Mortality among people who inject drugs – the interwoven roles of fentanyl and HIV: a community-based cohort study

Maris Salekešin; Sigrid Vorobjov; Don Des Jarlais; Anneli Uusküla

doi:10.1093/eurpub/ckad204

. 2023 Dec 1;34(2):329–334. doi: 10.1093/eurpub/ckad204

Mortality among people who inject drugs – the interwoven roles of fentanyl and HIV: a community-based cohort study

Maris Salekešin ^1,^2,^✉, Sigrid Vorobjov ³, Don Des Jarlais ⁴, Anneli Uusküla ⁵

PMCID: PMC10990538 PMID: 38041408

Abstract

Background

Excess all-cause mortality is a key indicator for assessing direct and indirect consequences of injection drug use and data are warranted to delineate sub-populations within people who inject drugs at higher risk of death. Our aim was to examine mortality and factors associated with mortality among people who inject drugs in Estonia.

Methods

Retrospective cohort study using data from people who inject drugs recruited in the community with linkage to death records. Standardized mortality ratios were used to compare the cohort mortality to the general population and potential predictors of death were examined through survival analysis (Cox regression). The cohort include a total of 1399 people who inject drugs recruited for cross-sectional surveys using respondent driven sampling between 2013 and 2018 in Estonia. A cohort with follow-up through 2019 was formed with linkage to national causes of death registry.

Results

Among 1399 participants with 4684 person-years of follow-up, 10% were deceased by 2019. The all-cause mortality rate in the cohort was 28.9 per 1000 person-years (95% confidence interval 25.3–35.3). Being HIV positive, injecting mainly opioids (fentanyl), living in the capital region and the main source of income other than work were associated with greater mortality risk.

Conclusions

While low-threshold services have been available for a long time for people who inject drugs, there is still a need to widen the availability and integration of services, particularly the integration of HIV and opioid treatment.

Introduction

Up-to-date data are needed on cause-specific mortality and identification of risk factors among people who use drugs to assess the effect of, and tailor prevention, care and harm reduction efforts for reducing morbidity and mortality.^1–3 Data on mortality and related factors from countries in Europe that witnessed explosive HIV and HCV epidemics at the beginning of the 2000s are scant. The most recent systematic review on this topic² located three cohorts from Central Europe and no studies were found from Eastern Europe.²

North America’s opioid crisis underscores how changes in opioid availability and use can profoundly impact public health. There, potent illicitly produced fentanyl derivatives have largely replaced prescription opioids and heroin, driving a widespread epidemic of opioid-related deaths.⁴

In Estonia, studies conducted among people who inject drugs (PWID), highlight the strong connection between injection fentanyl use,^5–7 overdose risk⁸^,⁹ and HIV in this population,¹⁰ but the cumulative impact remains unquantified. The widespread fentanyl injection in Estonia, which has resulted in a two-decade-long epidemic of high HIV seroprevalence¹⁰ and significant overdose mortality,⁸^,⁹ predates the situation in North America by two decades.^5–7 This makes it a unique and unprecedented occurrence globally.

Methods

We report findings from a retrospective cohort study on all-cause mortality, causes of death and factors associated with mortality among community-recruited people who inject drugs in Estonia.

Participants

People who inject drugs were recruited to the community-based cross-sectional surveys in the capital, Tallinn (2013, 2016, 2017, 2018); and in the north-eastern cities of Kohtla-Järve (2016) and Narva (2018). Participants were defined as current injection drug users if they had injected drugs within the past 2 months.

Data collection

These surveys used respondent driven sampling for subject recruitment. Potential participants were eligible to be included in the study if they were 18 years of age or older, were Estonian or Russian language speakers, reported having injected in the previous 2 months, and were able and willing to provide informed consent.

In all studies, an interviewer-administered structured questionnaire based on the World Health Organization (WHO) Drug Injecting Study Phase II survey (version 2b (rev.2))¹¹ that elicited information on respondent demographics, drug use (including age at initiation of injecting), and risk behaviours was used for data collection. Venous blood was tested for the presence of HIV antibodies using commercially available test kits [ADVIA Centaur CHIV Ag/Ab Combo (Siemens Healthcare Diagnostics, Inc., Erlangen, Germany)]. Participants received pre- and post-test HIV counselling. Data on the Estonian personal identification code (PIC) was collected. In Estonia, unique 11-digit PIC’s are assigned to all residents at birth or at the time of immigration.

Variables

Demographic data for age and sex were collected. Variables for analysis were selected from those used in (mortality) studies of people who inject drugs which appear to be prognostic indicators. There were indicators concerning: educational attainment (years of education); main source of income (work versus other, including social benefits, theft, etc.); prison experience; methadone agonist therapy (receiving treatment or not); injecting behaviour [main drug, duration of injecting, age and time of first injection, overdose (non-fatal overdose in the last 12 months)]; HIV serostatus (positive versus negative); and place of residence (capital region versus North-East Estonia).

Outcome

The primary outcome—survival time—was time since the first injection to death. Follow-up time (person-years) was calculated from the year of participating in the source studies to the date of death or at censoring if they remained alive until 4 April 2020.

Dates and causes of death were obtained from the Estonian Causes of Death Registry (ECD), a population-based registry that covers the entire country. ECD captures all deaths registered in Estonia and information collected includes a diagnostic code (ICD-10) for the underlying cause of death and the nature of injury. The registry conforms to the quality assurance criteria of the statistical office of the European Union (EU), EUROSTAT.

Causes of death among people who inject drugs were grouped into ICD-10 code combinations recommended by Santo et al.¹² Based on ICD 10 codes, causes of death included: HIV-related diseases (B20-B24), drug use related deaths (X40-X44, X60-X64), liver diseases (K70-K77), accidents (V01-V99, W00-W19, W65, W78, Y10-Y34), suicide (X60-X84), other. Due to few episodes, other diagnoses are grouped together (under ‘other’) (C00-C97, I10-I15, I30-I52, J09-J18, N10-N16, R95-R99). PIC, as single unique identifier, was used to link source survey data with mortality data.

