Effectiveness of needle and syringe Programmes in people who inject drugs : an overview of systematic reviews

RESEARCH ARTICLE Open Access Effectiveness of needle and syringe Programmes in people who inject drugs – An overview of systematic reviews Ricardo M Fernandes 1,3 , Maria Cary 2 , Gonçalo Duarte 1 , Gonçalo Jesus 1 , Joana Alarcão 1 , Carla Torre 2 , Suzete Costa 2 , João Costa 1,3 and António Vaz Carneiro 1,3* Abstract Background: Needle and syringe programmes (NSP) are a critical component of harm reduction interventions among people who inject drugs (PWID). Our primary objective was to summarize the evidence on the effectiveness of NSP for PWID in reducing blood-borne infection transmission and injecting risk behaviours (IRB). Methods: We conducted an overview of systematic reviews that included PWID (excluding prisons and consumption rooms), addressed community-based NSP, and provided estimates of the effect regarding incidence/prevalence of Human Immunodeficiency Virus (HIV), Hepatitis C virus (HCV), Hepatitis B virus (HBV) and bacteremia/sepsis, and/or measures of IRB. Systematic literature searches were undertaken on relevant databases, including EMBASE, MEDLINE, and PsychINFO (up to May 2015). For each review we identified relevant studies and extracted data on methods, and findings, including risk of bias and quality of evidence assessed by review authors. We evaluated the risk of bias of each systematic review using the ROBIS tool. We categorized reviews by reported outcomes and use of meta-analysis; no additional statistical analysis was performed. Results: We included thirteen systematic reviews with 133 relevant unique studies published between 1989 and 2012. Reported outcomes related to HIV (n=9),HCV(n= 8) and IRB (n= 6). Methods used varied at all levels of design and conduct, with four reviews performing meta-analysis. Only two reviews were considered to have low risk of bias using the ROBIS tool, and most included studies were evaluated as having low methodological quality by review authors. We found that NSP was effective in reducing HIV transmission and IRB among PWID, while there were mixed results regarding a reduction of HCV infection. Full harm reduction interventions provided at structural level and in multi-component programmes, as well as high level of coverage, were more beneficial. Conclusions: The heterogeneity and the overall low quality of evidence highlights the need for future community-level studies of adequate design to support these results. Trial registration: The protocol of this systematic review was registered in Prospective Register of Systematic Reviews (PROSPERO 2015:CRD42015026145). Keywords: Needle and syringe programmes, Harm reduction interventions, People who inject drugs, HIV/Aids-Hcv-Hbv * Correspondence: [email protected] 1 Center for Evidence-Based Medicine, Faculty of Medicine, University of Lisbonl, Av. Prof. Egas Moniz, 1649-028 Lisbon, Portugal 3 Portuguese Collaborating Centre of the IberoAmerican Cochrane Network-Cochrane Portugal Faculty of Medicine, University of Lisbon, Av. Prof. Egas Moniz, 1649-028 Lisbon, Portugal Full list of author information is available at the end of the article © The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Fernandes et al. BMC Public Health (2017) 17:309 DOI 10.1186/s12889-017-4210-2 Background People who inject drugs (PWID) experience high levels of morbidity and mortality. Drug-related harms include overdose, drug-related deaths, and blood-borne infections such as Human Immunodeficiency Virus (HIV), Hepatitis C (HCV), Hepatitis B (HBV), and bacteremia/sepsis. HCV is currently the most prevalent infectious disease affecting PWID, while HIV prevalence rates are lower. In an estimated total of 12.7 to 16 million PWID worldwide, it is believed that 1.2 million are infected with HIV [1] and 10 million are infected with HCV [2]. Sharing needles and syringes, as well as other injecting paraphernalia, is a key route of transmission of these infections [3]. Needle and syringe programmes (NSP) are thought to be a critical component of harm reduction interventions among PWID [3, 4]. The first NSP was established in the 1980s [5], in response to the global HIV epidemic, with the goal of providing access and encouraging the use of sterile injection paraphernalia by PWID. Since then, provision of these services has grown rapidly. Importantly, a shift in paradigm has favoured NSP as components of harm reduction or