Response rates for CBT for anxiety disorders: Need for standardized criteria. (December 2015)

• Record Type: Journal Article
• Authors: Loerinc, Amanda G.; Meuret, Alicia E.; Twohig, Michael P.; Rosenfield, David; Bluett, Ellen J.; Craske, Michelle G.
• Is Part Of: Clinical psychology review. Volume 42(2015)
• Page Start: 72
• Page End: 82
• Publication Date: 2015-12
• Subjects: CBT -- Anxiety disorders -- Response rates -- Effectiveness; Clinical psychology -- Periodicals; Psychology, Pathological -- Periodicals; Psychotherapy -- Periodicals; Psychology, Clinical -- Periodicals; Electronic journals; 616.89
• DOI: 10.1016/j.cpr.2015.08.004
• Languages: English
• ISSNs: 0272-7358

Abstract

Abstract: Full appreciation of the effectiveness of cognitive behavioral therapy (CBT) requires both effect size data and individual rates of positive response. Response rates are particularly helpful for clinicians when choosing among treatment options. However, systematic reviews on cross-study response rates have not been conducted, possibly due to the absence of a standardized metric for calculating response rates. We conducted a systematic review of the treatment outcome literature to determine overall response rates to CBT for anxiety disorders and whether current methods of defining treatment response influence overall response rates. Our database search (2000–2014) resulted in 87 studies that reported response rates and included at least one CBT condition. Results showed that overall treatment response rates across anxiety disorders averaged 49.5% at post-treatment and 53.6% at follow-up. Response rates varied significantly as a function of the properties used to define them. Measures that incorporated more than one criterion, the combination of a reliable change index with a clinical cutoff (a clinically significant change), and intent-to-treat samples yielded lower response rates at post-treatment. Blinded independent assessors yielded higher response rates than unblinded assessors. Based on previous empirical and theoretical work, we recommend that future studies use a clinically significant change index, in an intent-to-treat analysis (using a mixed-model approach), reflecting multiple modalities, and assessed by independent blinded assessors. Our results indicate that such measures are likely to reduce response rates, but may result in a less biased and more accurate representation of improvement and achievement of normative functioning. Highlights: Response across disorders averaged 49.5% at post-treatment and 53.6% at follow-up. Response rates varied as a function of the properties used to define them. We make recommendations for specific properties for operationalization of response.