Taking action to reduce deaths in custody

I have spent most of my 30-plus-year career focused on corrections and reentry reform, working to change the climate inside prisons and jails so that conditions are safer and healthier for people who are incarcerated and for the professionals who work in those facilities. And at the Office of Justice Programs (OJP), we continue to invest resources to make considerable progress on many fronts. Through grants, research, training and technical assistance:

• We are helping to transform corrections culture, supporting training, research, technical assistance and demonstration projects designed to make correctional facilities safer and more humane.
• We’re investing in research to better understand the factors that contribute to in-custody deaths and identify opportunities to intervene and prevent fatal encounters.
• We have taken steps toward reducing reliance on solitary confinement and protecting people in confinement facilities from sexual assault.
• We’ve expanded job and educational opportunities and opened avenues to treatment for individuals reentering their communities from incarceration.
• And de-escalation training is being adopted and deployed in both corrections and law enforcement settings to reduce the chances of harm during encounters with justice system professionals.

These actions are providing a greater measure of protection and a wider path to opportunity for people who come into contact with our criminal legal system. But admittedly, we have a long way to go when it comes to collecting data about and ultimately reducing the number of fatalities that occur when individuals are taken into custody by the police or incarcerated in correctional facilities.

Our work to implement the Death in Custody Reporting Act—specifically, to understand the nature and scope of the problem and identify strategies for reducing in-custody deaths—is central to achieving this critical goal. And while we have much more to do, we are making significant headway and we are deeply committed to finding ways to prevent and reduce deaths in custody.

Not every death in custody is preventable. But the power of the state confers upon it a responsibility to account for the lives lost while under its care and control in the clearest terms possible. The Death in Custody Reporting Act, signed into law in 2000 and reauthorized 14 years later as the Death in Custody Reporting Act of 2013 (DCRA), is intended to bring a measure of transparency to the circumstances surrounding in-custody deaths and, even more important, to help prevent and reduce future tragedies.

Collecting Complete and Accurate Data

The first step toward prevention is understanding the scope of the problem, and we are committed to collecting complete and accurate data. When DCRA first became law, the Bureau of Justice Statistics (BJS) served as the primary collector of information about deaths in custody, and its methods yielded a robust response from corrections institutions. Ninety-eight percent of jails and 100 percent of state prisons submitted data on in-custody deaths to BJS.

But DCRA of 2013 set in motion changes that, perhaps unintentionally, limited BJS’s role in the collection of federal DCRA data. The new legislation also required reporting from a single state-level agency instead of directly from local entities, another major shift. The Department of Justice proposed a technical legislative fix to explicitly enable BJS to resume its data collection activities for state and local agencies, but the legislative fix has not yet moved forward. The state and local data collection role is now in the hands of our Bureau of Justice Assistance (BJA), a grantmaking office of OJP.

BJA has since been working on many fronts to substantially and urgently improve data collection. But the challenge is significant: A report from the Government Accountability Office comparing publicly available information to the data reported to BJA in fiscal year 2021 found that more than 1,000 in-custody deaths were unaccounted for in the states’ submissions. Moreover, 70 percent of the submitted records were missing at least one required element (information about the deceased; date, time and location of the death; the agency involved; and circumstances surrounding the death).

This gap in data pains me: The gathering and reporting of this information are fundamental responsibilities owed to the public, and ones that OJP takes seriously. And yet with more than 18,000 distinct law enforcement agencies, more than 50 state- and territory-run prison systems and over 3,000 autonomous jails in the U.S., this is no simple task. BJA must encourage and assist these independently governed institutions to gather the relevant data and report it through a single point of contact within each state. Many states report major challenges to collecting and consolidating data from multiple sources and ensuring the completeness and accuracy of data prior to reporting to BJA. These issues are especially difficult to overcome without an appropriated funding source or in-state infrastructure.   

Efforts to Improve Data Collection

We are redoubling our efforts to provide the tools we’ve got—technical assistance and support—to jurisdictions to improve the completeness and quality of reporting, from local agencies to the state, and from the state to the federal government. 

