Table 1 provides a description of the baseline characteristics of each of the three classes for each of the three measures. Latent class analysis identified three groups of individuals for each of the three daily measures. For alcohol craving, 17 subjects reported elevated cravings during the entire treatment stay, 37 subjects reported initially elevated but then a slight improvement in craving, and 41 subjects reported relatively low craving from the time of admission to the end of residential treatment. Fig. 1 shows the trajectory of the alcohol craving classes over the course of the residential stay. Positive and negative affect graphs were not shown because the outcomes were similar to the craving class. Overall, the high alcohol craving class was associated with more alcohol related problems (SIP score), greater depression symptomotology, and worse physical health functioning. There were no differences in quantity or frequency of drinking, past treatment seeking, or the number of DSM symptoms. While antidepressant use was common prior to admission (52% of subjects), there were no associations between class assignment and use of an antidepressant. Other psychotropic
