Half a Million People, 55 Studies: Spirituality's Protective Effect on Substance Use Is Real

More than 20,000 studies examining spirituality and health have been published since 2000. Most are cross-sectional - they compare spiritual and non-spiritual individuals at a single point in time and cannot tell us whether spiritual practice precedes changes in health behavior, or whether healthier people gravitate toward spiritual communities. The question of whether spirituality actually influences substance use over time, rather than merely correlating with it, requires following the same people across months and years.

That is what longitudinal studies do. And from among those 20,000 published works, researchers at Harvard T.H. Chan School of Public Health identified 55 that met rigorous criteria for longitudinal design and sample size. Together, those 55 studies followed more than half a million people. The meta-analysis, published in JAMA Psychiatry, represents the first systematic synthesis of this longitudinal evidence base.

A 13% Reduction, Remarkably Consistent Across Studies

The central finding: broad spiritual practice was associated with a 13% reduction in the risk of hazardous or harmful use of alcohol, tobacco, marijuana, or illicit drugs. Among individuals attending religious services at least once per week, the risk reduction was 18%. These figures represent the aggregate effect across all included studies after statistical pooling.

What makes the finding credible is not just its magnitude but its consistency. Senior author Tyler VanderWeele, John L. Loeb and Frances Lehman Loeb Professor of Epidemiology at Harvard, noted that all but a few of the 55 studies showed a protective rather than detrimental association - including more than a dozen conducted outside the United States, spanning different religious traditions, cultural contexts, and definitions of spiritual practice. The results held across all four substance categories studied.

"The consistency of the results across all the studies was striking," VanderWeele said. "This is a sort of once-in-a-decade advance."

What Counts as Spirituality

The study defined spirituality broadly - deliberately so. Qualifying practices included formal religious service attendance, personal prayer, meditation, and other forms of engagement through which people report finding ultimate meaning, purpose, or connection beyond themselves. The breadth matters for interpretation: the protective association is not specific to any single religion, nor to formal religious participation alone. Individual private practice showed protective effects, as did community religious involvement, though the magnitude was somewhat higher for regular service attendance.

This breadth also raises measurement challenges. Spiritual practice is not a single clearly defined variable across 55 studies conducted by different research groups in different countries over more than two decades. The meta-analytic approach attempts to synthesize these varied measurements, but the heterogeneity across studies means the pooled effect size carries more uncertainty than a single large study with consistent measurement would provide.

Why the Association Might Exist

The literature proposes several overlapping mechanisms. Religious and spiritual communities provide social connection and accountability that can reduce isolation, which is among the factors most reliably associated with substance misuse. Spiritual practices - whether meditation, prayer, or service attendance - offer structured coping strategies for stress and negative affect that may compete with substance use as an emotional regulation tool. Many spiritual traditions carry explicit normative discouragement of excessive substance use. And the cultivation of meaning and purpose that participants in spiritual communities often report may reduce the vulnerability to substances that accompanies a sense of purposelessness.

Lead author Howard Koh, Harvey V. Fineberg Professor of the Practice of Public Health Leadership, noted that for many individuals, spiritual resources are already embedded in daily life. The question the findings raise for clinicians is whether those resources are being actively engaged in discussions about substance use risk.

Observational Evidence and Its Limits

Longitudinal design improves on cross-sectional comparisons but does not establish causation. People who choose to engage in spiritual practices differ from those who do not in numerous ways that observational studies cannot fully account for. Family background, neighborhood characteristics, personality traits associated with both religious participation and lower substance use risk, and access to social support structures all represent potential confounders. The 55 included studies varied in how thoroughly they controlled for these factors.

The researchers are explicit that the findings apply to individuals for whom spirituality is already meaningful. Prescribing spiritual practice to patients who have no prior spiritual engagement would be both clinically inappropriate and unsupported by this evidence base. The practical application the authors describe is more targeted: for patients who already report spiritual engagement, clinicians can explicitly acknowledge and draw on that resource as part of a broader approach to substance use prevention and recovery support.

Implications for Public Health Practice

At the population level, the research opens a different conversation. Public health organizations working on addiction prevention often operate in parallel with religious and spiritual communities that serve overlapping populations, without formal collaboration. The scale of the evidence assembled here - half a million people, consistent effects across cultures and substance types - makes a case for more systematic partnership between public health infrastructure and spiritual communities, particularly in settings where clinical resources are scarce.

The study was supported by the Templeton Religion Trust and the Lee Family Fund. The full author list includes Howard K. Koh, Donald E. Frederick, Tracy A. Balboni, Samantha M. O'Reilly, John F. Kelly, Keith Humphreys, Michael Botticelli, Maya B. Mathur, Constantine S. Psimopoulos, Katelyn N.G. Long, and Tyler J. VanderWeele.