harm minimization policies, which focus on reducing all drug-related harms, i.e. preventing HIV, HBV and HCV infection, minimizing needle and syringe sharing and reuse, reducing the volume of discarded needles and syringes in the environment, and facilitating access to sterile paraphernalia. Furthermore, NSP may also promote the use of condoms and provide opportunistic relevant health information and services [3]. NSP are complex health interventions with several interacting components, such as behavioural changes in PWID and providers, a complex operating framework (users, providers, setting, health systems), and some degree of flexibility of interventions [6]. There is considerable variability among regions and countries in service provision, coverage and range of harm reduction interventions offered by NSP. In particular, a variety of measures have been developed to improve access to and use of sterile injecting equipment and to increase users’ choice. These include several methods of distribution or sale such as conventional NSP in fixed-sites, pharmacybased distribution, dispensing machines and outreach programmes –often using a mobile van or bus, or through home-visits [3]. Generally, pharmacy and specialist needle exchange provides a wide range of harm reduction information and advice, along with clean needles and syringes and possibly injecting paraphernalia. The legal framework in which NSP operate also varies at country level [7]. Pharmacy-based NSP are known to be in place in at least Australia [8], Belgium, France, Ireland, Kyrgyzstan, the Netherlands, Portugal, Spain, Slovenia, Ukraine [7, 9], UK [10], and New Zealand [11]. Many systematic reviews on the effectiveness of interventions providing injecting equipment have been conducted to date. Overviews of reviews which compile information from these multiple systematic reviews, have also been published. The aim of these overviews is to provide end-users with a comprehensive and critical summary of the available evidence on this intervention. Overviews are of particular interest in this field because existing systematic reviews have focused on disparate questions, regarding either specific populations (e.g. country-specific evidence), specific types of interventions (e.g. different types of NSP provision) or, most commonly, specific outcomes (e.g. HIV or HCV transmission) [12–14]. Furthermore, there is variability in methods used in these systematic reviews, including the assessment of possible biases and the use of quantitative synthesis (meta-analysis). Nevertheless, to the best of our knowledge, the published overviews in this field have focused only on specific outcomes, i.e. transmission of HIV and/or HCV. Objectives Our primary objective was to conduct an overview of systematic reviews that evaluated the evidence of the effectiveness of NSP for PWID across a range of different relevant outcomes, i.e. blood-borne infection transmission and injecting risk behaviours (IRB). Our secondary objective was to assess how different aspects of NSP provision, including provider, setting, coverage and any related component delivered in parallel, such as harm reduction services and opiate substitution therapy, modified the effect of NSP, with a particular focus on pharmacy-driven NSP. Methods We followed current guidance on the conduct of overviews of reviews, including recommendations from the Cochrane Collaboration [15] and guidance on public health intervention reviews by the Centre for Reviews and Dissemination of the University of York [16]. We also followed the recommendations from the PRISMA-P statement regarding reporting items that we considered applicable to this overview [17]. The protocol for this overview was registered at PROSPERO (CRD42015026145) available at http://www.crd.york.ac.uk/ PROSPERO/display_record.asp?ID=CRD42015026145. Eligibility criteria For inclusion in this overview, studies had to meet the following criteria: 1. Study design: systematic reviews, operationally defined as studies reporting a clearly stated set of objectives, eligibility criteria, a systematic search using two or more sources, a systematic presentation of the characteristics and findings of the included studies, Fernandes et al. BMC Public Health (2017) 17:309 Page 2 of 15 and estimates of the size and direction of the effect of interventions presented as numerical data, on an individual study-level basis and/or with quantitative synthesis (meta-analysis); 