Predictors of sexual recidivism: did meta-analysis clarify the role and relevance of denial?

Abstract

Seven studies selected for meta-analysis of the denial variable in a large-scale meta-analysis of predictors of sexual recidivism are briefly reviewed and a number of methodologic issues are identified that relate to interpretation of the failure of meta-analysis to find an effect for the denial variable. These include variability in definition of the denial variable, variation in treatment access and exclusion for deniers, low base rates of recidivism, small sample sizes, low power, and high probability of type II error across individual studies. Despite the apparent objectivity and power of this methodology, a careful examination of the individual studies suggests that meta-analysis did not clarify the role and relevance of denial as a predictor of sexual recidivism. Further work is necessary to clarify the role and relevance of denial in treatment success, risk assessment, and risk prediction.