Back

Associations between ZIP code-level alcohol outlet density and binge drinking among people who inject drugs in 22 US Metropolitan Areas

Peddireddy, S. R.; Beane, S.; Yarbrough, C.; Ibragimov, U.; Cummings, J. R.; Haley, D. F.; Linton, S. L.; Cooper, H. L.

2025-06-23 public and global health
10.1101/2025.06.23.25329062 medRxiv
Show abstract

BackgroundAlthough alcohol outlet density (AOD) is associated with drinking behaviors in the general population, little evidence exists about this relationship among people who inject drugs (PWID). Establishing this connection is particularly important, as alcohol use exacerbates the risk of opioid overdose. This study investigated 1) the association between ZIP code-level AOD and binge drinking among PWID who use opioids, and 2) the potential moderating effect of race/ethnicity on the AOD-binge drinking relationship. MethodsThis analysis linked 2018 National HIV Behavioral Surveillance (NHBS) data and 2016 ZIP code-level AOD data from the ZIP Code Business Pattern survey. Hierarchical generalized linear models quantified the association between AOD and binge drinking, overall and by race/ethnicity. ResultsOf 9,660 PWID, 27% reported recent binge drinking. For Hispanic/Latinx and White PWID, increases in AOD were associated with a 1% and 3% increase in the odds of binge drinking, respectively (p < 0.05); AOD was unrelated to binge drinking among Black PWID. However, when AOD was set to the median, binge drinking probability was highest among Black PWID (PP: 0.28, 95% CI: 0.24-0.33). ConclusionsReducing AOD may decrease binge drinking among White and Latinx PWID. Our results suggest, however, that this strategy alone may be not be effective among Black PWID. Holistic strategies targeting structural determinants of alcohol and opioid co-use among Black PWID are needed to address inequities in drug-related harms.

Published in Drug and Alcohol Dependence (predicted rank #1) · training set

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.