) 1 1 1 1 1 1 1 1 1 1 1 1 1 1 14 Other* 3 1 1 2 2 3 1 13 Pharmacist 1 1 1 1 1 1 1 1 1 1 1 1 12 Exercise physiologist 1 1 1 1 1 1 1 1 1 1 10 Nurse anesthetist 1 1 1 1 1 1 1 1 1 1 10 Nurse midwife 1 1 1 1 1 1 1 1 8 Chiropractor 1 1 1 1 1 1 1 7 Naturopath 1 1 1 1 4 Physician assistant 1 1 2
No. of provider types at each institution 10 1 9 10 11 6 10 12 8 12 5 11 13 9 10 10 10 5 8 170
Credentialing Requirements and Policies Our survey asked respondents to confidentially provide their credentialing policies. The policies we received re-quired evidence of competence, such as some combina-tion of a valid state license (where available by state to the specific provider), specified limits of malpractice li-ability insurance, evidence of training (eg, documenta-tion of nadocumenta-tional certificadocumenta-tion, or of highest certificadocumenta-tion available; or a specified number of hours of theory and of clinical training), and evidence of experience (eg, let-ters of reference from patients and/or providers). Addi-tional requirements included interview and skills assess-ment, evidence of completion of mandatory continuing education units, health clearance, ongoing peer and com-mittee review, and background check.
PROFESSIONAL LIABILITY AND RISK MANAGEMENT Liability Insurance Requirements
There was no consistent pattern of liability insurance, either by provider type or between academic and non-academic centers (Web Appendix 4 [Minimum Re-quired Professional (Malpractice) Liability; Per Claim/ Cost per Aggregate]). The most frequent response was a requirement of $1 million per claim/$3 million per ag-gregate in liability insurance (74 of 170 providers), al-though there was no response for 66 providers, suggest-ing that this information may not be readily accessible or may be unknown at many institutions. The remainder of institutional requirements for “other” providers repre-sented in Web Appendix 4 reflected varied responses with regard to malpractice liability insurance: $5 million/ unspecified (n = 10); $3 million/unspecified (n = 4); $300000/$600000 (n=4); $1 million/$1 million (n=4); $4 million/$16.5 million (n=2); $600000/$600000 (n=2); $2 million/$2 million (n=1); $1 million/$2 million (n=1); $1.2 million/$1.2 million (n=1); “brings own” (n=1).
Informed Consent
While survey respondents differed in whether and how they required providers to obtain patients’ informed consent and to document consent to CAM therapies, most required risk-benefit disclosure in some form. Specifically, of 18 respon-dents, 8 (5 academic and 3 nonacademic institutions) in-dicated that their intake forms or medical records contained informed consent language specifically geared to use of CAM therapies; 8 indicated that they did not. Five of the 8 (3 academic and 2 nonacademic institutions) with CAM-specific informed consent forms also required that provid-ers delivering CAM therapies, including dietary supple-ments, verbally discuss the risks and benefits of these therapies with their patients and document these discus-sions in the medical record. An additional 6 of 18 indi-cated that they required discussions only. Four of 18 in-stitutions (2 academic and 2 nonacademic) indicated that they required neither written informed consent nor ver-bal discussions.
Other Risk Management and Human Resources Policies
Nine of 19 institutions indicated having no specific comprehensive risk management policies regarding their CAM providers other than the stated minimum requirements for malpractice liability insurance. Some of the institutions indicated that they addressed risk management more informally (Web Appendix 3). When asked what key issues institutions believed needed to be addressed as institutional policy, varied responses included a mix of legal, clinical, and adminis-trative concerns (Web Appendix 5 [Key Issues of Insti-tutional Policy]).