2. Participants: PWID, defined as people who inject some form of drug at the beginning of the study. We excluded studies focusing exclusively on participants whose consumption was confined to prisons and consumption rooms, since these are populations with distinct characteristics from our target population; 3. Interventions: systematic reviews had to evaluate community-based NSP, defined as the supply of at least needles and syringes, with or without other injecting paraphernalia for the preparation and consumption of drugs; 4. Outcomes: required reported outcomes included the incidence and/or prevalence of blood-borne infections (HIV, HCV, HBV and bacteremia/sepsis), and/or measures of IRB (including but not limited to syringe re-use, borrowing, sharing, renting and lending). When more than one review included exactly the same studies, the review that reported the most complete presentation of results was selected for inclusion in the overview. Search strategy and screening Searches were undertaken on MEDLINE®In-Process & Other Non-Indexed Citations (from 1946 to 12th of May 2015), EMBASE (from 1974 to 12th of May 2015) and PsycINFO (from 1806 to 12th of May 2015) via the OVID SP interface. In addition, the following databases were searched: the NHS Economic Evaluation Database (NHS EED), the Database of Abstracts of Reviews of Effects (DARE), the Cochrane Database of Systematic Reviews (CDSR), the National Institute for Health and Clinical Excellence (NICE), the Campbell Library of Systematic Reviews and the Database of Promoting Health Effectiveness Reviews (DoPHER). The search strategies for each database are shown in a supplementary file (Additional file 1). No language or other types of restrictions were applied. In addition to the database searches, handsearching of the references of the included reviews was undertaken to identify further relevant studies. Titles and abstracts of articles identified were screened independently by two authors (MC, JA), and classified as include, unclear or exclude. The full reports of all articles that classified as include or unclear were then obtained, and two authors (MC, GD) examined compliance of reviews with eligibility criteria, with a third author acting as an arbiter (RF). Data extraction Data from reports of all included systematic reviews were extracted by two authors (MC, GD) and validated by a third author (RF), using a data extraction form designed and pre-piloted for this overview. The following general characteristics were extracted from each systematic review: publication details; study objectives; eligibility criteria (population, intervention, comparators, outcomes, study designs); any reported protocol; and methods used for search, screening, data extraction and synthesis. For each systematic review, we listed all included studies and evaluated whether they matched the eligibility criteria for this overview regarding population, interventions and outcomes. We then extracted the following results from the eligible group of studies, whenever reported at the systematic review level: study design; countries involved; characteristics of included participants (demographics, prevalence of HIV/HCV); description of interventions and any co-intervention, duration of intervention and follow-up; effect estimates from meta-analysis (if available) or at individual studylevel, for each relevant outcome; and any subgroup or sensitivity analyses. The authors’conclusions for each relevant outcome were also collected. We contacted the authors of reviews for relevant missing data. Assessment of methodological quality At a study level, we extracted data on any risk of bias assessments of primary studies when performed and reported by reviewers in each systematic review, including tools used and summarized results. We also collected data on any reported evaluations of the quality of evidence concerning our outcomes of interest in included reviews, particularly those using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) tool [18], as well as data on assessments of publication bias. At a systematic review level, we assessed the methodological quality of each included systematic review using the Risk Of Bias In Systematic Reviews (ROBIS) tool [19]. Two review authors (MC, RF) performed quality assessments independently, using piloted decision rules. Disagreements regarding overall assessments were resolved through discussion, with a third reviewer serving as the