The spiritual and religious identities, beliefs, and practices of physicians are beginning to be explored generally and as they relate to clinical relationships and decision making. A recent study found that religious physicians who for moral reasons oppose certain medical treatments such as administering sedation to dying patients are less likely to refer patients or disclose relevant information.1 In studies about withdrawal of life support, abortion, and other issues, religion has been associated with the decisions physicians make.2–6 In a study of Pennsylvania internists, for example, after controlling for other independent variables, Catholic and Jewish physicians were less willing to withdraw life support, whereas younger clinicians and those practicing in tertiary care centers were more likely to do so.7 Despite these findings, it is only recently that descriptive information has been gathered about the spiritual and religious identities, beliefs, and practices of physicians in the United States.8 A survey based on a random national sample of physicians found that 55% of surveyed physicians felt that their religious beliefs influenced their medical practice.9 Physicians have become aware that a patient’s spiritual and religious beliefs and practices may play a role in coping with disease, medical decision making, and other health-related processes.10–11 Interest has also recently increased about the religious beliefs and practices of general and specialist physicians and about how these belief systems may influence how they care for patients.
Pediatricians care for nearly 100 million U.S. children, often playing an important role in the development of children and families.12 Data exist showing that religion and spirituality seem to influence pediatricians’ approaches to care. A single institutional survey of pediatric residents and faculty showed that those with strong religious and/or spiritual orientation demonstrated more positive attitudes toward incorporating religion and spirituality into their pediatric practice.13 A survey of pediatricians concerning the care of critically ill newborns found that religious affiliation influenced certain treatment decisions. Catholic pediatricians, for example, were less likely than other study participants to be swayed by parental opinions.14 A multinational survey of neonatologists found that those who rated religion as extremely or fairly important were less likely to have ever withheld intensive care or withdrawn mechanical ventilation.15 Although some evidence and observations16,17 suggest that pediatricians’ religious and spiritual backgrounds may influence, in subtle or more overt ways, their approach to patient care, little is known about their personal spiritual and religious identities, beliefs, and practices.
This study builds on previous research as the first detailed survey about personal religion and spirituality among academic pediatricians in 13 American academic medical centers. We gathered information from this subset of pediatricians because of their influential positions as teachers, researchers, clinicians, medical writers, and pediatric opinion leaders.
Physicians included in this study were selected from departments of pediatrics at 13 “honor roll” hospitals, as defined by the 2004 US News and World Report.18 Preference was given to hospitals associated with medical schools and large research universities. The methodology to determine “honor roll” distinction combined hospital reputation, mortality data, and patient-care-related factors. When six or more specialty areas showed “exceptional breadth of excellence,” the hospital placed greater than two standard deviations above the mean and was granted honor role status. The hospitals we studied were Stanford Hospital and Clinics, The Johns Hopkins Children’s Center, UCLA Medical Center, the University of Michigan Medical Center, Duke University Medical Center, University of Washington Medical Center, Mayo Clinic, Cleveland Clinic, New York-Presbyterian Medical Center, Massachusetts General Hospital, Hospital of the University of Pennsylvania, University of California San Francisco Medical Center, and Barnes-Jewish Hospital.
We compiled a physician population that included every faculty-level academic general pediatrician listed on the Web pages of the departments of pediatrics in hospitals in the sample, for a total of 565 physicians. Because we were relying on a Web-based survey, that sample was then narrowed to the 458 physicians who had e-mail addresses we could access through their departments and/or hospitals. These physicians were then assigned a random number, the random numbers were sorted numerically, and 208 physicians (45% of the total sample) were selected for participation in the survey, a manageable sample size based on the resources available. The study was approved by Rice University’s institutional review board. During a seven-week period from May through June 2005, an initial contact letter was sent to each participant containing a $15 cash preincentive. Each subject received a unique identification code with which to log onto a Web site and complete the survey. After five reminder emails, the research firm commissioned to field the survey, Schulman, Ronca, and Bucuvalas, Inc., called physicians requesting participation over the phone or Web up to a total of 20 times, as is the norm in social scientific survey research.
