The opioid crisis represents a multifaceted public health emergency, characterized by high rates of addiction and overdose mortality [¹]. It is commonly believed to have its roots in the general over-prescription of pharmaceutical opioids that started in the 1990s and planted the seeds of dependency in communities nationwide. This initial wave of policy was inconsistent as it alternated between punitive criminal justice and traditional abstinence-based treatment models [²,³]. These strategies were not effective, being unable to control the increasing death rates, as well as being unable to reach out to the increasing number of people who are not part of the traditional healthcare systems.

Since then, the crisis has passed through several phases, each of which has been very deadly. The shift to heroin was followed by a second wave, which was characterized by the rise of the illicit synthetic opioid market, which, as of today, is characterized by the presence of fentanyl. This powerful, unpredictable drug has reached unprecedented levels of overdose deaths, causing a societal cost that has never been seen before [⁴,⁵].

This progression has exposed significant, systemic vulnerabilities in healthcare delivery, social services, and economic policy. The current epidemiological context underscores the need to implement evidence-based interventions that reduce mortality and engage marginalized populations who remain at greatest risk [⁶,⁷].

For this review, "community-based" harm reduction programs are operationally defined as services delivered outside of inpatient or acute hospital settings, within or directly accessible to the communities where individuals who use drugs reside. This encompasses fixed-site facilities (such as SSPs, OPCs, and community health centers), mobile outreach services, peer-led distribution networks, and harm reduction components integrated into primary care or low-barrier clinic settings.

In response to this burden, community-based harm reduction programs have emerged as a pragmatic public health approach. These programs are designed to meet individuals where they are, both geographically and in terms of their readiness to engage, without preconditions [⁸]. Their guiding principle is to reduce immediate harm and preserve life, serving as a first point of contact with the healthcare system and a foundation for longer-term recovery pathways. This will involve making syringes clean so that no one gets sick, delivering the overdose-reversal medication naloxone directly to the most at-risk individuals to witness an overdose, offering low-barrier access to life-stabilizing drugs for opioid use disorder, and ensuring that there is a safe place where drugs can be used under supervision. They succeed by building trust with people who use drugs, a group often failed by traditional systems [⁹,¹⁰].

Beyond syringe exchange, SSPs frequently represent the primary point of contact between individuals who use drugs and the healthcare system, where participants may also receive wound care, hepatitis testing, or referrals to additional services [¹¹,¹²]. More importantly, these interdependent connections lead to a longer-term path of treatment and recovery assistance. The most obvious way to realize this life-saving mission, perhaps, is to lend naloxone to friends, family, and peers. This approach expands the effective responder network into communities, and available evidence is consistent with meaningful reductions in overdose mortality [¹³,¹⁴]. Similarly, overdose prevention centers – staffed by trained personnel – have demonstrated the capacity to manage acute overdose events on-site, facilitate referrals to care, and operate without documented adverse effects on surrounding neighborhoods [¹⁵,¹⁶].

Collectively, these programs provide a continuum of care grounded in principles of safety, dignity, and low-barrier access, serving as a critical link between individuals at highest risk and the broader healthcare system [¹⁷].

Community-based harm reduction is susceptible to scaling and integration due to very serious impediments, even though it has proven to be an effective intervention. The first drawback is the lack of implementation science and cost-effectiveness evidence across various geographical and socio-political settings, especially in rural and underserved areas, where stigmatization and logistical barriers are urgent, as emphasized in the latest global recommendations of the World Health Organization [¹⁸,¹⁹,²⁰]. Although urban models are more widely documented, there is limited evidence on the adaptation and maintenance of services in urban regions.

The evidence base for newer interventions, such as overdose prevention centers and fentanyl test strip distribution, requires further longitudinal and comparative effectiveness research [²¹,²²]. Additionally, more robust data is needed to quantify the broader public health impact of integrated harm reduction hubs, including effects on healthcare utilization, criminal justice involvement, and community well-being.

