This explainer is based on Prison Mortality & Deaths in Custody: Data Gaps, Misclassification, and Accountability Failures. All statistics and findings are drawn directly from this source.
News Lead
The federal government has lost track of thousands of people who died behind bars. A 2022 Department of Justice report identified more than 5,000 uncounted in-custody deaths missing from the national mortality database — a staggering gap that renders the country’s official record of who dies in prison and jail fundamentally unreliable. A separate Marshall Project investigation found nearly 700 additional individuals who died in law enforcement custody but never appeared in the federal dataset. Entire states, like Mississippi, had reported almost zero deaths in their prisons or jails despite holding substantial incarcerated populations.
Georgia is part of the problem. Original research by Georgia Prisoners’ Speak (GPS) found that the Georgia Department of Corrections misclassified at least 44 deaths — labeling drug overdoses as “natural causes” or “undetermined” — effectively erasing the true circumstances under which people in its custody died. Only one state prison system in the entire country (Iowa) provides complete and timely death data to the public.
The data crisis is not a bureaucratic inconvenience. When the state fails to accurately count and classify how people die in its custody, it becomes impossible to hold anyone accountable for preventable deaths — deaths that an Illinois study suggests may account for 36% to as many as 73% of all prison fatalities.
Key Takeaway: The Department of Justice identified more than 5,000 uncounted deaths in custody, and GPS research found Georgia misclassified at least 44 deaths, revealing a nationwide accountability collapse.
Quotable Statistics
National Scope of the Crisis:
– More than 5,000 deaths in custody went uncounted in the national mortality database, according to a 2022 Department of Justice report.
– Nearly 700 individuals who died in law enforcement custody were absent from the federal DCRA dataset (Marshall Project investigation).
– More than three-quarters of a random sample of approximately 1,000 DCRA entries did not meet the federal government’s own criteria for how a death should be recorded.
– The federal government reported 6,725 deaths in custody in FY 2023 — a figure widely understood to be a significant undercount.
Transparency Failures:
– 21 of 54 prison systems (38%) release no individual death data at all.
– Only 1 system — Iowa — releases complete and timely death data.
– Only 8 states have their own laws requiring regular reporting of deaths in custody to state authorities.
– 13 systems release incomplete data slower than one year after a death occurs.
Georgia-Specific Findings (GPS Original Research):
– At least 13 deaths the Georgia Department of Corrections reported as “natural causes” were later determined by medical examiners to be accidental drug overdoses.
– 31 additional deaths GDC labeled “undetermined” were later ruled accidental drug overdoses by medical examiners.
– At least 44 deaths total were misclassified by GDC.
– Georgia saw at least 49 drug overdose deaths in its prisons between 2019 and 2022, up from just 2 in 2018.
Preventable Deaths:
– A court-appointed medical expert in Illinois found that of 33 prison deaths studied, 12 were preventable, 7 might have been preventable, and 5 could not be determined because the deaths were not adequately documented — yielding a 36% confirmed preventable rate and up to 73% potentially preventable rate.
– Over 20% of state prisoners with persistent medical conditions go without care (Vera Institute, 2025).
– Nearly 3,000 incarcerated people died from COVID-19 since March 2020.
Key Takeaway: From the national level to Georgia specifically, the data reveals systemic undercounting, misclassification, and a failure to prevent deaths the state has a constitutional obligation to prevent.
Context and Background
What reporters need to know:
The federal reporting system is broken. The Death in Custody Reporting Act (DCRA), updated in 2013, requires states to report deaths in custody to the Attorney General as a condition of receiving federal criminal justice funding. DCRA requires reporting 10 data elements for each death within one quarter (3 months). But compliance has collapsed. When the Bureau of Justice Statistics (BJS) managed the data from 2000 to 2019, reporting was imperfect but functional. When responsibility shifted to the Bureau of Justice Assistance (BJA), data quality and transparency deteriorated significantly. The BJA now reports annual totals — FY 2020: 5,674 deaths; FY 2021: 6,909 deaths; FY 2022: 6,085 deaths; FY 2023: 6,725 deaths — but these are widely understood to be significant undercounts.
Independent sources are filling the void. The UCLA Law Behind Bars Data Project (formerly the COVID Behind Bars Data Project) is currently the most comprehensive independent source for prison mortality data. Reuters has built the largest database of jail deaths. The Third City Mortality Project, publishing in the Journal of Public Health Management and Practice (May/June 2024), assessed state-level compliance and found the data landscape deeply inadequate.
Georgia’s record is particularly concerning. Georgia does not proactively publish individual death data. GPS has tracked mortality through its own data collection, uncovering the misclassification of at least 44 deaths by the Georgia Department of Corrections. The Atlanta Journal-Constitution has conducted ongoing investigations into deaths in the Georgia prison system.
The legal standard is clear. The constitutional standard, rooted in the Eighth Amendment, requires that the state provide adequate medical care to people in its custody. When over 20% of state prisoners with persistent medical conditions go without care, and when up to 73% of studied prison deaths may have been preventable, that standard is being violated on a massive scale.
Key terminology: “Cause-of-death misclassification” refers to the incorrect categorization of what caused a death — for instance, labeling a drug overdose as “natural causes.” “Preventable death” means a death that would not have occurred with adequate medical care, proper procedures, or appropriate intervention. The Mortality in Correctional Institutions (MCI) data collection covers all 50 state prison systems and approximately 3,095 local jails.
Key Takeaway: The federal death-in-custody reporting system has collapsed since 2019, leaving independent researchers and advocacy organizations as the primary sources of reliable data on how people die in state custody.
Story Angles
1. “Georgia’s Hidden Overdose Crisis: 44 Deaths the State Didn’t Want to Call What They Were”
GPS original research found at least 44 deaths misclassified by the Georgia Department of Corrections — drug overdoses labeled as “natural causes” or “undetermined.” Georgia saw at least 49 drug overdose deaths between 2019 and 2022, up from just 2 in 2018. This story investigates how misclassification conceals the true scale of the drug crisis inside Georgia prisons, prevents evidence-based policy responses, and denies families the truth about how their loved ones died. Reporters can request GDC records and compare them against medical examiner rulings.
2. “One State Gets It Right: Why Iowa Can Count Its Dead and 53 Other Systems Can’t”
Out of 54 prison systems in the United States, only Iowa releases complete and timely death data. Meanwhile, 21 systems (38%) release no individual death data at all, and only 8 states have their own laws requiring regular death reporting. This comparative investigation asks what Iowa does differently, what it costs, and why other states — including Georgia — choose opacity over transparency. The fact that one state proves full compliance is feasible makes the failures of 53 others a policy choice, not an impossibility.
3. “The 73% Question: How Many Prison Deaths Could Be Prevented?”
A court-appointed medical expert studying 33 deaths in Illinois prisons found that 36% were confirmed preventable and up to 73% were potentially preventable — with some deaths so poorly documented that preventability couldn’t even be assessed. Paired with data showing over 20% of state prisoners with persistent medical conditions receive no care, this story explores the human cost of medical neglect behind bars and asks whether similar studies should be mandated in Georgia and other states.
Read the Source Document
Other Versions
- Public Version — A plain-language summary for community members, families, and advocates
- Legislator Version — A policy brief with reform recommendations for Georgia lawmakers