Although religiosity has been found to be protective against hazardous alcohol and drug use in the general population , findings regarding this relationship are mixed in studies with sexual minorities . Understanding factors that may protect against hazardous alcohol and marijuana use is important in the context of persistent sexual identity-related disparities in substance use . Examining factors that may affect alcohol and marijuana use among sexual minority women is particularly important given research documenting higher rates of hazardous drinking and marijuana use among SMW compared to heterosexual women, and disparities by sexual identity that are generally more pronounced among women than among men. Religiosity and spirituality constitute separate yet related phenomena. Religiosity represents involvement in the rituals, cultural traditions and practices of a particular religious institution or community . Spirituality represents an individual’s beliefs and practices related to a higher power, search for meaning, or sense of transcendence, which may be secular or linked with religion. Research suggests that sexual minorities generally consider spirituality as having greater importance in their life than religion . Relative to heterosexuals,cannabis grow racks sexual minorities are less likely to attend religious services or to consider religion as somewhat or very important in their lives , however these differences are smaller in relation to measures of spirituality .
Although general population studies have found small to medium positive health effects for religion and spirituality, using meta-analysis, Lefevor and colleagues found much smaller positive relationships with health outcomes among sexual minorities. Findings were also inconsistent among sexual minorities and relationships varied depending on how religion and spirituality were measured. Measures of spirituality were positively associated with health, but measures of religious attendance were not . The absence of a positive relationship between religious attendance and health among sexual minorities may be partially explained by exposure to unique stigma related stressors in religious contexts . Research with general population samples has found strong associations between higher religiosity and negative attitudes toward sexual minorities . Furthermore, close to one-third of sexual minority adults in a U.S. survey reported feeling unwelcome in a place of worship . Exposure to religious heteros exist stigma is, in turn, associated with negative health, mental health, and substance use outcomes among sexual minorities . This complexity of relationships between religiosity and spirituality underscores the importance of multiple measures of religiosity and spirituality in research with sexual minority adults. Literature on the relationship between religion and/or spirituality and alcohol or marijuana use among SMW is limited and shows mixed results . One study with SMW found that neither religiosity nor spirituality predicted past-year substance use outcomes of hazardous drinking or drug use, including marijuana.
Another study found that religiosity was protective against hazardous drinking and drug use among both SMW and heterosexual women . One study found that religiosity was protective against heavy episodic drinking among heterosexual women; however, it was not protective for lesbian women and it was associated with increased drinking among bisexual women . The authors hypothesize that relative to religious lesbian women, religious bisexual women may have less social support in lesbian and gay communities to counteract potential stigmatizing experiences. They may also have fewer role models for positive bisexual identity and experience greater pressure to adhere to heteros exist scripts. In the context of these mixed findings, research exploring the relationships between religion and spirituality and substance use outcomes, and in particular disaggregating findings for lesbian and bisexual women, is warranted. This study used data from SMW recruited from two web-based panels and from a sample of heterosexual women participating in a nationally representative study of alcohol use. We tested: whether spirituality was differentially associated with alcohol and marijuana use by sexual identity; whether religiosity was differentially associated with alcohol and marijuana use by sexual identity; whether observed differences between spirituality or religion and substance use by sexual identity persisted after adjusting for religious environment, defined as the degree to which women reported affiliation with churches that were welcoming of lesbian, gay, bisexual and transgender individuals.
SMW participants were recruited from two national online panels: a general population panel and an LGBT-specific panel. Eligibility for participation in the panel samples was restricted to participants ages 18 or older and who identified as lesbian, bisexual, or other non-exclusively heterosexual identity; resided in the U.S.; and identified as women at the time of the screening. The LGBT-specific panel was drawn from a diverse panel of over 50,000 LGBT participants across all states in the U. S., including 20,000 SMW, who were originally recruited through partnerships with over 300 LGBT websites, publications, organizations, apps and social media. The general population panel included approximately 2.5 million active participants in the U.S, recruited using a wide range of methods to obtain a geographically and demographically diverse sample of participants over age 13, in all 50 states, who own a smartphone and are registered to receive and respond to survey opportunities through an app. To over-sample SMW who identified as African American or Latinx, each wave of recruitment targeted a random sample that was one-third African American/Black, one-third Latinx, and one-third unrestricted by race/ethnicity. Participants were compensated through the panel companies following their standard payment protocols. The participation rate for the general panel sample was 45 % and the participation rate for the LGBT sample was 28.7 %. Heterosexual participants were recruited from a pool of former participants in the National Alcohol Survey , a national probability survey. The NAS is a cross-sectional probability survey of adults ages 18 or older in the U.S., conducted approximately every-five years that used computer-assisted interviewer with a random sample of both landlines and cell phones with oversampling in low-population states and oversampling in Black- and Latinx-dense areas. Participants from the probability survey were eligible for random selection in the present study if they selected “female” as their gender and “only heterosexual or straight” in response to a question asking them to choose the category that best described their sexual orientation. A random sample of 1,961 heterosexual women who participated in the 2015 NAS were invited to participate in the current study. Computer assisted telephone interviews were completed with 623 respondents . The general panel sample and the national probability survey sample included only binary “male/female” response options and did not assess whether respondents were assigned female at birth. The LGBT-specific panel allowed participants to select multiple sex and gender identities; however,cannabis grow system to be consistent with categorizations in the general panel and probability samples, only participants from this panel were included in the current study if they selected “female” as their gender . Although we refer to participants as “women” in this paper, we acknowledge that study participants may have endorsed other gender categories had they been provided such options. The SMW samples were initially screened based on endorsing sexual minority identity and the heterosexual comparison sample was selected based on prior endorsement of heterosexual identity; the few participants who selected “mostly heterosexual” in the current study were combined with those endorsing heterosexual or straight identity. As shown in Table 1, 44.6 % of the study sample was from the national population-based survey, one-quarter was from the LGBT panel sample, and close to one-third was from the general panel sample. Approximately 46 % of the sample identified as heterosexual; 23 % identified as bisexual and 31 % as lesbian. The majority of the sample was<50 years old , college-educated , currently employed , and currently partnered ; just under one-half identified their race as White.
