Books Reviewed
Koenig, Harold G., M. D. The Healing Power of Faith: Science Explores Medicine's Last Great Frontier.
New York: Simon & Schuster, 1999.
331 pp.

In the past quarter century epidemiologists have shown an increasing interest in the relationship between religion and health. While studying the determinants and distribution of disease in human populations some epidemiologists have inquired whether religious beliefs and behaviors are statistically significant factors in the causation of disease and death. No one has been more productive in this new project than Harold Koenig, a psychiatrist and professor at Duke University and the director of that university's Center for the Study of Religion/Spirituality and Health. Although a frequent contributor to research journals and the author of several previous books on aging, religion, and health, The Healing Power of Faith is Dr. Koenig's first book directed specifically toward an educated lay audience. It is informative, responsible, and highly readable.

Much has been published on the topic of religion, spirituality, and health recently, most simply anecdotal accounts of how an individual's religious faith promoted dramatic healing. Likewise, Dr. Koenig uses anecdotal narratives to introduce his topics. For instance, in a chapter entitled "Religious People Cope Well with Stress" is introduced by an account of an elderly man's faith helping him to stay healthy while caring for his wife suffering from Alzheimer's disease. What is distinctive, however, about his approach is that it provides information about religion and health in human populations and not just in exemplary individuals, and it describes the religion/health relationship with scientific rigor rather than with anecdotal narratives.

Since the studies summarized by Dr. Koenig deal with human populations rather than individuals their results are stated and interpreted as statistical claims. The author states the epidemiological findings he discusses with appropriate caution. Given the nature of religion as an exposure factor, the relevant methodological standards are controlled observational studies rather than randomized experimental studies. (Experimenters cannot randomly assign subjects to religious and nonreligious groups in the same way that they can randomly treat some with pharmaceutical drugs and others with placebos.) Specifically, the studies that Dr. Koenig discusses are ones in which great care has been taken to measure accurately the religious exposures and the health outcomes of interest. Also, in the design, implementation, and interpretation of such studies serious attention is given to controlling for nonreligious factors-i.e., to ensure that an association between a religious exposure and a health outcome is not, in fact, the result of some other factor jointly and independently associated with the exposure and outcome.

The Healing Power of Faith summarizes the various dimensions of human health that religious beliefs and behaviors have been shown to effect. According to Koenig, religious people generally have stronger families and healthier lifestyles than do nonreligious persons, and religious people suffer depression less frequently and endure stress more effectively. Evidence that religious beliefs and behaviors are protective against heart disease and immunological dysfunction is less compelling; Dr. Koenig notes this fact in his presentation of the relevant data.

The expository portions of Dr. Koenig's book have two shortcomings. First, the anecdotal stories of healing are apt illustrations of a topic but they rarely are so vivid or detailed as to give independent insight as to how religion has a positive impact on people's health. The account of the elderly man who cared for his wife suffering from Alzheimer's notes that he attended church regularly, prayed daily, and participated in church social ministries. Koenig renders a touching portrait of another man making toy wooden trucks for disadvantaged children, but similar specificity is not given to the man's prayer life or his church activities. Readers learn little about the regular practices that may have differentiated him from a similarly compassionate nonreligious person.

Second, alternative interpretations of the results are not as forcefully presented as they might have been. It is widely agreed that people who attend religious services regularly and describe themselves as highly religious have healthier lifestyles in the sense that they smoke less tobacco, drink less alcohol, and engage in less risky sexual activities. Dr. Koenig interprets this as evidence that religion promotes health. Others argue that religion in these circumstances is epiphenomenal: any population, religious or not, that smokes less will have less lung cancer. The fact that religious people generally have sounder marriages and larger networks of social support simply confirms that most people with relatively high levels of social support suffer less illness and experience quicker recoveries. How one interprets mediating factors like healthy behaviors, social support, and optimistic dispositions (trust and confidence are associated with less depression and less noxious stress) is one of the most controversial issues in this area of epidemiological research and Dr. Koenig's readers would have benefited from a more thorough exploration of it.

