This evaluation of coping strategies using data from a randomized clinical trial of computerized CBT as an adjunct to substance abuse treatment suggests the following: (1) the CSS-17 is significantly associated with observer ratings of quality of coping skills both pre- and post-treatment; and (2) increases in coping strategy use during treatment, as measured by the CSS-17, are related to decreases in drug use in participants who engaged in computerized CBT. To date, this is the first study to show that self-reported strategy use is correlated with observer ratings of coping skills based on a role-play task. This finding suggests that the higher frequency of use is associated with higher skill quality ratings.
It was expected that participants in the CBT4CBT group would show increases in coping strategies compared to the TAU group. Surprisingly, there were no group differences in coping strategies across time. It is possible that both groups were using coping strategies at the same rate, but the CBT4CBT group might be using them more effectively. Evidence from the Aptitude Treatment Interaction modeling suggested that as coping strategies increased, frequency of drug use decreased. In particular, for participants in the CBT4CBT condition, drug use outcomes were more highly correlated with change in strategy use than in the TAU condition. These findings suggest that computerized CBT may affect coping strategy use more strongly than TAU, and this increase in coping strategy use is related to decreased drug use. One possibility is that that for the TAU group participants, their substance use may be affected by other behaviors that are emphasized in TAU (e.g. 12-step recovery). Although previous studies have found that coping skill use was predictive of improved treatment outcome, this study is unique in providing evidence that links computerized CBT specifically with a correlation of coping strategies with reduced drug use.
We acknowledge several limitations of this study. First, these were exploratory analyses derived from an existing data set. Second, as is common with substance abuse research, attrition reduced the sample size, and therefore power, for these analyses. Although we present data on the factor structure of the CSS, the small sample size decreases the stability of these analyses, and they should be interpreted cautiously. Third, the reliance on a self-report measure to assess coping strategies may have affected the results. Specifically, the majority of the items on the CSS are desirable behaviors and individuals in treatment may feel that they should being doing these behaviors, which could bias their responses. However, the analyses comparing the self-reported CSS-17 scores with the DRRT role-play measure show that participants’ self-report corresponds significantly to independent ratings of coping skills responses. Finally, due to missing data, the ATI analyses could only examine data from 31 participants. Replication of these findings is necessary with a larger sample.
Despite these limitations, this study has several strengths and important implications. This is the first study to examine treatment-related change in the use of coping strategies following a computerized, and hence highly standardized, version of CBT. Moreover, unlike previous research, this study utilized a clinically representative sample that was not limited to one specific substance use diagnosis. This study also identified a set of coping strategies directly related to the effects of CBT treatment on drug use. Thus, one fruitful area for future coping strategy research may be to more closely link self-reported coping strategy use to those specifically targeted and delivered in the treatment. Moreover, this study was the first to find a significant relationship between self-reported use of coping strategies with observer ratings of quality of coping skills on a role-play measure. Thus, this study extends the literature by suggesting that for individuals who receive the addition of computerized CBT to treatment-as-usual, their substance use is more strongly linked to their use of coping strategies. Future research should examine if this finding is specific to computerized CBT or to CBT in general.