Matthew May’s study explores how different types of religious affiliation—or disaffiliation—correlate with depressive symptoms over time. Research suggests that only about one-third of people who consider leaving a religion actually do. What happens to the other two-thirds?

Using data from 2006-2012 derived from the PALS study, which tracks 2,610 American adults, May identifies distinct mental health patterns across four groups:
1. Those who remain active in religion
2. Those who leave religion
3. Those who stay despite considering leaving
4. Those who never identified as religious.

One of May’s key observations is the heightened mental health risk for those who seriously consider leaving their religion but ultimately remain, a group he calls “stayers.” Stayers reported higher levels of depressive symptoms over time compared to individuals who either never considered leaving (stable affiliate), who did leave, or who never identified as religious. In fact, while stayers experienced a slight increase in depressive symptoms, stable affiliates and those who left reported a decrease.

These findings suggest that the decision to stay within a religious institution, even amid doubts or conflicting beliefs, may create an inner conflict that negatively impacts mental health. The data suggest that religious doubt and a sense of religious disconnection may overshadow the traditional mental health benefits often associated with religious affiliation. These benefits can include a sense of stability and purpose, health-enhancing lifestyle choices (such as abstinence from substances), and social-emotional support. May’s research aligns with the idea that the stress of internal conflict, especially when it concerns one’s values or beliefs, can diminish the mental health support that religion is traditionally thought to provide.

These insights underscore the importance of exploring the complex intersection of identity, belief, and mental well-being in the context of religious affiliation and disaffiliation. Future research could investigate these effects further, examining how factors like specific religious traditions and community dynamics influence mental health outcomes, as well as the motivations behind choosing to stay or leave religious communities.

Q&A WITH THE AUTHOR: DR. MATTHEW MAY

Q: What motivated you to pursue this research?
A: My own experience growing up in a conservative Protestant environment (and never feeling closely
connected with the beliefs or community that came with that environment) sparked my initial interest in
religious doubt. After looking at the existing research on religious doubt and mental health, it was clear there were many areas for additional research. One such area was the need for longitudinal data on people with religious doubts. The PALS dataset presented an opportunity to examine the impact of religious doubt on mental health over time, but it also presented an opportunity to compare people who drop out of religion altogether and people who stay.

Q: Were there any findings you would highlight?
A: The key finding in my paper is that an (in)congruence between beliefs and affiliation is closely tied to mental well-being. When individuals consider dropping out of religion, they generally experience better mental health outcomes when they leave (compared to those who stay). People who never consider dropping out of religion also tend to have better mental health outcomes than people who consider dropping out but decide to stay. It is the incongruence between belief (i.e., thinking about dropping out) and affiliation (staying) that creates the worst health outcomes on the measures available in the PALS dataset (feelings of depression, hopelessness, and worthlessness).

Q: Were there any findings that surprised you?
A: My findings are largely consistent with prior research showing a curvilinear relationship between religion and mental health. That is, the most religious are generally fine. The least religious are generally fine. And the people in the middle tend to have the worst health outcomes. Of course, earlier studies were cross-sectional and only focused on religious doubt. My study is longitudinal and focuses on behaviors (staying v. leaving), yet the pattern is largely the same.

Q: What were the main challenges/limitations you encountered when conducting your research?
A: There are few longitudinal studies with sufficient data on religion. The PALS dataset is not without its own limitations: a 6 year gap between waves, a predominantly Christian sample, and limited measures of mental health. Nevertheless, it was/is the best dataset available to answer my research question, and the research is an important part of the growing body of research on religion and well-being generally, and religious disaffiliation and mental health, more specifically.

Q: How might future research build on your findings?
A: We need more qualitative research on the “why?” behind my findings. Do people who stay experience worse
mental health outcomes because of their own doubts? Are they treated differently in their religious
communities? Do people who leave find social connections in new communities that may help improve their
mental health? Will the experiences of women and men who drop out of religion altogether improve as the
number of unaffiliated Americans continues to grow? How might the experiences of people in non-Christian
groups differ? There is limited quantitative data available to answer these questions, but before a large survey
can even address these questions, we need more qualitative work