final arbitrator (AVC). The rationale behind assessments was documented. We calculated measures of agreement and reliability between raters for each ROBIS domain. Data synthesis We stratified the included systematic reviews by: (i) type of outcomes assessed (blood-borne infections, IRB), and (ii) type of analysis (with or without meta-analysis). We summarized data from the included reviews both in text and in summary tables and figures. When meta-analysis was performed, we report pooled estimates using the models and measures of effect reported by systematic review authors, with 95% confidence intervals (95% CI); Fernandes et al. BMC Public Health (2017) 17:309 Page 3 of 15 we did not perform any additional statistical analysis. When reported, the accompanying I 2 values, which describe the percentage of total variation across studies that is due to heterogeneity rather than chance, were collected [20]. Results The search and screening process is summarised in Fig. 1. A total of 667 citations were identified through the various database searches. Three additional records were identified in the reference lists of screened studies. After 37 duplicates were removed, we obtained 633 citations, which were screened by title and abstract. We excluded 582 citations as they did not meet the inclusion criteria, and the remaining 51 were screened full text. Thirty-five citations were further excluded, and two reports [21, 22] were unobtainable. Fourteen reports were thus included, corresponding to 13 systematic reviews, as two records referred to the same study [12, 23]. Within those publications three were reports from National Institutes/Expert Commitees in the US [24], UK [12] and Canada [25], while 10 were regular papers published in scientific journals. Description of included reviews Table 1 lists the key characteristics of the 13 included systematic reviews that evaluated the evidence of the effectiveness of NSP for PWID in the community setting. We classified reviews by type of outcomes reported and type of data analysis (Fig. 2). The number of databases searched per review varied between two and 15, with the three most common sources being MEDLINE, EMBASE and PsycINFO. The dates of last search varied between 1995 [26] and 2012 [27]. Of 287 studies included in these 13 reviews, 200 studies matched the eligibility criteria for this overview, corresponding to 133 unique studies (a matrix with included studies by review is available in a supplementary file: Additional file 2). We included the complete set of studies from four reviews, as in the remaining reviews at least one study did not fulfill our eligibility criteria [27–30]. The number of relevant studies included per review ranged from three [31, 32] to 43 [24]. While there was some overlap in the included studies between reviews, the majority (68%) was included in one review only. We found substantial variability in the criteria used by review authors to define and classify study designs, as well Fig. 1 Screening decisions up to date as of 30Jun2015. Original search date 12May2015 Fernandes et al. BMC Public Health (2017) 17:309 Page 4 of 15 Table 1 Key characteristics of the included systematic reviews and Risk of bias of each review Authors, date of publication a PICOS b Dates of last search Databases searched (n) Risk of Bias and quality of evidence tools Total number of studies/ Number of studies relevant to the overview (designs) c,d Risk of bias of the review (assessed using ROBIS tool) Quantitative synthesismeta-analysis Aspinall, 2014 [27] P: PWID I: NSP (Specialist sites, pharmacies or outreach services) C: No, rare, or less exposure to NSP O: HIV incidence S: any 2012 CINAHL, Cochrane Library, EMBASE, IBSS, MEDLINE and PsycINFO (6) Newcastle-Ottawa each study & GRADE to assess quality of evidence as a whole 12/12 (11 CC/Co; 1 OTH) Low Hagan, 2011 [34] P: individuals who could conceivably have acquired HCV infection via administration of an illegal drug (PWID among them) I: interventions that could prevent HCV infection C:NR O: HCV prevalence or incidence rates, and measures of association with prevalence or incidence S: any 2010 Current Contents, Dissertation Abstracts, ERIC, NIH CRISP, MEDLINE, PsycINFO, Sociological Abstracts (7) Developed their own quality scale 26/7 (7 CC/Co) High Turner, 2011 [32] P: PWID in the UK I: NSP or OST (or combination) C: OST vs non OST; high vs low coverage of NSP O: HCV infection and IRB S: any NR PubMed and Web of