provides descriptive information by gender on the 954 randomized subjects who completed at least one follow-up and answered the baseline RAPI. The participants were equally divided between males (49%) and females (51%). The sample was predominantly non-Hispanic white (91%), and represented a wide range of ages and class levels, including graduate students (16%).
Baseline Characteristics of High Risk College Drinkers in CHIPS, a Brief Alcohol Intervention Trial (n=954)
gives alcohol use, memory blackout frequency, and injury rates for the study participants. Male subjects consumed 81.8 drinks on average in the past 28 days, while females consumed significantly less, averaging 58.7 drinks (p<.001). Men drank significantly more drinks on a typical drinking day than women (7.4 vs. 5.6, p<.001). Females in the study reported significantly fewer heavy drinking days, defined as 5+ drinks for men or 4+ drinks for women, than did males (p=.01). More than half of the subjects experienced one or more memory blackouts in the 12 months prior to the study. Seven percent reported 6 or more past year memory blackout episodes prior to study entry. Despite significant differences in amounts of alcohol consumed, males and females in the study reported similar frequencies of memory blackouts (p=.94). In addition, there were no significant differences in alcohol-related injuries by gender at baseline (p=.48) or during 24-month follow-up (p=.51). The overall 24-month alcohol-related injury rate was 25.6%.
Alcohol Consumption, Memory Blackout Frequency, and Alcohol-Related Injury Among High-Risk College Drinkers (n=954)
Memory blackouts were negatively associated with age (r=−0.24, p<.001). Students 24 years or older reported the lowest baseline memory blackout rates. Students 18 to 20 years of age reported the highest memory blackout rates. Memory blackout frequency was positively correlated with number of heavy drinking days (r=0.19, p<.001). Subjects reporting no heavy drinking days at baseline averaged 0 memory blackouts, while subjects reporting 9 or more heavy drinking days at baseline averaged 1–2 memory blackouts in past 12 months.
provides alcohol-related injury rates by baseline blackout frequency. The overall 24-month alcohol-related injury rates were 18% for subjects who reported no memory blackouts at study baseline, but jumped to 49% for drinkers reporting frequent (6+) memory blackouts at baseline. At the first 6-month follow-up post-randomization, injury rates were 6% for students reporting no memory blackouts, going up to 23% for study enrollees reporting frequent (6+) memory blackouts. At 12 months, alcohol injuries differed from 7% with no memory blackouts to 22% for frequent memory blackouts sufferers. Eighteen-month rates were similar, varying from 6% for students with no memory blackouts to 22% for participants with frequent memory blackouts. Injury rates were lower at 24-month follow-up, but still showed a positive association with memory blackouts at baseline. The correlations between memory blackouts and alcohol-related injuries were highly significant at 6-month (r=0.17, p<.001), 12-month (r=0.14, p<.001), and 18-month (r=0.14, p<.001) follow-up, and statistically significant (r=0.09, p=.04) at 24-month assessment.
Follow-Up Alcohol Injury Rates among High-Risk College Drinkers by Baseline Memory Blackouts from 2004 to 2009
The multivariate GEE analysis results presented in model odds of alcohol-related injury during 24-month follow-up. The blackout-only GEE model shows a strong association between memory blackouts and alcohol-related injuries before adjusting for covariates. In the full GEE model, memory blackouts at baseline still exhibited a significant dose-response on alcohol injuries. Statistically significant baseline covariates for predicting future injury were prior alcohol-related injury, younger age (18 to 20 years), sensation-seeking disposition, and heavy drinking days. Experimental group status was not a significant predictor of injury during follow-up. Odds of alcohol-related injury in the 24-month follow-up steadily increased from 1.57 (95% CI: 1.13–2.19) for subjects reporting at least one memory blackout to 2.64 (95% CI: 1.65–4.21) for students acknowledging 6+ memory blackouts at baseline. Heavy drinking days were associated with an 8% increase (95% CI 1.01–1.16) in injury risk for each reported day. Notably, the effect size for the memory blackout measure was larger than for the heavy drinking measure.
Repeated Measures GEE Model for Alcohol-Related Injury during 24-Month Follow-Up among High-Risk College Drinkers