Public policy requires striking a balance between protecting the public’s health and respecting individual freedoms. Recent research has shown that states that implement effective policies have lower rates of alcohol consumption. However, only a few studies have looked at the role enforcement plays in the relationship between alcohol control policy and consumption. This week The DRAM reviews a study by Darin Erickson and his colleagues at the University of Minnesota School of Public Health that examined the relationships among state-level alcohol control policies, policy enforcement, and alcohol consumption.
What was the research question?
How does a state’s alcohol control policy environment relate to alcohol consumption, and what role does policy enforcement play in this relationship?
What did the researchers do?
Erickson and his colleagues assessed the strength of 18 state-level alcohol control policies included in the Alcohol Policy Information System database for 2009 in each of the 50 states. The researchers organized the policies into 4 domains: 1) underage use; 2) underage compliance checks; 3) drinking-driving; and 4) overservice. They rated each state’s policies on a scale from most to least restrictive. To measure enforcement, they surveyed one officer most familiar with alcohol control enforcement at local law enforcement agencies across the US. Finally, to assess how many of each state’s respondents drank any alcohol, binge drank, or drank heavily in the past month, they conducted secondary analysis of the 2009 Behavioral Risk Factor Surveillance System (BRFSS) data. The researchers performed latent class analysis to group states by their alcohol policy environment. Finally, Erickson and his colleagues completed multi-level logistic regressions to examine the relationships among policy classes, policy enforcement, and alcohol consumption.
What did they find?
As Table 1 shows, Erickson and his colleagues categorized the states into 4 groups or “classes” based on their alcohol control policies: 1) States with strong alcohol server policies, but weak policies in other domains; 2) States with average policies across all domains; 3) States with strong underage use policies, but average policies in other domains; and 4) States with strong policies in all four domains. After controlling for a variety of covariates, states with strong underage use policies had lower rates of drinking across all three drinking measures compared to states with average policies. States with strong alcohol server policies had lower rates of past-month drinking only. People who lived in states with strong policies in all four domains were not different, in terms of their drinking, from people who lived in states with average policies. When the researchers added enforcement to the model, they found it did not reduce or mediate the relationship between the policy groups and alcohol consumption.
Table 1. Alcohol consumption by alcohol control policy class, controlling for sex, age, race marital status, education, unemployment rate, and religiosity.
|1||Strong server policies, weak in other domains||DE, NV, NM, NY, VT, WY||
|2||Average strength for all policy domains||AK, AR, CA, CT, FL, GA, HI, IL, IN, IA, KY, LA, ME, MD, MA, MN, MS, MT, NE, NJ, ND, OH, OK, RI, TX, VA, WA,WV, WI||
|3||Strong underage use, but average in other domains||AZ, CO, ID, KS, MI, MO, NH, OR, PA, SD, TN||
|4||Strong for all policy||AL, NC, SC, UT||-0.09||-0.10||0.01|
Note. *Significantly different from the referent group. Regression coefficients measure the relationship between policy class and alcohol consumption.
Why do these findings matter?
This study shows that states can be grouped by their alcohol control policy profiles. State alcohol control policies that focus on restricting underage consumption appear to reduce alcohol consumption. Unfortunately, in this study, only a small fraction of states were considered to have strong policies to prevent underage drinking. The alcohol control policy environment appears to play a more significant role than enforcement in reducing alcohol consumption. Future studies should explore these concepts further using a longitudinal study design that expands the alcohol control policies under investigation and more thoroughly explores the role of enforcement.
Every study has limitations. What about this one?
Erickson and his colleagues used a cross-sectional study design,which means they cannot conclude that alcohol policies changed drinking levels. Other relationships are possible. For example, states with citizens concerned about underage drinking might be less likely to allow their children to drink and be more likely to support strong underage drinking policies. The authors descriptions’ about enforcement might not be generalizable because they relied on the beliefs and experiences of 20 to 40 law enforcement officers in each state.
For more information:
If you are concerned about yours or a loved one’s drinking, visit our website for the First Step to Change: Drinking, a free, anonymous toolkit.
-- John H. Kleschinsky
What do you think? Please use the comment link below to provide feedback on this article.