Formerly incarcerated persons (FIPs) have high prevalence of substance use and mental health disorders and are at high risk of opioid overdose death (OOD) after release from incarceration. In a previous North Carolina study, we found that from 2000 to 2015 the OOD rate among FIPs at 2 weeks and 1 year after release was 40 and 10.5 times that in the North Carolina general population, respectively. Because of the dynamic OOD epidemic from commonly prescribed opioids (2011) to heroin (2014) and now to fentanyl and its analogs, it is unclear how the OOD rates have changed among FIPs. This analysis adds 2016–2018 North Carolina incarceration release data to compare OOD rates among FIPs with the general population and examines effect measure modification by race, sex, and age.

METHODS

We conducted a retrospective cohort study among North Carolina FIPs released from incarceration from 2000 to 2018 to estimate postrelease OOD rates relative to the North Carolina general population OOD rates and examined modification by age, sex, and race.

Data Sources

We linked the North Carolina Department of Public Safety’s incarceration release data from January 1, 2000, to December 31, 2018, with North Carolina death records from the same period, using a deterministic algorithm including exact matches of Soundex for last and first names, date of birth, and sex. The North Carolina Department of Public Safety data included dates of prison entry and exit and demographics. The death records included death date and cause documented with International Classification of Diseases, 10th Revision, codes (ICD-10; Geneva, Switzerland: World Health Organization; 1992). We used the National Vital Statistics System’s bridged-race intercensal population estimates from 2000 to 2018 for annual North Carolina demographic and population estimates.

FIPs, aged 18 years or older at release, contributed person-time to OOD rates from their release date until reincarceration, death, or end of study, whichever occurred first. For the general population, each North Carolina resident contributed 1 person-year (PY) per calendar year. Between 2000 and 2018, there were 451 453 releases from North Carolina prisons among 259 861 unique individuals. Those released accrued 2 458 639 PYs. During this time, 17 422 individuals suffered out-of-prison deaths.

We defined OOD using underlying and contributing causes of death with an ICD-10 code of X40‒X44, T40.0 (opium), T40.1 (heroin), T40.2 (other opioids), T40.3 (methadone), and T40.4 (other synthetic narcotics, commonly fentanyl or its analogs). Upon recent Centers for Disease Control and Prevention (CDC) guidance, ICD-10 code T40.6 (unspecified narcotics) was also added. We examined all OODs combined and for specific opioids, including heroin, methadone, prescription opioids, and fentanyl. Covariates included age (18–24, 25–34, 35–44, 45–54, 55–64, and ≥ 65 years), sex (female or male), race (White or non-White), and calendar year of incarceration release among FIPs.

Statistical Analysis

We calculated FIPs’ OOD rates by dividing the number of OODs by the PYs contributed at 2-week, 1-year, and complete follow-up after incarceration release, and for North Carolina general population by dividing total OODs among North Carolina residents by the state population for each calendar year. Complete follow-up indicates all available postrelease person-time.

Using indirect standardization, we calculated standardized mortality ratios (SMRs) and 95% confidence intervals (CIs) at 2-week, 1-year, and complete follow-up after incarceration release, by comparing observed OODs among FIPs with expected OODs, had they had the same age‒race‒sex distribution as the North Carolina general population. We also calculated SMRs and 95% CIs comparing FIPs’ observed versus expected OODs from heroin, commonly prescribed opioids (T40.2+T40.3), and other synthetic narcotics and measured effect measure modification by age, sex, and race. We also present directly standardized rates.

RESULTS

Of the 17 422 out-of-prison deaths among FIPs released between 2000 and 2018 in North Carolina, 2078 (11.1%) were OODs. The OOD rate in the North Carolina general population decreased from 22.8 per 100 000 PYs in 2017 to 20.7 per 100 000 PYs in 2018 (10.1% decline). However, the OOD rate among FIPs increased by 32% from 362 per 100 000 PYs in 2017 to 479 per 100 000 PYs in 2018 (Figure 1), largely attributable to other synthetic narcotics, mostly fentanyl (Table A, available as a supplement to the online version of this article at http://www.ajph.org). As a result, whereas the SMR remained stable from 2014 to 2017, it increased significantly in 2018 (Figure 1).