A critical and persistent gap is the lack of systematic, multi-level interventions to reduce stigma among healthcare providers, policymakers, and the public. An empirical investigation into effective educational and contact-based strategies is essential to shift attitudes and build the political support necessary for policy change. Addressing these gaps is fundamental to translating these life-saving interventions into standardized, scalable components of the public health infrastructure [²³,²⁴,²⁵].

This review aims to consolidate the current evidence on the role of community-based harm reduction programs in addressing the opioid crisis. Its primary objectives are to:

By synthesizing this evidence, this review aims to inform public health policy, clinical practice, and future research, building an evidence-based response to the current public health emergency.

A systematic literature search was conducted to identify articles published from January 1, 2012, to April 30, 2023. This period was chosen to embrace the time frame that would best reflect the emergence of the illicit fentanyl market and how it has been responsible for a great number of overdose deaths [²⁶,²⁷]. This temporal restriction was selected deliberately to capture the period most relevant to the contemporary illicit fentanyl supply, which began reshaping the overdose landscape from approximately 2013 onward and has dominated the crisis in the period since. Readers should note that this restriction may exclude pre-2012 foundational evidence for long-established interventions, such as syringe service programs and naloxone distribution, which have extensive evidence bases predating the fentanyl era. The conclusions of this review therefore most directly pertain to the effectiveness of harm reduction interventions in the context of the contemporary illicit drug supply, rather than constituting an exhaustive historical appraisal of these programs.

The search has been conducted in the following large electronic databases: MEDLINE (through PubMed), PsycINFO, CINAHL Complete, Scopus, Web of Science Core Collection, and Sociological Abstracts. The search strategy used standardized subject headings, as well as free-text keywords of three key concepts, which were The Opioid Crisis, e.g., "Opioid-Related Disorders," fentanyl, overdose, Harm Reduction Interventions, e.g., "Harm Reduction," naloxone, syringe services, supervised consumption, Community-Based Settings, e.g., Community Health Services, grassroots, outreach. Key policy documents and seminal studies published after the formal search end date of April 2023, including some sources dated 2024 and beyond, were examined to provide contextual framing only in the Introduction, Discussion, and Recommendations sections. These post-search sources were not included in the formal narrative synthesis and are clearly used as contextual citations rather than synthesized evidence throughout the manuscript. All studies formally included in the synthesis were published between January 2012 and April 2023.

The complete search strategies for all six databases are provided in Appendix A. The PubMed strategy is presented there as Database 1. Complete search strategies for all six databases are provided in Appendix A (Supplementary Material). The following representative strategies were employed. Only English-language publications were searched. This review was not prospectively registered with PROSPERO or another protocol registry before commencement, which is a methodological limitation to consider when interpreting the findings.

Precisely defined eligibility criteria were used to guide the study selection process, and two independent reviewers conducted screening in two phases (title/abstract and full text). Any discrepancies were resolved through discussion and consensus between the two reviewers.

We have factored in the studies that dealt with individuals with opioid use disorder (OUD) or opioid users, and also the stakeholders, like staff in the program, individuals in the community, or even policymakers in the affected communities [⁴⁰,⁴¹]. The types of eligible interventions were community-based harm reduction programs, e.g., syringe service programs (SSP), overdose education and naloxone distribution (OEND), overdose prevention centers (OPC), drug checking services (e.g., fentanyl test strips), and low-barrier access to medications used to address OUD (MOUD) [⁴²,⁴³,⁴⁴]. The relevant outcomes required in studies included overdose rates, transmission of infectious diseases, connection with care, and indicators of program feasibility, acceptability, or cost-effectiveness. Original quantitative, qualitative, and mixed-methods research; systematic reviews; and program evaluations that included primary empirical data were all eligible as study designs [³²,⁴⁵]. The publications included were restricted to those in English and published between January 1, 2012, and April 30, 2023.

