A narrative review of meta-analyses on sexual recidivism and treatment factors: Implications for forensic examiners.

Abstract

We present a historical background and narrative review of meta-analyses on sex offender sexual recidivism and treatment, as anchor points for forensic mental health testimony in civil commitment hearings. Themes flowing from the meta-analyses included sexual recidivism for adults and juveniles, jurisdictional differences, treatment versus no treatment, empirically supported risk factors, operational definition of recidivism, offense type, motivation, research design, length of follow-up, diversity considerations, and treatment model, modality, specialization, and setting. We show where meta-analytic results appear to converge and diverge. We also present important null findings, findings from individual studies, and limitations. Our review suggests an examiner's opinion in favor of commitment may be more persuasive when anchored in the individual's history of repeat sexual offending, sexual deviancy, antisociality, and dropout from treatment. Procommitment opinions may be further supported via mention of longer follow-ups generally predicting more sexual recidivism, potentially higher numbers of unreported sex crimes, and the limitations of treatment and rehabilitation studies. An examiner's opinion in disfavor to commitment can be anchored in generally low rates of sexual reoffending, most meta-analyses supporting substantial treatment benefit even for high-risk individuals, studies suggesting rehabilitation failures may have much to do with poor quality treatment, inadequate research on cultural minority individuals, and perhaps null findings on voluntary versus involuntary treatment and the operational definition of sexual recidivism. Limitations of our recommended approach include that meta-analyses have a high proportion of redundant and older samples, often use nonequivalent statistical procedures, and may overlook important findings from individual studies. (PsycInfo Database Record (c) 2025 APA, all rights reserved).