Statistical analyses

For baseline characteristics, means were calculated for continuous variables and percentages for categorical variables using the data collected at the recruitment into respective cross-sectional studies.

We provide three mortality measures: crude and standardized mortality rates per 1000 years for all causes combined, and relative risk of all-cause mortality as measured by hazard ratios (HRs). We used all-cause crude mortality rates to determine absolute risk of mortality and all-cause age (and sex) standardized mortality ratios to compare the cohort mortality to the general population. Mortality was calculated using person-time methods. To calculate crude mortality rates (CMR), we summed the number of deaths by category and calculated a rate per 1000 person years.

To compare the cohort mortality to the general population, we calculated standardized mortality ratios (SMR) using 10-year age bands stratified by gender. Population data were obtained from Statistics Estonia, the Estonian government agency responsible for producing official statistics regarding the country.

Summary statistics were used to describe people who inject drugs. Survival curves of the cohort were constructed using the Kaplan–Meier method for left-truncated data. We used the normal log-rank test to compare survival between groups of people who inject drugs (i.e. by HIV status, main drug injected, region). The proportional hazard assumption was evaluated by Schoenfeld residuals.

We used Cox regression to determine predictors of all-cause mortality. Factors having a significant univariate Cox proportional hazard regression test at P < 0.20¹³^,¹⁴ were selected for the multivariable analysis. In multivariable Cox proportional hazard regression analysis, age was used as a time scale (to account for left-truncation in the data). One of the predictor variables (HIV status) displayed evidence of violation of the proportional hazards assumption (having a time-varying effect). Therefore, the multivariable model was stratified by this variable (HIV status).¹⁵ The final fitted multivariable model was used to identify independent predictors of survival and to estimate HR with 95% confidence interval (CI). Variables were considered independent predictors of survival when P < 0.05.

Given that stratification by a non-proportional variable precludes estimation of its strength and its test within the Cox model, the follow-up time was spilt (and fitted additional Cox models for two time-periods). Injecting 15≥ years was selected as a cut-off value (based on the Kaplan–Meier survival analyses) for periods. Fitted multivariable models (not stratified for HIV) were used to identify independent predictors of survival and to estimate HRs (95% CI) for time-to-death among people injecting for up to 15 years and for those injecting for 15 years or longer. The assumptions of the proportional hazards were tested and were not violated.

All analyses were conducted using Stata, version 14.2.

Results

Cohort overview

A compilation of the final sample for this analysis is presented in Supplementary figure. From July 2013 to April 2019, a total of 1673 people who inject drugs participated in the source studies. We excluded data on 274 participants for the following reasons: current status as a person who injects drugs not confirmed (n = 19); data on personal identifier (national identification code) missing (n = 11); and duplicate entries (participation in more than one source study) (n = 244). If there were duplicate entries, we kept the first occurrence. The resulting cohort included 1399 participants.

Data on sociodemographic and behavioural characteristics are shown in Supplementary annex 1. The median follow-up duration per study subject was 3.5 years (range 0.01–9.5 years) and a total of 4686 person-years. The median age at entry in the cohort was 34 years (range 18–61 years) and 75% (n = 1045) were men. During the follow-up period, 140 (10%) individuals in the cohort had died (n = 114 men and n = 26 women), and the mean age at death was 35 years (range 20–57 years).

Crude mortality rates and SMRs

Observed all-cause mortality rate in this cohort was 28.9 per 1000 person-years (95% CI 25.3–35.3). Among people who inject drugs, crude mortality was higher among HIV infected (versus uninfected: CMR 37.4 per 1000 person-years, 95% CI 30.7–45.6, and 20.1, 95% CI 14.8–27.3, respectively) and among those injecting opioids (CMR 36.5, 95% CI 30.14–44.11) versus injecting other than opioids (CMR 19.1, 95% CI 13.7–26.8).

Mortality among people who inject drugs significantly exceeded that of the Estonian general population (SMR 11.87 95% CI 10.06–14.01) both for men (SMR 10.55 95% CI 8.78–12.70) and for women (SMR 26.31 95% CI 17.91–38.64). Excess mortality was particularly prominent in the younger age group (aged 20–29: SMR 34.67 95% CI 25.43–47.62) and decreased somewhat with age (aged 30–39: SMR 15.14 95% CI 11.89–19.27; aged: 40>: 5.46 95% CI 3.90–7.64).

Survival analysis

The univariable analyses (Supplementary annex 1) showed that being HIV positive was associated with the highest risk of death. Injecting opioids, living in the capital region, age and (calendar) time of first injection, prison experience and reporting other than paid work as the main source of income were associated with higher mortality risk (Supplementary annex 1). Being a female was associated with a lower risk of death.

The median survival time for people who inject drugs was 23.5 years (95% CI 16.2–27.5) (since the first injection). The probability of surviving at 5, 10, 20 and 30 years from the first injection was 81% (95% CI 64.1–90.1), 73% (95% CI 58.7–83.3), 56% (95% CI 45.1–65.0) and 39% (95% CI 30.1–47.2) for the members of the cohort.

Figure 1 shows the survival curves by HIV status, main drug injected and region (variables associated with all-cause mortality at P < 0.05 in the univariable Cox proportional hazard model). The analyses revealed significantly shorter median survival time for HIV-positive (24 vs 32 years, P = 0.0009) than HIV-negative, injecting mainly opioids (17 vs 31 years, P = 0.0015) and living in the capital region (18 vs 31 years, P = 0.0296).

Kaplan–Meier survival curves by HIV status, (A) main drug injected; (B) and region; (C) among people currently injecting drugs in Estonia

Factors associated with mortality among current people who inject drugs

Factors in the multivariable analysis independently associated with mortality are presented in figure 2.

