Inadequate needle and syringe coverage among people who inject psychoactive drugs across England and Wales

Abstract Aim Needle and syringe (NS) provision is a proven intervention for reducing harms associated with injecting drug use, such as infections, but impact is coverage-dependent. We characterised people who injected drugs (PWID) in England and Wales who had insufficient NS in the past month to meet their injecting requirements. Methods This study utilised 2017–2019 data from the annual Unlinked Anonymous Monitoring (UAM) Survey of PWID recruited through specialist services. Logistic regression was used to identify factors associated with inadequate NS coverage. Findings Of 2,442 PWID surveyed who had injected in the past month, 34% reported inadequate NS coverage (or as high as 51% if including unsuccessful injections). Younger PWID (adjusted odds ratio: 2.18, 95% confidence interval: 1.35–3.52), those who began injecting in the past three years (1.46, 1.09–2.00), and those who reported sharing injecting equipment (1.46, 1.22–1.75) had greater odds of having inadequate NS coverage. PWID currently prescribed Opioid Substitution Treatment (OST) had lower odds of inadequate NS coverage (0.60, 0.49–0.73). Conclusion Given the poor reported coverage of NS provision in England and Wales, there is urgent need to address inequity in accessing sufficient NS and increase coverage among this vulnerable group to reduce injecting-related harms.


Introduction
People who inject drugs (PWID) are vulnerable to a wide range of poor health outcomes.Healthcare and harm reduction services can be difficult for this group to engage with, despite high rates of blood-borne viruses (BBVs), soft tissue infections, overdose, and other health hazards associated with injecting psychoactive drugs (Motavalli et al., 2021).Needle and syringe programmes (NSP) have been available in England since the mid-1980s, when they were introduced in response to the emerging HIV/AIDS epidemic, in an effort to limit transmission of BBVs among the PWID population (Jones et al., 2008).NSP provide sterile injecting equipment and advice, to reduce the reuse and sharing of injecting equipment and the risk of injection related harms (Hope et al., 2014).
There is ample evidence that, in combination with opioid substitution treatment (OST), provision of needles and syringes (NS) reduces BBV infections and associated risk behaviours, but their impact is coverage-dependent (Fernandes et al., 2017;Palmateer et al., 2022;Platt et al., 2017Platt et al., , 2018)).Additionally, contact with NSP provides the opportunity to engage with PWID who do not routinely access other health services and give harm reduction advice, offer testing and vaccination, provide treatment, and referrals for specialist support (Public Health England, 2020a).
Guidance from the National Institute for Health and Care Excellence (NICE) recommends extending and increasing the provision of sterile injecting equipment to ensure PWID have sufficient sterile NS for every injection (National Institute for Health and Care Excellence (NICE), 2014).However, the estimated coverage of NS provision remains insufficient, with just 33 NS provided per PWID per year globally in 2017, far from the recommended target of at least 200 NS per PWID per year by 2020, and 300 NS per PWID per year recommended by the World Health Organization to achieve hepatitis C virus (HCV) elimination by 2030 (World Health Organization, 2021).In Scotland, an average of just 47 NS were supplied per PWID between 2020 and 2021: a decrease from 53 the previous year (Public Health Scotland, 2021).In Wales, NS coverage is also suboptimal, with 80 syringes provided per PWID in 2020-2021(Public Health Wales, 2021).In North West England, the Integrated Monitoring System reported an average of 181 needles provided by services across Cheshire and Merseyside per PWID per year (Whitfield & Reed, 2021).There remains no reliable national estimate available for England (European Monitoring Centre for Drugs and Drug Addiction (EMCDDA), 2022).
While rates of HIV in the injecting population have remained relatively low in the UK (Croxford et al., 2022), more than half of PWID have been exposed to HCV, and chronic prevalence of HCV infection among recent initiates to injecting has remained stable around 28% over the past ten years, despite the introduction of direct-acting antivirals in 2015 (Public Health England, 2020b).Furthermore, cases of serious bacterial infections associated with injecting drugs, including invasive Group A streptococcus (iGAS) and Staphylococcus aureus, continue to be a public health issue (Public Health England, 2020c).Improving access to sterile injecting equipment to PWID at highest risk of having inadequate NS coverage would reduce the need for sharing and reusing, and therefore lower the risk of infection.
