In this issue, Miller and colleagues present data from a longitudinal study of offspring from a sample of depressed and nondepressed subjects to determine if religion or spirituality influenced the onset and course of major depression over the 10 years of follow-up (1). They found, among individuals who affiliated as either Protestant or Catholic, that subjects who reported religion or spirituality as highly important were 76% less likely to experience an episode of major depression during the follow-up. In contrast, religious attendance and denomination had no impact. The protective effect was experienced primarily among subjects at high risk because their parents experienced depression.

Though this study is the first long-term outcome study on the impact of religion or spirituality on the emergence of depression, it confirms a growing literature, including a previous study by the authors (2–5), that generally supports the benefit of religion or spirituality (usually religious participation) in decreasing the frequency and recurrence of depressive disorders. Studies to date have suggested three conclusions, all of which can be debated: 1) individuals with no religious affiliation are at greater risk for depressive symptoms and disorders, 2) people involved in their faith communities may be at reduced risk for depression, and 3) private religious activities and beliefs are not strongly related to risk for depression (6). Depression has been the most frequently studied of the psychiatric disorders in relationship to religion or spirituality, in large part because of the overlap in expression of both. For example, guilt associated with depression often is connected with a religious belief system, and apparent depressive symptoms (such as the “dark night of the soul”) are associated with religious experiences (7).

Given the strong and passionate views of American society regarding religion, such studies raise at least three questions. First, should studies that explore the association of religion or spirituality with health even be fielded and reported in the empirical scientific literature? Investigators have answered by voting with their feet (or, rather, with their questionnaires and computers). Hundreds of studies investigating the association of religion or spirituality with health have been published over the past three decades, summarized in two editions of a widely cited book, The Handbook of Religion and Health (6, 8). The question as to whether such studies should be published is still raised, but most major journals do not turn away research reports strictly on the basis of an evaluation that the subject matter of religion or spirituality is not appropriate for an empirically oriented platform that shares methodologically sound investigations.

Second, what are the strengths and weaknesses of the methodologies employed in such studies? This question is much more interesting and instructive. For example, studies of religious affiliation in relationship to a marker of depression (suicide) can be traced back over a century to Emil Durkeim. In Suicide (1897), he explored the differing suicide rates among Protestants and Catholics in German-speaking Europe and found that affiliation as Catholic was protective against suicide, a finding that was criticized methodologically even then because it was from an ecological study (9). He argued that stronger social control among Catholics results in lower suicide rates. Miller and colleagues did not find a difference between Protestants and Catholics, and perhaps this reflects an overall weakening of social control by denominations during the 115 years since Durkeim's work as well as a different social setting in New York. Religious affiliation with mainline religious groups today probably has much less influence on the individual than at the turn of the 20th century.

In addition, how does one even conceive of measuring such a nebulous topic as religion or spirituality? Some demarcate religion from spirituality, suggesting that religion implies a particular faith tradition that includes acceptance of a metaphysical or supernatural reality whereas spirituality does not and is not bound to any particular religious tradition. Yet in today's world, this distinction is perhaps artificial and a combined focus on religion and spirituality is warranted.

A short list of the dimensions of religion or spirituality include the following: religious belief (e.g., belief versus nonbelief), religious affiliation, religious participation (e.g., attendance at religious services or financial support of a faith community or larger effort), nonorganizational religiosity (e.g., engaging in private prayer or meditation), subjective religiosity (e.g., importance of religion), and spiritual well-being (e.g., feeling connected to a higher being or finding inner spiritual strength). Numerous scales have been developed, both brief and extensive. These scales have been used to determine the religious or spiritual “pulse” of the nation as well as to explore the relationship between religion or spirituality and health. For example, the Gallup Poll found in 2010 that 54% of those surveyed felt religion to be “very important” when respondents were asked about the importance of religion in their lives (10). That very question turned out to be a significant predictor of protection against recurrence of depression in the present study. Despite the complexity of religion and spirituality, relatively simple questions may exhibit strong power for predicting health outcome, just as simple questions about subjective physical well-being are strong predictors of health outcomes (11). With both, however, some mystery remains as to what we actually are measuring.

We must also recognize that this study focused on a sample with an average age of 29.3 years (SD=5.5). Most studies of depression and religion or spirituality have focused on older adults (3; 4; 6, pp. 118–135). And these studies have predominantly found that religious activities, especially attendance at religious services, are more important than “attitudes,” such as the expressed importance of religion. A relatively recent sociological survey by Smith and Snell of religious attitudes among young adults found them less active in structured religious activities but nevertheless continuing to view their spirituality as important. The spirituality they described, however, is quite different from that of their parents (12). Smith and Snell summarized the religion or spirituality of many adolescents and young adults in the United States today as “moral therapeutic deism,” a far cry from traditional Protestant and Catholic beliefs. “Moral” suggests an orientation toward being “good and nice,” “therapeutic” suggests being primarily concerned with one's own happiness, and “deism” suggests a view of God as distant and not normally involved in one's life. Therefore, the importance of religion or spirituality depends in large part on how it is conceived, and those conceptions may vary significantly across generations.

Finally, how is one to interpret these empirical findings given the strong and often nonempirical rationale for (or against) religious beliefs and their benefits (or their dangers)? Results from studies such as that by Miller and colleagues or studies that find no connection or a negative relationship between religion or spirituality and depression are at great risk of being overgeneralized. They may be falsely taken as proof of concept for interventions. Observational studies are not designed to support an intervention. The study under review is an empirical study and should be taken for what it is, no more and no less. Let the data speak for themselves. Nevertheless, the findings do suggest that clinicians should consider the religion or spirituality of their patients as part of the psychiatric evaluation, one more piece of the puzzle that makes up the person, whom we try to understand as well as possible so we can provide help to the best of our ability.