Faith and Healing
Annals of Internal Medicine, 18 January 2000.
Linda Gundersen
Studies suggesting that church going, religious beliefs, and
prayer can improve morbidity and mortality have increasingly
received attention in medical journals and the general media. One
study at Duke University concluded that steady church attendance
improves health and prolongs life (J Gerontol A Biol Sci Med Sci.
1999;54:M370-6). Some authorities, however, believe that
researchers often use flawed methods to study the relation between
health and religion. They point out that the cause-and-effect
relations are frequently unclear because religious persons, who
are often defined as regular attendees of worship services, are in
general already healthier than nonreligious persons. In fact,
several studies that report the beneficial effects of religion do
not adequately adjust for health behaviors or sociodemographic
factors, thereby casting substantial doubt on their conclusions.
However, some studies of intercessory prayer and some studies that
accounted for baseline health and other variables have produced a
modest amount of evidence that links religiosity with improved
health. Experts are divided: Many cannot explain these results,
and others continue to debate the possibility that a connection
exists at all. Hundreds of years ago, medicine and religion were
almost synonymous; in recent years, however, they have been almost
completely separate. Emerging controversy surrounds the ways in
which physicians should respond to this renewed tendency to link
the two domains.
Religious Practice and Health
It is important to distinguish between religiosity and
spirituality. Harold G. Koenig, MD, MHSc, lead author of the Duke
University study and associate professor at Duke University
Medical School, points out that most studies assess religiosity
rather than spirituality because religiosity is easier to measure.
Dale A. Matthews, MD, author of The Faith Factor, a recent book on
religion and health, defines religiosity as a person's adherence
to the beliefs and practices of organized religion. Only a few
studies have focused on spirituality, which is much more difficult
to gauge; Matthews defines it as a person's search for meaning or
for a personal relationship with a higher power. Studies of
religion's effect on health often assess religiosity by
determining frequency of attendance at worship services. This can
be easily measured; persons' perceptions of their relationships
with a higher power cannot.
The Duke University study used church and synagogue attendance as
a measure of religiosity. Baseline assessment included additional
health variables. During a 6-year period, researchers studied 3968
residents of northern California who were 64 to 101 years of age.
At follow-up, 29.7% of the participants had died. The relative
hazard of dying was 46% lower for those who attended religious
services frequently (at least once per week) than for those who
attended services infrequently. Even after such significant
variables as number and type of comorbid conditions and amount of
socialization were excluded, the mortality rate of churchgoers was
28% lower than that of persons who did not go to church.
Another study that used attendance of religious services as a
gauge of religiosity was conducted over a 28-year period in 5286
residents of Alameda County, California (Am J Public Health.
1997;87:957-61). In that study, Strawbridge and colleagues found
that the mortality rate was lower in persons who frequently
attended religious services than in those who attended
infrequently. The authors concluded that at least part of
religion's effect on the mortality rate could be attributed to the
fact that frequent attendees had less exposure to certain risk
factors for hastened death (for example, smoking, alcohol use, and
limited social contacts).
Prayer and Spirituality
In a study by Harris and coworkers (Arch Intern Med. 1999;159:
2273-8), attendance of worship services was not included in the
methods. The authors studied 990 patients who were admitted
consecutively to a coronary care unit and were randomly assigned
to receive intercessory prayer for 4 weeks. The patients did not
know that the prayers were being offered, and the intercessors
knew only the first names of the patients. Although length of
hospital stay did not differ between the prayer and nonprayer
group, the patients who received remote intercessory prayer had
lower coronary care unit severity "scores." Critics of
this study's findings point out, however, that the authors used a
flawed method to randomly assign patients to study groups and that
the method used to score severity was not validated.
Warfield and Goldstein (Counseling and Values. 1996;40:196-205)
did not examine religion and intercessory prayer but focused only
on the effect of spirituality as a treatment aid in patients with
alcoholism. They reviewed literature on Alcoholics Anonymous, the
first known 12-step program, and found that Alcoholics Anonymous,
the book that laid out the premise of the program and acted as a
springboard for other 12-step programs that followed it, suggests
that alcoholics "may be suffering from an illness which only
a spiritual experience will conquer" (Alcoholics Anonymous.
