Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Religious service attendance, divorce, and remarriage among U.S. nurses in mid and late life

  • Shanshan Li,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Writing – original draft, Writing – review & editing

    Affiliation Slone Epidemiology Center, Boston University, Boston, Massachusetts, United States of America

  • Laura D. Kubzansky,

    Roles Conceptualization, Resources, Writing – review & editing

    Affiliation Department of Social and Behavioral Sciences, Harvard T. H. Chan School of Public Health, Boston, Massachusetts, United States of America

  • Tyler J. VanderWeele

    Roles Conceptualization, Investigation, Methodology, Project administration, Resources, Supervision, Writing – review & editing

    tvanderw@hsph.harvard.edu

    Affiliations Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, Massachusetts, United States of America, Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, Massachusetts, United States of America

Religious service attendance, divorce, and remarriage among U.S. nurses in mid and late life

  • Shanshan Li, 
  • Laura D. Kubzansky, 
  • Tyler J. VanderWeele
PLOS
x

Abstract

Prior research has suggested religious participation can promote marital satisfaction and stability. However, current literature has mainly focused on early life divorce, and used cross-sectional data, leaving open the question of the directionality of effects. We evaluated the prospective associations between service attendance and marital stability in mid and late life considering either 1) divorce or separation; or 2) remarriage, as separate outcomes. Data were drawn from the Nurses’ Health Study, a large prospective cohort study that consisted of US female nurses in their 50s at study enrollment, with repeated measures of service attendance and marital status over 14 years of follow-up from 1996–2010. During follow up, among 66,444 initially married nurses who were mainly Christians, frequent service attendance was associated with 50% lower risk of divorce (95% CI: 32%, 63%), and 52% lower risk of either divorce or separation (95%CI: 37%, 63%). Among initially divorced or separated women, frequent service attendance was not associated with subsequent likelihood of remarriage; however, among widowed women, women who attended services frequently had 49% increased likelihood of remarriage (95% CI: 13%, 97%) compared to those women who did not. The study provides evidence that in this cohort of US nurses, frequent service attendance is associated with lower risk of becoming divorced in mid- and late- life, and increased likelihood of remarriage among widowed nurses, but not among divorced or separated nurses.

Introduction

The United States is a religious country. The Pew Research center reported that 92% U.S. adults in 2007 and 89% U.S. adults in 2012 believe in God, with a majority (71%) identifying as Christian [1]. Religion plays an important role in marriage and church attendance has been shown to be a strong predictor for remaining married versus divorcing [2]. Marriage, either marriage cohesiveness or divorce, are a special case of social groups cohesiveness [2], conferring psychological and economic benefits to couples and society [35]. However, despite these positive perspectives on marriage, the crude marriage rate declined from 8.2 per 1,000 total population in 2000 to 6.9 in 2015 [6]. Divorce rates have fallen somewhat over this time, with 4.0 per 1,000 reported in 2000 which declined to 3.1 in 2015 [6, 7]. Divorce rates in middle aged groups however doubled from 1990 to 2010 [8, 9]. There is some evidence that the tendency for divorced men and women, particularly aged 50 years older, to get remarried has increased in recent years [7, 10, 11]. Prior research has mainly focused on early-life divorces [7, 9, 1214]. There is limited evidence on subsequent re-marriage among the increasingly growing group of late life divorcees. The baby boomers in the U.S. are now in mid- to late- life. With late-life divorce rate doubling in recent years, there is a need to better understand divorce and remarriage for mid- and late- life women [13, 15].

Classic Durkheimian theory suggests that the institutions of religion contribute to human well-being, fostering and supporting marriage; and married couples who are religious may be happier and more satisfied with life [711, 1623]. Marriage may be viewed as an end in itself, but is also associated longitudinally with better physical and mental health and greater life satisfaction [4, 5, 2434]. Although marital conflicts can sometimes seem irreconcilable, interventions, counseling, or shared activity that might be available earlier in the relationship, when problems first begin to develop, could be effective ways to prevent divorce and promote well-being [12, 35]. Current evidence on service attendance and the likelihood of remarriage after widowhood or separation is sparse. It is unclear whether religious service attendance is associated with subsequent likelihood of remarriage generally, and whether this depends on whether the marriage ended or someone was widowed.

In addition, other research has suggested religious participation can play an important role in promoting strong marital bonds, happiness, and stability [3540]. On the other hand, religious attendance and affiliation may operate as a barrier to divorce [41, 42]. Religious service attendance might affect marital stability because of religious teaching of marriage as sacred, teachings against divorce, a place to meet other couples and families, and programs, counseling and retreats that support marriage and family life. Religious affiliation may affect family life because of different teachings and practices concerning marriage, divorce, and remarriage [35, 38]. Literature on remarriage has suggested that there are significant denominational subcultural variations for remarriage among Americans [15]. For example, conservative Protestants are most likely to remarry and least likely to cohabit [15, 43]. Moreover, remarriage after a divorce may be viewed less favorably than remarriage after widowhood or separation by some religious denominations [15, 35, 38]. Despite many qualitative studies that have examined religious participation, affiliation and marriage, the quantitative assessment of the joint effects of religious service attendance and religious affiliation on marriage is currently unclear. Here we will examine differences comparing Catholic and Protestant denominations.