Articles were excluded when they assessed interventions delivered exclusively in clinical or hospital environments without a community-based outreach component, reported results for non-opioid substances exclusively, or were commentaries, editorials, opinion pieces, letters to the editor, or dissertations not containing primary empirical data. Non-English language publications were also excluded.

This review incorporates both original primary empirical studies and existing systematic and scoping reviews, an approach adopted to provide comprehensive coverage across the range of intervention types and outcome domains under examination. To address the risk of evidence overlap arising from the inclusion of both review-level sources and their underlying primary studies, we applied the following mitigation approach: (1) findings attributed to included systematic or scoping reviews are reported at the review level and clearly identified as such throughout the synthesis; (2) where a primary study was both independently identified and captured within an included review, it is cited once as the primary source; and (3) cumulative quantitative claims, such as the number of studies addressing a given outcome domain, are based on primary studies only and do not double-count review-level sources. Despite these steps, a formal co-citation overlap matrix was not applied, and readers should consider residual evidence overlap as a potential source of overestimation in the apparent consistency of findings, particularly for SSP and OEND outcome domains where review-level sources are most prevalent.

All data from all included studies were extracted into a standardized, piloted format using the Covidence systematic review software. The lead reviewer performed the extraction, and each entry was confirmed by a second reviewer as accurate and complete. In each study, we extracted key characteristics, including author(s), year of publication, country, study design, and main objectives. More information was provided about the type of harm reduction program (e.g., a syringe service program [SSP] or an overdose education and naloxone distribution [OEND] program), the location where it was operated (e.g., a community, prison, school, etc.), and its intended audience [³⁹,³⁸]. Although the primary scope of this review is community-based harm reduction as operationally defined in the Introduction, a small number of studies conducted in institutional settings – such as correctional facilities or schools – were included where the harm reduction program in question incorporated community outreach components or was explicitly designed to facilitate participants’ transition to community-based services. These settings are identified in the extracted data and are discussed with appropriate contextual qualification, where findings are reported. We also included methodological information such as sample size, data sources, and the analytical approach. Any pertinent quantitative and qualitative research results were harvested, including the main findings and possible null findings, the initial author’s findings, and the mentioned methodological weaknesses.

Because of substantial heterogeneity across interventions, study populations, and outcome measures, formal meta-analysis was not feasible. A structured narrative synthesis was therefore employed, following the Synthesis Without Meta-Analysis (SWiM) reporting guidance to ensure transparency. Studies were first grouped by intervention type (SSP, OEND, OPC, drug-checking, integrated MOUD models), constituting the primary unit of analysis. Within each intervention group, studies were further organized by outcome domain (overdose mortality, infectious disease transmission, treatment linkage, program feasibility and acceptability, cost-effectiveness). Findings within each grouping were then compared directionally across studies, with attention to study design hierarchy, sample characteristics, setting, and methodological quality ratings. Where discordant findings were identified across studies in the same grouping, possible explanations, including differences in population characteristics, follow-up duration, intervention fidelity, or measurement approach, were considered and described narratively. Study quality, as assessed through the design-specific appraisal tools described in Section 2.4, was used to contextualize and weight findings descriptively; lower-quality studies were retained in the synthesis, but their conclusions were interpreted with explicit caution and appropriate qualification.

The methodological quality and risk of bias of included studies were evaluated using validated, design-specific appraisal tools. For quantitative observational studies (cohort and cross-sectional designs), the Newcastle-Ottawa Scale (NOS) was used to assess the selection, comparability, and outcome or exposure domains [⁴⁶,⁴⁷]. Qualitative and mixed-methods studies were appraised using the Critical Appraisal Skills Program (CASP) checklist, which evaluates the rigor, credibility, and relevance of qualitative findings [⁴⁸]. Program evaluations were assessed against criteria for methodological transparency, including clarity of study aims, description of the intervention, adequacy of data sources, and reporting of null or adverse findings. It is noted that the NOS and CASP are primarily reporting and quality-assessment instruments; they were used to characterize study quality and contextualize findings, not as formal risk-of-bias tools in the sense of the Cochrane RoB 2.0 framework.