The survey asked 34 closed-ended multiple-choice questions about spiritual and religious identities, beliefs, practices, ethics, and the intersection of religion and science in the respondent’s field. Many of these questions were replicated from the University of Chicago’s 1998 General Social Survey (GSS), the most recent GSS to include a detailed set of questions about religion, and report findings for several thousand participants, which enables comparisons with information about the general public.19 Analysis was conducted in two stages. First, the answers that respondents and members of the American public in the GSS gave to identical survey questions were compared using chi-square/t test. Missing data are noted in the tables. Second, logistic regression analyses were conducted to determine factors that might lead academic pediatricians to believe religion/spirituality influenced interactions with patients and colleagues, as measured through their degree of agreement with the statement, “My spiritual or religious beliefs have an influence on how I interact with colleagues and patients.” We controlled for the following independent variables: gender, marital status (currently married/not), number of children in the household under age 18, and racial background (white/nonwhite). We measured religiosity in terms of a closed-ended survey question about religious identity which offered a wide range of possible responses which matched the options available when the same question was asked in the GSS. We collapsed these responses into religious identification as Protestant, Catholic, Jewish, other, or none. Also included was a dummy variable indicating whether respondents had attended religious services monthly or more frequently in the prior year. We excluded from the logistic regression cases that were missing data on any of the included variables. The survey data were analyzed using the SPSS version 13 statistical computer package for Windows (SPSS Inc., Chicago, Illinois).
Overall, 116 pediatricians completed the survey (110 online, 6 by telephone), with a resulting response rate of 56%. Basic demographic information about survey respondents is summarized in Table 1.
Table 2 describes the spiritual and religious characteristics of respondents and the American public as described in the GSS in 1998. Physician respondents numbered 116, and the number of respondents from the general public ranged from 1,284 to 2,832. Of the 87% of physician respondents raised in a religious tradition, 48 (41.4%) were Protestant, 29 (25.0%) were Catholic, 18 (15.5%) were Jewish, and 7 (6.0%) were other. Only 9 (7.8%) were raised without a religious tradition. When they were growing up, the majority reported that religion was very important (31, 26.7%) or somewhat important (53, 45.7%) in their family. When asked about their current religious identification, however, many academic pediatricians reported their religious preference as none (32, 27.6%), followed by Protestant (27, 23.3%), Catholic (23, 19.8%), Jewish (20, 17.2%), and other (8, 6.9%), which included one Buddhist, one respondent who reported “another Eastern religion,” one Eastern Orthodox, two Unitarian Universalists, and one respondent who reported “other Christian religion.”
The religious identities of academic pediatricians were significantly different from those of the American public, as indicated by the GSS (Pearson chi-square 161.62, P < .001). Most notably, a larger proportion of academic pediatricians (32, 27.6%) than members of the broader American public (396, 14.0%) reported having no religious preference (t = 4.086, df = 2,946, P < .001). Further, fewer academic pediatricians were Protestant (27, 23.3%; t = −6.5, df = 2,946, P < .001) than were members of the American public (1,524, 53.8% Protestant), and many more academic pediatricians were Jewish (20, 17.2%) than in the American public (50, 1.8%; t = 10.92, df = 2,946, P < .001).
Although 32 (nearly 28%) academic pediatricians reported no religious preference, more than one half believed in God: 35 (30.2%) had no doubt about God’s existence, 24 (20.7%) believed in God but had some doubts, 6 (5.2%) believed in God sometimes, and 8 (6.9%) believed in a higher power that is not God. Only 13 (11.2%) reported not believing in God, and 17 (14.7%) said they did not know whether God existed and there was no way to find out. There were notable differences between the pediatricians and the American public in whether or not they believed in God (Pearson chi-square 71.54, P < .001). A lower proportion of academic pediatricians compared with members of the American public believed in God without doubts (30.2% pediatricians, 60.4% of the public, t = −6.392, df = 1,398, P < .001).
When asked about the Bible, a few similarities were observed in the responses of academic physicians and the public, but the answers each group provided also revealed markedly different views (Pearson chi-square 51.15, P < .001). A large proportion of both physicians (46.6%) and members of the public (45.6%) believed that the Bible is the inspired word of God but that not everything in it should be taken literally (t = 0.205, df = 1,398, P = .839). Nearly one third of physicians (35.3%) instead described the Bible as an “ancient book of fables recorded by man.” A much smaller portion of the public (16.1%) described the Bible this way (t = 5.24, df = 1,398, P < .001). Likewise, only 2.6% of physicians, compared with 27.9% of the public, described the Bible as the actual word of God and believed that it should be taken literally, word for word (t = −6.037, df = 1,398, P < .001).
When responding to questions about spirituality more generally, academic pediatricians and members of the public gave somewhat similar responses (Pearson chi-square = 8.99, P = .061). The physicians differed very little from the general public when asked to describe their level of engagement in spirituality. Specifically, 37.1% of academic pediatricians and 39.5% of the public described themselves as moderately spiritual (t = −0.519, df = 1,559, P = .604); 33.6% and 25.3%, respectively, as slightly spiritual (P > .05 (t = 1.961, df = 1,559, P = .050); and 15.5% and 11.8%, respectively, as not at all spiritual (t = 1.17, df = 1,559, P = .242). On the other hand, only 12.1% of academic pediatricians described themselves as very spiritual, compared with 21.7% of the general public (t = −2.46, df = 1,559, P < .05).