Institutions offered no discernible pattern (or consis-tently stated rationale for deciding) as to whether pro-viders were hired as employees or independent contrac-tors, across institutions, within institutions by provider type, or within the conventional or CAM domain (Web Appendix 6 [Employee Status]). Overall, 78 providers were hired as employees, 44 were hired as independent contractors, there were 42 nonresponses, and 6 re-sponses were ambiguous (eg, some indicated that the same provider type was hired both ways). There was no ap-preciable distinction between academic and nonaca-demic centers.
Responses to questions regarding contractual arrange-ments, income tax, and billing procedures also were var-ied and did not clarify the data regarding employee sta-tus. Survey results did not disclose whether decisions in this arena reflected liability vs administrative concerns (eg, benefits or costs, such as incentives to participate in clinical team and administrative meetings at the cen-ter). We did not collect information on reimbursement by third-party payers.
PHARMACY AND THERAPEUTICS COMMITTEE PRACTICES
As summarized in Web Appendix 7 (Policies and Prac-tices Involving Dietary Supplements), only 6 of 19 in-stitutions had formal policies regarding dietary supple-ments. For the purpose of the survey, dietary supplements were defined for the respondents to include vitamins, min-erals, and herbs (such as Ginkgo biloba, saw palmetto, and coenzyme Q10) that were neither approved nor classi-fied by the Food and Drug Administration as drugs. The category did not include liquid caloric supplements.
Outpatient Policies and Practices Involving Dietary Supplements
Information. Of the 18 centers responding to the ques-tion about a source of informaques-tion within the institu-tion concerning dietary supplements, 15 responded af-firmatively and indicated that the information typically was located in the hospital pharmacy, or text (eg, li-brary) and online references in an intranet library. Eleven of the 15 responding indicated that this source of information (eg, the hospital pharmacy) makes rec-ommendations to staff concerning patient use of dietary supplements. Three of the 11 indicated that such recom-mendations make distinctions between different brands of the same herb.
Outpatient Pharmacy. Six of 19 respondents (3 aca-demic and 3 nonacaaca-demic) reported having an outpatient pharmacy that sells dietary supplements (Web Appendix 7). Of these 6, 1 included only herbal products; 5 in-cluded herbal products and homeopathic remedies as well as vitamins, minerals, and supplements. An additional re-spondent indicated that such a pharmacy was in develop-ment and would include homeopathic remedies.
Inpatient Policies and Practices Involving Dietary Supplements
Confiscation; Home Supply. Seven of 19 institutions in-dicated that they confiscate any and all dietary supple-ments at the time the patient is hospitalized (Web Ap-pendix 7). Additional written responses are included in Web Appendix 8 (Policies Regarding Confiscation of Di-etary Supplements), and the criteria used to determine which dietary supplement products patients may con-tinue using during hospitalization are included in Web Appendix 9 (Criteria Regarding Home Supply of Di-etary Supplements).
Documentation of Inpatient Use. Eleven of 18 institu-tions responding to this question required documenta-tion of inpatient use of dietary supplements and 7 indi-cated that no documentation was required. Six respondents indicated that documentation was required only in the ad-mission note; 2 indicated that documentation was re-quired only in the patient’s hospital chart; and 2 respon-dents indicated that documentation was required in the admission note, hospital chart, and medication adminis-tration record. Eleven respondents specified that the docu-mentation must include the name of the dietary supple-ment; 9 of the 11 also required the dose, 5 required the indications, and 4 required product identification. Inpatient Formulary; Influence of Third-Party Reimbursement. Nine of 16 responding institutions re-ported having an inpatient formulary for dietary supple-ments. Eight of these 9 included vitamins, minerals, and supplements; the ninth included only vitamins. One also included herbal products. The criteria used to deter-mine which dietary supplement products and brands to include were quite varied (Web Appendix 10 [Criteria Used to Determine Which Dietary Supplement Prod-ucts to Include in Formulary] and Web Appendix 11 [Cri-teria Used to Determine Which Dietary Supplement Brands to Include in Formulary]). Two of 13
respon-dents reported that the availability of third-party reim-bursement would influence the decision to include di-etary supplements in the formulary (Web Appendix 7). Ten reported that this factor would have no influence, and one indicated “not certain.”
