Probation and parole professionals argue that supervision outcomes would improve if caseloads were reduced below commonly achieved standards. Criminal justice researchers are skeptical because random assignment and strong observation studies have failed to show that criminal recidivism falls with reductions in caseload sizes. One explanation is that caseload reduction by itself is insufficient; supervision must also be allotted and distributed to make the best use of supervision resources, a cornerstone of evidence-based practice (EBP). This study uses a regression discontinuity design (RDD) to test the null hypothesis that reduced caseload in an agency using EBP has no effect on criminal recidivism. Our conclusions are that reduced caseloads in this context probably reduce criminal recidivism and probably do not increase revocations for technical violations.KEYWORDS evidence-based practice, probation caseload size, probation outcomes, regression discontinuity designWe have received helpful comments from

No abstract

Objective This study estimated the prevalence of stimulant treatment among both adults and children at national, state, and county levels during 2008 and explored explanations for wide variations in treatment prevalence. Methods Records of 24.1 million stimulant prescriptions dispensed to insured and uninsured patients were obtained from approximately 76% of U.S. retail pharmacies. Data were weighted to estimate treatment prevalence on March 15, 2008, for all U.S. states and counties. Regression models were used to estimate the associations among the counties’ treatment rates and the characteristics of the counties and their resident populations. Results An estimated 2.5% of children ≤17 years of age (3.5% of males and 1.5% of females) and .6% of persons >17 years of age were being treated with stimulants in March 2008. Treatment prevalence among states varied widely, and variation among counties was even wider. Two-thirds of the variation among counties in treatment prevalence was associated with supply of physicians, socioeconomic composition of the population, and, among children, funding for special education. Rates of children and adults in treatment were highly correlated. Conclusions Wide variations in treatment prevalence signal disparities between established clinical practice guidelines and actual practice, especially for primary care, where most patients prescribed stimulants are managed. Better education and training for physicians may improve identification and treatment, thereby reducing disparities in care for attention-deficit hyperactivity disorder and other disabling conditions.

Probation practitioners argue that large caseloads limit their effectiveness, but prior research has not supported that assertion. In fact, with exceptions, strong studies of small caseload intensive probation supervision have produced undesirable findings: no reduction in recidivism and increased technical revocations. However, probation standards and practices have changed considerably since this earlier research, suggesting that new studies may yield different results. This study introduced reduced caseloads (approximately 54 medium-to high-risk probationers per officer) into an agency with officers fully trained in implementing evidence-based practices, a combination of supervision strategies based on known predictors of criminal recidivism. We evaluated the intervention's effect on recidivism and technical revocations of probation using a quasi-experimental design (difference-in-differences). We used survival analysis to estimate that the smaller caseload reduced the rate of recidivism by roughly 30%; technical violations increased by 4%. We conclude that reduced caseloads in agencies using modern supervision practices reduce recidivism.

Requiring prescribers to solicit patients' prescription histories is likely to be a more effective use of PDMP resources than proactive notification.