Science (2) None 4/3 (1 CC/Co; 2 OTH) High Cross, 1998 [26] P: PWID I: NSP or educational programmes C: NR O: IRB S: experimental or quasi-experimental design 1995 AIDSLINE, MEDLINE, PsycINFO, Social Science Citation Index (4) None 26/10 (1 RCT; 9 OTH) High Qualitative synthesis Abdul-Quader, 2013 [29] P: PWID I: Structural interventions that address external factors to reduce HIV and HCV C: any quantitative comparison O: Changes in HIV/HCV in relation to intervention S: controlled or before & after 2011 Cochrane Central Register of Controlled Trials, EMBASE, LILACS, PsycINFO, PubMed, and the Web of Science/Web of Social Science (6) WHO-Johns Hopkins 9-Point Rigour Scale. 15/15 (1 CC/Co; 14 OTH) Unclear Des Jarlais, 2013 [28] P: PWID in low-and middle income and transitional countries (LMICs) I: NSP C: any O: changes in prevalence or incidence of HIV, HCV, HIV/ HCV co-infection or changes over time in newly reported HIV or HCV cases of infection S: any 2011 EMBASE, NLM Gateway, PubMed, Web of Science (4) None 13/13 (13 OTH) High Fernandes et al. BMC Public Health (2017) 17:309 Page 5 of 15 Table 1 Key characteristics of the included systematic reviews and Risk of bias of each review (Continued) Hong and Li, 2009 [31] P: People with HIV/ AIDS in China I: Any behavioural intervention C: Any O: Rate of needle-sharing S: Any 2008 AIDSLine, EBSCO, FirstSearch, PsycINFO, PubMed (4) None 25/3 (1 RCT; 1 CC/Co; 1 OTH) High Jones, 2008 [12] P: PWID I: NSP C: any O: IRB, HIV, HVC S: RCT, controlled non-randomised studies, controlled and uncontrolled before and after studies, cross-sectional studies, cohort studies, case-control studies and ecological studies 2008 ACP Journal Club, ASSIA, CINAHL, Cochrane Library, EMBASE, EPPI-Centre databases, Health Information Management Consortium, IBSS, MEDLINE, National Research Register Archive, OpenSIGLE, Project CORK, PsycINFO, Sociological Abstracts, SOMED (15) Quality assessment tools developed by NICE Centre for Public Health Excellence, England and Wales and the Effective Public Health Practice Project, Canada 24/15 (2 RCT; 1 CC/Co; 12 OTH) Low Kall, 2007 [30] P: PWID I: NSPs C: control or comparison groups O: HIV incidence or prevalence S: any 2005 MEDLINE and PsyclNFO (2) None 16/16 (8 CC/Co; 8 OTH) High Wright, 2006 [33] P: PWID I: primary prevention intervention targeting injecting drug use C: any O: HCV prevalence and incidence S: intervention or observational studies 2003 Cochrane Library, CINAHL, EMBASE, MEDLINE and PsycINFO (4) Developed their own quality scale 18/12 (9 CC/Co; 3 OTH) High Tilson, 2006 [24] P: PWID I: NSP C: With and without OST; Placebo. O: IRB, HCV and HIV S: NR 2006 Cochrane Library, Grey Literature Report, EMBASE, PubMed, PsycINFO, Social Science Abstracts, Web of Science, WorldCat (8) GRADE 45/43 (28 CC/Co; 15 OTH) Unclear Gibson, 2001 [35] P:PWID I:NSP C: any O:HIV, HBV, HCV and IRB S:any 1999 Medline and PsycINFO (2) None 42/32 (19 CC/Co; 13 OTH) High Leonard, 1999 [25] P: PWID I: NSP C: Any comparator group (but none was not allowed) O: HIV, HCV and IRB S: NR 1999 AIDSLINE, CINAHL, Cochrane Library, EMBASE, MEDLINE, PHEffect (6) Quality assessment tools developed by the Effective Public Health Practice Project team, Canada 21/19 (15 CC/Co; 4 OTH) Unclear a Categorized by type of synthesis and listed by publication date b PICOS: acronym for Population, Intervention, Comparator, Outcome and Study Design c Classified by authors of the review d CC/Co: Case-Control/Cohort; RCT: Randomized Control Trial; OTH: other (time-series designs, before-after studies, ecological studies) Fernandes et al. BMC Public Health (2017) 17:309 Page 6 of 15 as types of designs of included studies. Only four reviews specified the types of study design in their eligibility criteria [12, 26, 29, 33]. Three reviews included randomized controlled trials (four RCTs overall) [12, 26, 31], while the remaining reviews included cohorts and case-controls (101 overall), as well as other study designs (time series, before/after studies or ecological studies) (95 overall). Three of the included reviews were restricted to studies conducted in specific countries or economic regions (China, UK, and lowand middle-income countries) [28, 31, 32]. Nine reviews reported data on HIV, eight on HCV, and six on IRB. Seven reviews evaluated more than one of these outcomes. Four reviews used meta-analysis [26, 27, 32, 34], one of which using individual-participant data [32]. The remaining nine