Anesthesia and Surgery Recommendations Regarding Discontinuance of Herbs, Vitamins, and Supplements. Six respondents (all academic centers) indicated that they make formal recommendations regarding the contin-ued use (or discontinuance) of dietary supplements be-fore elective surgery (eg, to discontinue those known or thought to interact with anesthesia or clotting; see Web Appendix 12 [Anesthesia and Surgery Recommenda-tions Regarding Discontinuance of Herbs, Vitamins, and Supplements] and Web Appendix 7). Thirteen other cen-ters did not respond to this question.
COMMENT
Our pilot data suggest that hospitals are using hetero-geneous approaches to address licensure, credentialing, and scope of practice of complementary care providers; malpractice liability; and dietary supplement use in ef-forts to develop models of integrative care. The standard-ization of provider education, training, and therapeutic approach facilitating consistent research across institu-tions and states is lacking.
LICENSURE, CREDENTIALING, AND SCOPE OF PRACTICE
We found no consistency in approaches to professional provider mix within the integrative care team. Nearly ev-ery center included a “mind-body” provider, for ex-ample, while only 4 included a naturopath; 16 included a massage therapist, while only 7 included a chiroprac-tor; some included both MD and non-MD acupunctur-ists, while others included one but not the other. Even within the category of “mind-body” expert, choices ranged from psychiatrist to psychologist to “healing touch thera-pist” and “spiritual counselor”; and the range of “other” providers included “child life specialist,” “biofeedback technician,” art therapist, movement therapist, and “en-ergy practitioner,” as well as physical therapist, reflex-ologist, and osteopath. Such diversity may reflect the un-even map of licensure for various CAM providers across states,2,3as well as potential liability concerns2,14; the lack of clear definitions of integrative clinical care and of “mind-body” care; institutional reluctance to include un-licensed providers (such as naturopathic physicians in some states)2; the present lack of an evidence base con-cerning the optimal (and most cost-effective) profes-sional provider mix in the integrative team; the effect of personal beliefs and practices (particularly involving herbal medicine) on CAM practices15; and the influence of decisional norms within institutions on local creden-tialing decisions.16
in-stitutions—regarding which providers are required to have a physician referral before delivering clinical services (40 yes and 92 no); which may recommend dietary supple-ments (71 yes and 63 no); and which may treat inpa-tients (99 yes and 43 no). Generally, conventional pro-viders on the integrative care team and their CAM counterparts had a fair amount of autonomy and author-ity across institutions; on the whole, there were few ex-plicit (or comparable) scope-of-practice restrictions. Such permissive practices arguably parallel a regulatory land-scape in which, in most states, chiropractors, massage therapists, and naturopathic physicians (where li-censed) are independent practitioners, not statutorily de-pendent on physician referral or supervision.2,3,17
PROFESSIONAL LIABILITY AND RISK MANAGEMENT
Although the most common requirement was $1 mil-lion per claim/$3 milmil-lion per aggregate, the require-ments were inconsistent across institutions and across providers within institutions. Generally, legal rules gov-erning malpractice liability involving CAM therapies are in flux2,4,5and present uncertainty to clinicians.18For some professions (such as acupuncture), the lack of signifi-cant published information concerning claims experi-ence makes it difficult to determine definitively what level of insurance should be required.6CAM therapies also ac-count for only approximately 5% of the total medical mal-practice insurance market; to date, both the number of claims against CAM providers and the average indem-nity paid per claim have been lower than claims against primary care physicians. However, this situation may change as integrative care expands.4-6,14
Concerning informed consent, respondents gener-ally seemed aware of the need for disclosure of risks and benefits concerning CAM therapies, although there was no standardized approach, and hospitals differed as to whether such disclosure should be written or verbal and how it should be documented. Inadequate informed con-sent can be grounds for malpractice liability apart from any negligent care.2,4Furthermore, new guidelines by the Federation of State Medical Boards require specified in-formed consent procedures, at least for physicians coun-seling patients regarding use of CAM therapies.19
