Past-month alcohol and nicotine use is most common during the first trimester of pregnancy , during which time women may not realize that they are pregnant. Although healthcare systems are well poised to screen women of reproductive age for substance use, it is challenging to predict which women are at risk for using nicotine and alcohol when they become pregnant, and health care systems have limited resources and require better data to prioritize whom to target with education about prenatal substance use prior to conception. Initial data from nationally representative studies indicate that lower socioeconomic status, lower education, White race, and serious psychological distress are associated with higher risk of nicotine use during pregnancy , while younger age, other substance use, depression, higher socioeconomic status, higher education, and being unmarried are associated with greater risk of alcohol use during pregnancy . Less is known about risk factors that are associated with continued use versus quitting alcohol or nicotine among those who use these substances prior to pregnancy. Given the health risks associated with alcohol and nicotine use during pregnancy, and the changing prevalence and patterns of use of these substances among women of reproductive age in the US ,indoor plant table research is needed to better understand trends in prenatal use of alcohol and nicotine and to identify factors associated with quitting versus continuing to use these substances during pregnancy.
The primary objective of this study was to examine trends in daily, weekly, and monthly or less self reported use of alcohol and nicotine in the year before pregnancy and during pregnancy from 2009 to 2017, among a diverse population of pregnant women within a large healthcare system with screening for substance use as part of standard prenatal care. We also examined whether frequency of alcohol and nicotine use in the year before pregnancy were associated with continued use of these substances during pregnancy. Kaiser Permanente Northern California is an integrated multi-specialty healthcare delivery system that provides care to more than 4 million diverse members who are representative of the Northern California region . During standard prenatal care, pregnant women are screened for frequency of substance use in the year before pregnancy and during pregnancy via a self-administered questionnaire at the first prenatal visit . Our study sample included all pregnant women aged 11 years and older who completed the self-administered questions about use of alcohol and nicotine in the year before and during pregnancy from January 1, 2009 to December 31, 2017. KPNC’s Institutional Review Board approved this study with waiver of informed consent. Pregnant women’s self-reported frequency of alcohol and nicotine use in the year before pregnancy and since the start of pregnancy were assessed using the self-administered questionnaire completed at the first prenatal visit. The frequency variable included four response levels [daily , weekly , monthly or less , none]. We created an additional variable to reflect whether women with self-reported alcohol use in the year before pregnancy quit alcohol use prior to pregnancy , decreased their frequency of alcohol use during pregnancy , maintained their frequency of alcohol use during pregnancy , or increased their frequency of alcohol use during pregnancy .
A similar variable was created to reflect changes in nicotine use frequency during pregnancy among women who self-reported nicotine use in the year before pregnancy. Socio-demographic variables that were available in the electronic health record that have been associated with prenatal alcohol or nicotine use in the literature were also examined, including age, self-reported race/ethnicity, and median neighborhood household income quartiles based on census data. We then used Poisson regression with a log link function to estimate the adjusted prevalence of self-reported daily, weekly, and monthly or less alcohol use in the year before pregnancy and during pregnancy annually. Each outcome was modeled separately. Socio-demographics were adjusted for in these analyses using the average covariate distributions across the study period. In order to estimate the annual prevalence of alcohol use in the year before and during pregnancy for the population, women with more than one pregnancy during the study period could contribute to the analysis more than once. Linear trends in alcohol use frequency before and during pregnancy were modeled using a linear term for calendar year to estimate the annual relative rate of change with 95% confidence intervals, and the significance of the trend was calculated using a Wald test.Next, among women who self-reported alcohol use in the year before pregnancy, we used generalized estimating equation models with a logit link to estimate the adjusted odds ratio and 95% confidence interval of any self-reported alcohol use during pregnancy across years by frequency of alcohol use in the year before pregnancy and by socio-demographic factors, accounting for the correlation among women who were pregnant more than once during the study. All models were adjusted for calendar year. We then extended this analysis to examine the categorical outcome of change in frequency of alcohol use from the year before pregnancy to during pregnancy, defined as 3 categories: quit during pregnancy, reduced frequency of use, and maintained or increased frequency of use.
We used GEE models for multinomial outcomes to estimate the aOR and 95% confidence interval for each category of use and by socio-demographic factors, accounting for the correlation among women who were pregnant more than once during the study. Using Poisson regression with a log link function, we also estimated the adjusted prevalence of any alcohol use during pregnancy, and changes in frequency of alcohol use during pregnancy, among those who self-reported use in the year before pregnancy. Analyses were conducted using SAS 9.4 and a two-sided P-value <0.05 was considered as statistically significant. Among the 238,138 women who self-reported any alcohol use in the year before pregnancy, 85.8% self-reported no use during pregnancy and 14.2% self-reported use during pregnancy. The odds of continued alcohol use during pregnancy were higher among women who reported daily or weekly alcohol use compared to those who reported monthly or less alcohol use in the year before pregnancy . In addition, demographic factors associated with higher odds of continued alcohol use during pregnancy included being age 11–24 , non-White race, having a median neighborhood household income in the first or second quartile, and having a pregnancy in an earlier study year . The adjusted prevalences in continued alcohol use are provided in eAppendix 4. However, demographic patterns among women with continued alcohol use during pregnancy slightly varied by whether they maintained or increased their frequency of alcohol use or decreased their frequency of alcohol use . Similar to the overall associations in Table 2, women aged 11–24 , non-White women, and those with a lower median neighborhood household income had slightly higher odds of increasing or maintaining their frequency of alcohol use versus quitting, although the association with income was only statistically significant for the second versus fourth quartile of income . In contrast, older women , White women, and those with a greater median household neighborhood income had significantly higher odds of decreasing their frequency of alcohol use versus quitting alcohol use during pregnancy,hydroponics flood table although the association with income was only significant for the third versus fourth quartile of income .