For example, BJA has launched a DCRA Training and Technical Assistance Center, managed by the Justice Information Resource Network (JIRN), which provides onsite and virtual training and technical assistance to state administering agencies and other DCRA reporters. The Center produces materials that identify and describe additional data sources that may help agencies provide complete and accurate reporting.

The Center’s efforts build on momentum from a December 2022 meeting with 118 representatives of state, territory and D.C. governments and several non-profit and professional member organizations. The meeting tackled many of the challenges around DCRA implementation and highlighted promising strategies for reporting. BJA and JIRN have continued to provide trainings, webinars and one-on-one outreach to support progress on data collection and reporting. 

BJA has also been using data from two open-source databases—Mapping Police Violence and The Washington Post’s Fatal Force—to identify otherwise unreported arrest-related deaths, and in the spring of 2023, BJA shared with states the data it gathered. Based on this information, state administering agencies have reported back over 200 new arrest-related deaths, a 47% increase in the number of reported deaths in this category. Twenty-five states back reported at least one new arrest-related death. BJA shared another round of data with states this spring, and while results are still being gathered, we expect a large increase in the number of reported deaths.

The goal here, of course, is to close the data gap and the open-source analysis is one additional tool. We are also working closely with the FBI’s National Use-of-Force Data Collection program to gain a better understanding of the scope of unreported arrest-related deaths.

BJA also developed DCRA Compliance Guidelines that lay out the steps states must take to meet DCRA reporting requirements. In addition to reporting complete and comprehensive quarterly data and evidence of quality improvements where gaps and challenges are identified, each state is obligated to submit a DCRA implementation plan for approval by BJA, describing the state’s data collection infrastructure, data collection methods and reporting methods. BJA and JIRN review these plans and use them to help target training and technical assistance resources to pressing challenges. State administering agencies are expected to submit either updated plans or implementation reports annually in subsequent fiscal years. Each state’s implementation plan is posted here. 

We aim to give state and local organizations the tools, training and assistance needed to vastly improve reporting. Most states are working hard to do so and taking advantage of the new guidance and assistance. For others, we are taking corrective actions that may include withholding from states a percentage of funds under the Edward Byrne Memorial Justice Assistance Grant Program[i], or they could be assessed up to a 10% penalty of their allocated amount (a provision known as “the DCRA penalty”). In 2024, BJA will conduct the first annual review of DCRA compliance for all states and make recommendations for corrective actions.

As noted earlier, BJS continues to collect, analyze and report information about deaths that occur in the custody of federal law enforcement agencies. In a recent report, Federal Deaths in Custody and During Arrest, 2021, BJS achieved a 100% response rate from federal agencies. The report describes decedent, incident and facility characteristics of deaths in federal custody and during arrest by federal law enforcement agencies during fiscal year 2021. Nine agencies reported at least one arrest-related death, and six agencies reported at least one death in custody.

Through the National Institute of Justice (NIJ), we are supporting rigorous research to analyze data collected through DCRA and other sources to produce data-driven recommendations for reducing deaths in custody. NIJ released an initial report titled Literature Review and Data Analysis on Deaths in Custody in 2022. This study reviewed existing research and data focused on the prevalence, patterns and contexts of deaths in custody and explored factors associated with deaths in correctional institutions. To build on the first study, NIJ contracted with RTI International in late 2021 to conduct a broader three-year study involving a national-level review and analysis of policies, practices (including management practices) and available data addressing deaths in custody, along with in-depth case studies of multiple sites and agency types. This research is ongoing and scheduled for completion in late 2024. 

Meanwhile, the President’s fiscal year 2025 budget proposes to strengthen states’ ability to carry out their DCRA responsibilities. The budget requests $5 million to support DCRA implementation which would provide states new funding to aid in collecting and reporting DCRA data, and for states to provide subawards to local entities to support their needs. 

Preventing and reducing deaths in custody is both a legal responsibility and a solemn moral obligation that the Office of Justice Programs is working hard to fulfill. By collaborating with our state and local partners to accurately report deaths in custody and assess their efforts to protect those under their charge, we can help save lives and uphold the principles of fairness, equity and safety on which our system of justice is built.