The respondents had no consistent approach to hir-ing status. The heterogeneity may reflect financial con-straints and/or legal ambiguity concerning institutional liability: health care institutions are vicariously liable for the negligence of their “employees” but not for the neg-ligence of affiliated “independent contractors.”2
PHARMACY AND THERAPEUTICS COMMITTEE PRACTICES
Dietary supplement policies also varied. Less than a third of the sample had a formal policy concerning dietary supplements. While 6 of 10 centers indicated selling di-etary supplements in their outpatient pharmacy (and 9 of 16 indicated selling in their inpatient pharmacy), there was no consistent rationale regarding which products and brands to include (or explicitly exclude); responses ranged
from “no specific criteria” to “cost.” Although approxi-mately a third had a practice of confiscating dietary supple-ments on the patient’s admission to the hospital, there were no consistent criteria across institutions as to which supplements to allow patients to continue using as home supply; several hospitals allowed ad hoc decisions by at-tending physicians. Practices also varied concerning re-quirements for documentation of dietary supplement use, a topic now being addressed by the Joint Commission on the Accreditation of Healthcare Organizations.20
Hospitals differed as to whether the availability of third-party reimbursement would influence their policy con-cerning dietary supplements; only 6 of 19 respondents made formal recommendations regarding the contin-ued use of supplements before elective surgery.
LIMITATIONS
The generalizability of this pilot study was limited by sev-eral factors, including the nonresponse rate (44%). Nu-merous attempts to reduce the nonresponse rate through follow-up telephone calls and e-mail were unsuccessful. However, it seems unlikely that data from the remain-ing sites would have significantly reduced the inconsis-tencies in the responses that were observed from the 19 responding sites.
The nonresponse rate across various questions may reflect the uncertainty that pervades this field, includ-ing concern about sharinclud-ing data. The length and com-plexity of the survey, and the lack of a centralized loca-tion within each center from which to retrieve the information, may have contributed to missing data. In addition, answers to surveys suggested that some of the follow-up questions simply did not apply to particular institutions, or required information that was unknown to the respondent. Finally, hospitals define and opera-tionalize “integrative care” differently—some create spe-cific centers for “complementary medicine” or “integra-tive medicine”; others have affiliated clinics devoted to “wellness,” “mind-body,” or similar concepts; and still others have CAM providers in various hospital depart-ments (eg, a non-MD acupuncturist in the pain center) but no “integrative care” unit. Therefore, respondents may not have included all relevant CAM providers in their sur-vey responses, while others may have been overinclu-sive.
CONCLUSIONS
acupuncturists, massage therapists, chiropractors, natu-ropaths, and “mind-body” therapists, and correspond-ing divergence of hospital practices concerncorrespond-ing creden-tialing, limitations on practice authority, and provider mix in the integrative team; (2) divergence of ap-proaches to liability management strategies, including minimum malpractice liability insurance, informed con-sent practices and documentation, and provider hiring status; and (3) variability of approaches to patient use of, and clinician recommendations involving, dietary supplements, including the presence (or absence) of rel-evant, formal (stated) policies, the presence (or ab-sence) of various combinations of dietary supplements in the formulary for outpatient and inpatient pharma-cies, varying criteria (or lack thereof ) regarding which products and brands to include (or explicitly exclude), varying confiscation practices and criteria for allowing home supply, and varying formal recommendations (or lack thereof ) regarding the continued use (or dis-continuance) of dietary supplements before elective surgery.
Furthermore, institutional inconsistency and ambi-guity complicate clinical decision making as well as re-search, and foster ethical issues.21Such difficulties would likely benefit from national consensus approaches to cre-dentialing, malpractice liability policies, and guidelines for dietary supplement recommendations.