FIGURE 1

_{Standardized Annual Opioid Overdose Death Rates and Standardized Mortality Ratio Comparing Formerly Incarcerated Persons With North Carolina Residents: 2000‒2018}

_{Note. FIPS = formerly incarcerated persons; NC GP = North Carolina general population; SMR = standardized mortality ratios calculated using indirect standardization for age, sex, race, and calendar year; SMR L95 = lower 95% confidence interval for SMR; SMR U95 = upper 95% confidence interval for SMR.}

The overall age-, sex-, race-, and calendar year‒standardized postrelease OOD rate among FIPs from 2016 to 2018 was 960 per 100 000 PYs (95% CI = 646, 1274) at 2 weeks after release, 359 per 100 000 PYs (95% CI = 314, 404) at 1 year after release, and 302 per 100 000 PYs (95% CI = 268, 335) at complete follow-up after release (Table A). The highest substance-specific OOD mortality rate at 2-week, 1-year, and complete follow-up after release during 2016 to 2018 was attributable to synthetic narcotics, followed by heroin and commonly prescribed opioids (Table A).

Similarly, the 2016–2018 SMRs comparing observed and expected OOD rates from FIPs were highest for synthetic narcotics. Compared to the general population, FIPs were 50.3 times (95% CI = 30.9, 69.6), 20.2 times (95% CI = 17.3, 23.2), and 18.2 times (95% CI = 15.9, 20.5) more likely to die from a synthetic narcotic overdose by 2-week, 1-year, and complete follow-up after release, respectively (Figure A, available as a supplement to the online version of this article at http://www.ajph.org). This is a large increase compared with 2000 to 2015 that is congruent with decreases for heroin and commonly prescribed OODs. Hence, the majority of the 2018 SMR increase is attributable to synthetic narcotics, commonly involving fentanyl.

Overall, compared with general North Carolina population, from 2016 to 2018, FIPs were 46.6 times (95% CI = 31.4, 61.8), 17.4 times (95% CI = 15.2, 19.6), and 14.6 times (95% CI = 13.0, 16.6) as likely to die from OODs by 2-week, 1-year, and complete follow-up after release from incarceration, respectively (Figure A).

While there was no substantial effect measure modification by age, the 1-year postrelease SMR for female FIPs (42.6; 95% CI = 31.7, 53.4) was higher than for male FIPs (14.7; 95% CI = 12.6, 16.8), and the SMR for White FIPs (20.9; 95% CI = 18.0, 23.8) was higher than for non-White FIPs (9.8; 95% CI = 6.9, 12.7).

DISCUSSION

While nationally and in North Carolina the OOD rates declined from 2017 to 2018, OOD rates among FIPs increased by 32%, largely attributable to synthetic narcotics like fentanyl and its analogs. While the risk of OODs from heroin and commonly prescribed opioids among FIPs decreased during 2016 to 2018, the risk of OOD from synthetic narcotics increased substantively. Compared with the 2000–2015 estimates, the 1-year and complete follow-up SMRs for OODs comparing FIPs to North Carolina residents have increased substantially, even after excluding unspecified narcotics. Although there is effect measure modification by sex and race, all FIP subgroup SMRs are 10 times or more higher than the general population.

The North Carolina correctional system began successfully piloting programs to link FIPs to medications for opioid use disorders in community correction in 2017. In addition, the North Carolina opioid action plan has increased naloxone access for everyone including community corrections officers. Despite these interventions, FIPs continue to be vulnerable to OODs, likely because our data predate these interventions; the 3-year period between 2016 and 2018 coincides with the 2017 peak of the synthetic opioid overdose epidemic in the United States, and FIPs face barriers to medication for opioid use disorders initiation and retention upon re-entry including lack of health insurance, housing, and employment, and stigma around drug use and incarceration. Furthermore, the North Carolina prison-based medication for opioid use disorders implementation has experienced COVID-19‒ related delays.