A number of methodological concerns limit research to date in the field of religion and marriage. Most existing studies on religion and marital status have used cross-sectional data [40, 44, 45], which may be subject to reverse causation issues with divorce itself affecting subsequent service attendance [46]. Data from prospective cohort studies on the interrelationship between religion and marital status over time are sparse [38, 39, 47]. Contemporary data with longer follow up are needed. Prior research has also included only a limited set of potential confounders when examining these associations, leaving open the strong possibility of unmeasured and residual confounding. For example, most covariates considered in prior studies focused on socio-economic related variables. Health and lifestyle related information, such as, parity (number of times that the nurse has given birth), depression, or good physical function, were not available. Various lifestyle factors may affect both service attendance and marital conflict and stability, thereby confounding the association [25, 4852]. In addition, existing literature has not taken into account prior religious service attendance, to further rule out issues of reverse causation.

To address these gaps in knowledge, we conducted a more rigorous test of the hypothesis that frequent religious service attendance is associated with a lower likelihood of divorce, and increased likelihood of remarriage among nurses in mid- to late- life. Using data from a large prospective cohort study of 66,444 US female nurses who were mainly Christians, we tested the following two hypotheses: 1) frequent religious service attendance is associated with lower subsequent odds of divorce or separation; and 2) frequent religious service attendance is associated with higher likelihood of remarriage. We also evaluated these associations by Catholic versus Protestant affiliation to assess the joint effects of attendance and affiliation.

Materials and methods

The Nurses’ Health Study

Our study sample was from the Nurses’ Health Study which began in 1976 when 121,700 married female registered nurses aged 30 to 55 years, completed a mailed questionnaire about their medical history, dietary and lifestyle habits. They resided in 1 of 11 US states with the largest number of registrants (New York, California, Pennsylvania, Ohio, Massachusetts, New Jersey, Michigan, Texas, Florida, Connecticut, and Maryland). Among those invited, 71.2% participants returned the questionnaire. The overall NHS in 1976 was 98% white, with a relatively higher socioeconomic status. While this limits generalizability, the homogeneity does enhance the interval validity of the NHS, as does also the high quality data as well as high follow-up rates [53].

Follow-up questionnaires were sent every 2 years, with follow-up for more than 90% the cohort remaining complete. Data on service attendance was available starting in 1992, 1996 and every four years subsequently; for these analyses we thus restricted the study population to nurses who were currently married in 1996 and also reported their religious service attendance information in 1996 (n = 66,444). For analyses on remarriage, we restricted the study population to nurses who were divorced (n = 7,311), separated (n = 986), or widowed (n = 21,372) in 1996 and also reported their religious service attendance information in 1996. Follow up data for these women is available through 2010. The Nurses’ Health Study was approved by the Brigham and Women’s hospital and Harvard T. H. Chan School of Public Health Institutional Review Board of (Boston, MA).

Ascertainment of religious service attendance and religious affiliation

Information on religious service attendance was collected from self-administered questionnaires in 1992 and 1996, and every four years subsequently. The study participants were asked, “How often do you go to religious meetings or services?” Response categories included “more than once a week, once a week, 1–3 times per month, less than once per month, never or almost never.” Previous literature suggested that even though religious service attendance tends to be over-reported, it is still a good indicator of communal religious practice [54]. Moreover, relative rank ordering across frequency may still be preserved.

Religious affiliation was self-reported, with a majority of the NHS participants being Protestant or Catholic. Due to small numbers in other religious group in our study sample, we restricted our analyses on religious affiliation to women who reported being either “Protestant” or “Catholic”.

Religious service attendance was modeled as four categories: “more than once a week, once a week, less than once per week, never or almost never.”

Ascertainment of divorce or separation

The NHS participants self-reported their marital status every four years from 1992 to 2010. We defined a divorce as the study participant self-reporting any divorce event during the follow up period from 1996 to 2010 (ever been divorced, yes or no) and evaluated both the presence and timing of such [55, 56]. Divorce or separation was defined by self-report as either divorce or separation during 1996–2010 (either being divorced or separated, yes or no). During the follow-up period, we identified 685 total events for self-reported divorce, and 924 total events for self-reported divorce or separation.

Ascertainment of remarriage

We defined a remarriage event as women who previously reported being divorced, separated, or widowed at baseline 1996, but self-reported being currently married in any of the follow-up questionnaires from 2000 until 2010. Widowhood information was obtained through women’s self-reported questionnaire. We categorized remarriage among four groups of women: 1) all previously married women who self-reported as unmarried for any reason (i.e., divorced or separated or widowed) in 1996; 2) only women who self-reported being divorced in 1996; 3) only women who self-reported being separated in 1996; and 4) only women who self-reported as widows in 1996.