In designing consensus, the regulatory landscape cates several conundrums. Since the US Constitution re-serves regulation of health, safety, and welfare (and there-fore licensing of health care providers) to the states, diversity of licensing and scope of practice rules across states is unlikely to change.2Nor are hospitals likely to increase standardization of integrative care teams in the absence of data concerning what combinations of pro-viders or therapies are most effective (and safe) for any given condition. Potential liability exposure is difficult to predict, particularly given changing case law and chang-ing levels of evidence for specific therapies,5and local in-stitutional politics are likely to continue playing a sig-nificant role in the emergence of models of integrative care.16In short, a national, standardized approach to li-censure, scope of practice, credentialing, and liability man-agement may not be feasible. In addition to regulatory issues and liability concerns,5ethical issues complicate matters.14,22,23
While licensure, credentialing, and scope of practice and liability issues are thorny, dietary supplement poli-cies may lend themselves more easily to research as evi-dence accumulates regarding safety and efficacy (or lack thereof) and mechanism of supplements, including their potential interactions with conventional medication.9Here greater consensus might be achieved, and hospitals may be able to abandon risky and unsupportable approaches that lack scientific justification.
We offer the following policy recommendations. First, while licensure and scope of practice are matters of state law, either an agency to coordinate federal policy con-cerning CAM research and practice,24the Council of Gov-ernors, or another appropriate governmental body could encourage CAM professional organizations to develop model legislation and guidelines (where feasible and
ap-propriate) to take the following steps: (1) increase the standardization in minimum required education and clini-cal training for licensure (and/or adopt a national certi-fying examination) across states; (2) create standard-ized, biomedical curricular components for their students; (3) facilitate reduction in scope of practice variations across states; (4) increase the proportion of educational institutions within the profession that are profession-ally accredited; (5) define or limit potentiprofession-ally mislead-ing uses of the term primary care provider within the CAM profession3; (6) enhance opportunities for licensees to participate in clinical residency in integrative care cen-ters, with the intent of receiving supervision by senior clinicians from both the conventional and CAM com-munities; and (7) create performance standards and as-sessment of clinical skills for conventional and comple-mentary care providers who work in integrative care settings.
Second, stakeholder groups could, individually and collectively, encourage relevant professional organiza-tions and agencies to consider creative ways of using in-tegrative care centers to help further consistent clinical care and research. Such centers could become key edu-cational hubs in which biomedical and CAM clinicians and researchers receive the cross-disciplinary training nec-essary for safe, evidence-based integrative care and in-novative investigation.
Third, regulations should encourage institutional re-view boards to include CAM providers and other exper-tise relevant to evaluation of protocols and risks for hu-man subjects in research studies involving study of CAM therapies. Such regulations could reduce potential bias against (or for) CAM protocols, and furnish institu-tional review boards a broader range of expertise with which to evaluate relevant research.
Fourth, in trying to achieve sufficiently standardized approaches to developing multisite research protocols across states, investigators may find it advantageous to define the modalities specified providers are allowed to offer in clinical trials.3Finally, the Consortium of Aca-demic Health Centers for Integrative Medicine plans to become a national repository of information concern-ing preferred institutional practices.11These steps, with-out intruding on states’ rights to be “laboratories for ex-perimentation,” and while respecting the diversity of institutional approaches and pluralistic development of models of integrative care, will encourage the develop-ment of more consistent policies relating to clinical de-livery and research involving integrative care.
Accepted for Publication: October 14, 2004.
Health, Bethesda, Md; and grant IG13LM07475-01 from the National Library of Medicine at the National Insti-tutes of Health. None of the entities sponsoring the project had any influence on the study design, analysis, or writ-ing of the manuscript.
Disclaimer: The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of the National Center for Complementary and Alternative Medicine, the National Library of Medicine, the National Institutes of Health, or any sponsor.
Additional Information: Copies of the survey instru-ment used for this study can be found on the Harvard Medical School Osher Institute’s Web site, http://www .osher.hms.harvard.edu.
