Osteopathy was one of numerous nontraditional approaches to medicine that originated in the 1800s in response to perceived inadequacies in orthodox medicine, most commonly the growing opposition to the ineffectual and often lethal pharmacopoeia of the day. Two of the more popular nontraditional approaches, homeopathic and eclectic medicine, initially developed into influential medical entities, but they eventually lost their unique identities when their key components were either coopted or absorbed by the dominant allopathic medical profession. Osteopathy, however, successfully resisted assimilation and has survived alongside allopathic medicine in modern medicine in the United States.1,2
Andrew Taylor Still was the driving force behind the establishment of osteopathic medicine. His philosophic approach to medicine viewed the human being as a unit consisting of body, mind, and spirit, and capable of self-regulation, self-healing, and health maintenance. Furthermore, he believed there was a reciprocal relationship between the functions of the body, and that, through the application of rational medical concepts, including manipulative treatment, the system could restore functional capacity, enhance wellness, and assist in recovery from injury and disease. Still's ideas were grounded in a distinctive palpatory diagnostic and manipulative therapeutic approach to health and disease now commonly known as osteopathic manipulative treatment (OMT).3 When Still's unique medical concepts were rejected by the existing medical establishment, he set an independent course that led to the establishment of the first osteopathic college in 1892. He determined that graduates would be known as Diplomates in Osteopathy (DOs) rather than as conventional medical doctors (MDs).1,2
Osteopathic medicine evolved during a period in which the growing influence of the American Medical Association (AMA) contributed to the unification of orthodox medicine.2 For nearly a century, through various groups, osteopathic medicine struggled in this repressive milieu to establish its professional identity and to achieve credibility. The osteopathic profession resolutely pursued these goals, and eventually its efforts resulted in full practice opportunities with license to diagnose, prescribe medications, and perform surgery in all 50 states, and DOs received reimbursement from workmen's compensation and various insurance programs. Moreover, DOs gained acceptance as practicing physicians in all branches of the military as well as the civil service. As the profession grew in stature, new osteopathic hospitals were established and accredited to meet training needs, educational standards were strengthened to support the burgeoning number of osteopathic colleges, and a full complement of specialties and certifying boards was initiated to accommodate referrals from osteopathic physicians. Finally, after years of presistent effort, DOs achieved a “separate and equal” status in terms of legislative and regulatory affairs.1,4
Although contemporary osteopathic medicine embraces a broad-based approach to medical practice, the emphasis on structure and somatic dysfunction associated with manipulative medicine is still its single most identifiable feature.1 In the 1970s, Stiles5 demonstrated both better health and shorter hospital stays for patients with both surgical and medical conditions when OMT was added to their hospital treatment regimens. As recently as 1997, the use of OMT in family medicine or specialty care as either a primary or an adjunctive treatment for maladies based on somatic dysfunction in all body systems was reaffirmed.6 However, OMT has frequently been a source of concern and controversy within the profession,7 and the osteopathic profession appears caught in an evolutionary flux regarding the specific role and appropriate emphasis of OMT as it pertains to training and clinical practice.1
In 1992, on the 100th anniversary of the osteopathic profession, Meyer and Price8 described how the profession had gained credibility and acceptance through “… a remarkable transformation from osteopathy, characterized by manipulative therapy, to osteopathic medicine, characterized by full-service health care ….” In their opinion, the profession's new status moved it away from traditional primary care and toward the medical specialties. Moreover, Meyer and Price postulated that the growing number of osteopathic graduates engaged in allopathic postgraduate training programs had obscured the difference between the osteopathic and allopathic approaches to medical practice. Furthermore, they suggested that the gradual “absorption” of osteopathic practitioners into allopathic health care settings had de-emphasized OMT, which was fast “becoming a lost art within the profession.”
Is OMT becoming a lost art as the osteopathic profession enters the 21st century? Although long suspected, an alarming decline in the use of OMT by family practitioners was confirmed by a national survey conducted in 1997.9 The ongoing evolution of the osteopathic profession toward mainstream medicine and the decreased use of OMT among family physicians prompted us to study the following research questions:
* To what extent do osteopathic family physicians and specialists use OMT in their practices?
* To what degree do attitudes, training, perceived barriers, and OMT practice protocol predict use of OMT by family physicians and specialists?
