This data set contains simulated effect sizes of randomized controlled trials evaluating the (relative) effectiveness of cognitive behavioral therapy for depression, using different delivery formats.

data("TherapyFormats")

Format

A data.frame with 7 columns.

author

character. The study label containing the first author and year of the study.

TE

numeric. The effect size of the treatment comparison. Negative values indicate results favoring treat1.

seTE

numeric. The standard error of the effect size.

treat1

character. Name of the first treatment (short label).

treat2

character. Name of the second treatment (short label).

treat2

character. Name of the first treatment (full name).

treat2

character. Name of the second treatment (full name).

Source

Cuijpers, P., Noma, H., Karyotaki, E., Cipriani, A., & Furukawa, T. A. (2019). Effectiveness and Acceptability of Cognitive Behavior Therapy Delivery Formats in Adults with Depression: A Network Meta-analysis. JAMA Psychiatry, 76(7), 700-707.

Details

This simulated data set is based on pooled results of a network meta-analysis examining the effects of different cognitive behavioral therapy (CBT) delivery formats in adults with depression (Cuijpers et al., 2019). Using network meta-analysis techniques on this data set will yield roughly similar, although not identical results to the ones reported in the original study.

For the simulation, pooled effects within designs as reported in Cuijpers et al. along with the estimated between-study heterogeneity variance were assumed as true population parameters. Based on these "true" population parameters, simulations were generated assuming a (random-effects) bayesian hierarchical model, using a non-central chi-squared distribution for the simulation of study-specific sampling errors. The number of effect size draws was chosen to (approximately) mimic the number of comparisons in the original analysis.