Acknowledgment: We thank Marian Osterweis, PhD (Association of Academic Health Centers, Washington, DC), Roger Wood, LMT, and Shannon Marcotte for their research and administrative assistance with this article, and for their help in providing a list of academi-cally affiliated medical centers. We also thank the par-ticipating centers: Alternative Medicine Division, Hen-nepin County Medical Center (South Minneapolis, Minn); Atlantic Mind Body Center for Complementary Medicine, Morristown Hospital (Morristown, NJ); Cen-ter for Complementary Medicine, Lutheran General Hospital (Park Ridge, Ill); Center for Complementary Medicine, University of Pittsburgh Shadyside (Pitts-burgh, Pa); Center for Integrative Medicine (Concord, NH); Complementary Medicine Clinics, Stanford Uni-versity School of Medicine (Palo Alto, Calif ); Medical University of South Carolina, Complementary and Alter-native Medicine Department (Charleston); Complemen-tary Medicine Program, University of Maryland School of Medicine (Baltimore); Continuum Center for Health and Healing, Beth Israel Medical Center (Albert Einstein Medical College, New York, NY); Georgetown Univer-sity Medical Center, Departments of Physiology, Bio-physics, and Medicine (Washington, DC); Institute for Health and Healing, California Pacific Medical Center (San Francisco); Mercy Hospital, University of New England Osteopathic Medical College (Portland, Me); New York Presbyterian Hospital, Milstein Hospital Building (Columbia University College of Physicians & Surgeons, New York, NY); Osher Center for Integrative Medicine, University of California at San Francisco; Pro-gram in Integrative Medicine, University of Arizona School of Medicine (Tucson); Scripps Center for Inte-grative Medicine (La Jolla, Calif ); Siegler Center for Integrative Medicine, St Barnabus Medical Center (Liv-ingston, NJ); Center for Integrative Medicine (San Jose, Calif ); and Yale-Griffin Preventive Research Center (Derby, Conn).
REFERENCES
1. American Hospital Association (Health Forum). 2000-2001 Complementary and Alternative Medicine Survey. Chicago, Ill: American Hospital Association; 2002. 2. Cohen MH. Complementary and Alternative Medicine: Legal Boundaries and
Regu-latory Perspectives. Baltimore, Md: Johns Hopkins University Press; 1998. 3. Eisenberg DM, Cohen MH, Hrbek A, Grayzel J, Van Rompay MI, Cooper RA.
Credentialing complementary and alternative medical providers. Ann Intern Med. 2002;137:965-973.
4. Cohen MH. Beyond Complementary Medicine: Legal and Ethical Perspectives on Health Care and Human Evolution. Ann Arbor: University of Michigan Press; 2000. 5. Cohen MH, Eisenberg DM. Potential physician malpractice liability associated with complementary/integrative medical therapies. Ann Intern Med. 2002;136:596-603.
6. Studdert DM, Eisenberg DM, Miller FH, Curto DA, Kaptchuk TJ, Brennan TA. Medical malpractice implications of alternative medicine. JAMA. 1998;280: 1610-1615.
7. Eisenberg DM. Advising patients who seek alternative medical therapies. Ann In-tern Med. 1997;127:61-69.
8. Walker PC. Evolution of a policy disallowing the use of alternative therapies in a health system. Am J Health Syst Pharm. 2000;57:1984-1990.
9. Boyer EW. Issues in the management of dietary supplement use among hospi-talized patients. Internet J Med Toxicol. 2002;5:1.
10. National Center for Complementary and Alternative Medicine. What Is mentary and Alternative Medicine? Washington, DC: National Center for Comple-mentary and Alternative Medicine; 2002.
11. Bravewell Collaborative. Academic Consortium of Health Centers for Integrative Medicine. Available at: http://www.pcintegrativemedicine.org/links/acadcon .asp. Accessed November 19, 2003.