Epidemiological studies can establish statistically significant associations between religious exposures and health outcomes, and, if data are collected over time, it can predict to a certain degree the future health status of groups of people on basis of their religious attributes. Epidemiologists can rarely speak definitively about causation and almost never do so in the absence of knowledge about the physiological mechanisms that relate an exposure to an outcome. Toward the conclusion of his book Dr. Koenig summarizes some very recent research that seeks to identify a physiological means by which religion may have a salutary impact. In a chapter entitled "Religious People May Have Stronger Immune Systems" the author describes his own study in which some evidence has been found that religious people have lower levels of interleukin-6 and other biochemical indicators of a weakened immune system. Natural body substances or functions that indicate the presence or absence of disease (often called biomarkers) are increasingly receiving epidemiologists' attention and Dr. Koenig's book well documents this trend as it applies to epidemiological research concerning religion. The author is, however, somewhat misleading about the results he reports. He quotes a medical colleague as saying, "This is one of the pioneering studies." I agree with this assessment but only for what the study attempts, rather than for the substantive results it reports. If one reads the article that Dr. Koenig and his colleagues published in a research journal, one finds that the effects of religion on levels of interleukin-6 and other biomarkers are quite a bit more equivocal than the book text suggests. This is one point at which the author's enthusiasm gives his readers a false impression of the scientifically established "healing power of faith."

While Dr. Koenig's book documents a movement from epidemiology to physiology in the study of religion, Herbert Benson's latest book recounts a reverse transition. As a professor at the Harvard Medical School, Dr. Benson first gained popular attention with his description of the physiological phenomenon he has called "the relaxation response." In brief, this phenomenon is the body's ability to attain a state of hypometabolism, or calm, in which muscle tension is relaxed, blood pressured is lowered, and heart and breathing rates are reduced. The relaxation response can be deliberately elicited by simple focusing techniques or meditation practices. It is the exact opposite of the fight-or-flight response delineated in the 1930's by Walter Cannon (also at the Harvard Medical School): here the body shifts into a higher rate of metabolic activity (hypermetabolism) in order to cope with a perceived environmental challenge.

Dr. Benson's research into the relaxation response led him to investigate a wide array of "top-down" events in which the mind influences the body. Prominent among them is the placebo effect in which patient and physician confidence in the restoration of health has an actual salutary impact. In contemporary medical practice this phrase refers especially to the fact that many persons given sugar pills (placebos) in clinical trials of drugs actually have better outcomes than people who receive no treatment at all (though their improvement is less than that experienced by persons receiving an effective pharmaceutical drug.) The limited efficacy of some premodern medical practices and contemporary faith healings have also been attributed by Dr. Benson and others to the placebo effect. A determination to explore and implement the benefits of "top-down" events in medicine led Dr. Benson in 1988 to establish the Mind/Body Medical Institute at Boston's Deaconness Hospital.

Dr. Benson claims that the relaxation response works in a wide variety of clinical and home care settings. He notes, however, that it usually works better when initiated under the guidance of a sympathetic caregiver and often works even better when practitioners use religious phrases and meanings as part of their meditation practice. Thus in Timeless Healing Dr. Benson begins with the physiology of the relaxation response and proceeds to discuss the epidemiology of religion; indeed, some of the same studies that Dr. Koenig features in his text are cited in this one. For instance, both authors cite the surveys of epidemiological research about religion compiled by David B. Larson, M.D., the President of the National Institute for Healthcare Research in Rockville, Maryland. The two authors are knowledgeable and appreciative of each other's work and their books are quite complementary: both are carefully researched and clearly written by major contributors to their respective fields.

Benson's Timeless Healing provides an insightful account of how beliefs- especially religious beliefs-effect embodied well-being. It makes some brief and nontechnical forays into the neurophysiology that underlies the relaxation response and the placebo effect.

At times Dr. Benson fails to articulate clearly the central ideas of his text. Even the idea of the relaxation response suffers from some ambiguity. Early on in the book he describes it as a "state" of bodily calm that results from sitting quietly, repeating a phrase, and ignoring other mental impressions. Later he suggests that it consists of the "physiologic changes" of reduced metabolism, circulation, and respiration that engender this state upon the consideration of meditative thoughts. This ambiguity regarding whether the relaxation response is the result or process of relaxation is reminiscent of one that accompanied Hans Selye's first formulation of the physiological notion of stress. At first it was unclear whether stress identified the cause of an organism's perturbation or the physiologic processes that constituted that perturbation. Eventually the word "stressor" was introduced to signify the cause of stress and the ambiguity was eliminated.

A second concept central to Benson's book is "remembered wellness." He says at one point that its components are the "belief and expectancy" of health engendered by the patient, physician, and by the relationship between them. The confusion here comes from a notion being posited as both a remembrance and an expectancy-terms which are temporal opposites. I take Benson to say that both memories of wellness and expectancies of wellness are thoughts of well-being and instances of the "top-down" physiologic process. (Readers must work a bit to attain full clarity on this point).

