| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIAL COMMENT |
Behavioral Medicine Program, Department of Psychiatry; Columbia University Medical Center; New York, NY
Studies examining the association of attendance at religious services and mortality provide, without question, the strongest evidence in the field that attempts to establish a connection between religious characteristics and health outcomes. However, in a field so riddled with poorly conducted studies, this is like complimenting a person for being the healthiest patient in a hospice.
Into the fray comes the methodologically sound report by Blumenthal and colleagues (1). Using data from a subset of 503 patients from the ENRICHD (Enhancing Recovery in Coronary Heart Disease) study of the impact of psychosocial intervention for postmyocardial infarction (post-MI) patients who were depressed or socially isolated, these researchers reported that there was no relationship between attendance and health outcomes, in this case death or nonfatal MI. In addition, neither frequency of prayer nor total spirituality was related to outcomes. The authors deserve credit for bringing a degree of methodological rigor to a field where it is so often lacking.
Because of the highly selective nature of the ENRICHD trial, the question of the generalizability of these findings naturally arises. To qualify for ENRICHD, patients had to be depressed or report having low social support and be willing to participate in a time consuming and demanding treatment protocol. Thus, some may be tempted to dismiss the negative findings because of the study’s narrow sample. Most research on the relationship of attendance at religious services and health outcomes is based on epidemiological investigations of community samples. Some have concluded that the evidence from these studies is "persuasive" (2) but even in the well-conducted studies, the findings are not consistent across populations and often hold only in subgroups.
For example, Hummer et al. found that after adjusting for confounders and covariates including functional status and social connection, frequency of religious attendance was inversely associated with mortality in a study of over 21,000 subjects (3). However, the protective effect was entirely absent for patients with cancer and only marginally significant for patients with heart disease, the two diseases that account for the bulk of deaths in the United States. Oman and Reed found that in a community sample of 2023 affluent, largely white adults over age 55 in Marin County, California (4), religious attendance was associated with reduced mortality in a multivariate model (RR = 0.76, 95% CI = 0.62–0.94), an effect seen for both men and women. However, the magnitude and significance of the effect varied depending on the definition of attendance they used and whether the model included other indices of social engagement. In the Tecumseh Community Health (5) and Alameda County (6) studies, frequency of attendance at religious services was inversely associated with mortality, but after controlling for all relevant covariates, this relationship held only for women. For men but not women in the Tecumseh study, frequency of attendance at meetings of volunteer organizations was associated with reduced mortality whereas religious attendance was not (5). In a study by Schoenbach (7), the effect of religious attendance on mortality was seen primarily only for white men. In the Duke cohort (N = 3968) of the Established Populations for Epidemiologic Studies of the Elderly (EPESE) study, the effect of religious attendance was significant in the multivariate model, but in the full model, the effect for men achieved only marginal significance (RR = 0.83, 95% CI = 0.69–1.00) (8). In the New Haven EPESE cohort, no such association was found (9) while more generally, social and productive activities were associated with reduced mortality (10).
Attendance at services may be the most ubiquitous metric in the field of religion and health, not because it is the best, but because it is so very easily measured. It is a simple quantifiable index that suffers from two principal problems that may explain the inconsistencies in these results.
The first problem is associated with construct validity, epitomized by a remark attributed to Garrison Keillor: "anyone who thinks that sitting in church makes you a Christian must also think that sitting in a garage makes you a car." Attendance at religious services is used as a proxy for religious devotion but people attend services for any number of reasons: out of habit, to make social contact, to promote business connections, to satisfy the demands of family members, and out of religious commitment, to mention only a few. The fact that many recent studies report associations between greater frequency of attendance and lower mortality cannot overcome the difficulties in interpretation associated with the multiple meanings of attendance.
The second problem is measurement error. Virtually all studies reporting relationships between attendance at religious services and health outcomes rely on self-reported estimates of attendance. Typically, participants are asked how often they attend services, and are given a series of responses from which to select: more than once a week, once a week, several times a month, etc. When collected during interviews, either in-person or over the telephone, responses to questions like this are subject to self-presentation bias and several lines of research have demonstrated convincingly that reports of attendance at religious services are correspondingly over-estimated (11–14). Although the impact of this measurement error is unclear, it nonetheless indicates the need for caution in interpreting these findings.
There is, in addition to the methodological concerns about this literature, significant subtext that relates to an American society increasingly enthralled with the irrational at the expense of science and reason (15). While this has emerged largely in the context of the "controversy" over the origins of the universe and of life on earth, it has spilled over into medicine as well. Curlin et al. recently reported that in a nationally representative survey of 1144 physicians, 54% indicated that they believed that God or another supernatural being intervenes in the health of their patients (16). In the same survey, 14% of physicians reported that their personal religious beliefs justify withholding from their patients information about the existence of a perfectly legal procedure, eg, abortion in the case of failed contraception, let alone referral to a provider who would perform the procedure (17).
Scientific research does not take place in a vacuum. Like it or not, claims that religious devotion is associated with better health are used to support efforts to transform American medicine in ways that violate ethical standards and conflate the roles of physicians and the clergy. As Curlin et al. point out, "If physicians’ ideas translate into their practices, then 14% of patients—more than 40 million Americans—may be cared for by physicians who do not believe they are obligated to disclose information about medically available treatments they consider objectionable. In addition, 29% of patients—or nearly 100 million Americans—may be cared for by physicians who do not believe they have an obligation to refer the patient to another provider for such treatments" (p. 597, (17)). This is not an idle warning. West Virginians for Vaccination Exemption, an advocacy organization, reports that 14 current vaccines, including those for hepatitis, chickenpox, measles, polio, and rabies, derive from aborted fetal tissue and therefore, parents should be permitted to conscientiously object and refuse to allow their children to receive them. According to a bill recently introduced before the Wisconsin legislature, a doctor could refuse not only to administer the chickenpox vaccine to a child but also could refuse to inform the child’s parents about it by claiming that the vaccine violated his or her religious beliefs (18). Such objection holds the wellbeing of the larger society hostage to the moral views of a small minority by risking the resurgence of epidemic diseases we conquered decades ago.
In addition to influencing the availability of medical procedures, religious intrusion into medical practice threatens to violate the norms of patient autonomy by manipulative or even coercive means. Already, the Christian Medical and Dental Association, a professional society half as large as the American College of Cardiology, publishes a handbook that instructs physicians on how to use their practices to evangelize (19). According to a recent article in the Des Moines Register, the Iowa City VA Hospital repeatedly attempted to convert a Jewish veteran to Christianity during hospitalizations over the past 2 years (20). In 2004, CBS News reported on a Colorado orthopedic surgeon who "requests" that patients pray with him while they are gowned and supine on the gurney, ready to be wheeled into surgery.
No one disputes that for a great many people, religion provides comfort in times of difficulty, illness related or otherwise. Whether medicine should be involved in this, however, is another matter entirely. Researchers who examine the relationship between religious characteristics and health should keep in mind the words of the Bhagavad-Gita uttered by physicist Robert Oppenheimer, expressing the dismay and regret he experienced as he recognized for the first time the potential terrifying misuse of his scientific handiwork: "I am become Death, the destroyer of worlds."
DOI:10.1097/PSY.0b013e31812f79da
 |
REFERENCES |
|---|
 TOP REFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  G Paul The remote prayer delusion: clinical trials that attempt to detect supernatural intervention are as futile as they are unethical J. Med. Ethics, September 1, 2008; 34(9): e18 - e18. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