Forest plot of hazard ratios of variables associated with mortality risk

Living in the capital region (HR = 1.68; 95% CI 1.10–2.58), opioid as a main drug injected (HR = 1.60; 95% CI 1.02–2.54) and main source of income other than permanent or temporary job with a contract (HR = 1.53; 95% CI 1.03–2.29) were associated with greater mortality risk.

Subgroup analyses

HIV status, as one of the predictor variables, had a time-varying effect according to the Kaplan-Meier curve (see figure 1A); therefore, we fitted another Cox model for two time-periods: injecting <15 and ≥15 years (table 1). Sub-group analyses of those injecting for 15 years and less and opioids as the main drug injected was the only variable associated with grater mortality risk (HR = 3.2; 95% CI 1.21–8.36; P = 0.02). For those injecting for 15 years or longer, being HIV-positive had greater mortality risk among those who have injected for less than 15 years (HR = 3.19; 95% CI 1.76–5.77; P < 0.00). Women who have injected for 15 years and less had smaller mortality risk.

Table 1.

Causes of death and factors associated with all-cause death, multivariable stratified (by injecting time) analysis

Variable	Those injecting <15 aHR (95% CI)	Those injecting ≥15 years, aHR (95% CI)
Age	1.07 (1.00–1.13)	1.02 (0.96–1.08)
Sex (female)	0.38 (0.14–0.98)	0.73 (0.40–1.32)
HIV (positive)	1.51 (0.67–3.39)	3.19 (1.76–5.77)
Age at first injection (20+)	0.59 (0.22–1.55)	1.60 (0.84–3.05)
Main drug injected (opioid)	3.19 (1.24–8.22)	1.35 (0.80–2.24)
Region (capital)	1.89 (0.72–4.96)	1.63 (1.00–2.27)
Ever in prison (yes)	0.92 (0.41–2.15)	1.23 (0.72–2.09)

Open in a new tab

Antiretroviral therapy (ART) status information was available for 76% (n = 589) of HIV-positive study participants. The subgroup analysis comparing HIV-positive study participants who received ART and those who did not receive ART yielded no difference in mortality (HR 1.51; 95% CI 0.65–3.51; P = 0.34). We also fitted other models to see the time-varying effect on HIV-positive study participants who received ART. The analyses did not show any differences in mortality among people who had shorter (<15 years) injecting history and received ART compared to those who received ART and had a longer (≥15 years) injecting history.

Causes of death among people who inject drugs

Analyses on causes of death showed that there were more drug use-related deaths among those who had injected for less than fifteen years (table 2). Also, there were differences in other ill-defined and unspecified causes of mortality among those who injected ≥15 years (n = 21; 24% alcohol poisoning deaths; 20% cancer; 20% cardiovascular deaths; 36% other).

Table 2.

Causes of death stratified by injecting time

	Among PWID injecting <15 years		Among PWID injecting ≥15 years		P-value^a
Causes of death	n	%	n	%	P-value^a
HIV-related deaths	3	7.3	19	19.6	0.081
Drug use-related deaths	28	68.3	40	41.2	0.005
Liver diseases	1	2.4	7	7.2	0.436
Accidents	3	7.3	8	8.2	1.0
Suicide	4	9.8	2	2.1	0.064
Other	2	4.9	21	21.7	0.022

Open in a new tab

Fisher’s exact test.

According to analysis stratified by the main drug injected, there were more drug use-related deaths (opioid users 56% vs 27% among non-opioid users, P = 0.005) among opioid users and deaths caused by liver diseases (opioid users 3% vs 15% among non-opioid users, P = 0.019) among those injecting other than opioids.

Discussion

In this study, we show high risk of excess mortality and observed notable risk factors for all-cause mortality among people who inject drugs in Estonia. To the best of our knowledge, this is the first study investigating the risk of dying among people who inject drugs in countries in Europe that are witnessing explosive injection drug use, illicit fentanyl use and related HIV epidemics since the turn of the century.

The crude all-cause mortality reported in this study (28.9 per 1000 person-years) was almost two times higher than previous estimates involving people who inject drugs from Western Europe,³^,⁴^,^16–18 North America³^,⁴^,¹⁹ and comparable to that reported among people who inject drugs using opioids in a recent systematic review (27.1 per 1000 person years).³

In earlier studies, other than gender and HIV infection, higher mortality has also been attributed to increasing age among people who inject drugs.²⁰^,²¹ The age of people who inject drugs in this study was notably younger (median age at baseline of 33 years) than people who inject drugs in other cohort studies (aged 37–43 years).¹⁷^,¹⁹^,²²

Mortality of PWID in Estonia was almost twelve times higher than the general population and remained elevated across all age groups.

The excess mortality was higher among women and in the younger age-group. These findings reflect the age and gender differences in mortality of the general population (younger age groups in comparison to older, and females in comparison to males have lower mortality in the general population).³ However, this could also imply that females are particularly sensitive to the consequences of injection drug use. Our findings are in good agreement with other studies reporting higher SMR’s for women who inject drugs.⁴^,²³

We saw a clear diverging pattern of factors contributing to all-cause death risk by the duration of injection drug use. Among people who inject drugs with a shorter injection career, injecting opioids was the sole significant contributor increasing risk of death. In studies from elsewhere in Europe, high mortality rates among people who inject drugs coincided with a reported high number of drug overdoses.²⁴ We were not able to replicate these results: self-reported past 12-month non-fatal overdose was not associated with greater mortality risk. However, the observed prevalence of non-fatal overdose within 12 months (30.1%) was high in comparison to the worldwide median rate of 16.8% (range 4–38%).²⁵^,²⁶ Up to 2017, for over a decade, Estonia has had the highest overdose mortality in Europe. Most of the overdose deaths have been caused by fentanyl and fentanyl derivates which have been the most common form of opioids used in Estonia. Since 2018–2021, overdose deaths declined, potentially related to decreased availability of fentanyl derivates and a scaling-up of interventions (e.g. the provision of take-home naloxone).⁵^,²⁷^,²⁸

