- An analysis of data from tens of thousands of people found that those who regularly attended religious services were, on average, less likely to become depressed, to smoke, or to drink heavily.
- Regular service attendance was also related to a lower likelihood of dying during a study's follow-up period.
- People who frequently attend services may benefit from higher life satisfaction, purpose in life, and other indicators of flourishing.
Religious and spiritual traditions have, for millennia, provided practices, rituals, and communities that help people come together to make sense of life, support one another, and seek the transcendent. Given the endurance and adaptability of these communities through history, we might anticipate that they have an important role to play in promoting human flourishing.
In recent decades, a growing body of rigorous longitudinal research has emerged showing just such a role. Past research, meta-analyzing longitudinal studies, has suggested an effect of religious service attendance on both greater longevity and lower psychological distress. There is also some evidence for an effect on a number of other health and well-being outcomes. However, as noted in both our earlier short-form and long-form reviews of the religious service attendance literature, the existing longitudinal evidence for other outcomes is often limited to only a few studies per outcome (and sometimes just a single study!). We need more evidence.
Thus, over the past couple of years the Human Flourishing Program has undertaken what we believe to be the most comprehensive study to date on the role of religious service attendance in shaping health and well-being in the United States. We have examined 22 different health and wellbeing outcomes in three large cohorts of adults at very different life stages.
Our study used data from three major epidemiologic cohorts, each of which covered a different life stage. To examine the role of religious service attendance among young adults, we used recent waves of the Growing Up Today Study (9,229 participants who were on average about 23 years old at the baseline). For middle-aged adults, we used data on 68,300 participants in the Nurses’ Health Study II (mean baseline age of about 47 years). And for older adults, we used a nationally representative sample of 12,549 participants from the Health and Retirement Study (about 69 years old on average at baseline).
Each of these studies had several features that allowed us to adhere to the most rigorous research practices in carrying out longitudinal data analysis (which we’ve recently summarized in an article in JAMA Psychiatry). Each of these studies followed the participants and collected data on a number of health and well-being outcomes over time (between 3 and 12 years, depending on the outcome). The studies included data on a rich set of socio-economic, demographic, and health factors at baseline. We were thus able to control for the effects of many potential confounders, as well as of past outcomes, and past religious participation. We further used sensitivity analysis to assess how robust the associations were to potential unmeasured confounding.
We looked across the various outcomes to see which of these were included in at least two of the three cohorts and we chose those as the primary outcomes for our study. We analyzed the data separately across the three cohorts and then also meta-analyzed the results of the three studies together. Even with these careful controls in place, we found strong evidence for effects of religious service attendance on a number of health and well-being outcomes.
What We Found
In some cases, our results closely replicated past work. For example, we found that, even after controlling for the factors above, individuals who attended religious services weekly or more were 16% less likely to become depressed, and showed a 29% reduction in smoking and 34% reduction in heavy drinking. These results match reasonably closely results from several prior studies. Somewhat strikingly, but again in line with prior analysis, weekly service attendees were 26% less likely to die during the follow-up period.
However, for a number of the social and psychological outcomes, because past research from rigorous longitudinal studies was often restricted to just one or two prior papers, our present study was able to double or even triple the existing evidence base. We found evidence for modest effects of religious service attendance on reducing hopelessness, anxiety, and loneliness; slightly larger effects on increasing positive affect and life satisfaction over time; and notably larger effects on social integration (even non-religious forms) and on purpose in life.
For the most part, the effect estimates looked similar across the different age groups. However, there were some exceptions. The effects on preventing depression seemed slightly larger for the young adults than for the other two age groups. Evidence for an effect on cancer likewise seemed only really to pertain to the younger adults. And the evidence for an effect of service attendance on purpose in life was stronger for middle-aged adults than for older adults. In some ways, it seemed that the psychological outcomes for older adults may have been less malleable. Such differences need to be interpreted somewhat cautiously, however, because the studies were designed in slightly different ways, measured the outcomes slightly differently, and involved differing lengths of follow-up.
Nonetheless, we did not find evidence for effects on all outcomes. There was little evidence that service attendance influenced sleep duration or physical activity, the likelihood of being overweight or obese, or the chances of suffering from stroke or hypertension. Also, the overall effect estimates on cancer or heart disease were relatively modest and involved considerable uncertainty. In some of our prior work, we, in contrast, found a fairly sizeable effect on survival from cancer and heart disease, even though there was more limited evidence for effects on the incidence of these diseases. It may be that in some cases, religion and spirituality have an even more powerful role in healing and recovery than in the maintenance of health. Religion will of course not always prevent suffering and illness, but it may enable individuals to find meaning within it, and to survive and heal from it.
In any case, the overall picture that emerges is that religious communities can play an important role in promoting longevity; in preventing depression; in encouraging healthy behaviors; in preventing hopelessness and loneliness; and in promoting happiness, social integration, and a sense of purpose in life. Religious communities help promote human flourishing.
Acknowledging Benefits, Addressing Harms
The implications of such research for public health of course need to be interpreted in a nuanced way. People do not generally become religious for health reasons. Such decisions are instead typically shaped by experiences, values, truth claims, systems of meaning, and relationships. One would not want to instrumentalize religion. However, for those who do already positively identify with a religious tradition, but do not attend services, we could envision these results as constituting an invitation back to communal religious life. The communal experience of religion clearly can powerfully shape health and well-being outcomes, and perhaps much else as well. This should be acknowledged both at the individual level and also in our thinking about the forces that shape population health.
Of course, religious communities have also sometimes hindered, rather than promoted, human flourishing. The ongoing sexual abuse crisis within Christian churches is a very clear example of this. These wrongs need to be acknowledged and addressed, and just outcomes ought to be sought. Given the tremendous good religious communities can do, the wrongs done by such communities create both direct harm to the victims and also indirect harm to others by driving them away. These issues need to be taken seriously, through a collective effort to strengthen prevention efforts and enhance pathways to healing.
Our next research update will take up the topic of prevention and healing in the face of sexual abuse. It will highlight a symposium we are hosting (April 8-10, 2021) to address these issues. Our keynote speaker, the 2018 Nobel Peace Prize Laureate, Rev. Dr. Denis Mukwege, has done important work to confront these problems, and our full set of speakers across numerous professions, backgrounds, and religious traditions will help us to find the right way forward towards prevention and healing. The conference will be virtual and, registration is free. You are very welcome to attend. We hope that the conference, and the discussions it provokes, will help bring healing, empower prevention, and will better enable religious communities to carry out their important work of promoting human flourishing.
Chen, Y., Kim, E.S., and VanderWeele, T.J. (2020). Religious service attendance and subsequent health and well-being throughout adulthood: evidence from three prospective cohorts. International Journal of Epidemiology, 49:2030–2040.
VanderWeele T.J. (2021) Can sophisticated study designs with regression analyses of observational data provide causal inferences? JAMA Psychiatry, 78(3):244-246.