The Bridge: See No Evil, Hear No Evil, Collect Too Little Data

Delivered to our subscribers on Thursdays, The Bridge doesn't get caught up in talking points from the Hill, the latest major headlines, or even the executive branch priorities. The Bridge is an email exclusive product, and we are offering our audience a sneak peek to see what The Bridge is all about. Sign up here.

In this week’s issue of The Bridge, we turn our lens on a dangerous case of neglect by the Department of Justice.


In 2000, Congress passed the Death In Custody Reporting Act (DCRA), which ordered the Department of Justice to track how many people die during interactions with law enforcement officials and collect data on the circumstances surrounding their deaths. The law was intended to help find ways to reduce deaths in custody. More than twenty years later, DOJ is still struggling to build an effective tracking program. A fix is overdue but possible.

Tragedy and the Bureaucratic Morass

Last month, a coalition of civil and human rights groups urged Attorney General Merrick Garland to make fully implementing DCRA a priority. Two weeks later, the chief coroner for Sedgwick County, Kansas, ruled that the September 2021 death of a Black Kansas teen was a homicide.

High school senior Cedric Lofton lost consciousness and had no pulse within four minutes of being cuffed, shackled, and forced facedown onto the floor of a juvenile detention center after a struggle with the facility’s staff. The 135-pound, 17-year-old died two days later. In reporting on his death, the Washington Post, the New York Times, and CNN all cited growing concerns about the dangers of restraining individuals in a prone, or facedown, position.

One of DCRA’s aims is to use data to develop strategies for reducing deaths. If DOJ had implemented an effective program two decades ago, maybe data would have led the department to work harder to prevent law enforcement officials from positioning restrained individuals facedown. Cedric Lofton might be alive today.

DOJ created two programs after Congress enacted DCRA, one to collect data on deaths of incarcerated persons and another to collect data on deaths that occur during arrests. DOJ continued both programs after the law expired in 2006. But the data it gathered was never complete. Every year in which DOJ collected data, some states failed to comply with the reporting requirements. Concerned about the quality of data it received, DOJ stopped collecting data on arrest-related deaths in 2014. An audit the next year estimated that the program had captured only about half of law enforcement homicides for most years – and DOJ collected too little data to even audit for 2010, 2012, and 2013.

In 2014, Congress passed a new version of DCRA. This time, the law came with teeth to incentivize compliance: Congress authorized DOJ to rescind up to 10% of a state’s funding under a grant program that accounts for the bulk of federal aid to state and local law enforcement agencies. The law gave DOJ until December 2016 to study the data and issue a report recommending “means by which such information can be used to reduce the number of such deaths.” By the end of 2016, DOJ had not made much headway collecting data, much less issue the required report.

Instead, in December 2016, Attorney General Loretta Lynch published a plan to begin collecting and studying data in 2017. The plan contemplated holding states accountable by undertaking an “open source review” through which DOJ would search news reports and other public information to identify deaths states omitted. The plan also called for states to submit compliance plans explaining how they would improve the quality of their data.

The Trump administration scrapped Lynch’s plan and further delayed data collection. Attorney General Jeff Sessions decided to take state submissions at face value, despite their history of submitting egregiously incomplete data. The department wouldn’t conduct an open source review or require state compliance plans. Sessions also never adopted a pledge Lynch had made to release anonymized data proactively.

The Project On Government Oversight (POGO) and a broad coalition of groups objected to this change, and DOJ’s inspector general echoed our concerns. A group of senators later asked Attorney General Bill Barr to change course, without success. When collection finally began in 2020, the lack of any data validation produced foreseeably bad results.

Attorney General Garland Can Fix This Problem

Garland needs to make the death in custody reporting program a priority. POGO sounded the alarm in a letter to him last July. We wrote to DOJ again in August when the department proposed to renew a poorly designed data collection form. Both of POGO’s letters offered recommendations for reforms that are within his power to implement.

Problems with DOJ’s approach are obvious. The department hasn’t revived Lynch’s ideas of validating data through open source reviews and requiring states to submit compliance plans. Though the department has now said it will release “general data related to deaths in custody,” POGO pointed out in its August 2021 letter that this statement falls short of an “affirmative commitment” to release anonymized data proactively. The state data collection form doesn’t seek information about known disabilities of the deceased, despite indications that disabilities, especially mental disabilities, are relevant to the study of deaths in custody. The form includes space for brief descriptions of circumstances surrounding deaths but provides minimal prompts indicating the type of information sought.

The department knows better. A DOJ factsheet says that ascertaining the circumstances of deaths is “crucial to developing policies and program changes that could reduce the number of in-custody fatalities.” A separate form for deaths in federal custody specifically asks if the individual was in a prone position at the time of death. But nothing in the statedata collection form would signal to Kansas officials any need to mention that Cedric Lofton died after they laid him out prone.

DOJ’s approach could compound the consequences of systemic racism in the United States. One May 2021 study showed that 60% of reported deaths in custody involve people of color, who represent only 40% of the U.S. population — and the study estimated that their deaths may be underreported by as much as 50%. This is a crisis the nation can’t afford to ignore.

The good news is that Garland can begin working on solutions without waiting on Congress to change the law. He could implement Lynch’s 2016 plan, improve data collection tools, and ensure that DOJ analyzes data quickly. But Garland has to want to do that. It’s up to Congress and the public to pressure him to act before more lives are needlessly lost.