reviews used narrative synthesis [12, 24, 25, 28–31, 33, 35]. The reasons reported by the authors for conducting or not conducting meta-analysis varied between reviews, even when there was a degree of overlap between included studies. Methodological quality Seven reviews reported having assessed the methodological quality of included studies [12, 24, 25, 27, 29, 33, 34]. In two reviews the authors developed ad hoc tools [33, 34], while the remaining reviews used previously existing instruments, often with modifications (tools used: Newcastle-Ottawa, GRADE, WHO-Johns Hopkins 9Point Rigour Scale, Quality assessment tools developed by NICE Centre for Public Health Excellence, England and Wales and the Effective Public Health Practice Project, Canada) [12, 24, 25, 27, 29]. Only two instruments were used more than once, and all other instruments varied. Not all reviews reported the results of these assessments; when reported, most studies were considered to have low methodological quality, given methodological weaknesses and biases related to observational study designs. Only one quantitative review [27] incorporated quality in data synthesis, by conducting a sensitivity analysis. In reviews without meta-analysis, methodological quality and risk of bias was presented descriptively in the interpretation of study results. Two reviews [24, 27] reported using the GRADE tool to assess the quality of the body of evidence. One considered the overall qualityofevidenceaslow[27],whiletheother,using an adaptation of GRADE, evaluated the quality of Fig. 2 Studies selected for inclusion classified by outcome(s) reported and strategy for data synthesis. *Grey shaded boxes represent the outcome reported in the reviews indicated in the right column. “HIV”,”HCV”and “Injecting risk behaviours”is used to classify the reviews that reported the impact of NSP in the number of HIV infections, number of HCV infections and change in injection behavior. The reviews are also presented in the right column by strategy used for data synthesis Fernandes et al. BMC Public Health (2017) 17:309 Page 7 of 15 evidence as modest to moderate for different outcomes [24]. Using the ROBIS tool, only two of the included reviews were considered to have low risk of bias, whereas the remaining were considered to have either high (n=8)or unclear risk of bias (n= 3) (summary of results available in a supplementary file: Additional file 3). The majority of reviews were rated as having high or unclear risk of bias across all ROBIS domains, i.e. study eligibility criteria, identification and selection of studies, data collection and study appraisal, synthesis and findings. Percentage of agreement between ROBIS raters varied between 77% and 92%, and weighted kappa (quadratic) between 0.82 and 0.95. The results from the PRISMA-P statement applied to this overview are presented at Additional file 4. Effects of NSP HIV prevalence and/or incidence Of nine reviews that reported data on HIV outcomes, two were at low risk of bias [12, 27], and only one performed meta-analysis [27]. The proportion of studies included in only one of these nine reviews was 71%. Reviews with meta-analysis The most recent review by Aspinall et al. included twelve studies (10 cohort, one case-control and one cross-sectional study), and was evaluated as having low risk of bias. The review reported a 34% “risk”reduction of HIV transmission (pooled effect estimate of 0.66; 95% CI: 0.43 to 1.01; n=10 studies; I 2 = 76%) in individuals exposed to NSP, compared with those who were not, or were less frequently, exposed to NSP. This estimate was obtained by pooling both adjusted and unadjusted effect sizes regardless of study design, in a random effects model, and pooling all types of effect measures (odds ratio, risk ratio, hazard ratio). Sensitivity analyses supported these results, with a statistically significant reduction in HIV transmission associated with NSP exposure in both higher quality studies (0.42; 95% CI: 0.22 to 0.81), and studies reporting relative risks or hazard ratios (0.60; 95% CI: 0.37 to 0.97). Subgroup analyses showed that studies using sequential follow-up (i.e. groups exposed to NSP and groups not exposed followed sequentially, and not concurrently) and studies recruiting post-1990 had lower effect estimates (0.21 [95% CI: 0.11 to 0.41] and 0.52 [95% CI: 0.28 to 0.95], respectively). Further, studies comparing 100% NSP coverage (i.e. clean needle and syringe used for 100% of