Coverage of NS provision can be measured using two approaches.Population-level measures are widely used for coverage monitoring and are based on the total amount of NS provided in relation to the estimated size of the whole PWID population, i.e., the number of NS distributed per PWID (Public Health England, 2017;World Health Organization, 2021).Alternatively, individual-level measures present coverage of adequate NS among PWID, based on the NS received by individuals in relation to their personal requirements, i.e., the number of NS a PWID has available per injection (Bluthenthal et al., 2007).
In this study, we characterise PWID who report inadequate NS coverage at the individual level, describing their demographics, social factors, risk behaviours, and uptake of health care services compared with those who have adequate NS provision.We also investigate factors associated with inadequate coverage of NS to focus future actions.This national data is particularly important given recent changes to funding and harm reduction provision in the UK (Department of Health and Social Care, 2021).

Methods
This study utilised data from the UK Health Security Agency's (UKHSA) Unlinked Anonymous Monitoring (UAM) Survey of PWID.More detailed methodology has been published previously (Public Health England, 2020c).Briefly, this annual survey recruits people who have ever injected psychoactive drugs via a range of specialist drug and alcohol services (e.g., services that provide treatment and harm reduction) across England, Wales, and Northern Ireland.Participants are asked to give a dried blood spot sample and self-complete a short questionnaire on their demographic and social factors, and risk behaviours.Those who provide consent and then participate are offered an acknowledgement (a retail voucher).Dried blood spot samples are tested for HIV, hepatitis B and C, and are linked with a unique identifier to the respondent's questionnaire data.All results are anonymous and cannot be traced back to the individual participant.The UAM Survey has ethics approval from London Research Ethics Committee (98/2/051) and UKHSA.
The data used were from three survey waves: 2017, 2018 and 2019.The survey questionnaire is regularly reviewed, and all data items needed for this analysis have been collected since 2017.Data for 2020 were not included as public health measures introduced in response to the COVID-19 pandemic limited face-to-face contact between PWID and drug and alcohol services, thus negatively affecting recruitment to the UAM Survey.Data from Northern Ireland were excluded due to small sample size.Any questionnaires received by respondents in 2018 or 2019 who indicated by self-report that they had already taken part in the survey earlier in the three-year period (N = 153) were excluded (i.e., the analysis only used someone's first participation during the three-year period).Analyses were restricted to participants reporting injecting drugs within the past month (28 days), for the purposes of calculating NS coverage.This calculation used four questions asked in the survey: 'How many times do you visit a needle exchange in a typical month?,' 'How many needles do you typically collect during each visit?,' 'In the last month, on how many days have you injected drugs?,' and 'On the last full day that you injected, how many times did you inject drugs?' NS coverage was calculated as the proportion of 'needles required' (injections per day x days injected per month) divided by 'needles collected' (needles collected per visit x visits per month).PWID who collected 100% or more of the needles required were categorised as having 'adequate' NS coverage, whereas those collecting less than 100% of their required needles were categorised as having 'inadequate' NS coverage.
Needing to insert a needle multiple times before being able to access a vein can increase the risk of developing injection site infections; ideally, PWID should have enough NS for all injection attempts (Hope et al., 2016).To calculate NS coverage including unsuccessful injection attempts, answers to the question 'Last time you injected how many times did you insert a needle before getting a 'hit?,' which were limited to '1′, '2′ '3′ or '4+' , were multiplied (1, 2, 3 or 4x) by the number of times the participant injected on the last day they injected drugs.
All statistical analyses were performed using STATA v17.Firstly, descriptive analyses were carried out, presenting differences between those with inadequate and adequate NS coverage.Differences between groups were compared using χ 2 and 2-sample t-tests for categorical variables, and a Mann-Whitney U test for continuous variables.Univariable analyses were then conducted to identify factors associated with inadequate NS coverage, and a multivariable model was created with variables that best predicted inadequate NS coverage using a backward stepwise methodology, adjusted for potential confounders including gender, age, region and year of survey (statistical significance p < 0.05).
Only questionnaires with answers to all questions used to calculate NS coverage were included in the dataset.To test whether this sample was representative of PWID participants, χ 2 tests and 2-sample t-tests were used to compare the proportion of PWID within key variable subgroups included in the analysis with the proportion who were not.