Overall evidence grades were assigned at the intervention-outcome level based on three considerations: (1) the study design hierarchy (e.g., RCT/meta-analysis > observational cohort > cross-sectional/program evaluation > qualitative), (2) the consistency of findings across multiple studies, and (3) the overall methodological quality ratings from the appraisal process. Grade A denoted evidence from RCTs, meta-analyses, or high-quality observational studies with large samples and consistent findings. Grade B denoted evidence from well-designed cohort studies, program evaluations with transparent methodology, or multiple consistent lower-quality studies. Grade C denoted evidence from single studies, uncontrolled evaluations, or studies with significant methodological limitations.

All quality assessments were conducted independently by two reviewers, with discrepancies resolved by consensus. No studies were excluded solely on quality grounds; lower-quality studies were retained in the synthesis but interpreted with appropriate caution and explicitly noted. The results of all quality assessments are summarised in Table 2 (Quality Assessment of Included Studies) and presented alongside the study designs and primary outcomes. Overall, 46 studies (59%) were rated as having a low-to-moderate risk of bias; the most common methodological limitations identified were reliance on self-reported outcomes, the absence of control groups in program evaluations, and potential selection bias in SSP-based cohort studies.

The study selection process followed the PRISMA 2020 guidelines, as shown in [⁴⁹,⁵⁰].

The systematic review identified 78 studies for final inclusion (see PRISMA flow diagram). The evidence base was predominantly North American, with 52 U.S. studies and 15 Canadian studies; the remaining studies were from Australia (n = 6) and Europe (n = 5). By study design, the corpus comprised quantitative observational studies (n = 35), qualitative and mixed-methods studies (n = 28), and formal program evaluations (n = 15).

By intervention type, included studies addressed syringe service programs (SSP, n = 32), overdose education and naloxone distribution (OEND, n = 29), overdose prevention centers (OPCs, n = 12), and integrated multi-service models (n = 17); several studies evaluated more than one intervention type simultaneously. Evidence grades were assigned at the intervention-outcome level according to the criteria described in Section 2.4; a representative selection of included studies, together with their assigned evidence grades, is presented in . The complete quality appraisal ratings for all 78 included studies are provided in .

Across 23 studies evaluating OEND programs, the evidence consistently indicated that naloxone distribution scale-up was associated with measurable reductions in opioid overdose mortality at the county and state level [⁵²]. These findings derive predominantly from ecological studies and interrupted time-series analyses conducted in the United States. Overdose prevention centers (12 studies) uniformly reported zero on-site fatal overdoses across a combined volume of thousands of supervised consumption events, with trained staff successfully managing acute overdose episodes in the large majority of cases without requiring emergency department transfer.

Syringe service programs (32 studies) demonstrated consistent associations with reduced HIV and Hepatitis C virus (HCV) transmission [⁵³]. Longitudinal cohort studies documented HCV seroconversion rates that were 30 – 76% lower among regular SSP participants than among non-participants. SSPs also functioned as platforms for infectious disease testing, hepatitis vaccination, and linkage to HIV and HCV care, with multiple studies reporting substantial proportions of participants receiving their first HIV or HCV test through an SSP.

Across 41 studies, harm reduction programs were consistently identified as effective entry points to the broader healthcare system [³⁵]. Integrated SSP-MOUD service models – in which low-barrier buprenorphine or other medications for opioid use disorder were co-located with or directly linked to SSP services – were associated with meaningfully higher rates of MOUD initiation and retention compared to non-integrated models. Qualitative and mixed-methods studies highlighted that non-coercive, trust-based relationships between participants and program staff were central to facilitating treatment engagement, particularly among individuals with histories of marginalization within conventional healthcare settings.