Covariates

Potential confounders for the association between religious service attendance and marital status (divorce, remarriage) were selected a priori based on the literature and subject knowledge [57, 58]. We included the socio-demographic characteristics of the study participants, including husband’s education, median family income, geographic region, and unemployment in the past two years. We also evaluated lifestyle and health characteristics that associated with service attendance and marriage, self-reported prior religious attendance frequency (never, < 1/week, > 1/week in 1992), and prior divorce history (yes, no). We included the following various health-related lifestyle factors and health conditions because health may affect both service attendance and marital conflict and stability: alcohol consumption and healthy eating index measured by a validated food frequency questionnaire, good physical or function measured by the 36-Item Short Form Health Survey (SF-36), depression measured by the Center for Epidemiologic Studies Depression -10 Scale, parity (number of times that the nurse has given birth), physical exercise, hypertension, hypercholesterolemia, type 2 diabetes, menopausal status and postmenopausal hormone use, physical exam in the past 2 years, smoking status, pack-years, and body mass index (BMI). Prior history of divorce was defined as “ever reported divorce in the past 20 years (1976–1996 (our analytic baseline)).

Statistical analyses

The primary exposure was religious service attendance in 1996, with prior attendance in 1992 controlled for as a covariate in an effort to mitigate concerns about reverse causation. Cox proportional hazard model and multivariate logistic regression was used to estimate the longitudinal association between the 1996 service attendance and likelihood of divorce/separation over the follow-up period. Model 1 was age adjusted. Model 2, the fully adjusted model, included baseline religious service attendance level in 1992, demographics (age, husband’s education, median family income, geographic region, unemployed in past two years, prior history of divorce), health conditions (parity (Number of times that nurses has given birth to a fetus with a gestational age of 24 weeks or more), hypertension, hypercholesterolemia, type 2 diabetes, menopausal status, postmenopausal hormone use, physical function, depression status), health lifestyle behaviors (smoking status, pack years, physical exercise, alcohol consumption, healthy diet, physical exam in past 2 years, BMI). For Cox model, we also adjusted for calendar year and questionnaire cycle. We tested proportional hazard assumption by adding an interaction term between services attendance and time.

We performed interaction subgroup analyses comparing self-identified Protestants and Catholics. Specifically, we evaluated interaction both on multiplicative and additive scales (using the relative excess risk due to interaction) [59, 60]. As a sensitivity analyses, we further used propensity score subclassification by quintiles to obtain effect estimates (comparing ever vs. never service attendance). We also assessed how substantial unmeasured confounding (for example, unmeasured marital status of the nurses’ parents in childhood) would have to be to explain away the associations [46].

Results and discussion

Association of religious service attendance with subsequent divorce

Of 66,444 women who were currently married at baseline year 1996, 924 of them became either divorced (n = 685), or separated (n = 239) from 1996–2010. Compared to women who never attended religious services, those who attended more than once per week in 1996 had lower prevalence of depression, drank less alcohol in midlife, were less likely to be current smoker, and less likely to be nulliparous (Table 1). The majority of our study participants were Caucasian (98%, S1 Table).

thumbnail
Table 1. Age-adjusted baseline characteristics of study participants by frequency of religious service attendance in 1996.

https://doi.org/10.1371/journal.pone.0207778.t001

Compared to women who never attended religious services, those who attended services more than once per week were 42% less likely to get divorced, or 47% less likely to get either divorced or separated (HR = 0.58, 95% CI: 0.44–0.74, p trend <0.0001 for divorce only, HR = 0.53, 95% CI: 0.42–0.67, p trend <0.0001 for divorce or separation, Table 2). Results were similar using propensity score stratification by quintiles (HR = 0.53, 95% CI: 0.41–0.69 for divorce only, HR = 0.49, 95% CI: 0.39–0.62 for divorce or separation). We also found similar results when using logistic regression mode (OR = 0.50, 95% CI: 0.37–0.68, p trend <0.0001 for divorce only, OR = 0.48, 95% CI: 0.37–0.63, p trend <0.0001 for divorce or separation, S2 Table).

thumbnail
Table 2. Multivariate adjusted association between religious services attendance and subsequent divorce or separation in the Nurses’ Health Study, 1996–2010.

https://doi.org/10.1371/journal.pone.0207778.t002

In our study, we did not have information on duration of current marriage, which may be a potential unmeasured confounder. This is so because there is some evidence that religious people tend to marry young, and divorce rate varies by marriage duration. However, for unmeasured confounding to explain away the odds ratio estimate of 0.50 for divorce, an unmeasured confounder associated with both divorce and regular religious service attendance by risk ratios of 3.4-fold each, above and beyond the measured confounders, could do so but weaker confounding could not. For an unmeasured confounder to bring the upper confidence limit of 0.68 for this estimate above 1, an unmeasured confounder that was associated with both divorce and regular service attendance by risk ratios of 2.3-fold each could do so, but weaker confounding could not. Similar substantial confounding would be needed to explain away the other estimates.

We further examined the joint effects of service attendance and religious affiliation. We found multiplicative interaction for divorce. Attending religious services once or more per week was associated with 52% lower (95% CI: 40%-66%) likelihood of divorce for Catholics, but only a 32% lower (95% CI: 10%-43%) likelihood of divorce for Protestants (p-value for multiplicative interaction = 0.02). The large effect size for Catholics was in part due to the higher likelihood of divorce for those not attending (Table 3, S3 Table).

thumbnail
Table 3. Joint effect of religious service attendance in 1996 and religious affiliation and subsequent divorce or separate.

https://doi.org/10.1371/journal.pone.0207778.t003

Association of religious service attendance with subsequent remarriage

During the follow-up from 1996–2010, we identified 1567 remarriage events among women who had divorced, separated, or widowed in 1996, with a total of 723 events among women divorced in 1996, 197 events among women separated in 1996, and 695 events among women widowed in 1996.