Results
Of the 2,442 PWID included in the analyses, 34% reported inadequate NS coverage.This proportion remained similar across the three survey years (p = 0.468; Table 1).When the participants reporting previous participation within the period were included (N = 153), the difference in NS coverage by year remained nonsignificant (p = 0.568).The median coverage among PWID with inadequate NS was 48% (IQR: 19%-71%), while among PWID with adequate NS, median coverage was 250% (IQR: 150%-500% Around three quarters (74%) of respondents were male, and there was no significant difference in NS coverage by gender (p = 0.310; Table 2).Those with inadequate NS coverage were slightly younger (median: 38 years [IQR: 32-44] vs. 39 years [IQR: 34-45]; p = 0.016) and a higher proportion initiated injecting within the past three years than among those with adequate NS coverage (12% vs. 8% p = 0.001; Table 2).Difference in NS coverage by region of England and Wales was not statistically significant (p = 0.152; Figure 1).
Receptive and/or distributive sharing of injecting equipment (including needles, syringes, containers, spoons and filters) in the past month was more commonly reported among PWID with inadequate NS coverage than those with adequate NS coverage (41% vs. 32%; p < 0.001; Table 2).A lower proportion of PWID with inadequate NS coverage reported having ever had a blood test for HCV (87% vs. 90%; p = 0.008) or HIV (79% vs. 85%; p < 0.001), or were currently prescribed OST (68% vs. 79%; p < 0.001; Table 2) than those with adequate NS coverage.
After adjustment through multivariable analysis, factors associated with inadequate NS coverage were younger age group, initiation of injecting within the past 3 years, current OST prescription and sharing injecting equipment (Table 3).PWID aged under 25 years had more than twice the odds of having inadequate NS coverage than those aged 35 years and over (adjusted odds ratio [aOR] = 2.18 [95% confidence interval (CI): 1.35-3.52],p = 0.001).Those who had initiated injecting within the past three years had 46% greater odds of having inadequate NS coverage (aOR = 1.46 [95% CI: 1.09-2.00],p = 0.012).PWID with a current OST prescription had 40% lower odds of having inadequate NS coverage (aOR= 0.60 [95% CI: 0.49-0.73],p < 0.001), while PWID who reported sharing injecting equipment had 46% greater odds of inadequate NS coverage (aOR = 1.46 [95% CI: 1.22-1.75],p= <0.001).Injecting and non-injecting use of powder cocaine did not significantly contribute to the multivariable analysis, so were dropped from the final model.
More than half (56%) of all PWID surveyed reported more than one insertion of a needle before successfully accessing a vein on their last injection attempt.A greater proportion of PWID with inadequate NS coverage reported more than one needle insertion than those with adequate NS coverage (60% vs. 54%; p = 0.006).When inadequate NS coverage was calculated, adjusting for the number of needle insertions (i.e. one NS per injection attempt rather than per successful injection), the proportion of PWID with inadequate NS coverage rose to 51%.
The dataset used for calculation of NS coverage included just 60% of the of the PWID who had injected in the past month in the UAM dataset for the three-year period, due to missing data for at least one data item needed to calculate coverage.Despite this, the data were broadly demographically representative of the wider UAM dataset, except for PWID born outside of the UK, who were underrepresented (5% of study dataset; 8% of UAM dataset; p < 0.001; Supplementary Table 1).

Discussion
This is the first study using data from a nationally reflective sample of PWID in contact with services for people who use drugs in England and Wales to investigate coverage of NS provision.We estimate that one-third (34%) of PWID in England and Wales report inadequate NS coverage.However, this could be as high as half (51%) of PWID if multiple needle insertions are accounted for, under the premise that sterile needles should be available for each injection attempt.These results are particularly concerning given the implications for BBV transmission and the risk of skin and soft tissue infections.We found that those with inadequate NS coverage tended to be younger, to be more recently initiated injecting, to report sharing of injecting equipment, and not to be currently prescribed OST, than those with adequate NS.This highlights particular vulnerability among this subgroup of PWID and indicates a need for targeted harm reduction interventions.Population-level measures of NS coverage are used for international programme monitoring and guidance, but they may mask inequity in distribution or accessibility between groups of PWID (O'Keefe et al., 2019).Australia, for example, has among the highest population-level NS coverage worldwide (Larney et al., 2017); however individual NS coverage remains insufficient for 20-37% of PWID (Iversen et al., 2012).Our findings are similar to those of Bryant et al. (2012), where despite a median NS coverage of 150% among all PWID recruited in New South Wales, coverage was inadequate for the requirements of 37% of individuals.Population-level and individual-level coverage  c People who began injecting in the last three years.d Drug(s) other than heroin, crack cocaine, powder cocaine or amphetamine, not specified.e sharing of needles, syringes, mixing containers or filters among those who had last injected in the four weeks preceding survey participation.f reported visit to general practice (gP), pharmacy, accident and emergency (a&E) or sexual health services in the last year.g current or previous year.h at least one dose.
