|Home | About | Journals | Submit | Contact Us | Français|
The aim of this study was to compare religious characteristics of general internists, rheumatologists, naturopaths, and acupuncturists, as well as to examine associations between physicians' religious characteristics and their openness to integrating complementary and alternative medicine (CAM).
The design involved a national mail survey. The subjects were internists, rheumatologists, naturopaths, and acupuncturists.
Physician outcome measures were use of and attitudes toward six classes of CAM. Predictors were religious affiliation, intrinsic religiosity, spirituality, and religious traditionalism.
There was a 65% response. Naturopaths and acupuncturists were three times as likely as internists and rheumatologists to report no religious affiliation (35% versus 12%, p < 0.001), but were more likely to describe themselves as very spiritual (51% versus 20%, p < 0.001) and to agree they try to carry religious beliefs into life's dealings (51% versus 44%, p < 0.01). Among physicians, increased spirituality and religiosity coincided with more personal use of CAM and willingness to integrate CAM into a treatment program.
Current and future integrative medicine will be shaped in part by religious and spiritual characteristics of providers.
Religion, spirituality, and complementary and alternative medicines (CAM) are related to one another in complex ways. Religious practitioners and CAM supporters are critical of conventional biomedicine's1 reductionism and impersonality.2–4 Snyderman and Weil, early advocates of integrative medicine,5 suggest “there are many complex clinical conditions that are neither well understood in mechanistic terms nor effectively treated by conventional therapies.”5 It is notable that CAM has flourished in the treatment of chronic refractory conditions such as musculoskeletal disorders and pain syndromes.6–10 Additionally, reflecting religious traditions, CAM supporters see patients as “whole persons—spiritual beings.”5 Thus, we might expect those who are more religious to be attracted to CAM practices.
Yet religious persons may be skeptical of some CAM practices. CAM proponents often promote a universal notion of spirituality over particular and concrete practices of religion. In Astin's landmark study, a strong predictor of CAM use was belonging to a subculture called “cultural creatives.”8 In contrast to “moderns,” cultural creatives criticized conventional medicine because of its dependence on technological consumerism and its reductionist, materialist illness paradigms. In contrast with “traditionalists,” cultural creatives were committed to individualized and “open-minded,” “non-churched” forms of spirituality rather than to formal or “organized” religion.11,12 In fact, the notion of “openness”—whether spiritual openness or a more basic “openness to new paradigms”—is related to use of CAM and is also espoused as a corrective to conventional medical orthodoxy.13,14 Thus, those with this spiritual “openness” may be drawn to CAM practices that have spiritually oriented roots,15 whereas some religious persons may be repelled by what they perceive as connections to foreign religious and spiritual sources.
We found no published studies of the religious characteristics of CAM practitioners. One study16 found that religiosity/spirituality was more important to holistic physicians (who tended to use CAM practices) than to family practitioners (who tended not to use CAM). Studies have found that physicians who believe CAM practices are legitimate and effective are more likely to recommend them to patients,10,17 but little is known about the characteristics of physicians associated with such recommendations. A study of physicians in Sweden found that those who believe the scientific worldview is limited are more likely to promote CAM practices.18 An “integrative medicine questionnaire” for U.S. physicians found that openness to new ideas and paradigms was associated with openness to CAM practices.13 A study of users and practitioners of CAM in the United States found that they differ from others in “philosophy, orientation, and worldview.”14 And Barrett has argued that “While political and economic forces are clearly involved, it is the underlying belief system, worldview, or philosophical orientation of individuals—replete with values, preferences, prejudices and desires—that will control the direction and scope of this social process” (i.e., the integration and assimilation of CAM).19 These observations lead us to hypothesize that physicians who identify themselves as highly “spiritual” will be more likely to endorse CAM practices than those less spiritual, and that self-identified religious traditionalism will be associated with less willingness to integrate CAM practices. To test these hypotheses, and to compare the religious characteristics of CAM providers and physicians, this study examines data from a national survey of two groups of CAM practitioners, acupuncturists and naturopaths, and two groups of conventionally trained physicians, internists and rheumatologists.