injections) with <100% NSP coverage generated a pooled effect estimate of 0.58 (95% CI: 0.22 to 1.57). Quality of evidence was considered low using GRADE, and most of these results had substantial heterogeneity. Reviews without meta-analysis The quantitative results presented above are supported by most reviews with qualitative summaries of the evidence only, all of which published between 1999 and 2013. Differences between reviews could be found regarding eligibility criteria (e.g. type of NSP, populationvs individual-level outcomes, study design), search and screening methods, quality assessments and interpretation of study results. In most of these reviews, the authors described and contextualized results from each individual study or group of studies regarding HIV outcomes. Some reviews used vote counting to compare the number of positive studies with reduction in HIV transmission, with the number of negative studies. In the following paragraphs, we present the main results of these reviews, starting with those focusing on the impact of different aspects of NSP provision followed by results of reviews focusing exclusively on the comparison between NSP andno NSP. Regarding aspects of NSP provision, Jones and colleagues reported results from a recent systematic review classified as low risk of bias, that focused on level of coverage, syringe dispensation policies, type of NSP, provision of additional harm-reduction services and of opiate substitution therapy [12]. Authors judged that the range of study designs, intervention approaches examined and outcomes precluded the use of meta-analysis. Only three studies focused on HIV outcomes. A low-quality study showed no significant trend for HIV prevalence when comparing primary sources of needles (pharmacies, fixed site NSP and van-based NSP), although HIV prevalence was lower among pharmacy users than in participants who reported using van or fixed site NSP (16% vs. 21% and 25%, respectively, p= 0.16) [36]. Another study [37] included in this review examined the impact of dispensation policies, and noted a decrease in HIV prevalence (based on testing or self-report) between the period of legal pharmacy syringe purchase and when up to five needles could be exchanged at newly established NSP (35% to 22%; p< 0.05). Finally, one moderate quality cohort study [38] also included in the review by Jones and colleagues evaluated different levels of harm reduction and found that a full harm reduction strategy (combination of methadone treatment and full participation in NSP) reduced the incidence of HIV when compared to incomplete or no harm reduction (incidence rate ratio: 0.32; 95% CI: 0.17 to 0.62). In line with these results, a review with an overall unclear risk of bias by Abdul-Quader et al. [29] identified studies with structural-level NSP, ie interventions in which changes in policy and legal environment have facilitated an increased availability of sterile syringes, and focused on population-level outcomes. The operational definition of structural-level NSP was a minimum 50% coverage of PWID and distribution of 10 or more needles/syringe per PWID per year. Nine studies Fernandes et al. BMC Public Health (2017) 17:309 Page 8 of 15 included in this review reported decreases in HIV prevalence, and three reported decreases in HIV incidence. All studies were non-randomized before-after comparisons or interrupted time series analyses, and most showed evidence of potential selection biases. The authors concluded that these results support NSP as a structural-level intervention to reduce population-level infection. Four older published reviews with qualitative synthesis and an overall unclear or high risk of bias reported mixed findings regarding the impact of NSP on HIV incidence/prevalence. These reviews provide a historical perspective on the evolution of NSP implementation and evaluation. The first published review by Leonard et al. [25], was an update of a previous systematic review. Gibson et al. [35] included six studies published up to 1999 with a range of study designs, participants and settings, reporting on HIV outcomes. The Institute of Medicine’s evidence report [24], identified 12 relevant studies, including cohort, case-control and ecological designs, as well as studies using mathematical models. Finally, the review by Kall et al. [30] included 16 studies published up to 2005, only two of which were not included in other reviews from this overview. All