The Autopsy / Case File · Disparities
The Two Exceptions
The intuitive story about dying in custody is that it's more dangerous than dying free — violence, neglect, a system that doesn't prioritize the people inside it. The Bureau of Justice Statistics' own age-adjusted comparisons tell a stranger and more specific story: people in state prison, matched for age, sex, and race against adults outside, are actually less likely to die of any cause in a given year. Except two.
In 2018, BJS found state prisoners had an age-adjusted mortality rate of 319 per 100,000, against 419 per 100,000 for the general adult population matched to the same demographic profile. Local jail inmates that same year had a rate of 146 per 100,000 against an adjusted 322 per 100,000 outside. Both custody populations were less likely to die of nearly every cause BJS tracks — except suicide and homicide, where the pattern reverses, sharply.
Join the discussion →
The counterintuitive finding isn't really a mystery once the mechanics of who ends up incarcerated, and how, are accounted for. Jails and prisons don't admit a random cross-section of the American public; they admit a population that's already been screened, however imperfectly, by intake medical exams, and one that — inside prison walls specifically — has no access to firearms, motor vehicles, illicit street drugs of unknown purity, or most of the other leading external causes of death in the general population. Substance use, high-risk driving, and gun violence account for a disproportionate share of preventable deaths among free adults in the same age range as the incarcerated population; remove access to cars, guns, and an unregulated drug supply, and a lot of the causes that would otherwise kill a demographically similar person outside simply can't happen the same way inside. What can't be engineered away by a locked door, and what BJS's own statistical tables show rising rather than falling: suicide, and to a lesser extent homicide, committed against or by other people in custody.
Two Institutions, Two Very Different Death Profiles
Jails and prisons are, functionally, different institutions serving different populations, and their mortality data reflects that split cleanly. Local jails hold people who are usually awaiting trial, newly arrested, or serving very short sentences — a population in the acute, disorienting first hours and days of losing their freedom, often without medication continuity, without their support network, and frequently in active withdrawal from alcohol or drugs. Suicide is the leading cause of death in local jails, and BJS's most recent statistical tables, covering 2000 through 2019, put the 2019 jail suicide rate at 49 per 100,000 inmates — the single largest contributor to a total jail mortality rate that rose from 151 per 100,000 in 2000 to 167 per 100,000 in 2019, an 11 percent increase over two decades. Total jail deaths in 2019 reached 1,200, up more than 5 percent from 1,138 the year before. Deaths from drug or alcohol intoxication hit a 20-year high the same year: 184 deaths, surpassing the previous peak of 180 set just one year earlier — a trend that tracks closely with the broader national overdose crisis, arriving inside jail walls via people detained while already in a life-threatening substance-use crisis that intake screening didn't catch or couldn't stabilize in time.
State and federal prisons hold a longer-serving, older, post-conviction population, and their death profile looks correspondingly like an aging-in-place problem rather than an acute-crisis one. Illness accounted for roughly 87 percent of the 65,027 state prisoner deaths and 89 percent of the 7,125 federal prisoner deaths recorded from 2001 through 2019 — cancer, heart disease, liver disease, the same causes that dominate mortality among older adults generally, arriving in a population that has aged substantially behind bars as sentence lengths grew and parole grants didn't keep pace. State prison mortality reached 330 per 100,000 in 2019, federal facilities 259 per 100,000; the state prison system recorded 3,853 deaths that year, down 284 from 2018, while federal facilities saw a slight rise to 381, up from 378. The gap between the two systems' rates likely reflects real differences in the populations each holds — the federal system's mix of offense types and its historically shorter average sentences relative to some state systems both being plausible partial explanations, though BJS's own tables don't isolate a single cause.
Jail inmates were less likely than adult U.S. residents to die from all causes of death except suicide and homicide.
The Trend Line That Doesn't Match the Overall Story
If the age-adjusted comparison were the whole story, incarceration would read as a strange, involuntary form of harm reduction — which is, bluntly, an uncomfortable way to describe a system built around punishment and confinement, and not remotely what the suicide-specific trend data shows. BJS's dedicated report on suicide in custody, covering 2000 through 2019, found the suicide rate climbed 85 percent in state prisons and 61 percent in federal prisons over that period, against a much smaller 13 percent rise in local jails — meaning the one cause of death where custody populations already fare worse than the outside world is also the cause getting rapidly worse over time, even as overall mortality in the same facilities trends in the opposite direction. That divergence — falling relative risk for nearly every cause of death, rising absolute risk for the one cause where custody was already worse — is the actual disparity this data documents, and it's a considerably narrower, more specific finding than "prison is dangerous" or "prison is safe," either of which oversimplifies what BJS's own numbers show.