Fentanyl use led to Estonia having the highest fatal drug overdose rate in Europe up to 2017. In Estonia, fentanyl emerged in 2003, swiftly replacing heroin in the illicit drug market. Between 2003 and 2006, 3-methylfentanyl dominated confiscations. Until 2015, only fentanyl and 3-methylfentanyl were detected by the Estonian Forensic Science Institute.⁵ Since then, various analogues (car-, acryl-, cyclopropyl-, furanylfentanyl) have been seized. Although, there was a fentanyl drug market drought in 2018, recent years have seen the emergence of a new type of synthetic opioids called nitazenes (proto-, meto-, isotonitazen), not infrequently mixed with xylazine (veterinary drug for sedation, anaesthesia, muscle relaxation, analgesia). These changes have led to a sharp increase in drug-related overdose mortality since 2021.²⁹

A similar fentanyl-overdose situation has developed in the USA. Over the last several decades, the USA has experienced different waves of opioid-related overdose deaths. First was the opioid prescription analgesic wave (roughly 1999–2009), followed by a heroin wave (roughly 2010–2012), followed by the fentanyl wave (roughly 2013 to the present).³⁰ These different waves represent changes in the supplies of opioid drugs in the illicit market. During this time the numbers of fatal drug overdoses rose from 16 849 in 1999 to 108 000 in 2021, which set a new record for fatal overdose deaths in the USA.³¹ One interesting difference in fentanyl use in the USA versus Estonia, is that in the USA, fentanyl is often sold not as fentanyl, but mixed with other drugs (heroin primarily, but also with stimulants such as cocaine and methamphetamine) and sometimes sold as other drugs (heroin and opioid analgesics in particular). The similarity between the USA and Estonian situations indicates there are needs for greater treatment for opioid use disorder and for greater distribution of naloxone in the community in both countries.

In the present study, where the main drug injected were opioids, only 19% (n = 262) were on medication-assisted treatment (methadone maintenance) during enrolment to the source study. Although drug treatment is considered a protective measure against mortality among people who inject drugs,²^,³² the low coverage observed in our study and other studies from Estonia³³^,³⁴ do not show a population-based effect.

HIV prevalence is high in our cohort (55%) compared to other cohort studies around the world (18–27%)¹⁹^,²²^,³⁵ but is similar to countries of eastern Europe where the prevalence of HIV has been reported to be over 40% among people who inject drugs.³⁶ Among people who inject drugs with a longer duration of injecting (injecting for over 15 years and escaping overdose death), HIV-infection emerges as a significant risk factor for death. This finding potentially indicates that the positive impact of ART remains largely inaccessible for people who inject drugs in Estonia. The argument is further confirmed by our subgroup analysis showing no difference in mortality among those on ART in comparison to those who are not. These results may indicate to stigmatization, lack of integrated services and suboptimal HIV care. Surveys have shown that ART adherence among people who inject drugs in Estonia is relatively high, based on self-reports.⁶^,⁷

The main source of income, other than work, increased mortality risk (HR = 1.53; 95% CI 1.03–2.29). These results are similar to other studies where higher mortality risk among people who inject drugs was associated with unemployment and the inability to work due to disability.³⁷^,³⁸ Studies have found a significant effect of the unemployment rate on opioid mortality; a 1% increase in unemployment increases the opioid death rate by 3.6% per 100 000.³⁹

Living in the capital region was also a factor associated with a doubled mortality risk. Regional differences in risk structure among people who inject drugs are well known. Higher numbers of mortality of people who inject drugs may be driven by opioid use, higher age and lower drug treatment coverage.⁴⁰

Strengths and limitations

The main strengths of this study include its large sample size, long follow-up period, valid measurement (laboratory confirmed) for HIV verification and linkage to the national causes of death registry using personal identifiers.

Our study benefits from rigorous measures for key variables, i.e. outcome (the national causes of death registry), the status of a person who injects drugs (by trained staff) and HIV status. We were able to control for the relevant known confounders (region, age, education, source of income, prison experience, receiving opioid agonist treatment (OAT)).

We are aware of biases based on analysis on non-random samples. The impossibility of obtaining probability samples in ‘hidden populations’ has led the authors of source studies to utilize respondent driven sampling to recruit people who were currently injecting drugs.

There are, however, limitations to our study. To minimize selection bias errors, we used the left truncation method to diminish the possibility of excluding from the sample people who had died before the source studies were conducted. Still, mortality among PWID who have been injecting for several decades may be overestimated, as our sample does not account for individuals who have ceased injecting. Secondly, some errors caused by self-reported questions may exist as people may give socially desirable answers. Thirdly, HIV coinfection, particularly coinfection with HCV, information was not available for all study participants; therefore, these factors could not be considered in the data analyses. However, these limitations seem unlikely to have caused the clear patterns observed in this study.

Conclusions

Our findings show the impact of drug policy and practice. A SMR of 12 underscores the significant impact of illicit drug use on excess mortality, emphasizing it as a major public health concern.

Although the low-threshold services for people who use drugs have been available for a long time in Estonia, there is a need to improve the services. To reduce harms caused by injecting illicit drugs, is important to make the services for people who inject drugs more accessible by integrating infectious disease care and substance use treatment. It is also important to increase the coverage of the services.

Supplementary Material

ckad204_Supplementary_Data

ckad204_supplementary_data.docx^{(31.5KB, docx)}

Contributor Information

Maris Salekešin, Department of Risk Behavior Studies, National Institute for Health Development, Tallinn, Estonia; Institute of Family Medicine and Public Health, University of Tartu, Tartu, Estonia.

Sigrid Vorobjov, Department of Risk Behavior Studies, National Institute for Health Development, Tallinn, Estonia.

Don Des Jarlais, School of Global Public Health, New York University, New York, USA.