A self-administered questionnaire was mailed to a stratified random sample of 1200 practicing U.S. physicians (600 general internists, 600 rheumatologists) under the age of 65, and 1200 licensed U.S. CAM providers (600 acupuncturists, 600 naturopaths). Between May 20 and June 1, 2007, each sampled provider received a cover letter, an informational sheet, a letter of support from a relevant professional society, and the questionnaire. A $20 incentive accompanied the initial mailing. After 6 weeks, a second packet was sent to nonrespondents.
Physician samples were obtained from the American Medical Association Physician Masterfile. We included general internists and rheumatologists because they are likely to encounter common chronic conditions for which patients often use CAM therapies.8 CAM provider samples were obtained from national databases of licensed acupuncturists and naturopaths compiled from state boards. At the time of sampling (Fall 2006), 13 states licensed naturopathic physicians; 44 states licensed acupuncturists. Eligibility was defined as holding an active license for the designated CAM profession and current involvement in patient care.
Details of questionnaire development are reported elsewhere.20 Primary criterion measures included physicians' report of having used for their own health and/or having recommended to patients each of six different classes of CAM: spinal manipulation (e.g., chiropractic), acupuncture, energy medicine (e.g., Reiki), meditation practices (e.g., yoga), glucosamine ± chondroitin, and body work (e.g., massage, Shiatsu, etc.). These CAM therapies are commonly used for chronic musculoskeletal conditions and correspond to a spectrum of CAM treatment categories as defined by the National Institutes of Health.21 We also asked physicians, “How likely would you be to recommend each of these therapies as one component of a comprehensive treatment package for patients with chronic back pain or joint pain?” Response categories ranged from “Very likely” to “Would never recommend.” We chose the clinical scenario of chronic back or joint pain both because multiple studies have indicated that CAM is often used for chronic musculoskeletal pain conditions,6–10 and because these are conditions that most general internists and rheumatologists will frequently treat in their clinical practices.
Primary predictors were physicians' religious affiliation, intrinsic religiosity, and spirituality. Affiliation is the most widely used metric of religion22,23 and names the tradition that a physician follows or to which the physician belongs. Intrinsic religiosity is a construct intended to measure the extent to which a person embraces their religion as the “master motive” that guides and gives meaning to their life.24 In this study, it was measured as level of agreement with the statement, “I try hard to carry my religious beliefs over into all my other dealings in life” (strongly agree to strongly disagree). This question is similar to one included in the Duke Religion Index.25 Spirituality was measured by asking physicians, “To what extent do you consider yourself a spiritual person?” (very spiritual to not spiritual at all). Respondents' age, sex, region, and specialty were included as controls in multivariate analyses. Because some religion categories had few members, affiliation was collapsed into None, Protestant, Catholic, Jewish, and other. Finally, to explore the influence of religious traditionalism, we divided Catholic, Jewish, and Protestant physicians as follows: Traditional Catholic physicians were those who identified themselves as traditional Catholics (n = 111), Traditional Jewish physicians were those who identified themselves as Orthodox or Conservative (n = 72), and traditional Protestants were those who identified themselves as Evangelical (n = 75). Catholic, Jewish, and Protestant physicians who did not put themselves into the above respective categories were considered nontraditional for the respective groups.
All data were double entered and 100% verified. We first generated overall population estimates for providers' religious characteristics and for physicians' agreement with each of the criterion measures. We then utilized the Pearson χ2 test (for unordered predictors), and Mantel-Haenszel χ2 test (for ordered predictors) to examine the bivariate associations between each predictor and each criterion measure. Finally, we created a scale indicating physicians' willingness to integrate CAM into their practices by averaging their responses for each of the six CAM practices (from “Very likely” to “Would never recommend”), with the resulting scale ranging from 1 to 4. The sample mean score was substituted for missing responses to each question. The α coefficient was 0.78 and the mean was 2.56 (standard deviation = 0.25). This score served as the dependent variable in multivariate regression analysis. All analyses were conducted using the statistical software Stata/SE 10.0 (Stata Corp., College Station, TX, 2007). This study was approved by the National Institutes of Health Office of Human Subjects Research and the University of Massachusetts–Boston Institutional Review Board.