Women who attended religious services once per week or more, and who were widows in 1996, had significantly higher likelihood of remarriage (HR = 2.06, 95% CI: 1.52–2.80, p for trend ≤ 0.001, Table 4), but those who were divorced or separated did not. Results were similar using logistic regression model (S4 Table) or propensity score subclassification by quintiles (OR = 1.50, 95% CI: 1.17–1.93 for remarriage after widowhood; OR = 0.92, 95% CI: 0.70–1.20 for remarriage after divorce, OR = 0.74, 95% CI: 0.44–1.25 for remarriage after separation). In examining the joint effects of religious service attendance and religious affiliation on remarriage, there was perhaps some indication that Catholics regularly attending religious services were less likely to remarry than Protestants when results were aggregated over divorced, widowed, and separated women (Table 5, S5 Table). Among women who attended services once or more per week, Catholics had 34% lower likelihood of remarriage than Protestants (95% CI: 22%-45%; p value for multiplicative interaction = 0.09, p value for additive interaction = 0.06).

thumbnail
Table 4. Multivariate adjusted association between religious services attendance and subsequent remarriage in the Nurses’ Health Study, 1996–2010.

https://doi.org/10.1371/journal.pone.0207778.t004

thumbnail
Table 5. Joint effect of religious service attendance in 1996 and religious affiliation on subsequent remarriage.

https://doi.org/10.1371/journal.pone.0207778.t005

Discussion

In this large prospective cohort study of 66,444 married U.S. nurses with 14 years of follow up from mid to late life, and repeated measurements of service attendance to allow for control of prior service attendance, nurses who attended religious service more than once per week had a 50% lower likelihood of subsequent divorce or separation, compared to those who never attend. Among widowed nurses, those who attended services more than once per week had a 49% higher likelihood of remarriage, compared to those who never attended services. However, for divorced or separated -nurses, religious service attendance was not significantly associated with the likelihood of remarriage. We found the associations between religious service attendance divorce, and remarriage were stronger for Catholics than Protestants.

Existing research on religion and marriage has mainly focused on early-life divorces [7, 9, 1214]. Currently little is known regarding late life marriage, divorce and remarriage. Given the increasingly growing group of late life divorcees and with the late-life divorce rate doubling in recent years, there is a need to advance our understanding regarding divorce and remarriage for mid- and late- life women [13, 15]. Social science literature on religiosity and marital status has been based mostly on two measures of religious participation: service attendance [16, 47, 6164] and religious affiliation [16, 57, 58, 65]. Religious service attendance might affect divorce and remarriage because of religious teaching of marriage as sacred, marriage counseling, and providing a place to meet other couples and families, and support family life. There is evidence that religiosity reduces the likelihood of marital infidelity among these couples, and may further indirectly reduce the likelihood of subsequent divorce by increasing levels of marital satisfaction [36]. Religious service attendance had previously been shown to be predictive of stronger marital bonds, greater marital happiness, and stability [38, 66]. In our study, we found that among a group of US nurses in their 50s, those who attended religious services more than once per week had a 50% lower likelihood of subsequent divorce or separation, compared to those who never attend. Our results on associations between service attendance and divorce are consistent with findings from previous research [7, 38, 67]. We expanded prior literature on this by focusing on mid and late life marriage and divorce.

Evidence on religious participation and late life divorce is sparse, but the literature on studies of remarriage after divorce or widowhood are even more limited [15, 6870]. Remarriage is a complex and understudied process [71, 72]. Bohannan proposed six important developmental tasks of remarriage including emotional, psychic, community, parental, economic, and legal aspects of remarriage [71]. Compared with first marriages, remarriage has its own distinct assortative mating patterns and gender specific dynamics [73, 74]. For a first marriage, both men and women emphasize spousal economic status as a valuable trait; however, for remarriage, high-status men tend to marry less-educated, young women regardless of economic status [73].

To the best of our knowledge, there has been no prior study focused on the link between religion and remarriage in mid- and late- life, a period in which most divorces/remarriage occurs. Only one study has prospectively examined the effects of religion on remarriage, but was focused on young adulthood [15]. The authors reported a positive association between religious service attendance and likelihood of remarriage among American women aged 15–44 years, and reported differences in religious affiliations [15]. However, this study was based on pooling of two cross-sectional surveys in 1988 and 1994. It is unknown whether this observation is still true with longitudinal data, longer follow-up, and among contemporary elderly Americans. Our study provided evidence by using a prospective cohort design with contemporary data 1996–2010, repeated measurements of service attendance and marital status, and controlling for past service attendance in an effort to rule out possible reverse causation [46, 75]. In our study, we were able to examine these associations among US nurses in their 50’s, an age group of women that has previously been understudied. We found that among widowed nurses, those who attended services more than once per week had a 49% higher likelihood of remarriage, compared to those who never attended services. However, the associations were not significant for those women who were divorced or separated. Religion may place a certain stigma on divorce [67]. Catholic teaching prohibits remarriage after divorce, unless an annulment of the prior marriage takes place. Such teachings may in part contribute to the lower rate of divorce, but may also help explain the result that service attendance is only associated with high likelihood of remarriage for those who are widowed but not for those who are divorced or separated [76]. Another possible explanation is that women who experience a divorce or separation event may have been extremely unhappy in marriage, and thus less likely to enter a marriage again. In contrast, a woman who is a widow may benefit from the support from religious community, and may be more willing to remarry.