A potential limitation is the lack of death data for FIPs who move out of state. However, most incarcerated people are released on probation or parole, so the 1-year postrelease OOD rates should not be underestimated. A CDC health alert from December 2020 shows an increase in OOD beyond the 2017 highs. Future research will need to examine how the drug overdose epidemic under the shadow of the COVID-19 pandemic affects the health of FIPs.

Individuals previously incarcerated exhibit elevated rates of substance use and mental health disorders, along with a significant risk of opioid overdose death (OOD) following release. Earlier research in North Carolina indicated that between 2000 and 2015, OOD rates for formerly incarcerated persons (FIPs) at two weeks and one year post-release were substantially higher than the general population. Given the evolving nature of the OOD epidemic, this study sought to update OOD rates among FIPs by incorporating data from 2016–2018, comparing them to the general North Carolina population, and analyzing variations by race, sex, and age.

Methods

A retrospective cohort study was conducted involving formerly incarcerated persons released from North Carolina prisons between 2000 and 2018. Data from the North Carolina Department of Public Safety on incarceration releases were linked with state death records for the same period. This linkage used a deterministic algorithm matching names, birth dates, and sex. Demographic and population estimates for the general North Carolina population were obtained from the National Vital Statistics System. Opioid overdose deaths were identified using specific International Classification of Diseases, 10th Revision (ICD-10) codes related to opioid poisoning. The study population included FIPs aged 18 or older at release, with person-time tracked until reincarceration, death, or study end. The general population contributed one person-year per resident annually. This study included 451,453 releases from North Carolina prisons involving 259,861 unique individuals, who collectively accumulated over 2.4 million person-years of follow-up. Among these, 17,422 deaths occurred outside of prison.

Statistical Analysis

Opioid overdose death rates for formerly incarcerated persons were calculated by dividing the number of OODs by person-years contributed at two-week, one-year, and complete follow-up periods post-release. For the general North Carolina population, OOD rates were derived by dividing total OODs by the state population for each calendar year. Standardized Mortality Ratios (SMRs) with 95% confidence intervals were computed through indirect standardization. These SMRs compared observed OODs among formerly incarcerated persons to the expected number of OODs if this population had the same age, race, and sex distribution as the general North Carolina population. SMRs were also determined for specific opioid types, including heroin, commonly prescribed opioids, and synthetic narcotics, to assess variations by age, sex, and race.

Results

Among 17,422 out-of-prison deaths for formerly incarcerated persons between 2000 and 2018, 2,078 (11.1%) were opioid overdose deaths (OODs). While the OOD rate in the general North Carolina population declined from 2017 to 2018, the rate among formerly incarcerated persons increased by 32% in the same period, primarily due to synthetic narcotics such as fentanyl. From 2016 to 2018, the standardized post-release OOD rate among formerly incarcerated persons was 960 per 100,000 person-years at two weeks post-release and 359 per 100,000 person-years at one year post-release. During this period, the risk of OOD from synthetic narcotics was notably higher, with formerly incarcerated persons 50.3 times more likely to die from a synthetic narcotic overdose within two weeks of release compared to the general population. Overall, from 2016 to 2018, formerly incarcerated persons were 46.6 times more likely to die from an OOD within two weeks of release than the general population. Although age did not significantly modify this effect, the one-year post-release OOD risk was higher for female formerly incarcerated persons (SMR 42.6) than for male formerly incarcerated persons (SMR 14.7), and higher for White formerly incarcerated persons (SMR 20.9) than for non-White formerly incarcerated persons (SMR 9.8).