Among the 2400 providers sampled, 1561 responded (65% overall response rate): 440 (73%) acupuncturists, 442 (74%) naturopaths, 334 (56%) general internists, and 345 (58%) rheumatologists. Provider groups varied in background and demographics (Table 1).A one-way analysis of variance of age across provider groups was statistically significant (F(3,1489) = 34.66, p < 0.001). Post-hoc tests showed that naturopaths are significantly younger than the other three groups (all p's < 0.001) and rheumatologists are the oldest (all p's < 0.02). Acupuncturists are older than naturopaths (p < 0.001), but acupuncturists and internists do not differ significantly in age (p = 0.57). Women make up a larger proportion of CAM providers (63%) than physicians (27%, p < 0.001). Acupuncturists and naturopaths were concentrated in the West. Internists were distributed about evenly across regions, while rheumatologists were concentrated in the South and Northeast.
Acupuncturists and naturopaths differed in religious background (p < 0.01). The modal religious affiliation for both is “none” (42% for naturopaths, 29% for acupuncturists), but 19% of acupuncturists identified as Buddhist, compared to 5% of naturopaths. Physicians also differed in religious background (p < 0.01). The modal group for rheumatologists was Jewish (31%), whereas internists most frequently identified as Catholic (29%). CAM providers did not differ in spirituality, but a higher percentage of acupuncturists strongly agreed with the statement that they tried to carry their religious beliefs into the rest of their lives (27%) compared to naturopaths (20%, p < 0.01). Physicians did not differ in spirituality or religiosity.
Compared to physicians, CAM providers were three times as likely to report having no religious affiliation (35% versus 12%, p < 0.001). Yet, as seen in Figure 1, CAM providers were more than twice as likely as physicians to consider themselves “very spiritual” (51% versus 20%, p < 0.001) and to strongly agree that they try hard to carry their religious beliefs over into all other dealings in life (23% versus 11%). With respect to our measure of religious traditionalism (data not in tables): 33% of Catholic CAM providers considered themselves traditional versus 54% of Catholic physicians (p = 0.005); 40% of Protestant CAM providers considered themselves evangelical compared to 22% of Protestant physicians (p = 0.003); and 15% of Jewish CAM providers were Orthodox or Conservative, compared to 41% of Jewish physicians (p < 0.001).
Compared to CAM providers, physicians reported lower levels of personal use of CAM practices (mean across 6 practices, 1.12, compared to 4.8 for CAM providers; p < 0.01) lower levels of referral (mean across 6 practices, 3.5, compared to 5.2 for CAM providers, p < 0.01), and lower willingness to integrate (mean scale score across 6 practices, 2.56, compared to 3.34 for CAM providers, p < 0.02).
Tables 2 and and33 display measures of physicians' use, recommendation, and willingness to integrate each CAM practice, stratified by physician spirituality and intrinsic religiosity. Integration was dichotomized for presentation as follows. Thirty percent (30%) to 50% of physicians gave responses of “very likely” to five of the six CAM practices, so that category was used as the cut point for bivariate analyses. For energy medicine, the cut point was between physicians who were “somewhat” or more likely (34%) and those who were “not very likely” or “would never recommend.” Higher spirituality was associated with greater likelihood of use and willingness to integrate all six CAM practices (all p's ≤ 0.06); there were fewer associations with recommending CAM practices (p < 0.05 only for meditation and glucosamine ± chondroitin). Intrinsic religiosity had associations that were similar to those of spirituality in direction, but generally lower in magnitude.
