TY - JOUR
T1 - Rates and causes of death after release from incarceration among 1 471 526 people in eight high-income and middle-income countries: an individual participant data meta-analysis
AU - MARIC collaborators
AU - Borschmann, Rohan
AU - Keen, Claire
AU - Spittal, Matthew J.
AU - Preen, David
AU - Pirkis, Jane
AU - Larney, Sarah
AU - Rosen, David L.
AU - Møller, Lars
AU - O'Moore, Eamonn
AU - Young, Jesse T.
AU - Altice, Frederick L.
AU - Binswanger, Ingrid A.
AU - Bukten, Anne
AU - Butler, Tony
AU - Chang, Zheng
AU - Chen, Chuan-Yu
AU - Clausen, Thomas
AU - Christensen, Peer B.
AU - Culbert, Gabriel J.
AU - Cunningham, Ruth
AU - Degenhardt, Louisa
AU - Dolan, Kate
AU - Fazel, Seena
AU - Fischbacher, Colin
AU - Graham, Lesley
AU - Huang, Yen-Fang
AU - Huber, Florence
AU - Karaminia, Azar
AU - King, Paula
AU - Kouyoumdjian, Fiona G.
AU - Lim, Sungwoo
AU - Lopez, Derrick
AU - Moniruzzaman, Akm
AU - Morenoff, Jeffrey
AU - Pizzicato, Lia N.
AU - Proescholdbell, Scott K.
AU - Ranapurwala, Shabbar I.
AU - Slaunwhite, Amanda
AU - Somers, Julian M.
AU - Spaulding, Anne C.
AU - Stavseth, Marianne R.
AU - Stern, Marc F.
AU - Telfer, Kendra
AU - Viner, Kendra
AU - Wang, Nadia
AU - Zhao, Bin
AU - Zhu, Nanbo
AU - Kinner, Stuart A
PY - 2024/5
Y1 - 2024/5
N2 - Summary Background Formerly incarcerated people have exceptionally poor health profiles and are at increased risk of preventable mortality when compared with their general population peers. However, not enough is known about the epidemiology of mortality in this population—specifically the rates, causes, and timing of death in specific subgroups and regions—to inform the development of targeted, evidence-based responses. We aimed to document the incidence, timing, causes, and risk factors for mortality after release from incarceration. Methods We analysed linked administrative data from the multi-national Mortality After Release from Incarceration Consortium (MARIC) study. We examined mortality outcomes for 1 471 526 people released from incarceration in eight countries (Australia, Brazil, Canada, New Zealand, Norway, Scotland, Sweden, and the USA) from 1980 to 2018, across 10 534 441 person-years of follow-up (range 0–24 years per person). We combined data from 18 cohort studies using two-step individual participant data meta-analyses to estimate pooled all-cause and cause-specific crude mortality rates (CMRs) per 100 000 person-years, for specific time periods (first, daily from days 1–14; second, weekly from weeks 3–12; third, weeks 13–52 combined; fourth, weeks 53 and over combined; and fifth, total follow-up) after release, overall and stratified by age, sex, and region. Findings 75 427 deaths were recorded. The all-cause CMR during the first week following release (1612 [95% CI 1048–2287]) was higher than during all other time periods (incidence rate ratio [IRR] compared with week 2: 1·5 [95% CI 1·2–1·8], I2=26·0%, weeks 3–4: 2·0 [1·5–2·6], I2=53·0%, and weeks 9–12: 2·2 [1·6–3·0], I2=70·5%). The highest cause-specific mortality rates during the first week were due to alcohol and other drug poisoning (CMR 657 [95% CI 332–1076]), suicide (135 [36–277]), and cardiovascular disease (71 [16–153]). We observed considerable variation in cause-specific CMRs over time since release and across regions. Pooled all-cause CMRs were similar between males (731 [95% CI 630–839]) and females (660 [560–767]) and were higher in older age groups. Interpretation The markedly elevated rate of death in the first week post-release underscores an urgent need for investment in evidence-based, coordinated transitional healthcare, including treatment for mental illness and substance use disorders to prevent post-release deaths due to suicide and overdose. Temporal variations in rates and causes of death highlight the need for routine monitoring of post-release mortality. Funding Australia's National Health and Medical Research Council.
AB - Summary Background Formerly incarcerated people have exceptionally poor health profiles and are at increased risk of preventable mortality when compared with their general population peers. However, not enough is known about the epidemiology of mortality in this population—specifically the rates, causes, and timing of death in specific subgroups and regions—to inform the development of targeted, evidence-based responses. We aimed to document the incidence, timing, causes, and risk factors for mortality after release from incarceration. Methods We analysed linked administrative data from the multi-national Mortality After Release from Incarceration Consortium (MARIC) study. We examined mortality outcomes for 1 471 526 people released from incarceration in eight countries (Australia, Brazil, Canada, New Zealand, Norway, Scotland, Sweden, and the USA) from 1980 to 2018, across 10 534 441 person-years of follow-up (range 0–24 years per person). We combined data from 18 cohort studies using two-step individual participant data meta-analyses to estimate pooled all-cause and cause-specific crude mortality rates (CMRs) per 100 000 person-years, for specific time periods (first, daily from days 1–14; second, weekly from weeks 3–12; third, weeks 13–52 combined; fourth, weeks 53 and over combined; and fifth, total follow-up) after release, overall and stratified by age, sex, and region. Findings 75 427 deaths were recorded. The all-cause CMR during the first week following release (1612 [95% CI 1048–2287]) was higher than during all other time periods (incidence rate ratio [IRR] compared with week 2: 1·5 [95% CI 1·2–1·8], I2=26·0%, weeks 3–4: 2·0 [1·5–2·6], I2=53·0%, and weeks 9–12: 2·2 [1·6–3·0], I2=70·5%). The highest cause-specific mortality rates during the first week were due to alcohol and other drug poisoning (CMR 657 [95% CI 332–1076]), suicide (135 [36–277]), and cardiovascular disease (71 [16–153]). We observed considerable variation in cause-specific CMRs over time since release and across regions. Pooled all-cause CMRs were similar between males (731 [95% CI 630–839]) and females (660 [560–767]) and were higher in older age groups. Interpretation The markedly elevated rate of death in the first week post-release underscores an urgent need for investment in evidence-based, coordinated transitional healthcare, including treatment for mental illness and substance use disorders to prevent post-release deaths due to suicide and overdose. Temporal variations in rates and causes of death highlight the need for routine monitoring of post-release mortality. Funding Australia's National Health and Medical Research Council.
U2 - 10.1016/S0140-6736(24)00344-1
DO - 10.1016/S0140-6736(24)00344-1
M3 - Article
C2 - 38614112
SN - 0140-6736
VL - 403
SP - 1779
EP - 1788
JO - The Lancet
JF - The Lancet
IS - 10438
ER -