Religious and Spiritual Correlates of Mental Health in a Sample of 31,854 U.S. Adults
Aravind Sreeram, Department of Psychiatry and Behavioral Sciences, The Johns Hopkins University School of Medicine, Sean Tackett, Division of General Internal Medicine, Johns Hopkins Bayview Medical Center, and Margaret Chisolm and Khalid Elzamzamy, Department of Psychiatry and Behavioral Sciences, The Johns Hopkins University School of Medicine
Background: Religiosity and spirituality (R/S) are increasingly recognized as powerful correlates of lower mental health (MH) morbidity and mortality, predicting reduced depression, anxiety, and suicide risk. Despite the multidimensional nature of R/S, prior studies have largely relied on unidimensional measures in narrow samples, limiting understanding of which dimensions underlie these relationships. Identifying the most influential dimensions is important for developing effective, person-centered strategies for individuals experiencing mental health burden. The aim of this study was to investigate which R/S dimensions in a national sample, if any, show associations with improved self-rated MH and fewer self-reported symptoms of anxiety, depressed mood (DM), anhedonia, and worry.
Methods: The Global Flourishing Study (GFS) is a survey taking place across 22 countries that includes multidimensional R/S measures and MH outcomes. Using data from U.S. adults in the 2024 GFS cohort, we examined the relationship between a priori-defined MH outcomes (self-rated MH [primary outcome], anxiety, DM, anhedonia, worry) and R/S measures (affiliation, salience [importance in daily life], connectedness, religious service attendance), with age, sex, race, education, and income as covariates. MH symptoms were each measured on 4-point ordinal scales (from ‘Nearly every day’ to ‘Not at all’); self-rated MH was on a 0-10 scale (higher=better). R/S measures included connectedness (4-point scale, ‘Never’ to ‘Always’), religious affiliation (spiritual/religious/neither/both), salience (yes/no), and service attendance (5-point scale, ‘Never’ to ‘>1×/week’). We first ran univariable analyses, then fit multivariable linear regressions (β>0 = higher MH) and ordinal logistic regressions (OR>1 = more likely to have no symptoms). Models used extreme-category contrasts (e.g., “Not at all” vs. “Nearly every day”, “Always” vs. “Never”) for ordinal variables with >2 levels. All confidence intervals (CIs) are 95%.
Results: Our 31,854 adults were most commonly male (16,485; 52%), 66-80 years old (12,568; 39%), White (26,217; 82%), and Christian (20,540; 64%).
In univariable analyses, all dimensions of R/S were associated with higher self-rated mental health: connectedness (β=1.40; CI 1.34–1.46), service attendance (β=1.02; CI 0.94–1.10), salience (β=0.52; CI 0.48–0.57), and combined religious and spiritual (R+S) affiliation (β=0.52; CI 0.47–0.58). When looking at frequency of MH symptoms, connectedness was correlated with lower anxiety (OR 2.40), DM (OR 2.79), anhedonia (OR 2.67), and worry (OR 2.10). Service attendance mirrored these results for anxiety (OR 2.17), DM (OR 2.23), anhedonia (OR 2.27), and worry (OR 2.23). Salience also showed similar trends for anxiety (OR 1.37), DM (OR 1.37), anhedonia (OR 1.35), and worry (OR 1.22). Finally, R+S affiliation also was correlated with lower anxiety (OR 1.25), DM (OR 1.31), anhedonia (OR 1.33), and worry (OR 1.06).
In multivariable analyses, higher self-rated MH was associated with higher connectedness (β=1.54; CI 1.46–1.63) and service attendance (β=0.13; CI 0.04–0.22), while lower MH was now associated with salience (β=-0.23; CI -0.29–-0.17) and R+S affiliation (β=-0.31; CI -0.38–-0.24). Connectedness still correlated with fewer symptoms of anxiety (OR 2.96), DM (OR 4.45), anhedonia (OR 4.23), and worry (OR 3.17). Frequent religious service attendance similarly still correlated with fewer symptoms of anxiety (OR 1.30), DM (OR 1.18), anhedonia (OR 1.23), and worry (OR 1.50). Salience, however, was now associated with more symptoms of anxiety (OR 0.81), DM (OR 0.74), anhedonia (OR 0.75), and worry (OR 0.75). Similarly, R+S affiliation was associated with more symptoms of anxiety (OR 0.72), DM (OR 0.69), anhedonia (OR 0.70), and worry (OR 0.65).
