Article: Silver, Ian A., DeMichele, Matthew, Bechtel, Kristin, and Lattimore, Pamela K. 2025. The Potential Effects of Combining Pretrial Supervision Conditions. Federal Probation 88(2): 76–86. https://www.uscourts.gov/sites/default/files/document/88_2_9.pdf
Pretrial release decisions influence court and community outcomes for numerous individuals each year. While pretrial release without conditions is possible, court officials frequently impose various supervision conditions to maintain community safety and reduce failure to appears during the pretrial period. Common conditions include regular reporting, employment/educational requirements, electronic monitoring, treatment mandates, and sobriety requirements. These conditions are often combined within a single pretrial supervision plan, with little empirical evidence to guide whether these combinations enhance or diminish the likelihood of failure during pretrial supervision
Prior research demonstrates that pretrial supervision can sometimes improve compliance, but excessive or mismatched conditions may overwhelm individuals, leading to violations and renewed system involvement. Despite these concerns, few studies have systematically assessed how specific conditions—or combinations of conditions—affect the likelihood of new arrests during pretrial supervision. This study addressed the gap in knowledge by forecasting the effects of various supervision conditions, individually and in combination, on the probability of experiencing a new arrest during pretrial supervision .
2. Summary of Findings
The results of the study suggested that supervision conditions and, more importantly, the combination of conditions could differentially affect new criminal arrests during pretrial supervision. Some combinations of pretrial conditions were forecasted to increase the probability of an individual experiencing a new criminal arrest, while other combinations of pretrial conditions were forecasted to decrease the probability of an individual experiencing a new criminal arrest in comparison to not being supervised during pretrial release and not having any special conditions during pretrial release.
To summarize the key findings:
- The likelihood of a new arrest during pretrial was estimated at 21.7%.
- Regular reporting, treatment, or sobriety orders were predicted to slightly reduce the probability of arrest when it was the only condition required.
- Electronic monitoring and location restrictions did not yield clear preventive effects and, in some cases, were predicted to be associated with elevated risk compared to unsupervised individuals when it was the only condition required.
- Pairing treatment with other conditions (e.g., treatment + sobriety requirements) were predicted to lower the probability of arrest beyond treatment alone, suggesting additional benefits of this supervision condition combination.
- By contrast, pairing weekly reporting with additional restrictions was predicted to increase the likelihood of arrest, suggesting that heightened surveillance and violation opportunities could diminish success during pretrial supervision.
Notes: Results are in comparison to individuals who were not supervised during their pretrial period and individuals that received no conditions during their pretrial period. The predictions are derived from a Bayesian Logistic Regression model, which was estimated using normally distributed priors with a mean of 0 and standard deviation of 2.5, 10,000 iterations and four Monte Carlo Markov Chains. The forecasts are based on a hypothetical set of 1600 high-risk male individuals with identical criminal history and current offense information. Of the 1600 individuals, 200 were specified to be American Indian/Native Alaskan, 200 were Asian/Pacific Islander, 200 were Black, 200 were Hispanic, 200 were multiple races, 200 were other, 200 were unknown, and 200 were White.
3. Implications
This study underscores the complex and sometimes counterproductive effects of layering pretrial supervision conditions. Several implications emerge:
- More conditions do not necessarily yield better outcomes. Overly burdensome or mismatched conditions may inadvertently heighten the probability of a new criminal arrest during pretrial supervision.
- Treatment requirements, particularly when paired with supportive conditions like sobriety monitoring, appear more promising than surveillance-heavy strategies.
- Frequent reporting and electronic monitoring may strain individuals’ capacity to comply, especially when combined with other obligations. Such approaches could unintentionally contribute to a heightened likelihood of failure during pretrial supervision.
- The findings suggest that tailoring conditions to the needs of the individual, rather than stacking multiple supervision requirements, may improve compliance and reduce new arrests during pretrial supervision.
4. Data and Methods
The data for the study came from a county that participated in the Advancing Pretrial Policy and Research (APPR) project. The study drew on administrative data for 17,824 individuals facing pretrial supervision, including demographic characteristics, prior criminal history, and current charges. The sample includes individuals on pretrial supervision from January 1st, 2017, to December 31st, 2018. The outcome of interest was whether individuals experienced a new criminal arrest while on pretrial release.
Supervision Conditions: The analysis considered conditions such as regular check-ins, employment/education requirements, electronic monitoring, location restrictions, treatment mandates, sobriety orders, stay-away restrictions, and reporting frequency (monthly, bi-weekly, weekly).
Analytical Approach: Using Bayesian logistic regression models, the researchers forecasted the probability of a new arrest under different conditions. Forecasts were based on a hypothetical set of 1,600 high-risk male individuals matched on criminal history and current offense, with racial/ethnic composition balanced across groups. This approach enabled simulations that isolate the effects of specific conditions and their combinations.
5. Conclusion
Pretrial supervision conditions are intended to manage risk and ensure public safety, yet this study shows that success on pretrial supervision could depend on the combination of supervision conditions assigned to an individual. Some combinations of supervision conditions are predicted to be more effective at reducing new arrest than being not supervisions, having no special conditions, or each supervision condition individually. However, some combinations of supervision conditions are predicted to have no effect or increase the likelihood of an individual having a new arrest during pretrial supervision. The evidence highlights the importance of pretrial supervision case planning, suggesting that an evidence-based approach to assigning pretrial supervision conditions could reduce criminal behavior – an maintain public safety – while an individual is awaiting trial.
Disclosure: This research brief was prepared by ChatGPT and reviewed/edited by Ian A. Silver.
ianasilver.com 