Anneli Uusküla, Institute of Family Medicine and Public Health, University of Tartu, Tartu, Estonia.

Supplementary data

Supplementary data are available at EURPUB online.

Funding

This work was supported through grants R01-AI083035 and DP1-DA039542 from the US National Institute on Drug Abuse, USA, and National Health Plan 2009-2020, funded by Estonian Ministry of Social Affairs.

Conflicts of interest: None declared.

Key points.

Illicit drug use influences on excess mortality.
Being HIV positive and injecting mainly opioids were associated with greater mortality risk.
To decrease mortality there is a need to integrate services provided to PWID, particularly the integration of HIV and opioid treatment.

Data availability

There are legal restrictions on sharing a de-identified data. According to legislative regulation and data protection law in Estonia, the authors cannot publicly release the data received from the health data registers in Estonia. The data on causes of death can be requested by completing the application in order to carry out research or an evaluation of public interest (https://www.tai.ee/et/statistika-ja-registrid/surma-pohjuste-register). More information about data availability: maris.salekesin@tai.ee.

References

1. Degenhardt L, Bucello C, Mathers B, et al. Mortality among regular or dependent users of heroin and other opioids: a systematic review and meta-analysis of cohort studies. Addiction 2011;106:32–51. [DOI] [PubMed] [Google Scholar]
2. Larney S, Tran LT, Leung J, et al. All-cause and cause-specific mortality among people using extramedical opioids: a systematic review and meta-analysis. JAMA Psychiatry 2020;77:493–502. [DOI] [PMC free article] [PubMed] [Google Scholar]
3. 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;91:102–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
4. Ahmad FB, Cisewski JA, Rossen LM, Sutton P. Provisional Drug Overdose Death Counts. National Center for Health Statistics, 2023. https://www.cdc.gov/nchs/nvss/vsrr/drug-overdose-data.htm (28 August 2023, date last accessed).
5. Uusküla A, Talu A, Vorobjov S, et al. The fentanyl epidemic in Estonia: factors in its evolution and opportunities for a comprehensive public health response, a scoping review. Int J Drug Policy 2020;81:102757. [DOI] [PMC free article] [PubMed] [Google Scholar]
6. Salekešin M, Vorobjov S. HIVi levimuse ja riskikäitumise uuring Narva narkootikume süstivate inimeste seas 2018. Tallinn; Tervise Arengu Instituut, 2019. https://www.tai.ee/sites/default/files/2021-03/15693218519_HIVi_levimuse_ja_riskikaitumise_uuring_Narva_narkootikume_sustivate_inimeste_seas_2018.pdf (23 May 2023, date last accessed).
7. Salekešin M, Vorobjov S. HIVi levimuse ja riskikäitumise uuring narkootikume süstivate inimeste seas Kohtla-Järvel 2020. Tallinn; Tervise Arengu Instituut, 2021. https://www.tai.ee/sites/default/files/2021-11/KJ_narkootikume_sustivate_inimeste_uuring_2020.pdf (23 May 2023, date last accessed).
8. European Monitoring Centre for Drugs and Drug Addiction. European Drug Report 2019: Trends and Developments. Luxembourg: Publications Office of the European Union, 2019. https://www.emcdda.europa.eu/system/files/publications/11364/20191724_TDAT19001ENN_PDF.pdf (23 May 2022, date last accessed).
9. European Monitoring Centre for Drugs and Drug Addiction. European Drug Report 2021: Trends and Developments. Luxembourg: Publications Office of the European Union, 2021. https://www.emcdda.europa.eu/system/files/publications/13838/TDAT21001ENN.pdf (23 May 2022, date last accessed).
10. Lemsalu L, Rice B, Uusküla A, et al. How to Ascertain an Accurate Number of People Living With Diagnosed HIV and Fill Data Gaps From the Past—Lessons Learned From a High Prevalence Setting in Europe. Poster Present AIDS 2018, July 23-27, Amsterdam, 2018. https://programme.aids2018.org/Abstract/Abstract/38 (23 May 2023, date last accessed).
11. Des Jarlais DC, Perlis TE, Stimson GV, et al. ; WHO Phase II Drug Injection Collaborative Study Group. Using standardized methods for research on HIV and injecting drug use in developing/transitional countries: case study from the WHO Drug Injection Study Phase II. BMC Public Health 2006;6:54. [DOI] [PMC free article] [PubMed] [Google Scholar]
12. Santo T, Bharat C, Colledge-Frisby S, et al. Mortality Among People With Substance Use Disorders: A Toolkit for Classifying Major Causes of Death. Sydney: National Drug and Alcohol Research Centre, UNSW Sydney, 2022. [Google Scholar]
13. Maldonado G, Greenland S.. Simulation study of confounder-selection strategies. Am J Epidemiol 1993;138:923–36. [DOI] [PubMed] [Google Scholar]
14. Budtz-Jørgensen E, Keiding N, Grandjean P, Weihe P.. Confounder selection in environmental epidemiology: assessment of health effects of prenatal mercury exposure. Ann Epidemiol 2007;17:27–35. [DOI] [PubMed] [Google Scholar]
15. Bellera CA, MacGrogan G, Debled M, et al. Variables with time-varying effects and the Cox model: some statistical concepts illustrated with a prognostic factor study in breast cancer. BMC Med Res Methodol 2010;10:20. [DOI] [PMC free article] [PubMed] [Google Scholar]
16. Bahji A, Cheng B, Gray S, Stuart H.. Mortality among people with opioid use disorder: a systematic review and meta-analysis. J Addict Med 2020;14:e118–32. [DOI] [PubMed] [Google Scholar]
17. Gjersing L, Bretteville-Jensen AL.. Patterns of substance use and mortality risk in a cohort of ‘hard-to-reach’ polysubstance users. Addiction 2018;113:729–39. [DOI] [PubMed] [Google Scholar]
18. Mathers BM, Degenhardt L.. Examining non-AIDS mortality among people who inject drugs. AIDS 2014;28 Suppl 4:S435–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
19. Sun J, Mehta SH, Astemborski J, et al. Mortality among people who inject drugs: a prospective cohort followed over three decades in Baltimore, MD, USA. Addiction 2022;117:646–55. [DOI] [PMC free article] [PubMed] [Google Scholar]
20. McDonald SA, McAuley A, Hickman M, et al. Increasing drug-related mortality rates over the last decade in Scotland are not just due to an ageing cohort: a retrospective longitudinal cohort study. Int J Drug Policy 2021;96:103286. [DOI] [PubMed] [Google Scholar]