Multiple studies assessed program feasibility and acceptability across intervention types, generally reporting high levels of participant uptake, satisfaction, and sustained engagement [⁵⁴,⁵⁵,⁵⁶]. Cost-effectiveness analyses consistently found favorable ratios for SSP and OEND. They integrated SSP-MOUD programs and conducted modeling studies that estimated substantial cost savings relative to the projected costs of HIV treatment, emergency hospital utilization, and criminal justice involvement associated with no-intervention scenarios [⁵⁷].

Forty-seven studies addressed structural and policy barriers to program implementation and sustainability. Commonly reported barriers included punitive paraphernalia laws, restrictive zoning and siting policies, reliance on short-term or unreliable funding, regulatory constraints on buprenorphine prescribing, and limited rural service infrastructure. Community-level barriers included NIMBY (Not In My Backyard) opposition, misinformation, and staffing shortfalls. Participant-level barriers included fear of law enforcement presence near service locations, transportation difficulties, and generalized distrust of institutional health services, particularly among individuals with prior negative healthcare experiences [⁵⁸].

The following discussion is organized by outcome domain, recognizing that these domains reflect different inferential levels and that improvements in one area, such as facility-level safety, do not automatically establish improvements in another, such as population-level mortality reduction.

It is important to note that the interventions discussed in this review do not share equivalent evidentiary maturity. Syringe service programs and naloxone distribution have been studied for decades and are supported by an extensive body of longitudinal, large-sample, and in some cases quasi-experimental evidence. By contrast, overdose prevention centers, while increasingly evaluated in recent years, remain supported predominantly by observational, feasibility, and modeling studies, with a smaller evidence base and greater variation in study design quality. Drug-checking services, including fentanyl test strip distribution, represent the most nascent evidence domain, with most available evidence coming from short-term feasibility and acceptability studies. Throughout this discussion, claims are intended to be understood as calibrated to the depth and maturity of evidence specific to each intervention type, rather than as applying uniformly across all interventions evaluated.

The synthesized evidence suggests that community-based harm reduction interventions are associated with meaningful health outcomes across general health, particularly overdose prevention, infectious disease prevention, and participation in treatment [⁵⁹]. In 23 studies, overdose education and naloxone distribution (OEND) programs were associated with lower opioid overdose mortality based on opioid overdoses at the county and state level [⁶⁰,³⁰]. Studies evaluating overdose prevention centers consistently documented zero on-site fatal overdoses across thousands of supervised consumption events, with trained staff managing acute overdose episodes without emergency transfer in the large majority of cases [⁶¹]. It is important, however, to distinguish between these facility-level safety outcomes, which reflect what occurs within the OPC itself, and population-level effectiveness outcomes, such as reductions in community overdose mortality rates. The available evidence most robustly supports the former; evidence for the latter, while promising in modeling studies, is more limited and should not be conflated with demonstrated facility-level safety.

There were also strong outcomes in preventing infectious diseases. Syringe service programs (SSPs) were closely linked with the decrease in HIV and Hepatitis C (HCV) transmission, with longitudinal studies showing a 30-76% decrease in HCV seroconversion among regular SSP users relative to non-users [²⁸,²⁹,⁶²]. In addition to disease prevention, SSPs served as pivotal points for testing, vaccination, and care for infectious diseases [⁶³].

In 41 studies, a pervasive finding was the effectiveness of harm reduction programs as entry points to the healthcare system [³⁵,³⁴]. Service models that included SSP co-location and were integrated with low-barrier buprenorphine to provide services were linked with much higher initiation and retention into medications used to treat opioid use disorder (MOUD) [³¹,⁶⁴,³³]. The qualitative evidence also highlighted the focus on the non-coercive and trust-based relationships with the program staff as the key to treatment engagement, especially between those who were historically marginalized within traditional healthcare systems [³⁷].