Discussion

Despite a national and state decline in opioid overdose death (OOD) rates between 2017 and 2018, rates among formerly incarcerated persons (FIPs) notably increased, primarily driven by synthetic narcotics such as fentanyl. While interventions like linking FIPs to opioid use disorder medications and increasing naloxone access have been initiated by the North Carolina correctional system, the period of this study (2016-2018) largely predates the full impact of these programs and coincided with the peak of the synthetic opioid epidemic. Formerly incarcerated persons also face significant barriers to care upon re-entry, including lack of health insurance, housing, employment, and stigma. A limitation of the study is the absence of out-of-state death data for FIPs; however, the reliance on probation or parole for most releases suggests that one-year post-release OOD rates are likely not significantly underestimated. Future research should investigate the continued evolution of the drug overdose epidemic and its impact on the health of formerly incarcerated persons, particularly in the context of the COVID-19 pandemic.

Individuals released from incarceration often experience high rates of substance use and mental health disorders, placing them at significant risk for opioid overdose death (OOD) after release. Earlier research in North Carolina showed that between 2000 and 2015, the OOD rate for formerly incarcerated persons (FIPs) was 40 times higher than the general population within two weeks of release, and 10.5 times higher after one year. The nature of the opioid overdose epidemic has shifted over time, moving from commonly prescribed opioids around 2011, to heroin by 2014, and more recently to fentanyl and its related substances. This evolution makes it necessary to re-evaluate how OOD rates among FIPs have changed. The current study extends this analysis by including North Carolina incarceration release data from 2016 to 2018. It compares OOD rates in FIPs with the general population and investigates how these rates vary by race, sex, and age.

Methods

A retrospective cohort study was conducted focusing on formerly incarcerated persons (FIPs) released from North Carolina correctional facilities between 2000 and 2018. The study aimed to calculate opioid overdose death (OOD) rates after release and compare them to OOD rates in the general North Carolina population. It also investigated how these rates were influenced by age, sex, and race.

Data Sources

Information for the study was gathered by linking North Carolina Department of Public Safety records, which covered incarceration releases from January 1, 2000, to December 31, 2018, with North Carolina death records from the same timeframe. This linkage was performed using a precise matching system that included exact matches for names, birth dates, and sex. The Department of Public Safety data provided details on prison entry and exit dates, along with demographics. Death records included the date and cause of death, categorized using codes from the International Classification of Diseases, 10th Revision (ICD-10). Annual population estimates and demographic data for North Carolina residents were obtained from the National Vital Statistics System’s bridged-race intercensal population estimates for the period 2000 to 2018.

Formerly incarcerated persons (FIPs) who were 18 years or older at the time of their release contributed "person-time" to the OOD rate calculations. This person-time was counted from their release date until the earliest of three events: reincarceration, death, or the study's end. For the general North Carolina population, each resident contributed one person-year (PY) for each calendar year. During the 2000 to 2018 period, 451,453 releases occurred from North Carolina prisons, involving 259,861 distinct individuals. These released individuals accumulated a total of 2,458,639 person-years, during which 17,422 deaths occurred outside of prison.

Opioid overdose deaths (OODs) were identified using specific ICD-10 codes related to underlying and contributing causes of death. These codes included X40-X44, and T40.0 (opium), T40.1 (heroin), T40.2 (other opioids), T40.3 (methadone), and T40.4 (other synthetic narcotics, often fentanyl or its analogs). Following updated guidance from the Centers for Disease Control and Prevention (CDC), ICD-10 code T40.6 (unspecified narcotics) was also included. The study analyzed all OODs collectively, as well as those linked to specific opioid types: heroin, methadone, prescription opioids, and fentanyl. Factors considered in the analysis included age groups (18–24, 25–34, 35–44, 45–54, 55–64, and 65 years and older), sex (female or male), race (White or non-White), and the calendar year of release from incarceration for FIPs.

