Pastoral Report Archives:

March, 2005

February, 2005

January, 2005

December, 2004

November, 2004

October, 2004

September, 2004

August, 2004

July, 2004

June, 2004

The College of Pastoral Supervision & Psychotherapy is a theologically based covenant community, dedicated to "recovery of the soul" and promoting competency in the clinical pastoral field.

« Religion, Spirituality and Medicine by Richard P. Sloan, Ph.D. | Main | Christmas 2001 »

December 3, 2001

Religion, Spirituality and Medicine by Richard P. Sloan, Ph.D.

Introduction of Dr. Sloan by Raymond Lawrence to 2001 CPSP Plenary

Dr. Sloan is the director of the Behavioral Medicine Program at the Columbia Medical Center. He is an Associate Professor in the Department of Psychiatry at the College of Physicians and Surgeons, Columbia University. His principal work focuses on identifying risk factors such as depression, hostility, and anxiety to heart disease.

In his address to the CPSP community, 16 March 2000, Dr. Sloan argued for a more cautious and considered approach to research into the health benefits of religious activities. He elaborated on the critical flaws in the recent research into the links between religion and health that have been widely touted recently by some religious and medical professionals. His paper Religion, Spirituality, and Medicine with T. Powell and E. Bagiella) in the British medical journal, The Lancet, February 20, 1999, received wide international attention. His subsequent paper published in the New England Journal of Medicine (with Larry VandeCreek, et alia, July 22, 2000) sent shock waves through the community promoting the link between health and religion in this country.

Presented by
Richard P. Sloan Ph.D
2001 CPSP Plenary

The possibility that religious activity might be associated with health benefits is an intrinsically interesting topic. Practically every week a story appears in the popular media that reports the health virtues of religious activity. I started noticing piec

es on this topic about four years ago. There were so many articles appearing that it made me suspicious. I also learned about the development of medical-school courses that teach students how to incorporate religion into health care.

So some three years ago, as a "night job," I started to investigate claims of health benefits of religion. Emilia Bagiella, a biostatistician, and Tia Powell, an ethicist from Columbia-Presbyterian Medical Center, joined me in this research. Ultimately, our investigation culminated in articles published in The Lancet (February 1999) and the New England Journal of Medicine (June 2000) on the harmful effects of having doctors prescribe religion.

We began by looking at some of the studies that claimed to demonstrate an association between religious activity and medical health. We found that the vast majority of these findings were based on poorly conducted studies. We further recognized a range of negative ethical implications associated with involving doctors in the practice of religion for example:

1. Coercion
2. Invasion of privacy
3. Doing harm
4. Discrimination

In no way do I view our research as a criticism of religion. In fact, I regard it as a defense of religion against interference by medicine. When medicine combines with religion in this way, it produces weak science, poor patient care and trivialized religion. Patients should not be deluded into believing that religious activities will promote better health. If patients have health-related religious concerns, physicians should refer them to the appropriate expert clergy as an internists refers patients to cardiologists for angioplasties. Medicine should simply stay out of religion and let patients behave religiously the way that they want to. More than just a concern within the medical community, I think that members of the clergy are also largely opposed to the concept of doctors prescribing religion. Seven religious leaders--Catholic, Buddhist, Jewish, Greek Orthodox, Protestant and Muslim were co-authors of our article in the New England Journal of Medicine.

Why has increasing religion's role in medicine become such a popular topic of late?
I think there are three major reasons:

1. Patient dissatisfaction with the quality of medical interactions.
2. A general trend toward a greater presence for religion in secular life.
3. The availability of funding for programs that introduces religion into medicine.

Let me briefly discuss the impact of the first and last of these trends.

1. Patient Dissatisfaction

Patients often express dissatisfaction with the quality of their medical care in general, and of their interactions with physicians in particular. While managed care has certainly contributed to the problem, patient dissatisfaction with physician relationships long pre-dates managed care's arrival. Patients do not want to be treated like collections of organ systems; patients want to be treated like people. Addressing religion may be one way to treat a patient like a person. However, it's the wrong response.