Strengths of our study include large sample size, prospective cohort design, long duration of follow-up, availability of repeated measures of marital status and religious service attendance, extensive covariates control, with over a 90% follow-up rate. We provided long-term evidence on religion, divorce, and remarriage among contemporary Americans. In our study, we were able to examine these associations among US nurses in their 50’s, an age group of women that has previously been understudied. Our study sample included only U.S. nurses and consisted mostly of Caucasian healthcare professionals. Compared with the general population, they generally have higher levels of education and socioeconomic status. Our results may not therefore be generalizable to the overall general population in the US, and women in other race/ethnicity, age, and cultural groups. However, this relative homogenous group does ensure excellent internal validity of our study. Further, our sensitivity analysis suggested that the results were robust to potential unmeasured confounding, such as marital duration. Although our study was an observational study, we adjusted for a number of major potential confounders of the association between religious service attendance and divorce. The results from sensitivity analyses were relatively robust. Another strength of our study is that we have detailed lifestyle factor information and were able to additionally account for the confounding factors by these variables, such as depression, and good physical function, as literature has suggested that marriage promotes health and healthy behavior which may in turn affect marital stability.

Existing literature suggests the following variables are relevant predictors for divorce: marrying as a teenager, poverty, unemployment, low education, living with one’s future spouse or another partner prior to marriage, having a premarital birth, bringing children from a previous union into a new marriage, inter-racial marriage, and growing up in a household without two continuously married parents [2, 7, 77]. We did not have data on all of these and some of these might also be related to religious service attendance. Introducing unmeasured confounding For example, religious participation is also associated with earlier timing of marriage [78]: those who attend religious services more frequently and those who report religion being important are more likely to get married at a younger age [79], which is itself a risk factor for divorce. In our study, we did not have detailed information about the duration of a study participant’s current marriage, and age at first marriage. However, we did perform sensitivity analyses to assess how strong unmeasured confounding would have to be to explain away the observed association. For an unmeasured confounder to explain away the association of service attendance with divorce, an unmeasured confounder would need to be associated with both divorce and regular religious service attendance by risk ratios of 3.4-fold each, above and beyond the measured covariates. Such substantial confounding by unmeasured factors seems very unlikely given adjustment for an extensive set of covariates and relatively homogenous study population in the NHS study.

Our study is subject to certain other limitations as well. First, men and women perceive marriage differently. Previous studies have suggested that there is a gender difference in perceptions of martial problems [7] and attitudes toward remarriage. The association we observed between religiosity and divorce/remarriage might be gender-specific. Our study focused on wives’ religious participation only. Currently, very few studies have information regarding the religiosity of both spouses, and joint service attendance [38, 80]. Future studies on religious participation and divorce/remarriage specifically for men, and for men and women jointly are needed. Second, service attendance in our study was self-reported but this has been widely used in the sociology literature and may at least preserve relative ordering of attendance. In our study, marital status was measured through self-reported questionnaire. For women who experienced both divorce and remarriage events within a 4-year interval, we were not be able to capture these events and may have misclassified them as remaining married. However, such misclassifications are probably rare. In the NHS, information was not collected regarding whether the couples were of the same sex. At the time of the NHS study baseline, same-sex couple were likely not common among married women and such misclassification would likely be rare. Third, religiosity is a multi-dimensional construct. One single measure of service attendance does not capture other aspects of religiosity, such as beliefs in religious teaching, personal spirituality, and practices of prayer or joint prayer with one’s spouse. In our study, religious affiliation information was collected through self-report and we could not separate out Protestants into conservative and mainline Protestants, as more specific groups may have heterogeneous effects. Further, the majority of our study participants reported being “Protestant” or “Catholic”, and very few women reported being in other religious groups. Our results may not generalizable to people with different cultural backgrounds, and religious beliefs. Fourth, some have argued that even though religiosity and religious service attendance may increase marital quality, and decrease marital problems [81], it may also increase the stigma associated with divorce [7] and perhaps thereby further decreases the likelihood of divorce [82]. In our study among US female nurses, we did not have marital quality and satisfaction information and it is possible that some women remained married despite low satisfaction with their marriage [83]. Future studies are needed to examine the association between frequency of religious service attendance, divorce stigma, and marital quality and satisfaction. Future studies on both spouses are still needed.

