Preventable deaths, including those due to drug overdose (OD), are a significant and growing public health concern in New York City (NYC). The age-adjusted rate of unintentional drug OD death (IOD) in NYC increased 143% between 2010 and 2016, from 8.2 per 100,000 in 2010 to 19.9 per 100,000 in 2016.1 This local trend mirrors national trends in drug OD deaths, which have substantially increased since 2000.2
NYC has a large and long-standing HIV epidemic, with approximately 90,800 people living with HIV in NYC in 2018.3 A total of 14,826 or 11.6% of people diagnosed with HIV/AIDS, reported in NYC and presumed to be living as of the end of 2018 had a history of injection drug use.4 People with HIV (PWH) have an increased risk of drug overdose compared with people without HIV because of a range of clinical, behavioral, and sociostructural and environmental factors.5 Furthermore, PWH who inject drugs have increased mortality rates, including because of drug OD.6
In NYC in 2017, there were 1759 deaths among PWH, with an age-adjusted all-cause death rate of 9.2 per 1000 people living with HIV/AIDS.4 In NYC and elsewhere in the United States, rates of death due to HIV as the primary cause of death are decreasing, and rates of deaths due to non–HIV-related causes are increasing.4,7 Despite the sizeable population of PWH with a history of injection drug use and documented rising rates of OD death citywide, trends in drug OD death among PWH in NYC have not been described. Description of OD mortality among PWH would provide critical data to inform and tailor interventions for preventing overdose among PWH who use drugs, including, for example, dual promotion of antiretroviral therapy and opioid agonist therapy, expanded access to naloxone, and integration of care and services for people who use drugs into HIV primary care.5,8
HIV care-seeking by people living with HIV presents an important opportunity to avert preventable causes of death, including death because of drug OD. Surveillance-based analysis suggests that 87% of people living with HIV in NYC in 2018 received HIV care, and 77% were virally suppressed. In addition, PWH with a history of injection drug use had relatively high rates of linkage to and engagement in HIV care.3
In this analysis, we used data from the NYC HIV surveillance registry to describe and analyze trends in OD deaths among NYC PWH and to examine predeath HIV care engagement among PWH who died of a drug OD during 2007–2017 to identify opportunities for prevention.
The New York City Department of Health and Mental Hygiene (NYC DOHMH) maintains the HIV surveillance registry. The registry contains information on all people diagnosed and reported with HIV infection since 2000 and AIDS since 1981 and is continuously updated with information on HIV-related laboratory tests [HIV diagnostic tests, CD4 T-cell counts, HIV viral load (VL) tests, and HIV genotype tests] for people living with HIV and receiving care in NYC. New York State has had comprehensive, electronic reporting of all HIV-related laboratory tests since 2005, and NYC DOHMH staff investigate all reported positive HIV diagnostic tests to identify and confirm new HIV diagnoses made by NYC providers. Vital status is updated for PWH in the registry through quarterly data linkages with NYC death certificate data from the DOHMH Office of Vital Statistics and annual matches with the National Death Index and Social Security Master File for deaths among NYC PWH that occurred outside NYC.
For this analysis, we selected PWH from the registry with death dates during 2007–2017, who resided in NYC at death, and whose deaths were classified on the death certificate with drug overdose as the underlying cause. We measured the trend over time in the age-adjusted rate of OD deaths among NYC PWH from 2007 to 2017, with rates age-adjusted to the NYC Census 2010 population and per 100,000 mid-year PWH. Furthermore, we used ICD-10 codes from the death certificate to classify PWH as having an accidental drug OD death (AOD) if their underlying cause of death was assigned ICD-10 code X40-X44 or an IOD if their underlying cause of death was assigned ICD-10 code X60-X64. We characterized the entire cohort of decedents by demographic and other factors, including gender, race/ethnicity, age at death, and HIV transmission risk, and compared PWH who died of AOD with those who died of IOD.