2. Financial Support

As with any new venture, successfully introducing religion into medicine requires financial backing. The John Templeton Foundation, run by the retired investment banker Sir John Templeton, has given millions of dollars to groups that further this effort. A good portion of this money goes to the National Institute for Healthcare Research (NIHR), a privately funded advocacy organization. Cleverly named and located to suggest an affiliation with the National Institutes of Health, NIHR works with the Association of American Medical Colleges to develop and run programs that introduce religion and spirituality into medical-school curricula.

NIHR recommends that physicians incorporate religion into their medical care by taking an initial spiritual history of their patients and then periodically inquiring into their spiritual health. NIHR qualifies this recommendation by suggesting that doctors drop all spiritual inquiries if a patient expresses no interest.

NIHR also relies on funding from the Templeton Foundation to conduct
research reviews of questionable quality that support the medical benefits of religion. General media coverage of these reviews often becomes the source of statistics quoted in the press.

Advocates for religion as adjunctive medicine present three major justifications for their cause:

1. The existence of solid empirical evidence proving a medical benefit.
2. The comfort that religious practice provides to patients.
3. Patient demand for the incorporation of religion into their medical care.

The New England Journal of Medicine article demonstrates that each of these justifications is lacking. Proponents of religion in medicine often refer to the "over 1,200 studies" showing religion's beneficial effect on health. Unfortunately, the vast majority of these studies was poorly conducted and contains significant methodological flaws--rendering their conclusions suspect at best. In the Middle Ages, thousands of studies "proved" that the earth was completely flat. They were all wrong.

When we examined the empirical evidence, we discovered three major areas of concern:

1.Failure to control for confounders.
2.Failure to control for multiple comparisons.
3.Reliance upon epidemiological studies.

Our findings in all three of these areas confirmed that there is no basis for modifying current non-religious physician practice.

1. Confounders

One of the standard classes of methodological flaws stems from a failure to control for confounders. For example, George Comstock, a senior and well-respected epidemiologist at Johns Hopkins University, published a paper in the Journal of Chronic Disease in 1972 which concluded that church attendance was associated with increased longevity and reduced mortality. Proponents of religion as adjunctive medicine always cite that paper.

In 1977, Comstock published another paper, which essentially retracted his initial finding. He realized that the original research had failed to control for functional status--the degree to which people were already sick or disabled. Church attendance is influenced by one's capacity to go to church. If you're already sick or disabled, you're
less likely to be able to attend church and you're also more likely to die. Rather than church attendance having an impact on health, health has an impact on church attendance. Of course, proponents of religion in medicine rarely cite Comstock's retraction.

2. Failure to Control for Multiple Comparisons

Despite its flawed methodology, Comstock's paper represented a major attempt to address the impact of religion on health. However, the majority of the 1,200 studies frequently cited have little to do with religion. For example, a 1968 study by Friedman and Hellerstein was entitled "Occupational stress, law school hierarchy, and coronary
artery diseases in Cleveland attorneys" but includes only one measure of religious activity, even though it is cited as a study of the association of religion and health.
The problem with this study and many others "about" religion and health is that they often measure a great many variables, some of which may be significant merely by chance. If you make enough comparisons, sooner or later one or two will be significant. This chance phenomenon has been described as the sharpshooter's fallacy. The sharpshooter empties a revolver into the side of a barn wall and then draws the bull's-eye.

While it's unfair to criticize the authors of those studies for considering religion as a factor, it is fair to criticize secondary sources who use the studies to support their own case. If you fish through data long enough, you will always find something. Whenever
you employ multiple tests, your odds of discovering a statistically significant finding increase dramatically as a product of chance. Finally, you also have to bear in mind that studies conducted in the 1950s to the 1970s lack today's standards of methodological sophistication.