21. Sheehan DM, Trepka MJ, Fennie KP, et al. Individual and neighborhood predictors of mortality among HIV-positive Latinos with history of injection drug use, Florida, 2000-2011 history after controlling for HHS Public Access. Drug Alcohol Depend 2015;154:243–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
22. Deans GD, Raffa JD, Lai C, et al. Mortality in a large community-based cohort of inner-city residents in Vancouver, Canada. CMAJ Open 2013;May 241:E68–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
23. Gjersing L, Bretteville-Jensen AL.. Gender differences in mortality and risk factors in a 13-year cohort study of street-recruited injecting drug users. BMC Public Health 2014;14:440. [DOI] [PMC free article] [PubMed] [Google Scholar]
24. European Monitoring Centre for Drugs and Drug Addiction (EMCDDA). Mortality Related to Drug Use in Europe. Luxembourg: Publ Off Eur Union, 2011, 33. http://www.emcdda.europa.eu/publications/selected-issues/mortality (7 December 2021, date last accessed).
25. Martins SS, Sampson L, Cerdá M, Galea S.. Worldwide prevalence and trends in unintentional drug overdose: a systematic review of the literature. Am J Public Health 2015;105:e29–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
26. Handanagic S, Bozicevic I, Sekerija M, et al. Overdose mortality rates in Croatia and factors associated with self-reported drug overdose among persons who inject drugs in three Croatian cities. Int J Drug Policy 2019;64:95–102. [DOI] [PubMed] [Google Scholar]
27. European Monitoring Centre for Drugs and Drug Addiction (EMCDDA). European Drug Report 2022: Trends and Developments, Vol. 1. Luxembourg: Publ Off Eur Union, 2022, 4–6. https://www.emcdda.europa.eu/system/files/publications/14644/TDAT22001ENN.pdf (19 July 2022, date last accessed). [Google Scholar]
28. European Monitoring Centre for Drugs and Drug Addiction (EMCDDA). European Drug Report 2020: Trends and Developments. Luxembourg: Publ Off Eur Union, 2020. https://www.emcdda.europa.eu/system/files/publications/13236/TDAT20001ENN_web.pdf (19 July 2022, date last accessed).
29. The National institute for health development. Deaths Caused by Drug Overdose Saw a Sharp Increase in 2022. https://www.tai.ee/et/uudised/narkootikumide-uledoosist-pohjustatud-surmad-2022-aastal-huppeliselt-kasvanud (24 September 2023, date last accessed).
30. Ciccarone D. The triple wave epidemic: supply and demand drivers of the US opioid overdose crisis. Int J Drug Policy 2019;71:183–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Hedegaard H, Miniño AM, Warner M.. Drug overdose deaths in the United States, 1999–2019. NCHS Data Brief 2020;394:1–8. [PubMed] [Google Scholar]
32. Santo T, Clark B, Hickman M, et al. Association of opioid agonist treatment with all-cause mortality and specific causes of death among people with opioid dependence: a systematic review and meta-analysis. JAMA Psychiatry 2021;78:979–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
33. World Health Organization (WHO). How to Improve Opioid Substitution Therapy Implementation: Experiences from EU Countries, Part of a Join EAHC-WHO Project Involving Bulgaria, Estonia, Germany, Latvia, Lithuania, Poland, Portugal and Romania. Copenhagen: WHO Regional Office for Europe, 2014. https://www.euro.who.int/__data/assets/pdf_file/0015/241341/How-to-improve-Opioid-Substitution-Therapy-implementation.pdf (14 December 2021, date last accessed).
34. Larney S, Peacock A, Leung J, et al. Global, regional, and country-level coverage of interventions to prevent and manage HIV and hepatitis C among people who inject drugs: a systematic review. Lancet Glob Health 2017;5:e1208–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Lappalainen L, Hayashi K, Dong H, et al. Ongoing impact of HIV infection on mortality among persons who inject drugs despite free antiretroviral therapy. Addiction 2015;110:111–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
36. Mathers BM, Degenhardt L, Phillips B, et al. ; 2007 Reference Group to the UN on HIV and Injecting Drug Use. Global epidemiology of injecting drug use and HIV among people who inject drugs: a systematic review. Lancet 2008;372:1733–45. [DOI] [PubMed] [Google Scholar]
37. Aram J, Johnson NJ, Lee MLT, Slopen N.. Drug overdose mortality is associated with employment status and occupation in the National Longitudinal Mortality Study. Am J Drug Alcohol Abuse 2020;46:769–76. [DOI] [PubMed] [Google Scholar]
38. Davis JM, Suleta K, Corsi KF, Booth RE.. A Hazard analysis of risk factors of mortality in individuals who inject drugs in Denver CO. AIDS Behav 2017;21:1044–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
39. Hollingsworth A, Ruhm CJ, Simon K.. Macroeconomic conditions and opioid abuse. J Health Econ 2017;56:222–33. [DOI] [PubMed] [Google Scholar]
40. Van Amsterdam J, Van Den Brink W, Pierce M.. Explaining the differences in opioid overdose deaths between Scotland and England/Wales: implications for European Opioid Policies. Eur Addict Res 2021;27:399–412. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ckad204_Supplementary_Data

ckad204_supplementary_data.docx^{(31.5KB, docx)}

Data Availability Statement

[ckad204-B1] 1. Degenhardt L, Bucello C, Mathers B, et al. Mortality among regular or dependent users of heroin and other opioids: a systematic review and meta-analysis of cohort studies. Addiction 2011;106:32–51. [DOI] [PubMed] [Google Scholar]

[ckad204-B2] 2. Larney S, Tran LT, Leung J, et al. All-cause and cause-specific mortality among people using extramedical opioids: a systematic review and meta-analysis. JAMA Psychiatry 2020;77:493–502. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B3] 3. 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;91:102–23. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B4] 4. Ahmad FB, Cisewski JA, Rossen LM, Sutton P. Provisional Drug Overdose Death Counts. National Center for Health Statistics, 2023. https://www.cdc.gov/nchs/nvss/vsrr/drug-overdose-data.htm (28 August 2023, date last accessed).

