Alcohol Use and Punitive Parenting: A Chronicle of Research Failures and One Success

Background: Alcohol misuse is a risk factor for child physical abuse. Most studies are cross-sectional and investigate past-year drinking behavior along with past year parenting, making it difficult to assess the causal relationship between alcohol use and parenting. Social desirability bias, the low base rate of drinking behaviors, and possible difficulty in recall of events have made it difficult to assess causality. Here, we document our attempts over the past decade to determine the sequencing of drinking behaviors and child physical abuse.

Methods: Attempt #1: Qualitative study of 60 parents to develop questions for a telephone or web survey to understand contexts where drinking and parenting co-occurred. Attempt #2: In a multimodal telephone and web survey we asked 1,499 parents who reported they used physical abuse whether they were drinking before, during, or after the event. Attempt #3: We used a “harms to others” approach where we asked parents via a survey whether their or someone else’s drinking caused physical harm to their child. Attempt #4. Conducted qualitative interviews with 23 of those parents who reported their or someone else’s drinking did cause harm to their child. Attempt #5: Conducted daily Ecological Momentary Assessments (EMA) three times a day (10 a.m., 3 p.m. and 9 p.m.) to ask about parenting behaviors. At the 9 p.m. EMAs on days 7 and 14, we asked questions related to past week drinking behaviors. If they indicated they drank, we further asked about time frame, which corresponded to when the daily EMAs occurred.

Findings: Asking parents on surveys about specific drinking behaviors with specific parenting events resulted in a very low rates of alcohol-related drinking (~3% of parents). Asking about physical harm resulted in much higher rate (13.6% of parents), but when conducting follow-up interviews parents provided answers about events occurring more than one year, discussed drug use instead of alcohol use, or said they gave the wrong answer to the initial survey question. Using the EMA approach, we had 56.58% of people report drinking behaviors on at least one day during the EMAs including 90.65% of weekly drinkers, 53.70% of monthly drinkers, and 22.22% of yearly drinkers. These questions allowed us to link up drinking events to the specific parenting behaviors used during a specific day and time while minimizing the likelihood that parents will be reporting socially desirable responses. Finally, we are able to assess whether drinking occurs before, during, or after certain types of parenting behaviors. SUCCESS!

Conclusions: Research is a series of trial and error, but we often do not discuss those failed attempts. Our unsuccessful attempts may have been too transparent with our study participants, resulting in primarily socially desirable responses. However, with the advances in technology and study methods, we were able to mask our true intentions by not asking drinking and parenting behaviors together. In doing so, we have been able to identify a way to start to assess the causal relationship between drinking behaviors and punitive and aggressive parenting.