3. Epidemiological Studies

The third empirical problem stems from the failure to distinguish epidemiological studies from clinical trials. Epidemiological studies demonstrate associations between variables in large populations. As previously shown in the Comstock example, causality can be difficult to determine in such studies. Yet understanding causality is essential for drawing clear inferences about the impact of an intervention.

It is possible, however, to conduct well-executed epidemiological research on this topic. In recent years, two solidly conducted epidemiological studies have shown an association between church attendance and reduced mortality--acceptably controlling for confounders and variants.

But the existence of these studies in no way means that physician-recommended church attendance would have the same impact. Undoubtedly, an enormous difference exists between attending church on a voluntary basis and attending church as a medical prescription. There are no data whatsoever on the health impact of efforts by physicians to promote religious activity. The particular religious affiliation may also influence outcome--a Quaker meeting may have a different impact on a patient's health than a
Roman Catholic Mass, for example.

Unfortunately, researchers cannot conduct clinical trials on this topic. It is simply unethical to randomize a group of 4,000 people, assigning half to be religious and half to be non-religious for 15 years, and study the health effects.

Beyond empirical considerations lies a whole range of ethical implications associated with doctors prescribing religion. Concerns of religious coercion, patient privacy, doing harm and discrimination were foremost in our research.

The nature of the physician-patient relationship is inherently asymmetrical. Physicians make recommendations and expect patients to follow them. This arrangement derives from physicians' medical expertise. When physicians depart from their medical
expertise to pursue another agenda, they violate the rules of this arrangement, raising the possibility of coercion. The mental state of the patient must also be considered. People typically see a physician when they have a problem and are experiencing pain. Their
suffering may make them feel vulnerable, further opening the door to coercion.

We deem some coercion acceptable as long as the physician's recommendations derive from medical expertise--advising that a patient take antibiotics, stop smoking or have an angioplasty. It's not acceptable when that coercion is brought to bear on non-medical matters--recommending that patients go to church or pray more often. Coercion does not have to be overt; it can be very subtle. We are concerned with the confusion that patients will undoubtedly experience in attempting to distinguish between medical and religious recommendations from their physicians.

Asking doctors to prescribe religion also raises the issue of invasion of privacy. Epidemiological studies have shown that many characteristics of our lives are associated with health outcomes. Most evidence suggests that being married is good for your health; people who are married tend to live longer. Yet we would not expect physicians to recommend to single patients that they get married. Such advice would be considered an outrageous intrusion into a private arena.

Similarly, strong evidence exists that early childbearing confers reduced risk for certain kinds of cancers. In light of that evidence, should physicians recommend that young women have children--irrespective of marital status--to reduce their risk of cancer later in life? A personal decision, childbirth remains out of bounds for clinical medicine despite an association with health.

Clergy have repeatedly made the point that in many respects religion is a more sensitive topic than money or sex. When physicians start to engage in discussions of religion and spirituality with patients, they may quickly wind up out of their depth. Seven minutes later--as dictated by managed care--they have to terminate the conversation and see their next patient.

Clergy receive years of rigorous training to constructively handle sensitive subjects and intense discussions. Moreover, pastoral counseling sessions are not limited by the time constraints that currently characterize clinical appointments. Medical-school courses on incorporating religion into health care usually last for an hour a week over a six-week period. How can you expect physicians--even those "trained" by such courses--to deal with the same religious issues as professional clergy?

The third ethical issue arises from the possibility of actually doing harm. Even in these days of medical consumerism, patients still confront age-old folk wisdom about personal and moral responsibility for illness. By suggesting to patients that religious activity is associated with better health, you also imply the converse--that poor health is a product of insufficient religious activity.

I witnessed this dynamic firsthand while visiting an oncology patient for a research study. She shared a room with another young woman; both were with their respective families, both awaiting the results of biopsies to determine whether they had cancer. While I was in the room, the biopsy results for the other patient came back. They were negative; she didn't have cancer. I'll never forget what happened next. Her father said with a great sigh of relief, "We're good people. We deserve this."