3rd ed. New York: Alcoholics Anonymous World Services,
Inc.;1976:44). The authors aimed to discern the importance of the
spiritual aspect of Alcoholics Anonymous and concluded that a
"spiritual awakening" is vital to recovery from
alcoholism.
"If" and "Why": The Debate
Many studies, whether they measure the effects of religion,
intercessory prayer, or spirituality, have been refuted by experts
who dismiss the results as negligible. Richard Sloan, PhD, from
the Behavioral Medicine Program at Columbia-Presbyterian Medical
Center, the Department of Psychiatry at Columbia University, and
the New York State Psychiatric Institute, says that after the
confounding variables are removed from these studies, the results
almost always are statistically insignificant. In an article in
The Lancet (1999;353: 664-67), Sloan and others contest the
findings of several studies. Colantonio and colleagues (Am J
Epidemiol. 1992;163:884-94) found that rates of stroke were lower
in persons who attended a religious service at least once per week
than in those who did not attend at all. Sloan and colleagues
noted that after differing levels of physical function were taken
into consideration, the relation between stroke and attendance at
religious services was no longer statistically significant. A
study by Pressman and colleagues (Am J Psychiatry. 1990;
174:758-60) found that the ambulatory status of religious elderly
women who had surgery on broken hips was improved at discharge.
Sloan and colleagues believe that the results of this study are
unconvincing because the authors did not control for one critical
variableŅage.
Koenig believes that some variables should not be discarded, such
as the fact that churchgoers are in relatively better health than
persons who do not go to church because of a lower smoking rate
and other health factors. "If you controlled for those
factors and you find the relationship goes away, you simply
controlled for the mechanism of the effect. That doesn't diminish
the effect of religious involvement on the health outcome-it
simply explains it," he stated. Sloan does not dispute
Koenig's statement, but adds that many positive behavior
mechanisms are often unrelated to religion. For instance, a
smoking cessation group, which has no religious affiliation,
teaches behavior modification without raising ethical issues.
The positive factors associated with religion seemingly cannot
explain the results of studies of intercessory prayer, however,
which Sloan particularly disputes. Recently, Sloan and his
colleague, Emilia Bagiella, PhD, challenged the study of
intercessory prayer, which suggested that prayer "may be an
effective adjunct to standard medical care." Sloan and
Bagiella asserted that "the lack of construct validity raises
serious questions about this finding" and stated that the
prayer and control groups had no difference in length of stay in
the coronary care unit or in the hospital; did not differ on the
Byrd Scale; and in fact only differed on the MAHI-CCU scale, which
was constructed for the study's purpose.
Matthews believes that removing all of the health variables
associated with religion may diminish study results but does not
eliminate them. Good health behaviors of church attendees, in his
view, are totally unrelated to the effects of intercessory prayer.
Matthews believes that after confounding variables are removed,
the remaining evidence cannot be explained only by science. In
regard to Harris and coworkers' study on intercessory prayer,
Matthews stated that no concrete medical evidence could explain
why the persons for whom the prayers were offered had improved. In
Matthews' opinion, that improvement could be attributed only to a
supernatural force or to some unexplained type of energy.
Koenig asserts that it is "simply an epidemiological
error" to deny the relation between religion and health
because nonspiritual explanatory factors are involved. He believes
that it is impossible to reduce religion to something that is
devoid of such factors. "Religion is a package. It packages
all of these potential health benefits: social support, improved
coping, less depression, and a more optimistic outlook," he
stated. He believes that research should stop focusing on
determining whether religion is beneficial to health and instead
focus on why it is. This viewpoint is supported by Strawbridge and
colleagues, who conceded that they could not entirely account for
their study results after factoring out health variables. The
authors also recommended that additional studies be conducted to
clarify the ways in which religion affects health.
