Intensive Case Management in NYC’s Supervised Release Program: A Technical Assistance Pilot



On any given day, nearly 450,000 people in the United States are detained while awaiting the resolution of their criminal charges. Many of these individuals, still legally innocent, are in jail because they cannot afford to pay the bail amount set as a condition of their release. Pretrial detention takes a significant toll on the lives of affected individuals, putting them at increased risk of continued involvement with the criminal legal system as well as of losing their jobs, housing, and child custody. Systemic racial inequities throughout the criminal legal system mean that communities of color are disproportionately affected by cash bail and pretrial detention.

In 2016, New York City rolled out a citywide program known as Supervised Release (SR). SR offers judges the option of releasing arrested individuals under supervision in lieu of setting bail. Individuals released to SR are required to report to case managers regularly and are offered reminders of their court dates, case management support services, and voluntary connections to social services. The city developed the SR program to reduce the number of people detained in jail because they could not afford to pay bail, while at the same time maintaining court appearance rates and public safety.

Previous evaluation findings from MDRC offer strong evidence that SR achieved these overarching goals. However, there remain clients on SR with a high level of need, beyond the supports SR was designed to offer, who are frequently incurring new arrests or failing to appear for court dates. These high-need clients often struggle with stable housing, mental health, and substance use issues. In 2023, New York City allocated 13.1 million dollars to the SR program budget to develop and implement a new model called Intensive Case Management (ICM). The goal of the new model is to support these high-need SR clients and improve their pretrial outcomes.

In January 2024, two of the city’s Supervised Release providers—the New York City Criminal Justice Agency (CJA, Queens) and the Center for Alternative Sentencing and Employment Services (CASES, Manhattan)—began a one-and-a-half-year pilot of the new model. To support this effort, MDRC will serve as a data-driven technical assistance provider. In this role, MDRC will work with CJA and CASES to conduct rapid-cycle analyses to adapt and strengthen the model. The pilot will serve as a “proof of concept” for New York City, with the goal of offering the ICM model to all high-need clients at the end of the pilot.

Agenda, Scope, and Goals

The technical assistance analyses aim to iteratively answer the following guiding questions:  

  1. Does the Intensive Case Management (ICM) model support high-need supervision clients in returning to court and avoiding new arrests?
  2. What components of the program model work well to meet the needs of ICM clients? What program model components could be improved?​ 

Due to the nature of the pilot phase, not all ICM-eligible clients mandated to SR can currently receive ICM services. To distribute limited services fairly, ICM slots are allocated randomly. The technical assistance team will leverage this circumstance to assess differences in outcomes of eligible clients enrolled in ICM compared to eligible clients enrolled in standard Supervised Release. The team will also hold conversations with implementing staff members to identify process issues in delivering the model that could help us understand the results of our quantitative analysis and inform program improvement efforts.

Design, Sites, and Data Sources

MDRC will aid the NYC Mayor’s Office of Criminal Justice (MOCJ) in development of an Intensive Case Management (ICM) model intended to serve individuals assigned to pretrial supervision who have exhibited persistent patterns of failure to appear for court dates or of incurring new arrests while facing a pending criminal case. Our role will be to provide data-driven technical assistance by designing and operationalizing random assignment learning cycles at two Supervised Release provider locations (Manhattan and Queens). The team will use the Learn-Do-Reflect framework to inform data-driven decision-making, measure impact, and collaboratively adapt the ICM model at regular intervals to address challenges as they arise: 

LEARN: Identify and define an area for learning or improvement.

DO: Test out the proposed solutions while collecting data to help evaluate successes and roadblocks.

REFLECT: Review data with all stakeholders; examine manager, staff, and client experiences; propose course corrections; and collaboratively decide on next steps.

The team will process and analyze supervision and court data regularly to provide critical, timely information to MOCJ, CJA, and CASES regarding the details of service delivery and the effects of the new programming. With this information, the cross-agency team will iterate on the developing program model to identify the most effective components until a final model is settled upon.