* To what extent is the use of OMT by osteopathic physicians a function of gender, ethnicity, type of practice, year of graduation, type of specialty training (allopathic, osteopathic, mixed staff, or military), and family physician versus specialist?
In April 1998, we surveyed a random list of 3,000 osteopathic family physicians and specialists selected from the American Osteopathic Association's Physician Masterfile. Retirees, students, and interns were excluded. We used the term “specialist” to denote any specialty other than family practice, which has been considered the core of the profession.
The two-page questionnaire (similar to the questionnaire previously employed to measure OMT use by family physicians9) asked about the physician's age, gender, race or ethnic identification, osteopathic medical college, date of graduation, specialty or subspecialty, number of years of practice, and type of practice. Respondents were asked whether their specialty training had been in osteopathic institutions, allopathic institutions, or institutions with both types of affiliations. In addition, the physicians were asked the percentages of patients on whom they used OMT, the diagnoses for which they used OMT, and the relative frequencies with which various OMT procedures were used. Twenty-four items addressed attitudes toward OMT, adequacy of OMT training, perceived barriers to use of OMT, and OMT practice protocol. These 24 questions were declarative in nature, with responses given according to a five-point Likert-type scale (5 = strongly agree, 4 = agree, 3 = undecided, 2 = disagree, 1 = strongly disagree). Finally, three open-ended questions asked respondents to provide additional comments on the use of OMT in their practices as well as on the characteristics of osteopathic medicine that differentiate it from allopathic medicine.
The questionnaire, a cover letter explaining the study, an addressed, prepaid return envelope, and a return postcard indicating that the questionnaire had been returned under a separate cover were mailed in April 1998. The respondent's zip code (and name if more than one had the same zip code) were written on the postcard so that a followup mailing could be sent to non-responders while maintaining the anonymity of responders. A second mailing was sent in May 1998 to those who did not return the postcard.
The appropriate university commitee on research involving human subjects approved the survey instrument and the study protocol.
For the first phase of the analysis, we computed basic descriptive statistics for all questions. Next, the 24 Likert-type questions were grouped to form scales measuring attitudes toward OMT (six items), adequacy of OMT training (four items), perceived barriers to use of OMT (nine items), and protocol for use of OMT in practice (five items). For each of the scales, a composite variable was computed as the average of the items in the scale. In the process of forming the composite variables, negatively worded questions were first recoded so that higher scores corresponded to more positive attitudes toward OMT use, better OMT training, fewer perceived barriers to OMT use, and a more OMT-oriented practice protocol. The scales (groups of questions) were tested for internal reliability using Cronbach's alpha.
For the analysis, the percentage of patients on whom OMT was used was quantified as a grouped variable (1 = less than 5%, 2 = 5% to 25%, 3 = 26% to 50%, 4 = 51% to 75%, 5 = 76% to 100%), and the physicians' date of graduation was also grouped (1 = 1950 or earlier, 2 = 1951–1970, 3 = 1971–1984, 4 = 1985 to present).
We performed a linear regression analysis with the percentage of patients on whom OMT was used as the dependent variable, and physicians' attitudes toward use of OMT, adequacy of OMT training, perceived barriers to use of OMT, practice protocol for OMT use, and years of practice as independent variables. We computed separate regressions for family physicians and specialists. Subsequently, we used analysis of variance techniques to test for differences in use of OMT according to physicians' gender, practice type, ethnicity, year of graduation, osteopathic or allopathic orientation of specialty training, and practice as family physician or specialist.
From the 3,000 osteopathic physicians contacted, 979 responses (33.2%) were received (54 questionnaires were returned undelivered). Of these, 955 questionnaires were complete and usable for analysis. Of the respondents to the survey, 77.6% were men, 39.3% were family physicians, 60.7% were specialists, and the average age was 44.2 years, (range 27–86 years). The physicians were predominately Caucasian (92.6%), followed by Asian (3.0%), Hispanic (2.1%), African American (1.6%), and Native American (1.6%). All colleges of osteopathic medicine were represented in the sample, and there was diversity in terms of age, date of graduation, specialty, type of practice, and number of years in practice.