Sloan believes that researchers need to take a step back. "It
is entirely premature to consider religious activities as
adjunctive medical treatments, given the very poor quality of the
evidence purportedly linking religion to health outcomes and the
substantial ethical issues raised," he stated. David Larson,
MD, MSPH, chairman of the National Institute of Healthcare
Research, believes that more studies should focus on the link
between religion and health and try to determine what it is based
on (physiologic factors, for example) and what creates it. More
clinically relevant research is needed, "whether it's
cardiovascular or gastrointestinal. Let's slow down on the prayer
studies, and let's have better measures of religion and
spirituality as well, spirituality outside of religion," he
stated. Sloan fervently agrees that little clinically relevant
research validates the link between religion and health.
"Virtually all of the studies that have been conducted so far
are methodologically deficient," Sloan asserted.
The Search for Explanations
Although some researchers agree that a link exists between
religion and health, they continue to speculate about its origin.
Matthews, like Koenig, believes that religion itself promotes good
health behaviors and that religious doctrine incorporates many
health-enhancing factors.
The Navajos believe that negative thoughts can lead to illness;
perhaps the other side of the coin reflects the "faith
factor," as Matthews called it. A strong faith in a greater
power may improve health for various reasons. Koenig believes that
being part of a religious community and having a strong belief
system can benefit health in ways that cannot yet be measured. In
addition, the effect of religion and intercessory prayer on health
cannot be completely explained.
Koenig contends that religion encompasses both measurable and
inexplicable components. "On scientific grounds there is good
evidence, in my view, that religious commitment… is associated
with better health. And then there's this tantalizing idea that
there's something beyond the social support, beyond the lifestyle
and the meditation. Scientifically, you would say that's something
that we can't explain. A person of faith would say that's the hand
of God."
Why Now?
No one is sure what has caused the recent increase in research
examining the link between religion and medicine. Some studies
seem to suggest that the trend evolved out of patient need. For
example, in one survey (Arch Intern Med. 1999;159: 1803-6), 45% of
177 patients said that their religious beliefs would affect their
decisions upon serious illness; of this 45%, 94% felt that
physicians should ask patients about their religious beliefs. In
1996, USA Today Weekend and Time commissioned surveys to determine
whether patients thought that religion should be combined with
medicine. Two thirds of respondents wanted their physicians to
address religion with them. Dana King, MD, of East Carolina
University, Greenville, North Carolina, surveyed his patients and
found that 48% of those who were hospitalized wanted their
physicians to pray with them.
Why do patients feel that their physicians should address
religion, or more precisely, why do they feel that way now? One
reason could be the negative view of the health care system that
has developed in recent years. Many patients believe that medical
care is becoming increasingly depersonalized, primarily as a
result of managed care. Patients may believe that the
provider-patient relationship could become more personal if it
included religious aspects. In addition, the two healing
traditions of medicine and religion have become more segregated in
this century than at any point in history. Patients who have
realized that science cannot solve every problem or cure every ill
may be yearning for a reincorporation of religion and medicine.
Sloan concedes that religion has its benefits but continues to
dispute the role of religion in health.
Ethical Considerations
Beyond the question of the benefits of religion to health lies
another ethical concern: Can religion be harmful to patients?
Sloan points out that patients may be made to feel guilty or
"insufficiently faithful." Matthews, Larson, Koenig, and
many other physicians strongly advocate the blending of religion
and medicine on some level. However, others believe that
physicians should not become involved in areas outside of their
expertise, and that doing so can be a misuse of valuable treatment
time. It also has the potential for being downright unethical, as
physicians may unduly influence a patient's beliefs. Sloan and
Bagiella believe that "medicine should [not] take on
religious practices as adjunctive treatments. To do so flies in
the face of the vast majority of empirical evidence and raises
serious ethical issues." Additionally, other factors beside
religion have been identified as positively influencing health,
such as marital and economic status. The question then becomes:
Where should physicians draw the line? Sloan believes that
physicians recommending religious activities to patients is akin
to advising patients to marry or advising patients on their
financial portfolio to improve health. Sloan asserts,
"There's just no evidence that there's any impact apart from
the conventional pathways that we already know about. It's a
serious ethical intrusion."
