Attitudes Toward Alcohol Use During Pregnancy Among Women Recruited From Alcohol-Serving Venues in Cape Town, South Africa: A Mixed-Methods Study

Background: The Western Cape Province of South Africa has one of the highest documented rates of FASD globally. In order to establish FASD prevention interventions that can effectively reach women in this high-risk population, it is important to understand the attitudes that women hold towards alcohol use during pregnancy and examine reasons why positive attitudes may not necessarily translate to behavior in this setting. Study aims: The aims of this secondary analysis were to: describe the attitudes towards alcohol use during pregnancy, compare attitudes with alcohol use behaviors during past pregnancies, and build a logistic regression model to examine predictors of harmful attitudes toward alcohol use during pregnancy. These quantitative aims then set the stage for a qualitative exploration of reasons for harmful behaviors, in spite of attitudes that would support drinking cessation during pregnancy. Methods: This study is based on cross-sectional data from 200 women who were recruited from alcohol-serving venues in a single township in Cape Town, South Africa. A sub-set of 24 was selected to participate in-depth qualitative interviews. Measures of interest included: demographics, attitudes about alcohol use during pregnancy, history of abuse, depression, current drinking behavior, drinking during past pregnancy, and pregnancy intentions and attitudes. Quantitative analysis was conducted in three steps. First, descriptive statistics were used to describe the sample. Second, drinking behaviors in past pregnancies and attitudes toward alcohol use during pregnancy were examined separately and then together in order to describe any potential attitude/behavior gap in this population. Third, logistic regression models were built to examine predictors of attitudes toward alcohol use during pregnancy. Interviews were reviewed and coded for emergent themes under categories that were identified a priori: knowledge about risks of alcohol use during pregnancy, contributors to alcohol use during pregnancy, and contributors to resiliency against alcohol use during pregnancy. Results: The sample of 200 women ranged in age from 18 to 43, were all of Coloured (mixed-race) ancestry, and most had not completed secondary school. Most were not intending to become pregnant and most reported a history of abuse. Though approximately 83% (n=164) of women with a history of pregnancy did not report having harmful attitudes toward alcohol use during pregnancy, more than half of these women (n=84, 51.2%) still reported alcohol use during a previous pregnancy. This discrepancy revealed the existence of a stark attitude-behavior gap in which the holding of harmful attitudes toward alcohol use during pregnancy is not entirely predictive of alcohol use during pregnancy. The strongest predictors of holding harmful attitudes were a history of abuse (AOR=3.33, 95% CI 1.06-10.50) and drinking during a previous pregnancy (AOR=6.87, 95% CI 1.79-26.33). Qualitative data analysis revealed several factors that contributed to alcohol use during pregnancy: 1) Having an unplanned pregnancy; 2) Drinking because of stress or to cope; 3) History of abuse or trauma; 4) Reliance on the venue space for solace; 5) Recreation; and 6) Feelings of invincibility. Conclusions: These data further define the existence of an attitude-behavior gap in this population and highlight that in this setting, having non-harmful attitudes might not be enough to elicit healthy behavior. This points to a need for identification and implementation of policies or interventions that go beyond education to build intrinsic motivation to refrain from alcohol use during pregnancy.