The percentages of patients on whom the physicians used OMT (our first research question) are presented in Table 1. Over 50% of the physicians used OMT on less than 5% of their patients. Family physicians used OMT more than did specialists: for example, 69.9% of the family physicians indicated they used OMT on 5% or more of their patients, while only 31.3% of specialists used OMT on more than 5% of their patients. A total of 37.5% of the specialists never used OMT on their patients, whereas 7.9% of family physicians never used OMT on their patients.
Mean scores for the 24 questions are depicted in Table 2. The respondents had very positive attitudes regarding OMT, with 96% agreeing or strongly agreeing that it is an efficacious treatment, and over 70% agreeing or strongly agreeing that they personally received OMT or provided it to friends, colleagues, or relatives outside of their practice.
The responses showed a difference between how the physicians perceived themselves and how they believed their patients perceived them. Ninety percent of the family physicians and 86% of the specialists agreed or strongly agreed that they thought of themselves as osteopathic physicians; however, only 53% of the family physicians and 34% of the specialists agreed or strongly agreed that their patients saw them as osteopathic physicians.
In general, the responses of the physicians to the Likert-type questions indicated that they believed their OMT training was adequate, with well over 70% agreeing or strongly agreeing that they were well prepared to diagnose and treat structural problems. Only 40% of the specialists, in contrast to 69% of the family physicians, agreed or strongly agreed that they were prepared to integrate OMT into their practices.
Only about a fifth of the physicians agreed or strongly agreed that insufficient OMT training limited their practices, and 36% of the specialists versus 21% of the family physicians agreed or strongly agreed that they lacked confidence in their OMT abilities. Other barriers to the use of OMT with which physicians agreed or strongly agreed were lack of time (69% family physicians; 57% specialists), other professional interests (30% family physicians; 47% specialists), poor reimbursement (29% family physicians; 15% specialists), physical facility unsuitable for OMT (23% family physicians; 37% specialists), lack of a supportive philosophical environment (19% family physicians; 24% specialists). Only 16% of the physicians cited lack of patient interest as a factor limiting OMT use, but 84% agreed or strongly agreed that the public associates OMT with chiropractic medicine.
A total of 81% of all respondents agreed or strongly agreed that OMT was important to assess musculoskeletal function. Despite this conviction, only 50% agreed or strongly agreed that they routinely tested patients for somatic dysfunction. Nevertheless, if positive signs of somatic dysfunction were noted, 76% of the family physicians and 45% of the specialists agreed or strongly agreed that these would be recorded in the medical record.
Mean scores and reliability coefficients (Cronbach's alpha) for the four scales formed from the 24 questions were attitudes toward OMT (mean = 3.67, alpha = .78), adequacy of OMT training (mean = 3.56, alpha = .79), perceived barriers to use of OMT (mean = 3.18, alpha = .74), and OMT practice protocol (mean = 3.41, alpha = .72). Concerning our second research question, the regression analysis for the family physicians revealed that the level of OMT use in practice was quite well predicted (R2 = 0.43), and that the significant predictors were perceived barriers to use of OMT (beta = 0.35, p < .001), OMT practice protocol (beta = 0.26, p < .001), and attitudes toward OMT (beta = 0.24, p < .001). For the specialists, the prediction was somewhat better (R2 = 0.58), and the significant predictors were OMT practice protocol (beta = 0.47, p < .001), perceived barriers to use of OMT (beta = 0.30, p < .001), attitudes toward OMT (beta = 0.19, p < .001), and adequacy of OMT training (beta = −0.10, p = .003).
The analysis of variance (used to test our third research question) revealed that the physicians' use of OMT varied significantly according to practice type (F = 31.45, p < .001), graduation date (F = 4.92, p = .002), and type of residency training (F = 2.74, p = .043), and between family physicians and specialists (F = 21.03, p < .001). Post hoc pair-wise comparisons using Bonferroni's correction for multiple comparisons showed that, not surprisingly, the rate of OMT use in special manipulative practice settings was significantly higher than were the rates of use in any of the other practice settings. Similarly, rates of OMT use in solo office practice were significantly higher than were rates in residencies, partnerships, group practices, and other types of practice. Further, the use of OMT was significantly greater among osteopathic physicians who graduated in 1950 or earlier than it was for physicians who graduated during the periods 1951–1970, 1971–1984, or 1984 to the present. Physicians who had received osteopathic residency training used OMT significantly more than did either physicians who had received allopathic residency training or physicians who had received training in institutions with mixed staffs. Finally, as expected, the family physicians made significantly more use of OMT than did the specialists (mean = 2.01 and 1.64, respectively). A further breakdown of OMT use scores according to practice type and graduation date is presented in Table 3.