Medicine's Response
Although Sloan and Bagiella raise valid concerns, the trend is
clearly toward an intertwining of religion and health.
Approximately 30 U.S. medical schools have incorporated some kind
of religious or spiritual teaching into their curricula. Faculty
at Harvard Medical School, Boston, Massachusetts, are teaching
medical students how to obtain a spiritual history and how to
discuss related concerns with their patients. Washington
University School of Medicine in St. Louis, Missouri, offers
courses in which medical students can explore their own and their
patients' spirituality. Yale School of Nursing, in conjunction
with the Divinity School and the Yale-New Haven Hospital, New
Haven, Connecticut, is teaching both health and pastoral
professionals ways in which to appropriately address spiritual and
physical concerns.
Despite these changes in medical school curricula, King discovered
that although many patients want their physicians to pray with
them, less than 1% of physicians are actually doing so. The
physician's role regarding religion is quickly becoming a subject
of intense debate. Even when physicians agree that religion
positively affects health, they are divided on what, if anything,
they should be doing about it.
Koenig and Larson strongly urge physicians to address religion
with their patients. Koenig believes that "doctors have been
ignoring this because they felt it was unrelated or related in a
negative way to health." At the same time, however, the
subject needs to be addressed sensitively. Koenig suggests that
physicians ask patients how important religious beliefs are to
them but does not believe that physicians should "prescribe
religion." Larson thinks that it is imperative to respond to
patient need on this issue regardless of study results. "I'm
not saying we should address it because the research says it's a
factor that tends to be a healthy factor. We should address it
because the patient surveys are saying that we should be
addressing it," he remarked.
However, Brian Haynes, MD, professor of clinical epidemiology and
medicine at McMaster University and editor of the American College
of Physicians-American Society of Internal Medicine's ACP Journal
Club, adamantly disagrees. "Dictating what the medical
profession should and shouldn't do solely on the basis of patient
need isn't reasonable," he stated. He believes that patient
need should be considered, but he also feels that all
interventions should be supported by some evidence of benefit.
Sloan believes that religion should have an extremely limited role
in the physician's office. "In these days of managed care,
when you have such limited amounts of time with patients, surely
there are more important things to do in the interaction than
pursuing some issue that has very poor empirical support and
raises all of these ethical issues," he asserted. Koenig
agrees that time is short but believes that addressing religious
aspects with patients is crucial. "Now we're getting stuck in
this position where we're seeing patients in 10 minutes. I can
hardly address the medical and psychiatric issues, let alone the
spiritual issues-but it doesn't take that long to affirm the
patient's religious beliefs and how to cope," he commented.
Sloan does not believe that physicians should totally ignore the
subject of religion. Rather, he thinks that patients who have
spiritual concerns about their health should be encouraged to seek
the help of trained professionals, such as health care chaplains.
Sister Patricia Talone, PhD, vice president of mission services
and ethicist at Unity Health in St. Louis, Missouri, agrees with
Sloan on that point but adds, "We don't always have enough
chaplains to care for patients' needs. Two of our hospitals are
rural, and there isn't always a chaplain available."
Talone feels that physicians could be more involved in patients'
religiosity but that the level of such involvement should be
dictated by context. "I firmly believe that in many smaller
communities where the doctor knows the family well or the doctor
has had a long-standing relationship with the patient or family,
it is not at all inappropriate for the physician to bring up the
spiritual issue or even to take a moment to pray with the
patient."
Koenig strongly believes that physicians will benefit from adding
a little religion to their practices. "Religion is good
medicine," he stated. "Most of us became doctors because
we really wanted to help people. I think this can bring back what
medicine needs, what it's lost, and what doctors really want. It
will give their practices meaning."
Whatever their position on these issues, most people would agree
that a physician's own beliefs are important. That is, physicians
must be comfortable with the idea of introducing religion into the
practice of medicine before they take any further steps. They
should not feel pressured to do anything that conflicts with their
deepest beliefs.
back to top
|