The suicide data also contains a racial disparity that runs in the opposite direction from most of the mortality gaps tracked elsewhere in this series. Among local jail inmates from 2015 through 2019, the average suicide rate for white inmates was 93 per 100,000 — five times the rate for Black inmates (18 per 100,000) and more than three times the rate for Hispanic inmates (26 per 100,000). That inversion doesn't fit the pattern found in almost every other cause-of-death disparity this series has covered, where Black and Hispanic Americans typically carry the higher burden; researchers who study custody suicide have pointed to differences in offense type, prior justice-system contact, socioeconomic background at intake, and the disorientation of a first-time versus repeat detention experience as plausible contributing factors, though BJS's statistical tables document the size of the gap without adjudicating its full explanation.
What "Less Dangerous" Doesn't Mean
None of this age-adjusted comparison is an argument that incarceration is safe, or that conditions inside American jails and prisons don't contribute to the deaths that do happen there. It's a narrower and, in some ways, more useful finding: that the mortality disparity between incarcerated and free populations isn't diffuse across every cause of death the way a simple headline rate comparison would suggest, but concentrated specifically in suicide and, to a smaller degree, homicide — the causes most directly tied to the psychological weight of confinement, the loss of autonomy and privacy, and the proximity to other people in a closed, sometimes violent environment that a locked door can't screen out the way it screens out a car or a gun. A comparison built only on raw, unadjusted death counts would miss this distinction entirely, since raw counts don't account for the fact that incarcerated populations skew younger and, at least at intake, have already passed through a screening process the general population never does. The age-adjusted version BJS actually publishes is the more honest one, and it points policy attention toward the two causes where it's actually warranted rather than treating custody mortality as one undifferentiated problem.
And Then You're Dead
Almost everything that kills a person outside a locked door — the car, the gun, the fentanyl of unknown purity — can't reach someone behind one. What the door can't stop is the weight of being behind it, and BJS's own numbers show that weight getting heavier, year over year, even as the rest of the risk falls away.
That's it. That's the whole thing.
Sources
- Age-adjusted state prisoner mortality rate of 319 per 100,000 versus 419 per 100,000 for the general adult population, 2018, adjusted for age, sex, and race/ethnicity — Carson EA, "Mortality in State and Federal Prisons, 2001–2018 — Statistical Tables," Bureau of Justice Statistics, NCJ 255970, April 2021
- Age-adjusted local jail mortality rate of 146 per 100,000 versus 322 per 100,000 for the adjusted general population, 2018, and jail inmates less likely to die of all causes except suicide and homicide — Carson EA, "Mortality in Local Jails, 2000–2018 — Statistical Tables," Bureau of Justice Statistics
- 2019 jail mortality rate (167/100,000, up from 151/100,000 in 2000), 2019 suicide rate (49/100,000), total 2019 jail deaths (1,200, up from 1,138 in 2018), and drug/alcohol intoxication deaths reaching a 20-year high (184 in 2019) — Carson EA, "Mortality in Local Jails, 2000–2019 — Statistical Tables," Bureau of Justice Statistics, December 2021
- 2019 state prison mortality rate (330/100,000) and federal rate (259/100,000); state deaths (3,853) and federal deaths (381) in 2019; illness accounting for 87% of 65,027 state and 89% of 7,125 federal prisoner deaths, 2001–2019 — Carson EA, "Mortality in State and Federal Prisons, 2001–2019 — Statistical Tables," Bureau of Justice Statistics, December 2021
- Suicide rate increases of 85% (state prisons), 61% (federal prisons), and 13% (local jails), 2001–2019; racial disparity in jail suicide rates, 2015–2019 average (white inmates 93/100,000, Black inmates 18/100,000, Hispanic inmates 26/100,000) — Carson EA, "Suicide in Local Jails and State and Federal Prisons, 2000–2019 — Statistical Tables," Bureau of Justice Statistics, NCJ 300731, October 2021
- Photo: "Prison cell block.jpg," Bob Jagendorf, CC BY 2.0, via Wikimedia Commons
Comments
Loading…