Preventing In-Custody Deaths in Jails

IN SHORT

Most in-custody deaths are not random. They cluster in the first hours after booking, in medical and withdrawal events, in suicide, and in the unwitnessed time between rounds. The common thread is time that no one was watching. Prevention is earlier awareness in those windows, faster response, and an honest record of both.

FROM THE FIELD: MICHAEL RANES

Every in-custody death review I have been part of came down to time. Time between when something started and when anyone noticed. Time between noticing and responding. The person was almost never invisible the whole time; there was a window where a set of eyes would have changed the outcome, and no eyes were on it. That is what preventing a death actually means in a jail. Not predicting who, but closing the windows where a person is alone with a problem and no one knows yet.

Michael Ranes, Chief Corrections Strategy and Standards Officer, Virtual Patrol Technologies

// Where deaths cluster

The predictable windows

The risk is not spread evenly. It concentrates:

in the first hours to days after booking, especially for a first incarceration
in medical events and withdrawal that build quietly before they turn critical
in suicide, most often early in custody and in unobserved time
in the gaps between rounds, when no one is watching a person who is deteriorating

// The common thread

Unwatched time

Different causes, same mechanism: a stretch of time when a person was in trouble and no one knew yet. The footage usually shows it later. Prevention is about compressing that gap while it is still happening, not reconstructing it afterward.

// Prevention as a system

Awareness, response, record

No single step prevents a death. Screening, watch integrity, ligature-risk reduction, medical coordination, and earlier awareness between rounds each close part of the gap. Safety Intelligence adds the continuous set of eyes on the feeds no one can watch all at once, and an honest record of what was seen and done.

// Clearing up the confusion

What it is not

Not a guarantee. Nothing can promise no one will ever die in custody. This is about closing the windows where risk goes unseen.
Not a medical device. It does not measure vital signs or diagnose.
Not a replacement for medical or mental-health care. It gets the right person looking and responding sooner.
Not a certification of compliance. It supports awareness and an honest record, not a compliance guarantee.

// On the floor

How jailers use it

Intake first hours. earlier awareness in the highest-risk window in custody
Suicide watch. support the required cadence with a record of each observation
Medical and detox watch. a second set of eyes as a condition changes between rounds
Isolation and segregation. attention on people who are alone and least observed
Control room. route a clear alert to the officer who can go check now

// Source roundup, October 2026

The reporting floor is rising

Prevention and reporting are two sides of the same record. The federal Death in Custody Reporting Act (DCRA) sets the baseline: states that receive Byrne JAG funding report deaths in custody to the U.S. Department of Justice. States keep building on it.

Delaware, SB 291 (enacted July 2026). Puts DCRA reporting procedures into state law, requires compliance reporting, and makes a non-complying agency ineligible for certain funding until it reports.
California, AB 2257 (signed September 2026). Gives counties a formal role in who administers the county jail system.
Virginia, HB 80 (carried over to 2027). Would tie jail death-reporting compliance to notice to the Governor and to state jail funding. Passed the House in 2026.
New York, Jail and Prison Oversight Omnibus (enacted 2025). Mandatory annual inspections of every jail plus public posting of in-custody deaths.

The pattern is the same everywhere: report faster, report publicly, and show what happened and when. A facility that already keeps an honest, reviewable record of each round and each response is ready for that. Full list and status in the Legislation Watch; state-by-state oversight in the State Oversight Tracker.

FREE, NO LOGIN

See where the unwatched windows are in your facility before they matter. The 60-Day Audit Readiness Checklist walks the highest-risk gaps and how to close them.

Audit Readiness Checklist Camera Coverage Self-Assessment

// Common questions

Frequently asked

When are in-custody deaths most likely?

Risk concentrates in the first hours to days after booking, in medical and withdrawal events, and in suicide, especially during unobserved time between rounds.

What do most preventable in-custody deaths have in common?

A window of time when a person was in trouble and no one knew yet. Closing that gap is the core of prevention.

Can technology prevent an in-custody death?

No technology can guarantee it. Safety Intelligence helps by adding continuous attention on existing cameras and routing risk to staff sooner, so the unwatched windows get smaller.

Does this replace medical or mental-health staff?

No. It gets the right person looking and responding sooner. Clinical care stays with clinical staff.

Are the rules for reporting in-custody deaths changing?

Yes, and in one direction. Federal DCRA is the baseline, and states keep adding to it: Delaware put DCRA reporting into state law in 2026 with a funding consequence, New York requires public posting of in-custody deaths, and Virginia has a reporting bill carried over to 2027. Check your own state's rule.

Does it make a facility compliant?

No. It supports earlier awareness and an honest record. It does not make a facility compliant or certify any standard.

Virtual Patrol Technologies helps jails close the unwatched windows where in-custody deaths happen. See the method, or request a Safety Intelligence Audit.

Request a 60-Day Safety Intelligence Audit Catching a Medical Emergency Early Jail Suicide Prevention

Preventing in-custody deaths: the outcome everything else serves