[ckad204-B5] 5. Uusküla A, Talu A, Vorobjov S, et al. The fentanyl epidemic in Estonia: factors in its evolution and opportunities for a comprehensive public health response, a scoping review. Int J Drug Policy 2020;81:102757. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B6] 6. Salekešin M, Vorobjov S. HIVi levimuse ja riskikäitumise uuring Narva narkootikume süstivate inimeste seas 2018. Tallinn; Tervise Arengu Instituut, 2019. https://www.tai.ee/sites/default/files/2021-03/15693218519_HIVi_levimuse_ja_riskikaitumise_uuring_Narva_narkootikume_sustivate_inimeste_seas_2018.pdf (23 May 2023, date last accessed).

[ckad204-B7] 7. Salekešin M, Vorobjov S. HIVi levimuse ja riskikäitumise uuring narkootikume süstivate inimeste seas Kohtla-Järvel 2020. Tallinn; Tervise Arengu Instituut, 2021. https://www.tai.ee/sites/default/files/2021-11/KJ_narkootikume_sustivate_inimeste_uuring_2020.pdf (23 May 2023, date last accessed).

[ckad204-B8] 8. European Monitoring Centre for Drugs and Drug Addiction. European Drug Report 2019: Trends and Developments. Luxembourg: Publications Office of the European Union, 2019. https://www.emcdda.europa.eu/system/files/publications/11364/20191724_TDAT19001ENN_PDF.pdf (23 May 2022, date last accessed).

[ckad204-B9] 9. European Monitoring Centre for Drugs and Drug Addiction. European Drug Report 2021: Trends and Developments. Luxembourg: Publications Office of the European Union, 2021. https://www.emcdda.europa.eu/system/files/publications/13838/TDAT21001ENN.pdf (23 May 2022, date last accessed).

[ckad204-B10] 10. Lemsalu L, Rice B, Uusküla A, et al. How to Ascertain an Accurate Number of People Living With Diagnosed HIV and Fill Data Gaps From the Past—Lessons Learned From a High Prevalence Setting in Europe. Poster Present AIDS 2018, July 23-27, Amsterdam, 2018. https://programme.aids2018.org/Abstract/Abstract/38 (23 May 2023, date last accessed).

[ckad204-B11] 11. Des Jarlais DC, Perlis TE, Stimson GV, et al. ; WHO Phase II Drug Injection Collaborative Study Group. Using standardized methods for research on HIV and injecting drug use in developing/transitional countries: case study from the WHO Drug Injection Study Phase II. BMC Public Health 2006;6:54. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B12] 12. Santo T, Bharat C, Colledge-Frisby S, et al. Mortality Among People With Substance Use Disorders: A Toolkit for Classifying Major Causes of Death. Sydney: National Drug and Alcohol Research Centre, UNSW Sydney, 2022. [Google Scholar]

[ckad204-B13] 13. Maldonado G, Greenland S.. Simulation study of confounder-selection strategies. Am J Epidemiol 1993;138:923–36. [DOI] [PubMed] [Google Scholar]

[ckad204-B14] 14. Budtz-Jørgensen E, Keiding N, Grandjean P, Weihe P.. Confounder selection in environmental epidemiology: assessment of health effects of prenatal mercury exposure. Ann Epidemiol 2007;17:27–35. [DOI] [PubMed] [Google Scholar]

[ckad204-B15] 15. Bellera CA, MacGrogan G, Debled M, et al. Variables with time-varying effects and the Cox model: some statistical concepts illustrated with a prognostic factor study in breast cancer. BMC Med Res Methodol 2010;10:20. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B16] 16. Bahji A, Cheng B, Gray S, Stuart H.. Mortality among people with opioid use disorder: a systematic review and meta-analysis. J Addict Med 2020;14:e118–32. [DOI] [PubMed] [Google Scholar]

[ckad204-B17] 17. Gjersing L, Bretteville-Jensen AL.. Patterns of substance use and mortality risk in a cohort of ‘hard-to-reach’ polysubstance users. Addiction 2018;113:729–39. [DOI] [PubMed] [Google Scholar]

[ckad204-B18] 18. Mathers BM, Degenhardt L.. Examining non-AIDS mortality among people who inject drugs. AIDS 2014;28 Suppl 4:S435–44. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B19] 19. Sun J, Mehta SH, Astemborski J, et al. Mortality among people who inject drugs: a prospective cohort followed over three decades in Baltimore, MD, USA. Addiction 2022;117:646–55. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B20] 20. McDonald SA, McAuley A, Hickman M, et al. Increasing drug-related mortality rates over the last decade in Scotland are not just due to an ageing cohort: a retrospective longitudinal cohort study. Int J Drug Policy 2021;96:103286. [DOI] [PubMed] [Google Scholar]

[ckad204-B21] 21. Sheehan DM, Trepka MJ, Fennie KP, et al. Individual and neighborhood predictors of mortality among HIV-positive Latinos with history of injection drug use, Florida, 2000-2011 history after controlling for HHS Public Access. Drug Alcohol Depend 2015;154:243–50. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B22] 22. Deans GD, Raffa JD, Lai C, et al. Mortality in a large community-based cohort of inner-city residents in Vancouver, Canada. CMAJ Open 2013;May 241:E68–76. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B23] 23. Gjersing L, Bretteville-Jensen AL.. Gender differences in mortality and risk factors in a 13-year cohort study of street-recruited injecting drug users. BMC Public Health 2014;14:440. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B24] 24. European Monitoring Centre for Drugs and Drug Addiction (EMCDDA). Mortality Related to Drug Use in Europe. Luxembourg: Publ Off Eur Union, 2011, 33. http://www.emcdda.europa.eu/publications/selected-issues/mortality (7 December 2021, date last accessed).