12. Weeks J. Hospital-sponsored integrative clinics: an in-depth report. Integr Med Consult. 2001;(suppl)2nd quarter.
13. Association of Academic Health Centers Web site. Available at: http://www.ahcnet .org/. Accessed November 19, 2003.
14. Cohen MH. Future Medicine: Ethical Dilemmas, Regulatory Challenges, and Thera-peutic Pathways to Health and Healing in Human Transformation. Ann Arbor: University of Michigan Press; 2003.
15. Dougherty K, Touger-Decker R, O’Sullivan Maillet J. Personal and professional beliefs and practices regarding herbal medicine among the full time faculty of the Newark-based schools of the University of Medicine and Dentistry of New Jersey. Integr Med. 2000;2:57-64.
16. Cohen MH, Ruggie MC. Integrating complementary and alternative medical thera-pies in conventional medical settings: legal quandaries and potential policy models. Univ Cincinnati Law Rev. 2004;72:671-729.
17. Mitchell BB. Acupuncture and Oriental Medicine Laws. Gig Harbor, Wash: Na-tional Acupuncture Foundation; 2001.
18. Weintraub MI. Legal implications of practicing alternative medicine. JAMA. 1999; 281:1698-1699.
19. Federation of State Medical Boards of the United States Inc. Model Guidelines for the Use of Complementary and Alternative Therapies in Medical Practice. Dal-las, Tex: Federation of State Medical Boards of the United States Inc; 2002. 20. Joint Commission on the Accreditation of Healthcare Organizations.
Compre-hensive Accreditation Manual for Hospitals: Proposed Revisions to Medication Use Standards (to Take Effect Jan. 1, 2004), Standards TX.3.2-3.3, TX.3.5-3.6. Oakbrook Terrace, Ill: Joint Commission on the Accreditation of Healthcare Or-ganizations; 2003.
21. Adams KE, Cohen MH, Jonsen AR, Eisenberg DM. Ethical considerations of comple-mentary and alternative medical therapies in conventional medical settings. Ann Intern Med. 2002;137:660-664.
22. Kaptchuk TJ, Eisenberg DM. Varieties of healing, 1: medical pluralism in the United States. Ann Intern Med. 2001;135:189-195.
23. Institute of Medicine, Committee on Quality of Health Care. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academy Press; 2001. Quality of Health Care in America Project.
11. Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557-560.
12. Higgins JP, Thompson SG. Quantifying hetero-geneity in a meta-analysis. Stat Med. 2002;21: 1539-1558.
13. Fleiss JL. The statistical basis of meta-analysis. Stat Methods Med Res. 1993;2:121-145. 14. Thompson SG, Sharp SJ. Explaining
heterogene-ity in meta-analysis: a comparison of methods. Stat Med. 1999;18:2693-2708.
15. Marx A, Bucher HC. Numbers needed to treat
de-rived from meta-analysis: a word of caution. ACP J Club. 2003;138:A11-A12.
16. Din JN, Newby DE, Flapan AD. Omega 3 fatty ac-ids and cardiovascular disease—fishing for a natu-ral treatment. BMJ. 2004;328:30-35. 17. Expert Panel on Detection EaToHBCiA.
Execu-tive Summary of The Third Report of The National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA. 2001; 285:2486-2497.
18. Grundy SM, Cleeman JI, Merz CN, et al. Implica-tions of recent clinical trials for the National Cho-lesterol Education Program Adult Treatment Panel III guidelines. Arterioscler Thromb Vasc Biol. 2004; 24:e149-e161.
19. Bucher HC, Hengstler P, Schindler C, Meier G. N-3 polyunsaturated fatty acids in coronary heart disease: a meta-analysis of randomized con-trolled trials. Am J Med. 2002;112:298-304. 20. Oxman AD, Guyatt GH. A consumer’s guide to
sub-group analyses. Ann Intern Med. 1992;116: 78-84.
Correction
Omissions in Byline. In the Original Investigation by