As shown in Table 2, academic pediatricians and the general public seemed to attend religious services, outside of weddings, baptisms, and funerals, in different proportions (Pearson chi-square 32.93, P < .001). The physicians seemed to have attended less frequently than members of the general population. Only 40.6% of academic pediatricians reported attending services six or more times in the prior year compared with 58.2% of the public (t = −3.81, df = 2,946, P < .001).
Despite relatively low levels of traditional religious service attendance, though, academic pediatricians did report engaging in private spiritual or religious practices in the prior year. More than half (61, 52.6%) reported privately praying; other private spiritual practices included relaxation techniques (45, 38.8%), private meditation (34, 29.3%), reading a sacred text (31, 26.7%), yoga (22, 19.0%), and other spiritual exercises (10, 9.7%). Only 20 (19.4%) did not engage in any of these practices.
When asked whether their spiritual or religious beliefs influenced interactions with patients and colleagues, more than one half (68, 58.6%) of academic pediatricians believed that they do to some extent. Nine (approximately 8%) had no opinion, and 36 (31%) believed that religious and spiritual factors do not influence these interactions. To investigate the factors that might influence these perceptions, logistic regression analysis was performed and is summarized in Table 3. Controlling for gender, marital status, parental status, racial background, and religious service attendance, the odds of pediatricians thinking their spiritual or religious beliefs influenced how they interacted with patients and colleagues were more than five times higher for those who attended religious services monthly or more in the prior year when compared with those who attended less than once a month (P < .05).
Our findings about the differences between the personal religious identities and spirituality of academic pediatricians and the general public will have several implications, if they are replicated in future studies. These physicians described themselves as more spiritual than religious, much like the physicians Curlin and colleagues studied.9 Although a much lower proportion of academic pediatricians cited a personal religious identity compared with the public, they were quite similar to the public in how they described themselves spiritually. Of interest, both the American public and a random physician sample differed notably from the academic pediatricians we surveyed on the question of religious identity.9 As evident in Table 2, despite often being raised in households where religion was important, more than one in three academic pediatricians rejected a religious identity compared with 14% of the public. Our survey questions did not address why these academics did not integrate their childhood religious identification into their adult identities. One possibility is that an academic career attracts researchers and scholars whose individual strengths may not include the emotional, spiritual, and psychological skills of caregivers who self-select to practice medicine in the community. Another is that an academic career socializes physicians in these ways. Whether frequent confrontation with suffering and dying children makes a religious or spiritually based worldview seem incompatible for some academic pediatricians awaits future study. Another interesting finding is the relatively stable proportion of Jewish identification. Given the fact that self-identification as Jewish is not only a religious but deeply held cultural identity, the stable proportions measured for childhood and adulthood were not unexpected.
The finding that the academic pediatricians who most regularly attended religious services felt religion/spirituality influenced their clinical practice stands out. Integration of one’s religious/spiritual beliefs into clinical encounters may occur in relation to personal spiritual development, but our data did not include an analysis of spiritual development. Whether physicians with more secular belief systems feel that nonreligious self-identification influences their practice of medicine will be of interest in future studies.
The results of our survey of the spiritual and religious identities, beliefs, and practices of some academic pediatricians raise many questions relevant to medical education, other health care professionals, patient-centered care, and integrative medicine. We are aware of two existing programs, one offered in two medical centers, and the other ongoing in 130 hospitals nationwide. Both have been created for health care providers to facilitate spiritual and religious understanding and growth which is one part of integrative medical practice. Clinical Pastoral Education (CPE) for Health Care Providers is a fully accredited program; at the Massachusetts General Hospital it is supervised by a coauthor (A.A.Z.). Here, clinicians learn about fundamental aspects of spirituality and religiosity especially as they may apply to the care of hospitalized patients. Two important goals of the CPE for Health Care Providers program are (1) to raise clinicians’ awareness of religious and spiritual beliefs and values, as these may impact patient care and decision making, and (2) to nourish clinicians’ ability to empathize with religious traditions or secular beliefs different from their own. As such, practitioners may find our data useful in assessing similarities and differences between their spiritual and religious beliefs and those of their patients. The other program, Schwartz Center Rounds, is an interdisciplinary conference regularly occurring in multiple U.S. hospitals. One patient is the focus for each discussion; an emphasis is placed on all issues related to providing compassionate care.