The concern that OMT might become a lost art in the osteopathic profession is unquestionably supported by our data. Although OMT has been cited as a beneficial primary or adjunctive treatment for maladies in all body systems,6 over half of our respondents used OMT on less than 5% of their patients. Even more worrisome is that a quarter of the practitioners reported that they did not use OMT on any of their patients.
Our study is limited by the reliance on self-reported data. As with any such study, the respondents' interpretations of the researchers' intent, as well as their vested interests, may have influenced their responses. This may be particularly true in a study of OMT, a treatment approach that raises passionate differences within as well as outside the osteopathic profession.7 The response rate, 33.2%, may be indicative of this self-selection bias. We fully expected that osteopathic physicians at the time of the study were using OMT less frequently than they had in the past, especially specialists. If this were the case, there may have been reluctance to respond to a survey assessing use of OMT in practice. Thus, although the demographic profile of our sample is similar to the profile published by the American Osteopathic Association, if our sample is not truly representative it would more likely overestimate the use of OMT. Nevertheless, statistically significant trends are noted in the data, and are worthy of consideration by medical educators.
Given the trends in our findings, it is important to determine the causal factors that contribute to the diminished use of OMT among osteopathic practitioners. We found that physicians who graduated prior to 1950 used OMT significantly more than did those who graduated later. In recent years, as a growing number of osteopathic physicians have become integrated into mainstream medicine, the various practice expectations and constraints have created obstacles that preclude the practice of OMT. While solo practitioners used significantly more OMT than did those in other practice modalities, time and cost constraints associated with the burgeoning managed care phenomenon clearly do not facilitate or encourage the use of OMT.
Furthermore, a growing number of osteopathic physicians apparently do not strongly associate with their professional roots,10 a factor that might contribute to the diminished use of OMT. Interestingly, while almost 90% of our respondents readily identified themselves as osteopathic physicians, only about two fifths of these physicians felt that their patients associated them with the osteopathic profession. Two recent studies have supported a situation often referred to as the “back door phenomenon,”13 in which applicants who have not been admitted to allopathic medical schools apply to osteopathic colleges to become physicians. Shlapentokh and colleagues11 referred to these trainees as “allopathic ideologists” who place limited importance on OMT as a viable treatment modality, and Aguwa and Liechty12 found that those entering by the “back door” were less likely to use OMT on patients. These “allopathic ideologists” may feel ill prepared to diagnose and treat structural problems, and therefore are unmotivated to incorporate OMT into their clinical practices. In our study, we found that the diminished use of OMT among specialists was, in part, a function of a perceived inadequacy in OMT training. Interestingly, this ostensible cause-and-effect relationship between training and OMT use did not predict OMT use among family physicians.
While our data depict a general diminution in OMT practice among all osteopathic physicians, the decrease is most significant among specialists. Meyer and Price8 suggested that the trend in graduate osteopathic medical training has shifted from an emphasis on traditional family practice to an emphasis on specialty training. This major realignment in osteopathic graduate training has contributed to a growing ennui and disinterest in using OMT within the various specialties. Physicians who felt uncertain or unprepared to integrate OMT into their practices (as did 60% of the specialists in our study) and who were doubtful about their manipulative skills (as were 36% of our specialists) obviously were not inclined to incorporate OMT into their treatment regimens.
In recent years there has been a noticeable decrease in the number of functioning osteopathic hospitals due to financial failures, mergers with allopathic hospitals, etc., while at the same time there has been an increase in osteopathic graduates due to larger entering classes and the establishment of several new osteopathic schools. The osteopathic training institutions can no longer provide adequate postgraduate training opportunities for the growing number of graduates. As a result, osteopathic graduate medical education is occurring more and more in either allopathic or mixed-staff institutions; since 1985, more osteopathic graduates have trained in Accreditation Council for Graduate Medical Education—approved residency programs than in American Osteopathic Association residency programs.8
One result of this shift is that osteopathic educators are losing opportunities to relate the basic osteopathic principles to graduate trainees.14 Our data support the notion that the use of OMT has not been nurtured and promoted effectively as an important treatment component in all areas of osteopathic clinical practice. We found statistically significant differences in the use of OMT between family physicians and specialists based on the orientations of the institutions where they took their postdoctoral training. Graduates who trained in allopathic programs incorporated OMT less in their practices.
