Is there a sleeper effect of exposure-based vs. cognitive-only intervention for anxiety disorders? A longitudinal multilevel meta-analysis.

There is a longstanding debate in the cognitive behavioral literature whether exposure-based methods produce more sustainable outcomes relative to cognitive methods or vice versa. This debate concerns particularly the time after treatment termination (at follow-up assessments), also referred to as the sleeper effect. Therefore, the aim of the current meta-analysis was to examine the enduring efficacy of Exposure Therapy (ET) in comparison to Cognitive Therapy (CT) from treatment termination to follow-up in anxiety disorders. Available literature also allowed for the assessment of their long-term additive benefits relative to ET only. Traditional random effects analyses with restricted maximum likelihood estimators and multilevel longitudinal analyses were conducted on 39 randomized controlled trials (N = 1878). Traditional analyses revealed no differential efficacy at post-treatment or follow-up. Similarly, the multilevel longitudinal analyses identified no differential growth in efficacy from treatment termination to follow-up. The majority of the variables investigated did not moderate the results. However, there was evidence suggesting that CT was superior to ET when treatment was delivered individually, while ET was superior to CT when delivered as group therapy. Overall, the findings did not validate a number of assumptions, such as the existence of a sleeper effect. Several strengths and limitations are further discussed in the paper.