Effects, experiences and perceptions of harm reduction interventions among people who inject drugs

<p dir="ltr">Drug-induced mortality remains a persistent public health issue globally, with Sweden historically reporting some of the highest overdose death rates in Europe. While harm reduction interventions such as Take-Home Naloxone (THN) programs and Supervised Consumption Sites (SCSs) have gained traction internationally, their implementation in Sweden has been slow, with THN only introduced in 2018 and SCSs yet to be adopted.</p><p dir="ltr">This thesis brings together four studies that collectively examine the effectiveness, lived experience, and perceptions of harm reduction interventions among people who inject drugs (PWID) in Stockholm, using both quantitative and qualitative methodologies. The overarching aim was to inform and improve public health responses to morbidity and mortality in this patient group by assessing current strategies and identifying future possibilities.</p><p dir="ltr">Study I explored the implementation and outcomes of Sweden's first THN program, launched at the Stockholm Needle and Syringe Program (NSP). This prospective cohort study followed 1,295 participants between January 2018 and March 2022, collecting data on naloxone refills and overdose reversals. Participants were provided with naloxone following a brief training and could return for refills at which point they reported whether the kit had been used and under what circumstances. Nearly half (44%) of participants reported reversing at least one overdose, amounting to a total of 1,625 reversals. In 95.6% of cases, the individual reporting the event stated that the person who overdosed survived.</p><p dir="ltr">Zero-inflated Poisson regression identified stimulant use, benzodiazepine use, homelessness, and country of birth outside Europe as significant predictors of higher numbers of reported reversals in the multivariate model. The study demonstrated that an NSP infrastructure can effectively support large-scale THN distribution and reach high-risk individuals, many of whom were reversing overdoses within their social networks.</p><p dir="ltr">Study II took a qualitative approach to understanding the lived experiences of people who survived an opioid overdose with the help of peer-administered naloxone. Drawing on Zinberg's 'Drug, set and setting' theory, semi-structured interviews were conducted with 22 PWID recruited through the Stockholm NSP between November 2021 and May 2022. The analysis revealed that THN influenced each dimension of the 'Drug, set and setting' framework.</p><p dir="ltr">In terms of 'drug', participants described distressing withdrawal symptoms and emotional reactions from peers during overdose reversals. 'Set' involved complex feelings of shame, gratitude, and in some cases, denial of having overdosed. The 'setting' dimension highlighted how THN enabled overdose management within peer groups, often without the involvement of emergency services or police, a benefit in a context of pervasive drug-related stigma.</p><p dir="ltr">Participants widely valued THN, integrating it into personal and communal risk management strategies. Yet the study also underscored the limitations of THN as a standalone solution, pointing to broader structural issues such as the lack of safer environments for drug use.</p><p dir="ltr">Study III shifted focus to explore the potential role of SCSs in Sweden, a harm reduction measure not yet implemented here. This mixed-methods study used both survey and interview data to assess willingness among PWID to use an SCS, if one was available in Stockholm. The survey, administered to 219 PWID, captured a range of demographic and behavioural data, as well as opinions on SCSs. The results showed that 72.1% of respondents expressed willingness to use an SCS.</p><p dir="ltr">Factors independently associated with willingness to use an SCS in a multivariate logistic regression model included injecting opioids, having experienced an opioid overdose within the past year, and public injecting in the past month. The most frequently cited reason for wanting to use an SCS was "to avoid stress". Conversely, the main reason for reluctance was already having a private, perceived safe space to use drugs.</p><p dir="ltr">Interview data supported these findings and additionally highlighted concerns about privacy and restrictive rules, such as bans on assisted injecting or injecting into veins in the neck or groin. The findings underscore the high potential acceptability of SCSs among PWID in Sweden, while also pointing to the importance of tailoring services to user needs and preferences Study IV provided a broader epidemiological perspective by examining mortality among Stockholm NSP clients over a ten-year period (2013-2023), including a specific focus on the change in opioid overdose deaths after the introduction of the THN program in 2018. This register-based cohort study included 4,192 individuals enrolled in the Stockholm NSP. Data from the Swedish Cause of death register were linked with clinical and demographic data from the NSP's own database. All-cause and cause-specific mortality rates were calculated annually, and Fine and Gray regression was used to assess predictors of all-cause and opioid overdose deaths.</p><p dir="ltr">Of the cohort, 685 individuals (16%) died during the study period. Although crude mortality rates remained high compared to international levels, a significant downward trend was observed over time. Opioid overdose was the leading cause of death, but rates decreased during the study period. THN access and participation in opioid agonist therapy (OAT) were associated with a reduced hazard of opioid overdose death. The study suggests that THN, when integrated into harm reduction services, may be effective in reducing fatal opioid overdoses.</p><p dir="ltr">Taken together, these four studies demonstrate the multifaceted value of harm reduction interventions and their potential to reduce morbidity and mortality among PWID in Sweden. THN distribution through an NSP can be effectively scaled and reach high-risk individuals, enabling them to act as first responders within their communities. Qualitative findings highlight the emotional and social dimensions of overdose survival and the limitations of THN in isolation. SCSs are widely accepted among Swedish PWID and could address some of the gaps left by THN programs, particularly for those injecting in public. Finally, both cohort- and individual-level data support the protective effect of both THN and OAT in reducing opioid overdose deaths over time.</p><p dir="ltr">This thesis contributes important evidence to the evolving field of harm reduction in Sweden. By combining quantitative and qualitative methods, it offers an expanded understanding of how structural and interpersonal factors shape overdose risks and responses. The findings provide a strong argument for the continued and expanded implementation of harm reduction services in Sweden, including comprehensive THN access and the introduction of SCSs tailored to the needs of PWID.</p><h3>List of scientific papers</h3><p dir="ltr">I. <b>Holmen E</b>, Warnqvist A, Kåberg M. Sweden's first Take-Home Naloxone program: participant characteristics, dose endpoints and predictors for overdose reversals. (2023) Substance Abuse Treatment Prevention and Policy. 2023;18(1):24. <a href="https://doi.org/10.1186/s13011-023-00533-2">https://doi.org/10.1186/s13011-023-00533-2</a></p><p dir="ltr">II. <b>Holmén E,</b> Hammarberg A, Kåberg M, Storbjörk J. (2023) Take- Home Naloxone and risk management from the perspective of people who survived an opioid overdose in Stockholm - An analysis informed by drug, set and setting. International Journal of Drug Policy. 2023;115:104021. <a href="https://doi.org/10.1016/j.drugpo.2023.104021">https://doi.org/10.1016/j.drugpo.2023.104021</a></p><p dir="ltr">III. <b>Holmén E,</b> Kåberg M, Lundeberg E, Storbjörk J, Hammarberg A. (2025) Willingness and contextual considerations for supervised consumption sites: a mixed-methods study among people who inject drugs in Stockholm. International Journal of Drug Policy. 143:104866. <a href="https://doi.org/10.1016/j.drugpo.2025.104866">https://doi.org/10.1016/j.drugpo.2025.104866</a></p><p dir="ltr">IV. <b>Holmen E</b>, Kåberg M, Hammarberg A. All-cause mortality and overdose deaths among 4,192 people who inject drugs in Stockholm [Submitted]</p>