these reviews included landmark prospective cohort studies conducted in Montreal and Vancouver in the 1990s [39, 40], which found an association between NSP participation and higher risk of HIV seroconversion. This led Leonard et al. to conclude that there was methodologically weak evidence that NSP were not as effective as previously found in modifying HIV prevalence and incidence among PWID. Kall et al. used vote counting and reported that on most studies assessing seroincidence the effect of NSP was not significant, while four studies investigating seroprevalence at baseline were unfavourable to NSP. The authors also stated that in studies that found positive effects, confounders had not been adequately controlled for. However, based on epidemiological evidence that accumulated progressively, authors of the other reviews highlighted a number of selection biases that could account for these findings, including: the inclusion of high-risk cocaine injectors, who injected more often than heroin users; the limited number of needles and syringes that users could have access to in early NSP; and the ready availability of clean injecting equipment through pharmacies which could have attracted marginalized, particularly high-risk individuals. Follow-up studies in the same settings, conducted after expanding and adjusting NSP (e.g. by allowing unlimited distribution of needles/syringes, increasing the number of access points, and offering different distribution methods), found no such increase in risk, or a decrease in HIV prevalence. Gibson et al. [35] used vote counting and considered there was substantial evidence that NSP were effective in preventing HIV seroconversion. Tilson et al. [24] included four ecological studies that found an association between HIV prevention programmes that include NSP with reduced prevalence of HIV in urban settings. Based on the weakness of these studies designs, this evidence was considered modest using a modified GRADE approach. Further, moderate evidence was found that multi-component HIV prevention programmes that include NSP reduce intermediate HIV risk behaviour. The Institute of Medicine report highlighted how almost all published studies originate in North America, Western Europe, and Australia [24]. Two additional reviews were restricted to specific populations by geographical or economical source. Hong and Li summarized evidence from two studies conducted solely in China [31], while Des Jarlais et al. focused on 13 studies of 11 NSPs with high-coverage conducted in low/middle-income countries [28]. In both cases, results from included studies generally supported the effectiveness of NSP in reducing HIV. Des Jarlais et al. reported a reduction of HIV prevalence in four studies (from −3% to −15%), of estimated HIV incidence in three studies (from −11/100 to −16/100 personyears at risk), and of newly reported nationwide cases in three national reports (from −30% to −93.3%). Conversely, increases in HIV prevalence were found in two studies (from +5.6% to +15.8%) and one national report (+37.6%) included in the review by Des Jarlais et al. The authors considered that, if high coverage is achieved, NSP appear to be as effective in low/middle-income as in high-income countries [28]. HCV prevalence and/or incidence Eight included reviews synthesized the evidence on the use of NSP in preventing HCV prevention in PWID. One was rated as being at low [12], three at unclear [24, 25, 29] and four at high risk of bias [28, 32–34], including two reviews that used meta-analysis [32, 34]. The proportion of studies included exclusively in one of these reviews was 88%. Both reviews with quantitative and qualitative synthesis showed mixed results. Reviews with meta-analysis Two reviews [32, 34], both published in 2011 and rated as being at high risk of bias, used meta-analysis. Hagan et al. [34] included 7 studies focusing on NSP and HCV outcomes (6 cohort and 1 case-control study), all from North America. The pooled analysis of all studies, using random effects models and with all measures of effect converted to relative risks, showed an increase in the risk of HCV acquisition with NSP (relative risk, 1.62; 95% CI: 1.04 to 2.52). There was considerable heterogeneity (I 2 = 81%), but no subgroup or sensitivity analyses were performed, and study quality was not explicitly reported or considered in the analysis. Authors cautioned Fernandes et al. BMC Public Health (2017) 17:309 Page 9 of 15