What was the patient I was visiting supposed to say when her biopsy came back positive? Was she supposed to say, "I'm a bad person, that's why I got cancer. I haven't been religious enough. I haven't gone to church enough. I haven't been sufficiently devout, that's why I got cancer"?

It's bad enough to be sick. It's worse still to be gravely ill. It's simply unconscionable to add to that the burden of remorse and guilt for some supposed religious failure. Far from providing comfort to patients, you can actually cause harm to patients through such suggestions.

To reflect their sensitivity to the possibility of religious coercion, proponents of religion in medicine say they would never recommend religious activity to people who are not so inclined. They say that by taking a spiritual history, they can determine for whom such recommendations will and will not be appropriate.

This is like determining patients' views on antibiotics before recommending them to treat pneumonia. It creates two classes of patients: one for whom an effective and life-extending treatment is recommended and another for whom it is not. Proponents say that 1,200 studies support associations between religious activity and health. They also say that they won't recommend religious activity to people who are not so inclined. In other words, they are withholding a treatment that they suggest is manifestly effective. Physicians would never consider such an approach with antibiotics; it would be ethically untenable. The result is discrimination in favor of religious people. To do this surely is unethical, given the purported strength of the evidence.

Patient preference is often given as a major justification for incorporating religion into medical care. First of all, most of the research suggesting that patients want to bring religion into medicine comes from highly selected studies: family practice settings. In 2000, family practice represents an increasingly small fraction of clinical medicine. In addition, most of these studies are from conservative communities in the South. Thus, these findings are unlikely to generalize to broader clinical settings.

In addition, patients want all sorts of things that may not necessarily benefit them. They want to be discharged from the hospital against medical advice. They want to terminate chemotherapy early. They want shark cartilage. The physician must always balance the dual ethical principles of beneficence and autonomy--often a difficult judgment.

Treating religion as adjunctive medicine not only leads to poor medicine but also trivializes religion. Religious activity is not like a low-fat diet. Physicians cannot just talk about religion in the same manner in which they recently began to discuss diet and sexuality. Religion is much more than a health aid.

On a more philosophical level, touting the medical benefits of religion offers a justification for faith: Religion is good for you, therefore you should be religious. That kind of utilitarian approach to religion violates most religious tenets.

I think that the interest in religion as adjunctive medicine is consistent with the burgeoning interest in alternative medicine and self-help activities. In fact, religious treatment of disease is sometimes directly considered as an alternative medicine treatment. Both alternative medicine and religion in medicine also contain an element of anti-intellectualism; they both represent a general dumbing down of science. For example, many alternative-medicine treatments have received publicity despite a lack of supporting evidence. Proponents of these approaches urge us to keep an open mind about them. As Willem Van der Does at the University of Leiden in Holland commented to me, "It's good to be open-minded, but not so open-minded that your brains fall out."

While there has been widespread coverage on the benefits supposedly deriving from religious activity, there has been very little coverage on the negative ethical implications of doctors' involvement. I think that the press is far too uncritical. Under pressure to do too much in too little time, they rely on the poorly documented assertions of press releases. They fail to delve into the details of a story to see where a study is flawed. I'd like to see a higher profile for opponents of doctors prescribing religion.


The Pastoral Report is appreciative of Dr. Sloan's generosity in allowing the PR to publish his address to the CPSP Community. This is a very vital and important work that is of immense interest to the whole pastoral care community. It was through George Hull's efforts to have the addressed transcribed and his contact with Dr. Sloan that enabled the PR and our readers to enjoy the benefit of Dr. Sloan's provocative thinking.
Posted by Perry Miller, Editor at December 3, 2001 02:08 PM

Posted by Perry Miller, Editor at December 3, 2001 9:00 PM

Email this article to a friend

Email this entry to:

Your email address:

Message (optional):

Print this entry