Consideration of an upper-bound continuous maximum drinks measure for adolescent binge and high-intensity drinking prevalence (2021)

BACKGROUND: The degree to which binge and high-intensity drinking prevalence estimates vary from fixed threshold frequency and continuous maximum drinks measures is unknown. The current study compared prevalence estimates for adolescent binge and high-intensity drinking (5+ drinks, 10+ drinks, respectively) and sex-specific thresholds using fixed threshold frequency and continuous maximum drinks measures.
METHODS: Data were obtained from 7911 respondents participating in the 2018 and 2019 nationally representative Monitoring the Future 12th-grade surveys. Comparisons of frequency prevalence (e.g., any occasions of 5+ drinking using the frequency measure) versus maximum drinks prevalence (e.g., reporting 5 or more drinks using the maximum number of drinks measure) were made using all respondents and then separately within males and females.
RESULTS: Among the sample overall and within sex, binge drinking estimates from the 5+ frequency prevalence and 5+ maximum drinks prevalence measures evidenced overlapping confidence intervals (estimates were slightly higher for frequency prevalence); similar results were observed for high-intensity drinking 10+ frequency prevalence and 10+ maximum drinks prevalence. For example, among the sample overall, 5+ frequency prevalence was 11.4% [95% CI 10.3, 12.6]; 5+ maximum drinks prevalence was 10.7% [9.6, 11.8]; 10+ frequency prevalence was 5.1% [4.4, 5.8]; and 10+ maximum drinks prevalence was 4.1% [3.5, 4.7]. Using sex-specific thresholds (i.e., 4+ drinks for females and 5+ drinks for males), binge frequency and maximum drinks levels also had overlapping confidence intervals. Binge drinking prevalence estimates for females were approximately 1.5 times higher using sex-specific (4+) versus universal (5+) thresholds.
CONCLUSIONS: In this nationally representative sample of 12th-grade students, prevalence levels for 5+ and 10+ drinking did not differ significantly when using frequency versus maximum drinks measures. Among females, binge drinking prevalence was higher using sex-specific versus universal thresholds. Both the frequency and maximum drinks measures provided comparable estimates of binge and high-intensity drinking prevalence among older adolescents.