Conclusions

In conclusion, in this prospective cohort study of U.S. female nurses over a 14-year period of follow-up from mid- to late- life, we found that frequent religious services attendance halved the risk of divorce among married women. Additionally, frequent religious service attendance was also associated with increased likelihood of remarriage among widowed women, but not for divorced or separated women. Future studies on religion and divorce/remarriage are needed with other religious beliefs other than Christianity, with other dimensions of religiosity, information on both couples, and the incorporation of information on marital satisfaction and quality.

Supporting information

S1 Table. Characteristics of the Nurses’ Health Study participants included and excluded in the analytic sample.

https://doi.org/10.1371/journal.pone.0207778.s001

(DOCX)

S2 Table. Multivariate adjusted association between religious services attendance and subsequent divorce or separation in the Nurses’ Health Study, 1996–2010.

https://doi.org/10.1371/journal.pone.0207778.s002

(DOCX)

S3 Table. Joint effect of religious service attendance in 1996 and religious affiliation and subsequent divorce or separate.

https://doi.org/10.1371/journal.pone.0207778.s003

(DOCX)

S4 Table. Multivariate adjusted association between religious services attendance and subsequent remarriage in the Nurses’ Health Study, 1996–2010.

https://doi.org/10.1371/journal.pone.0207778.s004

(DOCX)

S5 Table. Joint effect of religious service attendance in 1996 and religious affiliation on subsequent remarriage.

https://doi.org/10.1371/journal.pone.0207778.s005

(DOCX)

Acknowledgments

The authors would like to thank Dr. Ying Chen for her assistance with database access.