To evaluate HIV outcomes in the 12 months before death which we term the “intervenable period,” we used HIV-related laboratory tests (CD4 counts and HIV VLs) reported to the NYC HIV surveillance registry by March 31, 2019, to construct an HIV Mortality Reduction Continuum of Care (HMRCC). The HMRCC is an adaptation of the traditional HIV care continuum to evaluate engagement in HIV care among deceased PWH using HIV surveillance data, with the goal of identifying opportunities to optimize HIV care and implement health care system–based interventions to reduce death from preventable causes, including HIV, among PWH.9
We used NYC HIV surveillance data to analyze outcomes in the HMRCC, including the proportion of patients who were: eligible for analysis (had a date of death >15 months (456 days) after HIV diagnosis to account for the 12-month intervenable period plus a 3-month immediate predeath period, which was excluded from analysis); ever linked to care (had a CD4/HIV VL test reported to surveillance ≥8 days after their HIV diagnosis); retained in care (had ≥2 CD4/VL tests ≥90 days apart in the intervenable period); ever prescribed antiretroviral therapy (calculated as 96.3% based on data from NYC's 2017 Medical Monitoring Project); and virally suppressed (had a last HIV VL during the intervenable period with value <200 copies/mL).
Trends in OD Death Among PWH
From 2007 to 2017, 870 PWH died of either AOD or IOD in NYC (Table 1). Although the age-adjusted rate of OD deaths among PWH declined during the full period, from 58.5 per 100,000 in 2007 to 47.9 per 100,000 in 2017, it increased from 2014 (30.9/100,000) to 2016 (53.3/100,000) and stayed high through 2017 (47.9/100,000). The OD death rate among NYC PWH in 2017 (47.9/100,000) was more than double the OD death rate in NYC overall that year (21.2/100,000). By race/ethnicity, although OD death rates generally declined during 2007–2011 among Black and Latino PWH, rates in these groups generally increased during 2012–2017. OD death rates among White PWH were highly variable during this period, although ultimately declined between 2016 and 2017.
TABLE 1. -
Deaths due to Drug Overdose Among People With HIV, New York City 2007–2017
||Total Overdose Deaths*
||Type of Drug Overdose (OD)†
Age group at death (yr)
HIV transmission risk║
| Men who have sex with men (MSM)
| Injection drug use history (IDU)
| Heterosexual contact
|Age-adjusted OD death rate/100,000 mid-year PLWHA
*Includes people with HIV/AIDS who were living in New York City at death and whose underlying cause of death was drug overdose (OD), as indicated by ICD10 codes X40-44 and X60-64.
†Accidental OD includes deaths with ICD10 codes X40-X64 listed as underlying cause of death and intentional OD includes deaths with ICD10 codes X60-64 listed as underlying cause of death.
‡Transgender men are included with men and transgender women are included with women.
§Other race/ethnicity includes Asian/Pacific Islander, Native American, and multiracial categories.
‖“MSM-IDU” includes men who report sex with men and a history of injection drug use. “Heterosexual contact” includes people who had heterosexual sex with a person they know to be HIV-infected, a person who injects drugs, or a person who has received blood products. For women alone, also includes history of sex work, multiple sex partners, sexually transmitted disease, crack/cocaine use, sex with a bisexual male, probable heterosexual transmission as noted in medical chart, or sex with a male and negative history of injection drug use. “Others/unknown” includes people identified as transgender by self-report, diagnosing provider, or medical chart review with sexual contact reported and negative history of injection drug use; people who received treatment for hemophilia; people who received a transfusion or transplant; children with a nonperinatal transmission risk; and people with an unknown HIV transmission risk.
PLWHA, people living with HIV/AIDS. Data as reported to the NYC DOHMH by March 31, 2019.
Decedents during 2007–2017 were predominantly men (70.8%), Black (38.0%) or Latino/Hispanic (38.7%), aged 40–59 years (73.2%), and people with a history of injection drug use (43.0%), or men who reported sex with men (MSM) (21.7%).
Comparison of PWH With Accidental vs. Intentional OD Death
Of the total OD deaths, 821 (94.4%) deaths were classified as AOD and 49 (5.6%) as IOD (Table 1). AOD decedents were also predominantly men (69.8%), Black (39.5%) and Latino/Hispanic (39.8%), aged 40–59 years (74.3%), and had a history of injection drug use (44.9%). By contrast, IOD decedents were nearly all men (87.8%), mostly White (63.3%), older (24.5% aged ≥60 years), and MSM (61.2%). MSM with AOD death were younger and living in lower-poverty areas of NYC than non-MSM with AOD death (data not shown). MSM with IOD death were mostly White and also living in lower-poverty areas compared with non-MSM with IOD death.