In parallel analyses, evangelical Protestants were somewhat less likely than nonevangelicals to report having referred for acupuncture (49% versus 67%, p < 0.05), but did not differ at p < 0.05 on other measures. Traditional Jews were less likely to report personal use of acupuncture (3% versus 13%, p = 0.03), but did not differ from nontraditional Jews on other measures. Traditional Catholics were less likely to report personal use of meditation (17% versus 41%, p < 0.01) and body work (35% versus 56%, p < 0.01), but did not differ significantly from nontraditional Catholics on other measures.
Table 4 shows results for multivariate models of physicians' willingness to integrate CAM practices into a comprehensive treatment program for back or joint pain. The first model included specialty, region, sex, and age. Internists compared to rheumatologists and women compared to men said that they would be more likely to incorporate CAM into their practice. Those living in the Northeastern and Southern regions of the country said that they would be more likely than those in the West to integrate CAM. Model 2 included all of the variables in Model 1 but added Religious Affiliation. The R2 increased significantly by 0.03 (F(5,605) = 23.68, p < 0.01). The only religious group with a significant coefficient was the Roman Catholic/Eastern Orthodox physicians who, compared to those with no religious affiliation, indicated they would be more likely to incorporate CAM into their practice.
Model 3 introduced the Religiosity variable. The R2 increased significantly by 0.01 (F(3,602) = 8.55, p < 0.01). When Religiosity was included, the effects of Specialty and Region remained statistically significant but the effect of Religious Affiliation was no longer significant. Those with the lowest level of intrinsic religiosity were the least willing to integrate CAM practices. Model 4 included everything from Model 3 and added responses to the Spirituality question. The R2 increased significantly by 0.01 (F(3,599) = 9.35, p < 0.01). When Spirituality was included in the model, the Religiosity coefficients were no longer statistically significant, whereas willingness to integrate CAM was higher for those who said that they were moderately or very spiritual, compared to those who said that they were not spiritual at all.
With increased demand for and use of integrative medicine, the medical profession should be aware of those characteristics of physicians that are associated with greater willingness to integrate CAM practices. We found that compared to general internists and rheumatologists, acupuncturists and naturopaths are three times as likely to have no formal religious affiliation, yet they score more highly on measures of religiosity and spirituality. Moreover, among the two physician groups, those who are more spiritual are more likely to have personally used and are more likely to say they would integrate different CAM practices. Religiosity, religious affiliation, and religious traditionalism do not appear to strongly influence physicians' willingness to integrate CAM practices, especially after accounting for spirituality.
This study finds that like patients who use CAM treatments,8 practitioners tend to embrace a self-conscious and intentional spirituality outside of organized religion. After finding this pattern among the general public a decade ago, Astin hypothesized that “growing interest in alternative medicine may represent a type of cultural (Kuhnian) paradigm shift regarding health beliefs and practices . . . part of a broader value orientation and set of cultural beliefs, one that embraces a holistic, spiritual orientation to life.”8 Our findings support this hypothesis. The spiritual dimensions of CAM practices may repel those who have a strict secular outlook: Those physicians who are least religious and spiritual are least likely to embrace CAM. In addition, these data suggest that interest in CAM is related to shifts away from religious orthodoxy and tradition toward more individualized notions of spirituality. For example, Catholic and Jewish acupuncturists and naturopaths are less likely to be traditional or Orthodox/Conservative, respectively, than are Catholic and Jewish physicians. Yet, the association for Protestants was in the reverse direction from what we had anticipated. Evangelicals tend to be more traditional theologically than nonevangelical Protestants; yet, Protestant CAM providers are more likely to be evangelicals than Protestant physicians. This may suggest that the innovativeness and eclecticism of evangelicalism resonates with the use of CAM in a way that outweighs any repellant effect of religious traditionalism.