Although the majority also reported being Protestant, Catholic, Jewish, or some other religion, a quarter of the sample reported not having a religious affiliation. Table 1 displays characteristics by sexual identity. Differences by sexual identity were found with respect to age, race/ethnicity, educational attainment, current employment, partner status, current religion, current religious environment, and study sample. Panel sample participants were invited in 2019 to complete an online survey that included a range of questions related to substance use and factors known to be predictive of hazardous drinking and drug use. Heterosexual women who previously participated in the NAS were again recruited in 2016 to complete a supplemental survey. The goal of recontacting heterosexual women participants in the NAS was to administer measures that were included in the panel surveys but were not asked in the original NAS survey. Data from these sources were merged for analysis in the current study. All procedures were reviewed and approved by the institutional IRB. Participants who endorsed attending religious services were asked about the environment where they attended services. Specifically, SMW respondents in the panel samples who attended religious services were asked whether the place they attend was welcoming of LGBT people. Respondents in the heterosexual resample were asked if their congregation had adopted a statement that officially welcomes gays and lesbians. A three-category variable was created for analysis: not a member of a religious organization, attended congregation welcoming of LGBT people, and attended congregation unwelcoming of LGBT people. We created a dichotomous indicator of whether participants met criteria for past year alcohol use disorder as set forth in the 5th edition of the American Psychiatric Association’s Diagnostic and Statistical Manual . Participants were asked about symptoms in 11 domains . Participants who endorsed two or more criteria were classified as positive for AUD. Participants were asked how often they used marijuana, hash, pot, THC, or ‘weed’ during the last twelve months. Response options included every day or nearly every day, about once a week, once every 2 or 3 weeks, once every month or two, less often than that, and never. Two dichotomous variables were constructed. Any use was constructed as any past year use vs none. Regular use was constructed as use once every month or two or more times a month vs less frequent or no use. All analyses were conducted in Stata using sample weights and variance estimation techniques that adjusted for the complex design features of the NAS recontact and panel surveys. We first conducted separate logistic regression analyses to test the independent effects of spirituality and religiosity measures and sexual identity on the alcohol and marijuana outcomes. Wald tests were performed to test the overall effect of variables with multiple categories. We then ran separate models including the interaction between the spirituality and religiosity measures with sexual identity to examine the differential effects of sexual identity on the relationship between spirituality and religiosity with the alcohol and marijuana outcomes. These models adjusted for the demographic and other covariates listed above. In these interaction models, contrasts tested the joint effects of the interaction. In addition to presenting model coefficients, we also graphically display predictive margins for models in which interactions were statistically significant. We then reran these models adding the covariate of religious environment to examine whether any observed differences between religion and substance use measures by sexual identity persisted. In the current study we examined differences in the associations of religiosity and importance of spirituality with AUD and marijuana use by sexual identity in a large sample of adult sexual minority and heterosexual women. Consistent with prior research, importance of spirituality, importance of religion, and participation in religious services were independently associated with lower odds of substance use. However, this protective effect varied by sexual identity, particularly in regard to AUD. In analyses of the interaction between spirituality and the study outcomes, we found that greater importance of spirituality was associated with greater odds of AUD among both lesbian and bisexual women, but was protective for heterosexual women. Among study participants who reported the highest levels religious importance odds of AUD were also greater among lesbian women than heterosexual women. These findings are consistent with minority stress theory, which suggests that religiosity and spirituality are less protective for SMW than heterosexual women and, in some cases, may contribute to greater risk of substance use. Findings are also consistent with research results indicating that religiosity is not protective against, and is sometimes associated with, increased heavy episodic drinking among adolescent and young adult SMW . Although our study did not explore participants’ subjective experiences and meanings of religiosity and spirituality, findings from qualitative research suggests that although religion or spirituality may provide support for some sexual minorities, it can also be associated with unique stressors which may contribute to adverse psychological and health outcomes .