Statistical Analysis

Opioid overdose death (OOD) rates for formerly incarcerated persons (FIPs) were determined by dividing the number of OODs by the total person-years contributed by FIPs at specific intervals: 2 weeks, 1 year, and complete follow-up after release. "Complete follow-up" refers to all available person-time after release. For the general North Carolina population, OOD rates were calculated by dividing the total OODs among residents by the state's population for each calendar year. Standardized mortality ratios (SMRs) and their 95% confidence intervals (CIs) were also computed for these intervals, using indirect standardization to compare observed OODs among FIPs to the expected number if FIPs had the same age, race, and sex distribution as the general population. SMRs were similarly calculated for specific substances such as heroin, commonly prescribed opioids, and other synthetic narcotics, with an assessment of how age, sex, and race modified these effects. Directly standardized rates are also presented.

Results

Out of 17,422 deaths among formerly incarcerated persons (FIPs) in North Carolina between 2000 and 2018, 2,078 (11.1%) were due to opioid overdose. While the opioid overdose death (OOD) rate in the North Carolina general population declined by 10.1% from 22.8 per 100,000 person-years (PYs) in 2017 to 20.7 per 100,000 PYs in 2018, the OOD rate among FIPs showed an increase. It rose by 32% from 362 per 100,000 PYs in 2017 to 479 per 100,000 PYs in 2018. This increase was primarily linked to other synthetic narcotics, predominantly fentanyl. Consequently, although the standardized mortality ratio (SMR) remained consistent from 2014 to 2017, it rose significantly in 2018.

For the period 2016 to 2018, the overall post-release OOD rate for FIPs, when standardized for age, sex, race, and calendar year, was 960 per 100,000 PYs (95% CI = 646, 1274) within two weeks of release. This rate was 359 per 100,000 PYs (95% CI = 314, 404) at one year after release, and 302 per 100,000 PYs (95% CI = 268, 335) for the complete follow-up period. During this same period (2016-2018), the highest substance-specific OOD mortality rate at all follow-up intervals (2 weeks, 1 year, and complete follow-up) was due to synthetic narcotics, followed by heroin and commonly prescribed opioids.

Standardized mortality ratios (SMRs) from 2016 to 2018, which compare observed and expected OOD rates for FIPs, were also highest for synthetic narcotics. FIPs were 50.3 times (95% CI = 30.9, 69.6) more likely to die from a synthetic narcotic overdose within two weeks of release, 20.2 times (95% CI = 17.3, 23.2) more likely after one year, and 18.2 times (95% CI = 15.9, 20.5) more likely over the complete follow-up period, when compared to the general population. This represented a substantial increase compared to the 2000-2015 period, coinciding with a decrease in OODs from heroin and commonly prescribed opioids. Therefore, the significant rise in the SMR in 2018 was largely due to synthetic narcotics, particularly fentanyl.

Overall, from 2016 to 2018, FIPs were significantly more likely to die from an opioid overdose compared to the general North Carolina population. This likelihood was 46.6 times higher (95% CI = 31.4, 61.8) within two weeks of release, 17.4 times higher (95% CI = 15.2, 19.6) after one year, and 14.6 times higher (95% CI = 13.0, 16.6) over the complete follow-up period following incarceration release.

While age did not significantly alter the effect measure, the one-year post-release SMR was notably higher for female FIPs (42.6; 95% CI = 31.7, 53.4) compared to male FIPs (14.7; 95% CI = 12.6, 16.8). Similarly, the SMR for White FIPs (20.9; 95% CI = 18.0, 23.8) was higher than for non-White FIPs (9.8; 95% CI = 6.9, 12.7).

Discussion

Despite a national and North Carolina-wide decline in opioid overdose death (OOD) rates from 2017 to 2018, OOD rates among formerly incarcerated persons (FIPs) rose by 32%. This increase was largely driven by synthetic narcotics, such as fentanyl and its related substances. While the risk of OODs from heroin and commonly prescribed opioids for FIPs decreased between 2016 and 2018, the risk from synthetic narcotics significantly increased. When compared to findings from 2000-2015, the standardized mortality ratios (SMRs) for OODs among FIPs relative to North Carolina residents have substantially increased for both 1-year and complete follow-up periods, even when excluding unspecified narcotics. Although sex and race influenced these measures, all FIP subgroups still demonstrated SMRs that were 10 times or more higher than the general population.