References

  1. 1. The Pew Research Center 2018. [Accessed 10/19/2018, at http://www.pewforum.org/religious-landscape-study/]
  2. 2. Levinger G. Marital cohesiveness and dissolution: An integrative review. Journal of Marriage and the Family. 1965;27:19.
  3. 3. Stutzer A, Frey BS. Does marriage make people happy, or do happy people get married? Journal of Socio-Economics. 2006;35:326–47.
  4. 4. Uecker JE. Marriage and Mental Health among Young Adults. Journal of Health and Social Behavior. 2012;53:67–83. pmid:22328171
  5. 5. Marks N, Lambert J. Marital status continuity and change among young and midlife adults. Journal of Family Issues. 1998;19:652–86.
  6. 6. National Center for Health Statistics. Change in the Reporting of Marriage and Divorce Statistics. [Accessed 10/19/2018, at https://www.cdc.gov/nchs/nvss/marriage-divorce.htm]
  7. 7. Amato PR. Research on divorce: Continuing trends and new developments. Journal of Marriage and Family. 2010;72:650–66.
  8. 8. Lloyd SA. Divorce at mid and later life: Does the empirical evidence support the theory? Journal of Divorce. 1986;9:89–102.
  9. 9. Brown SL, Lin IF. The gray divorce revolution: rising divorce among middle-aged and older adults, 1990–2010. The Journals of Gerontology, Series B. 2012;67:731.
  10. 10. Livingston G. Chapter 2: The Demographics of Remarriage 2014 [Accessed 10/19/2018, http://www.pewsocialtrends.org/2014/11/14/chapter-2-the-demographics-of-remarriage/.
  11. 11. Livingston G. Four-in-Ten Couples are Saying “I Do,” Again. Growing Number of Adults Have Remarried. [Accessed 10/19/2018, http://www.pewsocialtrends.org/2014/11/14/four-in-ten-couples-are-saying-i-do-again/]
  12. 12. Doss BD, Cicila LN, Georgia EJ, et al. A randomized controlled trial of the web-based OurRelationship program: Effects on relationship and individual functioning. Journal of Consulting and Clinical Psychology 2016.
  13. 13. Bowen GL, Jensen TM. Late-life divorce and postdivorce adult subjective well-being. Journal of Family Issues 2017;38:1363–88.
  14. 14. Wu Z, Penning MJ. Marital and cohabiting union dissolution in middle and later life. Research on Aging. 2017:016402751769802.
  15. 15. Brown SM, Porter J. The effects of religion on remarriage among American women: Evidence from the National Survey of Family Growth. Journal of Divorce & Remarriage. 2013;54:142–62.
  16. 16. Village A, Williams E, Francis L. Living in Sin? Religion and cohabitation in Britain 1985–2005. Marriage & Family Review. 2010;46:468–79.
  17. 17. Waite LJ, Lehrer EL. The benefits from marriage and religion in the United States: a comparative analysis. Population and Development Review. 2003;29:255. pmid:19129930
  18. 18. Xu X, Hudspeth CD, Bartkowski JP. The role of cohabitation in remarriage. Journal of Marriage and Family 2006;68:261–74.
  19. 19. Bramlett MD, Mosher WD. Cohabitation, marriage, divorce, and remarriage in the United States. Vital and health statistics series 23, Data from the National Survey of Family Growth. 2002:1.
  20. 20. Smith SS. National Center for Health Statistics. Trends in patterns in marriage, divorce, and cohabitation. Health insurance coverage for U.S. children improves. Public Health Reports 2002;117:194.
  21. 21. Schoen R, Canudas-Romo V. Timing effects on divorce: 20th century experience in the United States. Journal of Marriage and Family 2006;68:749–58.
  22. 22. Cherlin AJ. American marriage in the early twenty-first century. The future of children. 2005;15:33–55. pmid:16158729
  23. 23. National Center for Health Statistics. Cohabitation, marriage, divorce, and remarriage in the United States. Hyattsville, Md.: Dept. of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics; 2002.
  24. 24. Wilcox W.B. Why marriage matters: 30 conclusions from the social sciences. New York: Institute for American Values/National Marriage Project; 2011.
  25. 25. Wilson C, Oswald A. How does marriage affect physical and psychological health? A survey of the longitudinal evidence. IDEAS Working Paper Series from RePEc. 2005.
  26. 26. Wood R, Goesling B, Avellar S. The effects of marriage on health: A synthesis of recent research evidence. IDEAS Working Paper Series from RePEc 2007.
  27. 27. Lamb KA, Lee GR, Demaris A. Union formation and depression: Selection and relationship effects. Journal of Marriage and Family. 2003;65:953–62.
  28. 28. Horwitz A, White H, Howell-White S. Becoming married and mental health: A longitudinal study of a cohort of young adults. Journal of Marriage and the Family 1996;58:895.
  29. 29. Kim HK, McKenry P. The relationship between marriage and psychological well-being. Journal of Family Issues 2002;23:885–911.
  30. 30. Simon R. Revisiting the relationships among gender, marital status, and mental health. American Journal of Sociology 2002;4:1065–96.
  31. 31. Lillard L, Panis C. Marital status and mortality: The role of health. Demography. 1996;33:313–27. pmid:8875065
  32. 32. Williams K, Umberson D. Marital status, marital transitions, and health: a gendered life course perspective. The Journal of Health and Social Behavior. 2004;45:81. pmid:15179909
  33. 33. Manzoli L, Villari P, M Pirone G, Boccia A. Marital status and mortality in the elderly: A systematic review and meta-analysis. Social Science & Medicine. 2007;64:77–94.
  34. 34. Kaplan RM, Kronick RG. Marital status and longevity in the United States population. J Epidemiol Community Health. 2006;60:760. pmid:16905719
  35. 35. Vaaler ML, Ellison CG, Powers DA. Religious influences on the risk of marital dissolution. Journal of Marriage and Family 2009;71:917–34.
  36. 36. Tuttle JD, Davis SN. Religion, infidelity, and divorce: Reexamining the effect of religious behavior on divorce among long-married couples. Journal of Divorce & Remarriage 2015;56:475–89.
  37. 37. Lehrer EL. Religion, economics, and demography: the effects of religion on education, work, and the family. London New York: Routledge; 2009.
  38. 38. Call V, Heaton T. Religious Influence on on marital stability. Journal for the Scientific Study of Religion 1997;36:382.
  39. 39. Booth A, Johnson D. R. Branaman A S A. Belief and behavior: Does religion matter in today’s marriage. Journal of Marriage and the Family. 1995;57:661.
  40. 40. Glenn N, Supancic M. Social and demographic correlates of divorce and separation in the United States: An update and reconsideration. Journal of Marriage and the Family 1984;46:563.