Engagement in HIV Care Before Death Among NYC PWH
A total of 847 PWH with OD death (801 with accidental OD death and 46 with intentional OD death) and whose HIV diagnosis date was >15 months before death were included in the HMRCC analysis. We found that nearly all AOD and IOD decedents were linked to HIV care after diagnosis (98% and 98%, respectively) and over 3-quarters of both AOD and IOD decedents were retained in HIV care during the intervenable period before death (77% and 80%, respectively) (Figs. 1A and B). However, a substantially higher proportion of decedents with IOD were virally suppressed (74%) compared with decedents with AOD (51%). Compared with PWH who died during the same period but of non-OD causes (Fig. 1C), decedents with OD death of both types had similar linkage and retention outcomes in the intervenable period, but higher viral suppression (VS) proportions. Finally, compared with people living with HIV in 2017 (Fig. 1D), decedents with both types of OD death had lower retention in HIV care, and AOD decedents had lower VS.
We used population-level data on deaths and cause of death among people with HIV in NYC to analyze trends in and characteristics of NYC PWH who died of a drug overdose. We found that the rate of drug OD deaths among NYC PWH increased in the last few years and remains high, mirroring citywide and national trends in the general public. Furthermore, the OD death rate among NYC PWH in 2017 was more than double the OD death rate in NYC overall that year. Our analysis also identified important disparities in trends over time in rates of OD death by race/ethnicity among NYC PWH, with recent increases among Black and Latino PWH, and relatively higher rates and steeper increases for Latino PWH compared with other race/ethnic groups. Higher OD death rates for Black and Latino people have been reported in other analyses for NYC and the US as a whole.1,10
Overall, PWH who died of a drug OD during this period appeared to be engaged in HIV care in the year before death. Specifically, nearly half of decedents with drug OD death were virally suppressed before death, suggesting they had sustained interaction with providers and the health care system. This is comparable with the VS proportion among NYC PWH decedents with non-OD death (45%), but substantially lower than the 74% of living NYC PWH in 2017 who were virally suppressed. These findings reveal potential missed opportunities for engaging at-risk individuals in overdose prevention interventions. Furthermore, most drug OD deaths among PWH during our analytic period were classified as accidental. Although the absolute number of deaths due to intentional OD was small, we found important demographic differences between people with accidental OD death versus intentional OD death, including that decedents with accidental OD death were more commonly Black and Latino/Hispanic and people with a history of injection drug use, whereas decedents with intentional OD death were more commonly White and MSM. These and other differences in the demographic profiles of AOD and IOD decedents in NYC warrant further exploration. Although both groups had high rates of engagement in HIV care, decedents with intentional OD death may be managing their HIV more optimally, as evidenced by the higher VS rate. Given the demographic differences between these 2 groups, it is possible, though, that these differences in HIV outcomes reflect demographic and structural factors that lead to inequities in HIV outcomes, for example racism and socioeconomic hardship. Nonetheless, our data suggest the need to provide a range of services and interventions, including related to harm reduction and suicide prevention, to PWH at risk of overdose. Although more research is needed in the area of HIV and aging, the finding that intentional OD was more common in older MSM highlights the need for ongoing emphasis on providing trauma-informed care and mental health services for older adults with HIV, including increased focus on the needs of long-term survivors of HIV/AIDS.
NYC has a robust and well-established system of harm-reduction services for people who use drugs, including PWH. In response to these findings, DOHMH took steps to ensure that contracted Ryan White-funded HIV care providers and funded HIV prevention providers were aware of overdose deaths and how to address them. These efforts included supporting contracted service sites to become equipped to serve as overdose prevention centers, including training staff in mental health first aid, trauma-informed care, and the provision of Narcan for clients and their support networks. In response to the opioid epidemic more broadly, NYC has rolled out a number of new initiatives aimed specifically at reducing the risk of opioid overdose and associated sequelae, including death. For example, the City launched Healing NYC in 2017, which is a comprehensive strategy to reduce opioid overdose deaths by 35% by 2022. Expanding access to treatment for opioid use disorder is a critical strategy for preventing fatal overdose.11 Although coverage remains suboptimal in NYC, efforts are underway to improve provider prescribing and patient access.12,13 PWH should also benefit from citywide enhanced programming and availability of services related to drug OD prevention.
