It may also be possible that some SMW are turning to accessible coping outlets to deal with minority stress—including both spirituality and substance use. Findings from tests of interaction between religious attendance and sexual identity approached but did not reach statistical significance in relation to past year AUD or any marijuana use in the past year. Lack of significant differences may be related to the markedly lower levels of religious service attendance reported by SMW relative to heterosexual women, which may have reduced our ability to detect differential risk by sexual identity. These findings underscore the importance of future research considering religious behavior, such as attending religious services, in addition to religiosity or spirituality, given the potential of each to contribute to disparately negative behavioral health outcomes for SMW. We also explored potential differences in outcomes among participants based on religious environment—specifically, whether the study outcomes differed for participants involved in religious environments that were unwelcoming to LGBT people. Differences by sexual identity in interaction models were attenuated and no longer significant when we added this variable. Our ability to explore this question in greater depth was limited by the relatively small number of SMW participants who reported that they attended services in unwelcoming religious environments . Although the percentage of participants reporting attendance at LGBT welcoming environments was similar across sexual identity groups , over 80 % of SMW,marijuana grow system compared with 48.5 % of heterosexual women, described themselves as not affiliated with or attending services. These demographic differences are consistent with literature suggesting that sexual minorities are more likely than heterosexuals to dissociate from religious institutions entirely or seek alternatives to disaffirming religions .
Studies with larger samples of SMW who attend religious services that are both welcoming and unwelcoming of LGBT people are needed to explore the potential impact of the immediate religious environment on substance use outcomes. Findings should be interpreted in the context of study limitations. Although the SMW participants were drawn from two large national panel samples of SMW, they were not recruited using probability sampling methods, which may limit generalizability. As noted above, the great majority of SMW did not participate in religious services, which limited our ability to explore the impact of religious environment on substance use outcomes. There were also some limitations related to measurement. We assessed importance of religion, religious attendance, and importance of spirituality each with a single item. Although the use of single items are common in survey research, there are other measures that capture different dimensions of religiosity not captured in the current study, such as organizational, nonorganizational, and subjective religiosity ; daily spiritual experiences such as awe, inner peace, gratitude, transcendent experiences ; or facets of religiosity that may be particularly salient to health such as religious coping and religious social support . It is possible that a measure of religious coping or a multi-dimensional measure of spirituality would have yielded different results. Furthermore, it was not possible to assess the degree to which participants conflated religiosity and spirituality; multidimensional measures may have allowed for a more nuanced exploration of the impact of spirituality independent of religiosity.
Measures of religious environment also differed between the SMW and heterosexual women, which may have contributed to the different distributions of “non-affirming” attendance by sexual identity. Given research suggesting differences in perceived importance of religion and religious affiliation by race and ethnicity among SMW , future studies might examine possible subgroup differences in the associations of religiosity and spirituality to substance use outcomes. Finally, differences between the two panel samples may have influenced the findings in the current study. Although research suggests that substance use is typically greater among SMW relative to heterosexual women regardless of the sample or measures used , LGBT specific panels may reach individuals whose characteristics differ than LGBT peers recruited from general samples . Marijuana is a commonly used drug during pregnancy, and its use has increased among pregnant women in recent years . While more research is needed , there are significant concerns about the potential risks of prenatal marijuana use, and national guidelines strongly recommend that clinicians screen for and advise against marijuana use in pregnancy . Marijuana has antiemetic properties, and prenatal marijuana use is most prevalent in the first trimester of pregnancy when nausea and vomiting in pregnancy peaks . Available data from a small number of studies suggest that pregnant women may use marijuana to self-medicate their NVP symptoms. Cross-sectional data from 2009-2011 from the Hawaii Pregnancy Risk Assessment Monitoring System indicated that self-reported prenatal marijuana use was higher among those with versus without severe self-reported NVP . Similarly, our study of California women screened for prenatal marijuana use by self-report and urine toxicology tests using combined data from 2009-2016 found that first trimester marijuana use was elevated among those with severe and mild versus no NVP .