[ckad204-B25] 25. Martins SS, Sampson L, Cerdá M, Galea S.. Worldwide prevalence and trends in unintentional drug overdose: a systematic review of the literature. Am J Public Health 2015;105:e29–49. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B26] 26. Handanagic S, Bozicevic I, Sekerija M, et al. Overdose mortality rates in Croatia and factors associated with self-reported drug overdose among persons who inject drugs in three Croatian cities. Int J Drug Policy 2019;64:95–102. [DOI] [PubMed] [Google Scholar]

[ckad204-B27] 27. European Monitoring Centre for Drugs and Drug Addiction (EMCDDA). European Drug Report 2022: Trends and Developments, Vol. 1. Luxembourg: Publ Off Eur Union, 2022, 4–6. https://www.emcdda.europa.eu/system/files/publications/14644/TDAT22001ENN.pdf (19 July 2022, date last accessed). [Google Scholar]

[ckad204-B28] 28. European Monitoring Centre for Drugs and Drug Addiction (EMCDDA). European Drug Report 2020: Trends and Developments. Luxembourg: Publ Off Eur Union, 2020. https://www.emcdda.europa.eu/system/files/publications/13236/TDAT20001ENN_web.pdf (19 July 2022, date last accessed).

[ckad204-B29] 29. The National institute for health development. Deaths Caused by Drug Overdose Saw a Sharp Increase in 2022. https://www.tai.ee/et/uudised/narkootikumide-uledoosist-pohjustatud-surmad-2022-aastal-huppeliselt-kasvanud (24 September 2023, date last accessed).

[ckad204-B30] 30. Ciccarone D. The triple wave epidemic: supply and demand drivers of the US opioid overdose crisis. Int J Drug Policy 2019;71:183–8. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B31] 31. Hedegaard H, Miniño AM, Warner M.. Drug overdose deaths in the United States, 1999–2019. NCHS Data Brief 2020;394:1–8. [PubMed] [Google Scholar]

[ckad204-B32] 32. Santo T, Clark B, Hickman M, et al. Association of opioid agonist treatment with all-cause mortality and specific causes of death among people with opioid dependence: a systematic review and meta-analysis. JAMA Psychiatry 2021;78:979–93. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B33] 33. World Health Organization (WHO). How to Improve Opioid Substitution Therapy Implementation: Experiences from EU Countries, Part of a Join EAHC-WHO Project Involving Bulgaria, Estonia, Germany, Latvia, Lithuania, Poland, Portugal and Romania. Copenhagen: WHO Regional Office for Europe, 2014. https://www.euro.who.int/__data/assets/pdf_file/0015/241341/How-to-improve-Opioid-Substitution-Therapy-implementation.pdf (14 December 2021, date last accessed).

[ckad204-B34] 34. Larney S, Peacock A, Leung J, et al. Global, regional, and country-level coverage of interventions to prevent and manage HIV and hepatitis C among people who inject drugs: a systematic review. Lancet Glob Health 2017;5:e1208–20. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B35] 35. Lappalainen L, Hayashi K, Dong H, et al. Ongoing impact of HIV infection on mortality among persons who inject drugs despite free antiretroviral therapy. Addiction 2015;110:111–9. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B36] 36. Mathers BM, Degenhardt L, Phillips B, et al. ; 2007 Reference Group to the UN on HIV and Injecting Drug Use. Global epidemiology of injecting drug use and HIV among people who inject drugs: a systematic review. Lancet 2008;372:1733–45. [DOI] [PubMed] [Google Scholar]

[ckad204-B37] 37. Aram J, Johnson NJ, Lee MLT, Slopen N.. Drug overdose mortality is associated with employment status and occupation in the National Longitudinal Mortality Study. Am J Drug Alcohol Abuse 2020;46:769–76. [DOI] [PubMed] [Google Scholar]

[ckad204-B38] 38. Davis JM, Suleta K, Corsi KF, Booth RE.. A Hazard analysis of risk factors of mortality in individuals who inject drugs in Denver CO. AIDS Behav 2017;21:1044–53. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ckad204-B39] 39. Hollingsworth A, Ruhm CJ, Simon K.. Macroeconomic conditions and opioid abuse. J Health Econ 2017;56:222–33. [DOI] [PubMed] [Google Scholar]

[ckad204-B40] 40. Van Amsterdam J, Van Den Brink W, Pierce M.. Explaining the differences in opioid overdose deaths between Scotland and England/Wales: implications for European Opioid Policies. Eur Addict Res 2021;27:399–412. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Mortality among people who inject drugs – the interwoven roles of fentanyl and HIV: a community-based cohort study

Maris Salekešin

Sigrid Vorobjov

Don Des Jarlais

Anneli Uusküla

Abstract

Background

Methods

Results

Conclusions

Introduction

Methods

Participants

Data collection

Variables

Outcome

Statistical analyses

Results

Cohort overview

Crude mortality rates and SMRs

Survival analysis

Figure 1.

Factors associated with mortality among current people who inject drugs

Figure 2.

Subgroup analyses

Table 1.

Causes of death among people who inject drugs

Table 2.

Discussion

Strengths and limitations

Conclusions

Supplementary Material

Contributor Information

Supplementary data

Funding

Key points.

Data availability

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases