Prayer After Loss: Large Study Finds Religious Coping Rarely Shields Grieving Women

When a loved one dies, many people instinctively turn to prayer, scripture, and spiritual community in search of comfort. The assumption that faith buffers grief is deeply embedded in both popular culture and clinical practice, and decades of smaller studies have suggested that religious coping can ease the psychological toll of loss. But a new longitudinal analysis of thousands of bereaved older American women delivers a strikingly sobering verdict: religious coping, at least as measured by how often women prayed or sought comfort in their spiritual beliefs, offered essentially no measurable protection against depression or anxiety in the months following a death during the COVID-19 pandemic. In some women, it was even associated with slightly worse outcomes.

The study, published in SSM – Mental Health, drew on the Nurses’ Health Study II COVID-19 substudy, an offshoot of one of the longest-running and most influential women’s health cohorts in the world. The parent cohort began in 1989, enrolling 116,429 U.S. female registered nurses aged 25 to 42, who have completed biennial questionnaires ever since. In May 2020, as the pandemic surged, researchers invited roughly 56,000 active participants to join a dedicated substudy tracking mental health through monthly and then quarterly surveys. More than 39,000 women, with a mean age of about 65, enrolled. From this framework, the research team identified 6,868 women who reported having lost a loved one or someone important to them since March 1, 2020, and who had answered questions about religious coping.

The exposure of interest was captured with two deceptively simple items drawn from the Brief COPE inventory: whether, in the past seven days, the participant had tried to find comfort in her religion or spiritual beliefs, and whether she had prayed. Responses ranged from “not at all” to “a lot,” and the researchers constructed several versions of the exposure, including a composite measure, separate binary and continuous indicators for religious practice and prayer, and a high-religiosity subgroup of women who reported engaging in both activities frequently. The outcomes were assessed with the PHQ-4, a validated four-item screening instrument that yields a composite psychological distress score along with separate depression (PHQ-2) and anxiety (GAD-2) subscales, averaged across two follow-up waves in early 2021.

One of the study’s most eye-opening findings came before any outcome analysis: religious coping is astonishingly common in this population. Nearly 88 percent of bereaved women reported engaging in it, with prayer being the more prevalent form. Women who used religious coping also tended to report higher gratitude, hope, flourishing, mindfulness, community belongingness, and use of other coping strategies, painting a picture of faith as part of a broader resilient profile rather than an isolated behavior. Yet their baseline psychological distress scores were essentially identical to those of women who did not turn to religion, hinting that the apparent advantages of the religious group might reflect who they were rather than what their faith did for them.

That distinction is precisely where the study’s methodological ambition lies. Rather than simply comparing outcomes between groups, the researchers deployed inverse probability of treatment weighting, a causal inference technique that estimates each woman’s propensity to engage in religious coping based on a rich set of pre-exposure covariates, then reweights the sample so that the two groups are statistically balanced. The covariate set was unusually comprehensive, spanning demographics, neighborhood-level income, occupational status, social connectedness, loneliness, gratitude, sleep, physical activity, other coping methods, and critically, baseline mental health, including a pre-pandemic depression index built from 2017 cohort data using CES-D scores, physician diagnoses, and antidepressant use. Missing data were addressed through multiple imputation with chained equations, and the analysis was repeated under three different timing specifications to guard against reverse causality.

After all that statistical machinery, the headline result was null. Religious coping showed no significant association with subsequent overall distress (a beta coefficient of 0.01, with a confidence interval spanning zero), depression, or anxiety. The pattern held whether coping was treated as a binary yes-or-no behavior, a continuous frequency measure, or analyzed separately for prayer and religious practice. It held among women grieving immediate family members, extended family, and friends alike, and it held among the nearly 2,000 women who reported high religiosity. Sensitivity analyses using complete cases, separately analyzed timepoints, and alternative covariate windows all converged on the same conclusion: within the limits of what frequency-based measures can capture, religious coping neither protected nor harmed the average bereaved woman’s mental health.

But the stratified analyses revealed a wrinkle that complicates the tidy null. When the researchers split the sample by baseline depression severity, they found that among women with no depressive symptoms at baseline (PHQ-4 scores of 0 to 2), religious coping predicted a small but statistically significant increase in subsequent distress, with an overall PHQ-4 coefficient of 0.16. Among women with mild depression, the association flipped direction but was not significant, and among those with moderate to severe symptoms it was essentially null. A formal interaction test confirmed that the effect of religious coping differed meaningfully between women with and without baseline depression. The authors suggest several interpretations: for previously undepressed women, a sudden turn toward intensive religious coping might itself signal emerging distress, or it might displace more habitual and better-matched coping strategies, a possibility consistent with prior work linking coping inflexibility to worse outcomes.

The null findings in the most distressed subgroup echo one of the few prior prospective studies, an eight-week follow-up of university students in which baseline positive religious coping failed to buffer the effect of stress on later depression. The researchers also point to a conceptual limitation that may underlie much of the mixed literature: religious coping is not one thing. Decades of research, notably Kenneth Pargament’s work on the Brief RCOPE, distinguish positive religious coping, reflecting a secure relationship with the sacred, from negative religious coping, or spiritual struggle, marked by anger toward God, feeling abandoned, or questioning. These two dimensions are only weakly correlated and show divergent mental health associations, with positive coping generally linked to growth and well-being and negative coping to heightened distress. The current study could only measure how often women prayed or sought comfort, not the tone, content, or purpose of that engagement, leaving open the possibility that beneficial and harmful forms of religious coping canceled each other out in the aggregate.

There are other caveats worth weighing. The bereavement question captured any loss since March 2020 without pinpointing timing, so some women may have been surveyed long after their loss while others were in acute grief. The measure could not distinguish ongoing grief-related distress from more resolved mourning, and the sample, drawn from a cohort of predominantly white, highly educated nurses, limits generalizability. The authors also note that even positive religious coping may backfire under conditions of extreme, overlapping stress: conservation of resources theory suggests that when people invest coping resources such as prayer and hope without experiencing relief, the mismatch can amplify rather than relieve suffering, a dynamic that may have been especially relevant during the prolonged uncertainty of the pandemic.

For clinicians, grief counselors, and spiritual care providers, the practical message is one of calibrated expectations rather than dismissal. Religious coping is nearly universal among older women facing loss, and it may well provide meaning, community, and comfort that frequency-based survey items cannot detect. But the evidence here suggests it should not be assumed to be a protective intervention for mental health symptoms, particularly for women who were not depressed to begin with, in whom it may even mark or foreshadow emerging distress. Severe grief-related depression, the findings imply, likely requires more intensive support than faith practices alone can supply. The authors call for future research using instruments that separately capture positive and negative religious coping, along with the motivations behind prayer, so that spiritual and psychological support can be tailored to what each grieving person actually needs rather than to what tradition assumes helps.

Subject of Research: Religious coping and depression and anxiety outcomes among bereaved older women during the COVID-19 pandemic

Article Title: Religious coping and mental health outcomes among bereaved older women in the Nurses Health Study II

Article References: Sung, M., Cowden, R., Kang, J. H., & Denckla, C. (2026). Religious coping and mental health outcomes among bereaved older women in the Nurses Health Study II. SSM – Mental Health, 10, Article 100698. https://doi.org/10.1016/j.ssmmh.2026.100698

Image Credits: AI Generated

DOI: 10.1016/j.ssmmh.2026.100698

Keywords: religious coping, bereavement, grief, depression, anxiety, Nurses' Health Study II, COVID-19, mental health, prayer, older women, epidemiology, causal inference