Discordance between perceived alcohol dependence and AUDIT scores in a sample of adults in Lebanon: correlates and focus on religious prohibition and social desirability

The accurate assessment of substance use, particularly alcohol use, can be further complicated by social desirability bias and religious injunctions. This study explored correlates of a non-zero AUDIT score and of discordance between self-reported drinking status, self-perceived alcohol dependence, and AUDIT score, with particular focus on self-reported religious prohibition of alcohol and social desirability. A cross-sectional study was conducted between February and March 2025, enrolling 460 participants. Data related to sociodemographic characteristics, alcohol use (AUDIT), cigarette (LCD), and waterpipe dependence (LWDS-11), and social desirability (LSDS) were gathered using validated tools. Participants were recruited by snowball sampling, and Benjamini–Hochberg false discovery rate (FDR) correction was applied. Among 460 participants, 23.5% had a non-zero AUDIT score; among these 108 participants, the mean score was 6.66 ± 7.76 and 79 (73.1%) scored 1–7, below the conventional screening threshold of 8. Overall, 100 participants (21.74%) met the definition of discordance between perceived alcohol dependence and AUDIT score (DPADAS). DPADAS was observed in 86.7% of occasional drinkers, 66.7% of previous drinkers, and 58.8% of regular drinkers, compared with 8.6% of participants who described themselves as non-drinkers. Among self-described non-drinkers, 33 had a non-zero AUDIT score, of whom 32 met the DPADAS definition. In multivariable analyses, self-reported religious prohibition of alcohol was associated with lower odds of DPADAS (ORa = 0.040, 95% CI 0.019–0.084; q < 0.002), whereas consuming alcohol while smoking and higher LWDS-11 scores were associated with higher odds. Associations involving LSDS factors were inconsistent. This study identified associations between AUDIT categories, discordance between self-perceived alcohol dependence and AUDIT score, and several sociodemographic and behavioral factors in a non-probability sample of adults in Lebanon. Approximately one-fifth of participants met the DPADAS definition. Self-reported religious prohibition of alcohol was associated with substantially lower odds of DPADAS, which may largely reflect lower alcohol use, whereas associations involving social desirability were inconsistent. These findings are hypothesis-generating and should not be interpreted as evidence of alcohol underreporting, causality, or population-level prevalence. Further studies using more representative sampling, direct measures of alcohol consumption and religiosity, and longitudinal designs are warranted.