  41. 41. Tzeng OCS, Yoo S-I, Chataignier C. Theories of love dissolution and divorce. In: Tzeng OCS, Tzeng OCS, editors. Theories of love development, maintenance, and dissolution: Octagonal cycle and differential perspectives. Westport, CT, US: Praeger Publishers/Greenwood Publishing Group; 1992. p. 210–42.
  42. 42. Snyder DK. Toward a science of marriage and marital therapy. Contemporary Psychology 1992;37:423–5.
  43. 43. Wu Z, Schimmele CM. Repartnering after first union disruption. Journal of Marriage and Family 2005;67:27–36.
  44. 44. Mahoney A. Religion in families 1999 to 2009: A relational spirituality framework. J Marriage Fam. 2010;72:805–27. pmid:22102761
  45. 45. Heaton TB, Pratt EL. The effects of religious homogamy on marital satisfaction and stability. Journal of Family Issues 1990;11:191–207.
  46. 46. Ding P, VanderWeele TJ. Sensitivity analysis without assumptions. Epidemiology 2016;27:368–77. pmid:26841057
  47. 47. Wilcox WB, Wolfinger NH. Then comes marriage? Religion, race, and marriage in urban America. Social Science Research. 2007;36:569–89.
  48. 48. Joung IMA, van de Mheen HD, Stronks K, van Poppel FWA, Mackenbach JP. A longitudinal study of health selection in marital transitions. Social Science & Medicine. 1998;46:425–35.
  49. 49. Wright LK. The impact of Alzheimer’s disease on the marital relationship. The Gerontologist 1991;31:224–37. pmid:2044995
  50. 50. Flor H, Turk DC, Berndt Scholz O. Impact of chronic pain on the spouse: Marital, emotional and physical consequences. Journal of Psychosomatic Research.31:63–71. pmid:3820147
  51. 51. Hawley DJ, Wolfe F, Cathey MA, Roberts FK. Marital status in rheumatoid arthritis and other rheumatic disorders: a study of 7,293 patients. The Journal of rheumatology. 1991;18:654–60. pmid:1865410
  52. 52. Johnson CL. The impact of illness on late-life marriages. Journal of Marriage and Family 1985;47:165–72.
  53. 53. Bao Y, Bertoia ML, Lenart EB, et al. Origin, methods, and evolution of the three Nurses’ Health Studies. American journal of public health 2016;106:1573–81. pmid:27459450
  54. 54. Brenner PS. Identity importance and the over reporting of religious service attendance: Multiple imputation of religious attendance using the American time use study and the general social survey. Journal for the Scientific Study of Religion 2011;50:103–15.
  55. 55. Heaton TB, Albrecht SL, Martin TK. The timing of divorce. Journal of Marriage and Family 1985;47:631–9.
  56. 56. Teachman JD. Methodological issues in the analysis of family formation and dissolution. Journal of Marriage and Family 1982;44:1037–53.
  57. 57. de Graaf PM, Kalmijn M. Change and stability in the social determinants of divorce: A comparison of marriage cohorts in the Netherlands. European Sociological Review 2006;22:561–72.
  58. 58. Dempsey K, de Vaus D. Who cohabits in 2001? The significance of age, gender, religion and ethnicity. Journal of Sociology 2004;40:157–78.
  59. 59. Rothman KJ. Modern epidemiology. 1st ed. ed. Boston: Little, Brown; 1986.
  60. 60. Hosmer DW, Lemeshow S. Confidence interval estimation of interaction. Epidemiology 1992;3:452–6. pmid:1391139
  61. 61. Thornton A, Axinn WG, Hill DH. Reciprocal effects of religiosity, cohabitation, and marriage. American Journal of Sociology 1992;98:628–51.
  62. 62. Bahr H, Chadwick B. Religion and family in Middletown, USA. Journal of Marriage and the Family. 1985;47:407–14.
  63. 63. Maselko J, Buka S. Religious activity and lifetime prevalence of psychiatric disorder. Soc Psychiat Epidemiol. 2008;43:18–24.
  64. 64. Petrowsky M. Marital status, sex, and the social networks of the elderly. Journal of Marriage and the Family 1976;38:749.
  65. 65. Lehrer E. The role of religion in union formation: An economic perspective. Population Research and Policy Review 2004;23:161–85.
  66. 66. Strawbridge W, Shema S, Cohen R, Kaplan G. Religious attendance increases survival by improving and maintaining good health behaviors, mental health, and social relationships. Ann Behav Med 2001;23:68–74. pmid:11302358
  67. 67. Amato PR, Rogers SJ. A longitudinal study of marital problems and subsequent divorce. Journal of Marriage and Family 1997;59:612–24.
  68. 68. Cherlin A. Religion and remarriage: Reply to Halliday. American Journal of Sociology 1980;86:636.
  69. 69. Graaf PMD, Kalmijn M. Alternative routes in the remarriage market: Competing-risk analyses of union formation after divorce. Social Forces 2003;81:1459–98.
  70. 70. McDaniel S, Boco AG, Zella S. Changing patterns of religious affiliation, religiosity, and marital dissolution: A 35-year study of three-generation families. Journal of Divorce & Remarriage 2013;54:629–57.
  71. 71. Goetting A. The six stations of remarriage: Developmental tasks of remarriage after divorce. Family Relations 1982;31:213–22.
  72. 72. Carr D. The desire to date and remarry among older widows and widowers. Journal of Marriage and Family 2004;66:1051–68.
  73. 73. Shafer K. Unique matching patterns in remarriage. Journal of Family Issues 2013;34:1500–35.
  74. 74. Shafer K, Jensen T, Larson J. Relationship effort, satisfaction, and stability: Differences across union type. Journal of Marital and Family Therapy 2014;40:212–32. pmid:24320175
  75. 75. Vanderweele TJ. Re: "Religious service attendance and major depression: a case of reverse causality?". American journal of epidemiology. 2013;177:275–6. pmid:23292958
  76. 76. Catholic C. Catechism of the Catholic Church. Vatican City: Chicago, Ill. 2Ed.: Libreria Editrice Vaticana [distributed by] Loyola University Press; 2000.
  77. 77. Levinger G. Sources of marital dissatisfaction among applicants for divorce. American Journal of Orthopsychiatry. 1966;36:803–7. pmid:5971487
  78. 78. Uecker JE, Stokes CE. Early marriage in the United States. Journal of Marriage and Family 2008;70:835–46. pmid:20305796
  79. 79. Eggebeen D, Dew J. The role of religion in adolescence for family formation in young adulthood. Journal of Marriage and Family 2009;71:108–21. pmid:25473126
  80. 80. Brown E, Orbuch TL, Bauermeister JA. Religiosity and marital stability among black american and white american couples. Family Relations 2008;57:186–97.
  81. 81. Ellison C. Religious heterogamy and marital conflict: Finding from the National Survey of Families and Households. Journal of Family Issues 2002;23:551–76.
  82. 82. Lambert NM, Fincham FD, Lavallee DC, Brantley CW. Praying together and staying together: Couple prayer and trust. Psychology of Religion and Spirituality 2012;4:1–9.
  83. 83. Amato PR, Hohmann-Marriott B. A comparison of high- and low-distress marriages that end in divorce. Journal of Marriage and Family 2007;69:621–38.