Appendix 5 shows the adjusted prevalence of changes in alcohol use frequency.Using data from a large study of pregnant women screened for self-reported alcohol and nicotine use in the year before and during pregnancy as part of standard prenatal care from 2009 to 2017, we found that the adjusted prevalence of any alcohol use in the year before pregnancy increased slightly over time, driven by modest increases in weekly alcohol use. However, there were significant decreases in the adjusted prevalence of alcohol use during pregnancy and in the adjusted prevalence of nicotine use both in the year before pregnancy and during pregnancy, with significant decreases in daily, weekly, and monthly or less use. These findings are somewhat consistent with national data from 2002 to 2016 , indicating significant declines in any cigarette smoking during pregnancy, and non-significant decreases in any alcohol use during pregnancy. Notably, the prevalence of nicotine use in our sample was lower than national estimates, consistent with lower smoking rates in California in general . Health care systems could benefit from additional information about which reproductive aged women might be at greatest risk for using alcohol or nicotine during pregnancy. Results from the current study indicate that subgroups of women at risk for more frequent preconception and prenatal alcohol use include older women, White women, and those with higher neighborhood incomes. These demographic differences might be important for developing interventions to prevent or limit drinking during pregnancy. As these subgroups are not traditionally seen as “at-risk”, results emphasize the importance of universal substance use screening and highlight that women’s health providers should be aware of any unconscious bias or assumptions they make about which women are at risk for alcohol use during pregnancy. Most women who self-reported alcohol use in the year before pregnancy reported no alcohol use during pregnancy; however, 4.5% reported decreasing their alcohol use frequency during pregnancy, and 9.7% reported increasing or maintaining their frequency of use during pregnancy. Among those who drank in the year before pregnancy, continued alcohol use during pregnancy was most likely among women who drank daily or weekly versus monthly or less, consistent with prior research indicating that pre-pregnancy substance use is a strong predictor of prenatal use . Further, among those who drank alcohol in the year prior to pregnancy, younger women, non-White women, and those with a lower neighborhood income were more likely to increase or maintain their frequency of alcohol use. These findings indicate that the factors associated with more frequent alcohol use may be different than those associated with maintaining or increasing pre-pregnancy frequency of use during pregnancy, and subgroups with a lower prevalence of use may have greater difficulty cutting down or quitting use during pregnancy. Consistent with prior studies, more frequent nicotine use both in the year before pregnancy and during pregnancy was associated with younger age, Black and White race/ethnicity, and lower income . Among women who self-reported any nicotine use in the year before pregnancy, most self-reported no use during pregnancy; however, 9.5% reported decreasing their pre-pregnancy frequency of use during pregnancy, and 18.8% reported increasing or maintaining their pre-pregnancy frequency of use during pregnancy. It is notable that the percentage of women with continued use during pregnancy was twice as high for nicotine versus alcohol, reflecting the highly addictive and reinforcing nature of nicotine addiction. Factors associated with continued nicotine use during pregnancy among those who used in the past year were similar to those associated with greater frequency of prenatal use, including daily or weekly versus monthly or less prepregnancy nicotine use, adolescent age group, Black race/ethnicity, and having a lower neighborhood income. Although most women stop using substances when they learn they are pregnant, many continue to use in the weeks after conception before they realize they are pregnant, and some continue to use throughout pregnancy . Research has shown that earlier discontinuation of substance use may decrease risks to the woman and the developing fetus, particularly during the first trimester . Notably, 98% of women who self-reported alcohol or nicotine use during pregnancy also reported use in the year prior to pregnancy, and consistent with prior studies , pre-pregnancy substance use was a strong predictor of prenatal use. Given that few women initiate alcohol or nicotine use during pregnancy, our findings highlight an important window of opportunity to provide education about prenatal substance use to women of reproductive age prior to conception, as it is not possible to offer pregnant patients education about prenatal substance use and advice to quit during the early weeks of pregnancy if they do not know they are pregnant or have not yet initiated prenatal care. In particular, targeted education about the risks of prenatal alcohol and nicotine use, advice to cut down or quit use while trying to conceive, and referrals for substance use interventions and resources for women of reproductive age who report daily or weekly use of alcohol or nicotine could be beneficial.