This study has several limitations. These are preliminary analyses employing data from a larger study that was designed to examine relapse as a secondary endpoint rather than to be primarily an investigation of relapse predictors. Thus, aspects of the study such as small sample size, variable lengths of abstinence from alcohol and other substances, and lack of information regarding the precise time point at which relapsers first used alcohol or other substances are artifacts of the parent study design. An additional limitation is the unknown validity of retrospective self-reports of trauma history, which are susceptible to recall bias as well as underreporting due to the sensitivity of the topic.22 Also, the lower frequency or severity of some types of trauma (e.g., sexual abuse) among the men may have obscured our ability to detect statistically significant relationships with relapse in that group. Finally, although we controlled for within-gender differences in some of the clinical and demographic characteristics of the sample in our regression models, we were unable to rule out the possibility that one or more of these factors may have influenced the finding that trauma predicted relapse in women, but not in men.
