Objective: A low level of response (LR) to alcohol as measured through alcohol challenges is an early-appearing, genetically influenced characteristic that predicts the risk of heavier drinking and alcohol problems. A less expensive and more easily used measure of LR, the retrospective Self-Rating of the Effects of Alcohol (SRE) questionnaire, also relates to alcohol intake and problems but has not been evaluated for its ability to predict alcohol-related problems 5 years later. Method: At Time 1, 95 18- to 35-year-old (mean age: 25.9 years) subjects who were offspring from families participating at the San Diego site of the Collaborative Study on the Genetics of Alcoholism (COGA) were administered the SRE and evaluated regarding alcohol, drug, and demographic characteristics using the Semi-Structured Assessment for the Genetics of Alcoholism (SSAGA) interview. Follow-up interviews (Time 2) using the SSAGA were completed an average (SD) of 5.4 (1.34) years later for approximately 80% of the original sample. Results: The retrospective SRE score at Time 1 (the number of drinks for effects the first five times [First 5] of drinking) correlated with Time 2 maximum quantity and frequency, alcohol problems overall, the number of alcohol-dependence items endorsed, and the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, diagnosis of alcohol abuse or dependence. The relationships remained robust in hierarchical logistic regression analyses even in the context of age, gender, the number of SRE items endorsed, and alcohol use and problem variables at Time 1. The regressions explained between 21% and 43% of the variance of the outcomes overall, with the First 5 SRE score alone accounting for between 4% and 14%. Conclusions: These findings are consistent with the ability of SRE-based LR scores at Time 1 to predict alcohol-related outcomes over the subsequent 5 years.