Longitudinal Patterns of Maternal Alcohol Use by Age: A Longitudinal Latent Class Analysis

The transition to parenthood has been discussed as a potential turning point in reducing risky behaviors, one of them being the consumption of alcohol. Yet, the first few years of parenthood can be particularly stressful and may contribute to increasing substance use.  Binge drinking during pregnancy and early parenthood presents ongoing risks for both mothers’ and children’s well-being. It is thus important to examine such patterns and identify modifiable risk factors to inform intervention strategies. While maternal age at pregnancy has long been identified as an important factor influencing maternal alcohol use, past studies have not examined the role of age systematically nor has sufficient attention been paid to alcohol use post parturition. This study addresses these gaps by investigating how longitudinal patterns of maternal alcohol use may differ by maternal age and, more importantly, how other maternal characteristics may influence such patterns differentially across age groups. Analysis was conducted on a sample of 3,397 mothers from the Fragile Families and Child Wellbeing Study, which followed a cohort of mothers who gave birth between 1998 and 2000, representing births in U.S. cities with a population of 200,000 or more. Mothers were interviewed about their drinking behavior immediately after birth (baseline measure of drinking during pregnancy, used as a control in analyses) and followed up when their children were one, three and five years old.  Separate longitudinal latent class analysis (LLCA) was conducted for mothers of three age groups (age 20-25: N=1,717; age 26-35: N=1,367; age 36+: N=313). Results revealed different class structures for these groups. While two classes (regular drinkers and nondrinkers) were identified for the group of mothers who gave birth between age 20 and 25, three classes (binge drinkers, regular drinkers, and nondrinkers) were separately distinguished for the two older age groups. Additionally, class proportions and characteristics of each class are different across age groups. Results revealed that covariates measured at baseline and year one interview have differential effects on drinking across age groups. For example, while depression significantly increases the odds of being regular drinkers for the age 20-25 group and that of being binge drinkers for the age 36+ group, such effect failed to reach the conventional .05 significance level for the age 26-35 group. Findings highlight the importance of distinguishing risk factors by age groups of maternal drinking while parenting a young child, and stress the importance of designing intervention strategies that target particular ages.