Before mentioning the important implications of Drs. Koenig and Benson's research I will note that they both correctly refrain from attributing the salutary effect of religion to the intervention of supernatural agencies. Certainly many religious people confess that God is ultimately responsible for what healing they enjoy. Scientific methods, of course, are not able to address issues that purport to transcend the natural world, nor are they requisite for validating confessions of faith which presuppose trust and hope rather than factual certainty. Both authors acknowledge their own openness to religious faith and Benson speculates that because belief in God is so advantageous to human health it is probably "hard-wired" into the human nervous system-i.e., people are born with a neurological predisposition toward belief in an "almighty power" who offsets the psychological burden of bodily illness and death. (Apprehensions of death elicit the fight-or-flight response of metabolic activation, and in an opposite manner, belief in God engenders a relaxation response of calm and contentment). This speculation, however, like the authors' own religious beliefs, is not essential to the basic claim that being religious in belief and observance is beneficial for one's health.

These texts an important philosophical implication: if religious faith and certain associated practices tend to promote good health in people who embrace them, then this fact alone seems to evidence the rationality of religious beliefs and behaviors. Such good health does not necessarily confirm the cognitive contents of religious beliefs nor the efficacy of religious practices according to their religious meanings, but it does imply that, since it is generally rational to pursue health promoting activities, it is rational to pursue religious activities when they are shown to be health promoting. Data from the epidemiological study of religion is not directly relevant as to whether or not God exists-but it is relevant to the more pragmatic question prompting Pascal's famous Wager, i.e., whether or not it is rationally prudent to believe that God exists. Pascal's affirmative answer rests upon the infinite value of eternal salvation while the epidemiologist can only estimate the finite value of marginal health outcomes. Both Drs. Koenig and Benson mention the rational credibility that their research lends to religious beliefs and behaviors.

Both authors conclude by offering ways their results might be used for public health promotion. Dr. Benson provocatively opines that the relaxation response might profitably and appropriately be introduced into the public schools. Since it may be undertaken either with or without religious content he thinks it should not provoke constitutional challenge. I think careful introduction into the school day could both produce health benefits and elicit profitable discussion as to whether meditation can be undertaken in a way that is spiritual but not expressly religious. A valuable upshot of Dr. Benson's research is that it helps reshape a rigid dichotomy between religious spirituality and nonspiriutal secularity that has not served the American public well.

Dr. Koenig cautions that the health benefits associated with religious faith cannot be gained by an "extrinsic religiosity" that merely adopts the outward forms of religion for the sake of securing some nonreligious end like good health. (Pascal actually advised his skeptical readers to adopt the outward trappings of faith because habitual practice may eventually engender inward conviction). Assuming that one does have a sincere religious faith, Dr. Koenig offers advice for disease prevention and health maintenance: attend religious services more frequently, pray regularly, and donate more time to community service. Although these recommendations show much common sense and good will, they also reveal the limited range of religious beliefs and behaviors that he has studied. Even the traditional Christian healing practice of the laying on of hands and anointing with oil is not mentioned. As a specialist in geriatric psychiatry, Dr. Koenig has often studied elderly populations and his subjects are predominantly Protestant Christians. Absent are the generally younger representatives of the religious pluralism characteristic of California and large metropolitan areas like New York City. Muslim and East Asian religious traditions are seldom mentioned in Dr. Koenig's book, nor are the nonreligious varieties of spirituality that have become popular. The studies of yogic practices conducted by Indian researchers and others are not mentioned.

Will the health benefits of religion be as manifest in younger and more racially and ethnically diverse populations than have been studied thus far? Will religion enable people to bear up in contemporary contexts where stress is intensified by ecological degradation, economic competition, and social fragmentation? Will religion help people maintain healthy lifestyles in a culture where traditional morality is less commonly shared and religious groups are more sectarian? Some critics have argued that the health benefits of religion are minimal at best. It is possible that these benefits reflect not a general phenomenon attributable to religion, but a cohort effect characteristic of populations from certain eras and specific locations.

A second line of inquiry might ask whether more pronounced or rigorous forms of religious belief and behavior have an impact beyond those so far noticed from attendance at religious services, prayer, and community service.

For instance, if the moderate diets of American Jews and Christians show salutary effects then will the more abstemious diets and periodic fasting of Muslims Americans show greater benefits? If the relaxation response yields health benefits then will Hindu yogic meditation conjoined with bodily postures and breathing practices show yet better results? Finally, is the placebo effect compounded when the expectancy of successful treatment felt by patient and caregiver is motivated by both scientific and spiritual confidence as is the case for many alternative health care providers?

Drs. Benson and Koenig show commendable boldness. Pursuing epidemiological research on more diverse religious populations and concerning varied types of religious beliefs and behaviors will help ensure that genuine health benefits of religion are resources for preventing disease and promoting health in the next millennium.