Factors Associated with Self‐Estimated Breath Alcohol Concentration Among Bar Patrons

BACKGROUND: Few studies have examined the context in which drinkers underestimate their breath alcohol concentration (BrAC) in natural drinking environments. This study examined factors associated with bar patrons' self-estimated BrAC in high-risk college town settings. METHODS: Guided interview and BrAC data were collected from 510 participants recruited as they exited bars located close to large universities: 1 in Florida and 1 in Texas. RESULTS: Participants with the highest measured BrACs underestimated their BrAC levels the most. Findings from multivariable linear regression analysis indicated that BrAC (std β = 0.014, p < 0.001), number of alcoholic drinks consumed (std β = 0.006, p < 0.01), and perceived drunkenness (std β = 0.024, p < 0.001) had significant positive associations with BrAC self-estimates, where the regression coefficients were scaled by values approximately equal to each variable's interquartile range. Among the 321 participants with BrAC levels ≥ 0.08 g/dl, 21.2% believed their BrAC was below the legal per se driving limit of 0.08 g/dl. Results from a logistic regression analysis indicated that higher levels of perceived drunkenness were associated with better self-recognition that one's BrAC level exceeded the legal driving threshold (OR = 3.312, p < 0.001). Further, participants under 26 years of age had reduced odds of recognizing that their BrAC was greater than 0.079 g/dl (OR = 0.245, p < 0.05). CONCLUSIONS: These findings highlight the inaccuracy of self-estimated BrAC when drinking, particularly among younger drinkers. Adjusting for BrAC, situational factors were strongly associated with self-estimated BrAC. Future research is needed to better understand how altering drinking environments may improve accuracy of BrAC self-estimates and deter driving after drinking.