Results from two smaller surveys indicated that most pregnant women with ongoing prenatal marijuana use reported using marijuana to self-medicate nausea , and a majority of those who used it for this purpose rated it as effective in treating NVP . As public acceptance and availability of marijuana increase overall , pregnant women may be increasingly using marijuana for a variety of reasons unrelated to NVP . Understanding whether prenatal marijuana use has remained elevated among pregnant women with NVP in recent years is critical and of growing importance, as this information can be used by clinicians to better tailor discussions with pregnant patients and to inform interventions and education programs to reduce prenatal marijuana use. The current study extended our previous work using data from a large California healthcare system with universal screening for prenatal marijuana use via self-report and urine toxicology from 2009-2016 and is the first study to examine trends in prenatal marijuana use separately for women with and without clinical NVP diagnoses. We modeled the prevalence of prenatal marijuana use annually by NVP status using Poisson regression with a log link function controlling for age, race/ethnicity, median neighborhood household income,cannabis vertical farming prenatal marijuana use screening year, and parity. We estimated the covariate-adjusted prevalence using the direct method, standardized to the total study sample population across all years. We modeled linear trends of marijuana use and NVP by including a linear term for calendar year in the Poisson regression model, and we tested for statistical significance using a Wald test. We modeled marijuana use trends by NVP status by including cross product terms for year by NVP status in the Poisson regression model, and we tested for significance of trend differences using a Wald test. We repeated these analyses for self-report and toxicology results separately. Next, we additionally adjusted for self-reported marijuana use during the year before pregnancy to examine how results were affected by pre-pregnancy marijuana use. In a large diverse sample of pregnant women in California with universal screening for marijuana use via self-report and urine toxicology testing as part of standard prenatal care, women with NVP had a higher prevalence of marijuana use than those without NVP each year from 2009 to 2016. The adjusted prevalence of marijuana use increased at a similar rate regardless of NVP status, increasing from 6.5% to 11.1% among women with NVP and from 3.4% to 5.8% among those without NVP. The elevated prevalence of marijuana use across years among pregnant women with NVP is notable. Although national clinical management guidelines indicate that NVP can be successfully treated with dietary and lifestyle modifications and safe medically recommended interventions , pregnant women may instead choose to use marijuana to self-medicate NVP symptoms. Despite potential risks and national guidelines that advise strongly against marijuana use in pregnancy , pregnant women perceive a lack of evidence about the harms of prenatal marijuana use , and some believe there is little-to-no harm in using marijuana during pregnancy .
Women report searching online and seeking advice about prenatal marijuana use from friends, describing stories of others who used marijuana throughout pregnancy without apparent negative effects . Further, online media and marijuana dispensaries are touting marijuana as a harmless and effective treatment for NVP, which may contribute to elevated use among women with NVP. For example, a systematic content analysis of online media items about prenatal and postpartum marijuana use identified using Google Alerts between 2015 and 2017 indicated that more than one-quarter of online media items mentioned the treatment of NVP as a health benefit of marijuana use . Further, in a recent study of marijuana dispensaries in Colorado, 69% of dispensaries recommended marijuana products to treat NVP in the first trimester of pregnancy, and 36% of dispensaries endorsed the safety of marijuana use in pregnancy . In the current study and in the Hawaii PRAMS , women with NVP were also more likely than those without NVP to self-report marijuana use in the year before pregnancy. Thus, it is possible that marijuana use before pregnancy is related to increased risk of NVP. For example, withdrawal from marijuana among women who stop using it when they learn they are pregnant might lead to or worsen NVP symptoms, increasing the likelihood of an NVP diagnosis. However, we found a similar pattern of results after adjusting for marijuana use in the year prior to pregnancy, suggesting that pre-pregnancy use does not fully account for the elevated prenatal marijuana use associated with NVP. Additional research is needed to understand whether the relationship between marijuana use and NVP is bidirectional. Importantly, although the prevalence of prenatal marijuana use was higher each year among women with NVP, use also increased significantly over time among women without a diagnosis of NVP. This suggests that milder NVP symptoms that do not come to the attention of the healthcare system or factors other than NVP are also likely contributing to rising use of marijuana in pregnancy. With legalization of marijuana for recreational use in California in 2018, rates of prenatal marijuana use may increase even more rapidly in the future. Our sample included KPNC women who were screened for marijuana use in the first trimester of pregnancy. Findings may not generalize to women without healthcare or to those who enter prenatal care late. Provider diagnoses of NVP may not capture mild NVP. Further, our self-reported measure of marijuana use in pregnancy does not differentiate prenatal use before versus after women realized they were pregnant. While cannabis metabolites are detectable in urine for ~30 days, this varies with marijuana potency and heaviness of use, and toxicology tests may have picked up pre-pregnancy use in a small number of cases. Finally, our study did not examine whether frequency of marijuana use varies with NVP status, which is an important question for future studies. Dual marijuana and alcohol use is especially prevalent, with 47% of marijuana users reporting simultaneous use of alcohol . Furthermore, individuals who have a cannabis use disorder are at increased likelihood for the development of an alcohol use disorder , and rates of substance use disorders and treatment admissions are highest among individuals that use marijuana or alcohol compared to other substances . Approximately 68% of individuals with current CUD and over 86% of those with a history of CUD meet criteria for an AUD . Cannabis dependence doubles the risk for long-term persistent alcohol consequences and dual marijuana and alcohol users consume higher levels of alcohol and experience more alcohol-related consequences than only drinkers . Despite these additional risks, 60% of college students do not perceive regular marijuana use to be harmful .The combination of low perceived risk, policy changes surrounding marijuana legalization, and the rise in marijuana use over the past 10 years heightens the importance of effective interventions for alcohol and marijuana use. In the adult substance use treatment literature, it is relatively well-established that alcohol use negatively impacts treatment of other substances . In contrast, literature examining the impact of marijuana use on the treatment of other substances is mixed. With the exception of a few studies that do not show marijuana use to negatively influence alcohol or smoking cessation outcomes , many studies have demonstrated that using marijuana before or during alcohol treatment is associated with higher levels of drinking at follow-up . For example, among alcohol dependent individuals, those who used marijuana during alcohol treatment reported fewer days abstinent from alcohol one year following treatment than those who did not use marijuana . Thus, marijuana use seems to have a negative impact on alcohol treatment outcomes.