We were surprised to find, however, that religious traditionalists among Christians and Jews were as willing as their less traditional counterparts to integrate CAM practices.26 Use of CAM may be associated with “religious dissent.”27 Eskinazi noted that many unconventional healing practices include a sense of a “vital force” and a “belief in a unity underlying all diversity,”15 both of which are concepts more congruent with Eastern traditions such as Buddhism and Hinduism than with Abrahamic religions. Kaptchuk noted that unconventional healing practices parallel what in the past was called the “nature heresy,” and includes “a form of religiosity,” “liturgy,” and the promise of “transcendence.”28 Ayurvedic medicine is part of the Hindu tradition,15,26,29 and is based on a power called prana.28 Reiki uses “spiritual energy with innate intelligence” and can involve contact with “spirit guides.”29 Forms of meditation have roots in Buddhism and Hinduism,15 and yoga has been called “a theological practice that leads to unification with the divine.”26 Other practices are not as clearly rooted in religious traditions, but make tacit claims to being built on an understanding of spiritual reality. For example, New Age healing practices speak of “psychic,”28 “astral,”28 or “esoteric”30 energies and have been described as seeking “a ‘third way,’ a ‘spiritual science,’ between revealed biblical religion and ‘atheistic-materialism.’”30
Because of this, we hypothesized that Jews and Christians who are more traditional would tend to reject CAM practices regardless of the efficacy of those practices. We found, on the contrary, that religious physicians are more open to integrating CAM regardless of whether they are part of the more traditional or conservative elements of their traditions, though spirituality—at least as applied by physicians to themselves—seems to have the stronger influence on physicians' approaches to CAM. The openness of religious believers to CAM may reflect religious traditions' and CAM supporters' shared criticism of scientific reductionism and shared emphasis on the spiritual dimensions of the human condition.
This study is another in a growing body of research that underscores the important roles religion and spirituality play in shaping physicians' clinical judgments. Prior studies suggest that physicians' religious commitments influence their judgments about the legitimacy of different ethically controversial clinical practices. This study highlights a modest but real effect of physicians' religious and spiritual characteristics on their willingness to integrate CAM practices, suggesting that as physicians' familiarity with and experience of CAM practices grow, varied and eclectic patterns of integrating CAM will likely emerge, shaped in part by the self-conscious spirituality of physicians. Some proponents of integrative medicine have welcomed this prospect, suggesting that “[by incorporating] different approaches to treating patients . . . [physicians] can balance their own personal and professional values and develop individualized models of health and health care provision.”31 In light of these findings, proponents of CAM may want to focus their efforts on healthcare providers who are more self-consciously spiritual, and/or focus on spiritual themes in encouraging such providers to consider integration of CAM into their practices. Yet, others may be disturbed by the influence of physicians' “personal values” on their professional practices.
The study has limitations. Physician results apply only to internists and rheumatologists. The relationship between spirituality and CAM acceptance may vary for physicians of other specialties. Similarly, the religious characteristics of other CAM practitioners may be different from those of acupuncturists and naturopaths. Related to this, single-item indicators of spiritual and religious characteristics were used. More reliable estimates may have been obtained with multiple-item indicators. We chose six classes of CAM, but future studies should examine other classes of CAM and should look at differences within classes. For example, the relationship between religious traditionalism and one meditation practice (e.g., mindfulness meditation) may be different from the relationship with another (e.g., yoga). Finally, this study is subject to the limitations of cross-sectional surveys, including survey question and response bias, and no causal inferences can be drawn from our findings.
Physicians should remain conscious of the ways in which their own beliefs, attitudes, and commitments may shape their judgments of CAM. Current and future forms of integrative medicine will be shaped in part by physicians' religious and spiritual characteristics.
Funding was provided by the National Center for Complementary and Alternative Medicine (NCCAM) and the Department of Bioethics, National Institutes of Health, Bethesda, MD. NCCAM was not involved in data collection, analysis, or writing of the manuscript. Development and implementation of the survey was performed by the Center for Survey Research, University of Massachusetts, Boston, MA. Dr. Curlin is also supported by a grant from NCCAM (1 K23 AT002749).
No competing financial interests exist.