The North Carolina correctional system initiated successful pilot programs in 2017 to connect FIPs with medications for opioid use disorders within community corrections. Furthermore, the state's opioid action plan has improved access to naloxone for various groups, including community corrections officers. Despite these efforts, FIPs remain highly susceptible to OODs. This vulnerability is likely due to several factors: the study's data largely precedes the full impact of these interventions, the 2016-2018 study period aligns with the 2017 peak of the synthetic opioid overdose epidemic in the U.S., and FIPs encounter significant obstacles to starting and maintaining medication for opioid use disorders upon re-entry. These barriers include a lack of health insurance, stable housing, and employment, as well as societal stigma associated with drug use and incarceration. Additionally, the implementation of prison-based medication for opioid use disorders in North Carolina has experienced delays related to the COVID-19 pandemic.

A potential limitation of the study is the absence of death data for formerly incarcerated persons who relocate outside of North Carolina. However, given that most individuals released from incarceration are under probation or parole, the one-year post-release OOD rates are unlikely to be significantly underestimated. Recent health alerts from the CDC in December 2020 indicate that OODs have risen even beyond their 2017 peak levels. Future research will be essential to understand how the ongoing drug overdose epidemic, particularly in the context of the COVID-19 pandemic, impacts the health outcomes of FIPs.

Individuals released from prison often have high rates of substance use and mental health problems. They also face a high risk of dying from an opioid overdose shortly after leaving prison. Earlier research in North Carolina showed that from 2000 to 2015, the opioid overdose death rate for formerly incarcerated persons (FIPs) was much higher than for the general population: 40 times higher two weeks after release and 10.5 times higher after one year. Because the opioid overdose crisis has changed over time, from prescribed opioids to heroin and now to fentanyl, it was unclear how these death rates for FIPs might have changed. This study added North Carolina release data from 2016 to 2018 to compare opioid overdose death rates between FIPs and the general population, also looking at differences by race, sex, and age.

Methods

This retrospective study examined opioid overdose death (OOD) rates among formerly incarcerated persons (FIPs) released from North Carolina prisons between 2000 and 2018, comparing them to the general North Carolina population. Researchers linked data from the North Carolina Department of Public Safety with North Carolina death records, using specific medical codes to identify OODs from various opioids like heroin and fentanyl. Population estimates for North Carolina were also used. Individuals aged 18 or older were tracked from their release until reincarceration, death, or the study's end. OOD rates for FIPs were calculated based on the time at risk after release, such as within 2 weeks or 1 year. For the general population, rates were based on total deaths per state population each year. To fairly compare these rates, standardized mortality ratios (SMRs) were calculated, showing how much more likely FIPs were to die from an OOD compared to the general population, adjusted for age, sex, and race. Over 451,000 prison releases were included, with 17,422 FIPs experiencing out-of-prison deaths during the study period.

Results

Between 2000 and 2018 in North Carolina, 2,078 (11.1%) of the 17,422 deaths among formerly incarcerated persons (FIPs) after release were due to opioid overdose. From 2017 to 2018, the opioid overdose death (OOD) rate in the general North Carolina population decreased, but the OOD rate among FIPs increased by 32%, largely due to synthetic narcotics, such as fentanyl. As a result, the standardized mortality ratio (SMR), which compares FIPs' death risk to the general population, significantly rose in 2018. From 2016 to 2018, FIPs had a very high OOD rate of 960 per 100,000 person-years within two weeks after release, decreasing to 359 per 100,000 after one year and 302 per 100,000 for the complete follow-up period. Synthetic narcotics were the main cause of these deaths. Compared to the general population, FIPs were 50.3 times more likely to die from a synthetic narcotic overdose within two weeks of release, and 18.2 times more likely over the full follow-up. Overall, FIPs were 46.6 times more likely to die from any opioid overdose within two weeks of release, and 14.6 times more likely over the complete follow-up period. While age did not show a major difference, female FIPs faced a higher overdose death risk than male FIPs (42.6 vs. 14.7 times higher than the general population at one year post-release), and White FIPs had a higher risk than non-White FIPs (20.9 vs. 9.8 times higher).