Despite a solid body of evidence for their effectiveness, community-based harm reduction programs have not been implemented or sustained due to ongoing, multilevel barriers. Based on 47 studies, the underlying challenges are structural and policy barriers [⁶⁵,⁶⁶]. They are pervasive stigmas that are often institutionalized in the form of paraphernalia possession legislation and restrictive zoning policies, and long-term dependence on short-term or unstable sources of funds, which have a disproportionate constraining effect on the scale of programs in rural and underserved areas [⁶⁷,⁶⁸,⁶⁹].

Locally, implementation is often blocked by local resistance, misinformation, and the Not In My Backyard (NIMBY) principle, as well as real-world issues such as personnel shortages and inaccessibility [⁷⁰,⁷¹,⁷²]. There are also additional regulatory limitations to integrated care models, such as historical restrictions on prescribing buprenorphine and on federal funding for syringe services, which make service co-location and continuity of care difficult [⁷³].

Barriers at a participant level are also quite high. The fear of being monitored by the police around service locations, lack of transport, and the general suspicion of institutional healthcare services remain barriers to involvement in those individuals at the highest risk of overdose and infectious disease [⁷⁴,⁷⁵,⁷⁶]. Collectively, these obstacles lead to a continued widening of the gap between known effectiveness and reality on the ground.

This 78-study review supports the view that harm reduction is an evidence-based public health strategy rather than a marginal or alternative one. The identified associations between harm reduction programs and reduced overdose fatalities and infectious disease transmission reflect a practical approach that emphasizes immediate safety, dignity, and trust. Notably, the evidence repeatedly disproves the fact that the reduction of harm is a pretext for substance use [⁷⁷,⁷⁸]. Rather, such programs are low-barrier entry points to treatment, healthcare, and social services, enabling long-term recovery and stability. It is also important to note that harm reduction programs do not boost crime or disorder in the community [⁷⁹,⁸⁰]. Indicators of declines in community injecting, syringe littering, and emergency service use provide data-driven evidence for communities and policymakers, helping overcome a frequent barrier to greater adoption of such services as part of the broader public health infrastructure [⁸¹,⁸²,⁸³].

Theoretical Conceptual Framework: Proposed Pathways Between Community-Based Harm Reduction Program Implementation and Population-Level Opioid Overdose Mortality. This figure is a theoretical model constructed by the authors to summarise the proposed causal mechanisms, informed by the evidence synthesized in this review. It does not present empirical data, measurements, or quantitative estimates from any primary data source. Any axis labels represent qualitative directional constructs only and should not be interpreted as empirical scales or dose-response relationships. Readers seeking quantitative effect estimates are directed to the Results section and cited primary studies.

Some of the intervention models have a high potential, especially on a broader scale. Integrated SSP-MOUD models can be used to manage overdose risk in the short term and underlying opioid use disorder by using the available infrastructure and enhancing care continuity. Despite regulatory and financing challenges, these models offer a high-impact avenue for integrating systems at the system level [⁸⁹,⁹⁰,⁹¹].

Peer-led naloxone distribution is another highly scalable intervention because it is low-cost, requires minimal infrastructure, and can reach individuals at highest risk through existing social networks. Conversely, although evidence of the effectiveness of overdose prevention centers is compelling, their expansion is constrained by legal and political hurdles, underscoring the importance of policy advocacy and public education [⁹²,⁹³].

Overall, community-based harm reduction programs demonstrate consistent associations with beneficial outcomes across the included evidence base. The main issue is no longer the development of efficacy but the bridging of the implementation gap. Subsequent activities in implementation science, sustainable funding solutions, and structural changes should focus on ensuring these life-saving interventions are fully embedded in available and equitable public health systems [⁹⁴].

This review has several methodological limitations that must be taken into account when interpreting its findings. To begin with, the search was also limited to English-language publications, potentially excluding evidence relevant to other non-English-speaking countries, especially in Latin America and Eastern and Southeast Asia, where community harm reduction programs are increasingly active. This language barrier can create a reporting bias, skewing the reporting toward the North American and European program contexts.