The survey data and analyses from our study present a first descriptive portrait of spirituality and religion in the lives of academic pediatricians, but they are limited in several ways. This sample of academic pediatricians is small, so population estimates are imprecise, as are measures of association with the broader American public. The sampling frame included only academic pediatricians at specific highly ranked institutions. These respondents may be somewhat different from academic pediatricians at other institutions and from pediatricians more broadly, making the results not easily generalized. However, as opinion leaders, teachers of pediatricians-in-training, and authors of original research and textbooks, the beliefs and practices of these pediatricians are relevant. Nonetheless, detailed study of larger numbers of pediatric generalists and subspecialists are a necessary next step in refuting or supporting the findings presented.
Dr. Catlin received grant support from the Louisville Institute, Louisville, Kentucky. Dr. Cadge was supported by the Robert Wood Johnson Foundation Scholars in Health Policy Research Program at Harvard University. This data collection was funded by grant #11,299 from the John Templeton Foundation, Elaine Howard Ecklund, PI.
1 Curlin FA, Lawrence RE, Chin ME, Lantos J. Religion, conscience, and controversial clinical practices. N Engl J Med. 2007;356:593–600.
2 Laine C, Davidoff F. Patient-centered medicine: A professional evolution. JAMA. 1996;275:152–156.
3 Aiyer AN, Ruiz G, Steinman A, Ho GY. Influence of physician attitudes on willingness to perform abortion. Obstet Gynecol. 1999;93:576–580.
4 Crane D. The Sanctity of Social Life: Physicians’ Treatment of Critically Ill Patients. New York, NY: Russell Sage Foundation; 1975.
5 Imber JB. Abortion and the Private Practice of Medicine. New Haven, Conn: Yale University Press; 1986.
6 Kagawa-Singer M, Blackhall LJ. Negotiating cross-cultural issues at the end of life: To go where he lives. JAMA. 2001;286:2993–3001.
7 Christakis NA, Asch DA. Physician characteristics associated with decisions to withdraw life support. Am J Public Health. 1995;85:367–372.
8 Curlin FA, Chin MH, Sellergren SA, Roach CJ, Lantos JD. The association of physician’s religious characteristics with their attitudes and self-reported behaviors regarding religion and spirituality in the clinical encounter. Med Care. 2006;44:446–453.
9 Curlin FA, Lantos JD, Roach CJ, Sellergren SA, Chin MH. Religious characteristics of U.S. physicians—A national survey. J Gen Intern Med. 2005;20:629–634.
10 Koenig HG, Larson DB, Larson SS. Religion and coping with serious medical illness. Ann Pharmacother. 2001;35:352–359.
11 Robinson MR, Thiel MM, Backus MM, Meyer EC. Matters of spirituality at the end of life in the pediatric intensive care unit. Pediatrics. 2006;118:e719–e729.
12 Goodman DC; Committee on Pediatric Workforce. The pediatrician workforce: Current status and future prospects. Pediatrics. 2005;116:e156–e173.
13 Siegel BS, Tenenbaum AJ, Jamanka A, Barnes L, Hubbard C, Zuckerman B. Faculty and resident attitudes about spirituality and religion in provision of pediatric health care. Ambul Pediatr. 2002;2:5–10.
14 Todres ID, Guillemin J, Catlin EA, Nordstrom A, Marlow A. Moral and ethical dilemmas in critically-ill newborns: A 20 year follow-up survey of Massachusetts’ pediatricians. J Perinatol. 2000;1:6–12.
15 Cuttini M, Nadai M, Kaminski M, et al. End-of-life decisions in neonatal intensive care: Physicians’ self-reported practices in seven European countries. EURONIC Study Group. Lancet. 2000;355:2112–2118.
16 Rosenthal MS. Cultural competency. JAMA. 2006;296:23–24.
17 Barnes LL, Plotnikoff GA, Fox K, Pendelton S. Spirituality, religion, and pediatrics: Intersecting worlds of healing. Pediatrics. 2000;106(4 suppl):899–908.
18 U.S. News & World Report. July 12, 2004 edition. “Honor Roll Hospitals.”
19 University of Chicago, National Opinion Research Center. General Social Survey. Available at: (www.norc.uchicago.edu/projects/gensoc.asp
). Accessed August 12, 2008.
20 Curlin FA, Roach CJ, Gorawara-Bhat R, Lantos JD, Chin MH. How are religion and spirituality related to health? A study of physicians’ perspectives. South Med J. 2005;98:761–766.