Discussion

This study found that while opioid overdose death (OOD) rates for the general population in North Carolina decreased, rates for formerly incarcerated persons (FIPs) increased significantly, driven mostly by synthetic narcotics like fentanyl. Even though the risk of overdose from heroin and prescribed opioids went down for FIPs, the overall risk from synthetic narcotics rose sharply. The higher risk of death for FIPs compared to the general population has grown considerably since earlier studies, and even with differences based on sex and race, all FIP subgroups still face a risk at least 10 times higher than the general population.

Even with programs like increased access to overdose reversal medication (naloxone) and efforts to connect FIPs to treatment for opioid use disorder, this vulnerable group remains at high risk. This is likely because the study data mostly covers a period before these interventions fully took effect, coinciding with the peak of the synthetic opioid epidemic. FIPs also face many barriers upon release, such as lack of health insurance, stable housing, employment, and the stigma associated with drug use and incarceration, all of which hinder their access to and continuation of treatment. A limitation of the study is that it could not track FIPs who moved out of state. Future research should explore how the ongoing drug overdose crisis, especially with the impact of the COVID-19 pandemic, continues to affect the health of formerly incarcerated individuals.

Overview

People who have been in prison often have problems with drugs and mental health. They are also much more likely to die from an opioid overdose after they leave prison. A past study in North Carolina showed that in the years 2000 to 2015, these individuals were 40 times more likely to die from an overdose within two weeks of release, and 10.5 times more likely within a year, compared to other people in North Carolina. Since drug overdose trends have changed over time (from pain pills to heroin and now fentanyl), this new study looked at what happened from 2016 to 2018. It compared overdose deaths among people released from prison to others in North Carolina and looked at differences based on a person's race, sex, and age.

Methods

Researchers looked at information from people released from North Carolina prisons between 2000 and 2018. They matched these records with North Carolina death records to see who died from an opioid overdose. The study included people aged 18 or older when they left prison. Researchers checked how many of these individuals died from an overdose at 2 weeks, 1 year, and over a longer time after release. They compared these numbers to the general population in North Carolina. They used special math to make sure the comparisons were fair, considering things like age, sex, and race. Opioid overdose deaths were identified using specific codes from medical records.

Results

Out of all the deaths that happened outside of prison for these individuals, about 1 in 10 were due to opioid overdose. From 2017 to 2018, the number of opioid overdose deaths for people in North Carolina overall went down. But for people who had been in prison, the number of overdose deaths went up by 32% in the same period. Most of this increase was because of drugs like fentanyl. People released from prison were much more likely to die from an opioid overdose compared to the general population. For example, at 2 weeks after release, they were about 47 times more likely to die from an overdose, and at 1 year, about 17 times more likely.

When looking at specific types of opioids, the risk of overdose from fentanyl and similar drugs grew a lot. For these drugs, people released from prison were about 50 times more likely to die within two weeks of release, compared to the general public. While the risk of overdose from heroin and older prescription opioids went down, the risk from fentanyl greatly increased. The study also found that women who had been in prison faced a much higher risk of overdose death than men. White individuals who had been in prison also faced a higher risk than non-White individuals.

Discussion

Even though opioid overdose deaths went down for most people in North Carolina from 2017 to 2018, they went up for people released from prison. This was mostly due to fentanyl. The risk of overdose from fentanyl was much higher than from heroin or older pain pills. Overall, people released from prison are still at a much higher risk of dying from an overdose than other people, no matter their age, sex, or race. North Carolina has started programs to help people use medicine for opioid addiction and to get naloxone (a drug that can stop an overdose). However, these efforts might not fully show up in this study's data yet. People leaving prison also face many challenges like not having health insurance, a place to live, or a job, and also facing judgment from others, which can make it hard to get help. Future studies will need to look at how the COVID-19 pandemic has also affected this problem.