Second, the high heterogeneity in study designs, intervention types, target populations, and outcome measures across the 78 included studies prevented formal meta-analysis. Although the narrative synthesis method is suitable in this evidence base, it creates the risk of interpretive subjectivity. The results are to be interpreted as broad themes and directional clues rather than as accurate pooled effect estimates.

Several review-level threats to validity warrant explicit consideration. First, this synthesis includes both primary empirical studies and existing systematic and scoping reviews within a single evidence base. While this broadened coverage, it introduces the risk of double-counting evidence when the same primary studies are captured both directly and through included reviews. A formal overlap analysis was not conducted, which may cause the cumulative evidence base to appear larger or more consistent than it is in reality. Second, although the formal search was restricted to studies published through April 2023, several post-search sources (dated 2024 – 2026) were drawn upon in the Introduction, Discussion, and Recommendations for contextualization. The selective use of post-search literature – not subject to the same systematic identification process- may introduce a form of citation bias, and readers should note that claims drawing on these sources reflect contextual framing rather than synthesized findings. Third, the evidence base is heavily dominated by North American studies (52 U.S. and 15 Canadian studies, representing approximately 86% of the included studies), with limited representation from Europe and Australia, and no studies from low- or middle-income countries. This geographic concentration limits the generalisability of findings to settings with different structural, legal, and epidemiological contexts.

Additionally, this review’s central focus is on the opioid crisis; however, a portion of the included evidence, particularly in the infectious disease and SSP literature, derives from studies conducted in populations who inject drugs more broadly, rather than opioid-specific populations. For many infectious disease outcomes, this overlap is arguably appropriate given the shared risk behaviors involved. However, readers should note that the applicability of some findings to opioid-specific populations is assumed rather than directly demonstrated, and the conclusions in those domains should be interpreted with corresponding caution.

Third, the review includes program appraisals, grey literature, and peer-reviewed studies. Lack of methodological rigor: This is especially true of program evaluations that lack concurrent control groups, thereby limiting the strength of causal inferences. The majority of the included research is observational, and the selection bias in the SSP participants, as the most at-risk participants might be over-represented, might impact the external validity of the results.

Fourth, the search was conducted until April 2023. Harm reduction is a rapidly developing area, especially when speaking of overdose prevention centers, vending machines selling naloxone, and drug-checking services. This review does not reflect developments that have taken place after this date.

Lastly, the review was performed by two authors. Although independent double screening and consensus-based conflict resolution were used, the lack of a third independent arbiter is a weakness relative to the methodological ideal of systematic reviews. All findings should be understood in the context of these limitations, and future revisions of this review should aim to expand the author team and language coverage.

The obvious sign of the need is the urgent transition from proof of concept to a stable, available system of care. The way ahead is not only about investing in programs, but about creating a more sustainable public health infrastructure. This would involve developing long-term, consistent funding through planned opioid settlement provisions and Medicaid reform to avoid short-term, stuttering grants to programs and individuals, which leave them in a continual state of change [⁹⁵,⁹⁶,⁹⁷].

Simultaneously, we should be keen to eliminate the legal and logistical impediments that prevent individuals from accessing the care they require. This is achieved through the deregulation of clinical practice to empower more providers and the reform of obsolete laws that criminalize health and safety instruments [⁹⁸,⁹⁹,¹⁰⁰]. We also need to adopt and invest in flexible solutions, such as mobile units and mail-based services, to help close the geographic gaps in access, since no community should be left behind.

Lastly, these initiatives should be successful, and to achieve that, we need to gain widespread social and political legitimacy. This means there must be sustained communication about the demonstrated gains: these programs save lives without negatively affecting community security. More importantly, this legitimacy is created through the sincerity of collaboration and the payment of individuals with lived experience, so that their expertise informs the design and promotion of services intended to assist them [¹⁰¹,¹⁰²,¹⁰³].