That the use of OMT, the key distinguishing feature of the osteopathic profession, is rapidly becoming a lost art among osteopathic family physicians and specialists should not come as a surprise to anyone associated with the osteopathic profession. Other authors have written effectively about the lack of instruction and reinforcement of osteopathic principles and practices along the continuum of osteopathic education.1,9–13 Gevitz1 has faulted the profession for not clearly defining and teaching the unique aspects of what it means to be an osteopathic physician. Consequently, many DOs emerge from predoctoral training unable to differentiate between their chosen profession and the allopathic profession. Therefore, when making a decision with respect to postdoctoral education, it is understandable that many trainees believe they have nothing to lose, and perhaps more to gain, by taking an allopathic residency.
As Gevitz1 has pointed out, the ramifications of the “asymmetric expansion” of the osteopathic profession, increasing numbers of graduates that cannot be accommodated in approved osteopathic internship or residency programs, means that many graduates, irrespective of their wishes, are precluded from the opportunity for a seamless predoctoral and postdoctoral osteopathic medical education. In short, assimilation of young DOs into allopathic residency programs effectively truncates the reinforcement of basic osteopathic concepts and similarly inhibits the opportunity to practice and cultivate OMT skills.
There are long-term consequences when the unique qualities of the osteopathic profession are ill defined and lack reinforcement. DOs who either have been previously rejected for allopathic undergraduate study or have not completed an approved osteopathic residency program are more apt to eschew membership in osteopathic organizations, and moreover, these physicians use OMT less in their practices than do their counterparts.12 And, unfortunately, a growing number of those trained in allopathic residency programs drift away from the osteopathic profession and become formally affiliated with allopathic associations. Interestingly, as early as 1988, osteopathic physicians were the fastest growing segment within the AMA, nearly doubling their membership since 1984,15 and by 1996, almost 20% of all osteopathic physicians were members of the AMA.16
Our findings raise an important question: Is the ultimate demise of OMT a fait accompli, or can the profession generate new interest and broader usage of this treatment approach? If OMT is to remain a treatment alternative in the U.S. health care system, it is incumbent on health care policymakers and health care professionals to promulgate the health benefits and the cost benefits of OMT. These groups should work together to develop strategies to address barriers to the use of OMT described by the specialists and family physicians in our study, such as a lack of time, poor reimbursement, unsuitable physical facilities, and an unsupportive philosophical environment. Strong didactic and clinical educational components should re-focus on the quintessence of OMT and the integration of this treatment regimen, not only into ambulatory osteopathic primary and specialty training, but also into various hospital settings.17 Furthermore, given the large number of osteopathic graduates trained in allopathic postgraduate programs, it is important that the unique aspects of osteopathic medicine be included as a specific training component within allopathic residency training.18
While the use of OMT by osteopathic generalists and specialists is but one element of a multifaceted philosophical and medical spectrum14 that identifies osteopathic physicians, it has been the development and application of OMT in a variety of medical settings that has most often been viewed outside the profession as the core tenet of osteopathic medicine. But, as Fry19 has noted, osteopathic medicine's ongoing fight to achieve parity and nurture acceptance by the allopathic profession has resulted in a major casualty—the loss of OMT and its symbolic importance.
Osteopathic physicians historically have resisted assimilation by the allopathic profession. But in their search for acceptance and respectability within the field of medicine, many osteopathic physicians have embraced philosophic and treatment aspects of their allopathic counterparts at the cost of those functions reflected in Andrew Taylor Still's foundational precepts. As this trend continues, our evidence suggests that osteopathic family physicians and specialists are increasingly disenamored with the use of OMT in their practices. As OMT becomes a lost art, the future of osteopathic medicine in the 21st century is in jeopardy. To survive as a unique medical entity, the osteopathic profession must curtail the ongoing exodus of young graduates from the philosophical roots of the profession, and both academic and clinical components must be re-infused with the efficacious values inherent in the osteopathic approach to patient care. If major action is not taken in this regard, it is conceivable that, during the 21st century, the unique aspects of this proud profession will not survive.