This analysis is subject to several limitations. The use of laboratory tests from surveillance to measure retention in care instead of information on HIV outpatient medical visits could underestimate or overestimate actual HIV care. However, we expect the extent of misclassification of the care status to be minor based on a previous validation study we conducted of laboratory tests reported to surveillance as proxy for HIV medical care.14 In that study, the majority (86%) of patients considered to be “in care” based on laboratory test data in surveillance were confirmed to be receiving clinical care for HIV based on clinical data. Furthermore, this analysis was limited to people living in and/or accessing care in NYC given the data available to NYC HIV surveillance; care received outside NYC was not captured. Misclassification of cause of death on death certificates, including the classification of overdose deaths as accidental or intentional, is possible. However, the NYC Office of the Chief Medical Examiner conducts autopsy and toxicology on all OD deaths to confirm cause, which should limit misclassification. Finally, data on specific drug(s) involved in the overdose were not available for our analysis.
In conclusion, the rate of deaths due to drug overdose is increasing in recent years among people with HIV in NYC, mirroring citywide and national trends in OD deaths, and recent rates among PWH are substantially higher than for the NYC population overall. PWH who died of drug OD were engaged in HIV care before death, suggesting missed opportunities for intervention to reduce overdose and prevent death because of overdose. The HMRCC framework could be applied by other jurisdictions with high HIV morbidity and high rates of overdose and overdose death.
1. Paone D, Nolan ML, Tuazon E, et al. Unintentional Drug Poisoning (Overdose) Deaths in New York City, 2000–2016. New York, NY: New York City Department of Health and Mental Hygiene: Epi Data Brief (89); 2017.
2. Rudd RA, Noah Aleshire JD, Zibbell JE, et al., Increases in drug and opioid overdose deaths—United States, 2000–2014. MMWR. 2016;64;1378–1382.
3. New York City HIV/AIDS Surveillance
Slide Sets. New York, NY: New York City Department of Health and Mental Hygiene, 2018. Available at: https://www1.nyc.gov/site/doh/data/data-sets/epi-surveillance-slide-sets.page
. Accessed December 2, 2019.
4. HIV Epidemiology Program. HIV Surveillance
Annual Report. New York, NY; 2018: New York City Department of Health and Mental Hygiene.
5. Green TC, McGowan SK, Yokell MA, et al. HIV infection and risk of overdose: a systematic review and meta-analysis. AIDS. 2012;26:403–417.
6. Mathers BM, Degenhardt L, Bucello C, et al. Mortality among people who inject drugs: a systematic review and meta-analysis. Bull World Health Organ. 2013;91:102–123.
7. Schwarcz SK, Vu A, Hsu LC, et al. Changes in causes of death among persons with AIDS: San Francisco, California, 1996–2011. AIDS Patient Care STDS. 2014;28:517–523.
8. Tofighi B, Sindhu SS, Chemi C, et al. Perspectives on the HIV continuum of care among adult opioid users in New York City: a qualitative study. Harm Reduct J. 2019;16:58.
9. Braunstein SL, Robbins RS, Daskalakis DC. Missed opportunities: adapting the HIV care continuum to reduce HIV-related deaths. J Acquir Immune Defic Syndr. 2017;76:231–240.
10. Alexander MJ, Kiang MV, Barbieri M. Trends in Black and white opioid mortality in the United States, 1979–2015. Epidemiol 2018;29:707–715.
11. National Academies of Sciences, Engineering, and Medicine. Medications for Opioid Use Disorder Save Lives. Washington, DC: The National Academies Press. 2019.
12. Tofighi B, Isaacs N, Byrnes-Enoch H, et al. Expanding treatment for opioid use disorder in publicly funded primary care clinics: exploratory evaluation of the NYC health + hospitals buprenorphine ECHO program. J Subst Abuse Treat. 2019;106:1–3.
13. Welch AE, Jeffers A, Allen B, et al. Relay: a peer-delivered emergency department-based response to nonfatal opioid overdose. Am J Public Health. 2019;109:1392–1395.
14. Pati R, Robbins RS, Braunstein SL. Validation of retention in HIV care status using the New York city HIV surveillance
registry and clinical care data from a large HIV care center. J Public Health Manag Pract. 2017;23:564–570.