Conclusion: R/S are comprised of distinct dimensions, and MH assessments should disaggregate them rather than relying on a single, composite measure. Internal connectedness and religious service attendance may underpin apparent benefits of other R/S measures like religious affiliation and reported salience in daily life. Future work should focus on elucidating mechanisms of connectedness and participation, as well as developing identity measures beyond affiliations ascribed at birth.
Methods: The Global Flourishing Study (GFS) is a survey taking place across 22 countries that includes multidimensional R/S measures and MH outcomes. Using data from U.S. adults in the 2024 GFS cohort, we examined the relationship between a priori-defined MH outcomes (self-rated MH [primary outcome], anxiety, DM, anhedonia, worry) and R/S measures (affiliation, salience [importance in daily life], connectedness, religious service attendance), with age, sex, race, education, and income as covariates. MH symptoms were each measured on 4-point ordinal scales (from ‘Nearly every day’ to ‘Not at all’); self-rated MH was on a 0-10 scale (higher=better). R/S measures included connectedness (4-point scale, ‘Never’ to ‘Always’), religious affiliation (spiritual/religious/neither/both), salience (yes/no), and service attendance (5-point scale, ‘Never’ to ‘>1×/week’). We first ran univariable analyses, then fit multivariable linear regressions (β>0 = higher MH) and ordinal logistic regressions (OR>1 = more likely to have no symptoms). Models used extreme-category contrasts (e.g., “Not at all” vs. “Nearly every day”, “Always” vs. “Never”) for ordinal variables with >2 levels. All confidence intervals (CIs) are 95%.
Results: Our 31,854 adults were most commonly male (16,485; 52%), 66-80 years old (12,568; 39%), White (26,217; 82%), and Christian (20,540; 64%).
In univariable analyses, all dimensions of R/S were associated with higher self-rated mental health: connectedness (β=1.40; CI 1.34–1.46), service attendance (β=1.02; CI 0.94–1.10), salience (β=0.52; CI 0.48–0.57), and combined religious and spiritual (R+S) affiliation (β=0.52; CI 0.47–0.58). When looking at frequency of MH symptoms, connectedness was correlated with lower anxiety (OR 2.40), DM (OR 2.79), anhedonia (OR 2.67), and worry (OR 2.10). Service attendance mirrored these results for anxiety (OR 2.17), DM (OR 2.23), anhedonia (OR 2.27), and worry (OR 2.23). Salience also showed similar trends for anxiety (OR 1.37), DM (OR 1.37), anhedonia (OR 1.35), and worry (OR 1.22). Finally, R+S affiliation also was correlated with lower anxiety (OR 1.25), DM (OR 1.31), anhedonia (OR 1.33), and worry (OR 1.06).
In multivariable analyses, higher self-rated MH was associated with higher connectedness (β=1.54; CI 1.46–1.63) and service attendance (β=0.13; CI 0.04–0.22), while lower MH was now associated with salience (β=-0.23; CI -0.29–-0.17) and R+S affiliation (β=-0.31; CI -0.38–-0.24). Connectedness still correlated with fewer symptoms of anxiety (OR 2.96), DM (OR 4.45), anhedonia (OR 4.23), and worry (OR 3.17). Frequent religious service attendance similarly still correlated with fewer symptoms of anxiety (OR 1.30), DM (OR 1.18), anhedonia (OR 1.23), and worry (OR 1.50). Salience, however, was now associated with more symptoms of anxiety (OR 0.81), DM (OR 0.74), anhedonia (OR 0.75), and worry (OR 0.75). Similarly, R+S affiliation was associated with more symptoms of anxiety (OR 0.72), DM (OR 0.69), anhedonia (OR 0.70), and worry (OR 0.65).
Conclusion: R/S are comprised of distinct dimensions, and MH assessments should disaggregate them rather than relying on a single, composite measure. Internal connectedness and religious service attendance may underpin apparent benefits of other R/S measures like religious affiliation and reported salience in daily life. Future work should focus on elucidating mechanisms of connectedness and participation, as well as developing identity measures beyond affiliations ascribed at birth.