The Arrest-Related Deaths (ARD) program is a national census of persons who died either during the process of arrest or while in the custody of state or local law enforcement personnel. The Bureau of Justice Statistics (BJS) implemented the ARD program in 2003 as part of the larger Deaths in Custody Reporting Program (DCRP). The DCRP was initiated to fulfill the data collection requirement of the Death in Custody Reporting Act of 2000 (DICRA), reauthorized in 2014, and collects data on mortality during arrest and incarceration, providing national information on the deaths of suspects and offenders from the time they come in contact with law enforcement personnel through the time they are incarcerated in jail or prison. Although BJS continues to collect information on deaths occurring in jails and prisons through the DCRP, data collection for the ARD program was suspended March 31, 2014.

Defining deaths in the process of arrest
The DICRA legislation required the collection of data on deaths that occur in the process of arrest. Therefore, the ARD program was not limited to data on traditional notions of police use of lethal force or justifiable homicides. It collected information about homicides, suicides, accidental deaths, and deaths attributed to intoxication and medical conditions that occurred during an interaction with state or local law enforcement.

BJS uses the term "arrest-related" to describe the wide range of circumstances pertaining to an arrest, including events that occurred from the process of apprehension to detention. BJS defines a death as arrest-related when the event that caused the death (e.g., gunshot wound, cardiac arrest, or drowning) occurred during an interaction with state or local law enforcement personnel. Under the ARD program definition, an arrest-related death could occur (1) during an attempted arrest or in the process of apprehension, (2) while in custody (prior to transfer to jail), or (3) shortly after freedom to leave is restricted by law enforcement.

Program exclusions
Not all deaths that occurred during an interaction with state or local law enforcement personnel were reportable to the program. The ARD program scope excluded four general situations: (1) deaths of bystanders, hostages, and law enforcement personnel; (2) deaths by federal law enforcement agents; (3) deaths of wanted criminal suspects prior to police contact; and (4) deaths by vehicular pursuits without any direct police action.

BJS implemented the ARD program after passage of the DICRA. Through the program, State Reporting Coordinators (SRC) in all 50 states and the District of Columbia coordinated data collection efforts and reported the data to BJS. BJS compiled the data and produced national-level statistics on deaths that occurred in the process of arrest by, or while in the custody of, state or local law enforcement personnel.

The ARD program design specified that SRCs were responsible for identifying all reportable deaths in their state and for obtaining information about these deaths. BJS offered guidance to the SRCs on how to collect ARD data, but a standard methodology was not required. BJS recommended that SRCs use one or more of the following procedures to identify arrest-related deaths: implement legislation, contact state medical examiner and county coroner officers, involve state and local prosecutors' offices, contact state police, expand the role of state Uniform Crime Reporting reporters, and conduct open-source searches. SRCs developed data identification and collection strategies according to their preferences and resources; therefore, the ARD program design reflects many state-specific methodologies that are integrated into a single, national-level program.

SRC arrest-related deaths identification methods

SRCs use a variety of techniques to identify deaths consistent with the ARD program definition. BJS encourages SRCs to be proactive in identifying arrest-related deaths, such as surveying law enforcement agencies, mining existing data collections, and conducting open-source searches of media. Passive methodologies rely on other agencies to identify arrest-related deaths, such as collecting data through voluntary submissions of completed CJ-11A Incident Reports from other entities (e.g., law enforcement agencies and medical examiner/coroners' offices).

When a reportable death is identified, SRCs obtain information about the circumstances of the death and submit the information to BJS on a CJ-11A Incident Report. SRCs are advised to work with the law enforcement agency involved with the arrest-related death to obtain critical facts about the incident. In instances where the SRCs are unable to obtain full reporting from the law enforcement agency involved with the death, they will compile details of the event from official source documents and complete the CJ-11A Incident Report. If an SRC is not able to complete a CJ-11A from source documents, they will complete any items on the Incident Report that are available from the data source used to identify the death, and then submit the records to BJS.