i variable combining all current Hiv, Hcv, HBv results.A multitude of interacting structural and individual factors can prevent service users receiving adequate NS coverage beyond simply offering sufficient equipment (Strathdee & Vlahov, 2001).Structural and social barriers to accessing healthcare services include housing status, incarceration, and location, while psychosocial barriers include societal stigma, perceived bias, mistrust of the health care system, and fear of criminalisation (Motavalli et al., 2021).The effects of these barriers are often compounded by intersectional health inequalities determined by gender, housing status, mental and physical health, sexuality, sex work, first language, ethnicity, and socioeconomic status (Gibson & Hutton, 2021;Nambiar et al., 2014;Wood et al., 2005).
This study identified demographic factors and social determinants, including younger age and recent initiation to injecting drugs, to be associated with adequate NS coverage (Tables 2 and 3).Younger PWID, and recent initiates to injecting drugs, may be less aware of available services and may have less knowledge and experience of safe injecting practices (Grund et al., 2009).Higher rates of risky injecting behaviours, including assisted injecting, and BBV infection have been reported among young PWID and those recently initiating injecting (Cheng et al., 2016;Ghiasvand et al., 2018).Young PWID and recent initiates to injecting are therefore underserved and particularly important groups to engage and target for harm reduction interventions.
Following adjustment in multivariate analysis, cocaine injection was not a significant predictor of inadequate NS.This is despite the fact that injection of stimulants, such as cocaine, is known to be associated with a higher injection frequency, and so a higher needle requirement, due to the shorter half-life of this drug class (Tyndall et al., 2003).This finding may be due to the confounding effect of another variable in the model, such as age or recent initiation to injecting, but requires further investigation.Injection of crack and powder cocaine have increased in the UK over the past decade (Public Health England, 2020a).
If stimulant injection rates continue to rise, NS provision and access must meet the increased requirement to maintain sufficient coverage.This is particularly important given the context of the recent and, in some cases, ongoing HIV outbreaks among PWID found to be associated with increases in stimulant injection in Europe (Arendt et al., 2019;Croxford et al., 2022; European Monitoring Centre for Drugs and Drug Addiction (EMCDDA) & European Centre for Disease Prevention and Control (ECDC), 2011; Giese et al., 2015;McAuley et al., 2019), and the high prevalence of HCV and skin and soft-tissue infection among PWID in the UK (Public Health England, 2020a).
We found that inadequate NS coverage was associated with sharing injecting equipment (either receptive or distributive).Though this association is unsurprising, it is of concern due to the associated risk of acquiring and transmitting bacterial infections and BBVs, particularly in the context of the recent outbreaks mentioned above.Around 20% of PWID in England, Wales and Northern Ireland report direct sharing of needles or syringes, while 37% report sharing any injecting equipment including filters, spoons and containers, and these practices have not declined in the last decade (Public Health England, 2020a).Reducing the need for sharing and reusing injecting equipment by ensuring sufficient accessible NS and so adequate coverage, including provision of low dead-space syringes (LDSS), is critical.While harm reduction approaches that combine equipment provision and BBV testing are most beneficial, NS provision is among the most cost-effective interventions available (Sweeney et al., 2019;Wilson et al., 2015).
There was some indication that PWID with inadequate NS coverage were less engaged by healthcare services than those with adequate NS coverage, as PWID with a current OST prescription had lower odds of having inadequate NS than those without a current prescription (Table 3), and a smaller proportion had ever had blood tests for HIV or HCV, though other measures of engagement with healthcare (such as naloxone carriage, recent BBV testing, and vaccination) did not reach statistical significance (Table 2).Lack of engagement with healthcare services may reduce opportunities to collect injecting equipment.These findings have further implications for harm reduction and health outcomes at an individual level, as well as for limiting the spread of communicable disease across populations.
Since the data were collected for this study (2017-2019), many of the issues described here have been exacerbated by the ongoing coronavirus disease (COVID-19) pandemic.In 2020, 26% of PWID surveyed reported difficulties in accessing equipment for the safe use or injection of drugs (Croxford et al., 2021).In addition, 15% reported injecting more frequently than in 2019, and those reporting injecting any form of cocaine in the past month rose from 17% in 2019 to 25% (Croxford et al., 2021).Some PWID reported that changes in their drug usage were the product of increased social isolation and worsened mental health resulting from lockdown restrictions (Kesten et al., 2021).Many drug and alcohol services were not able to open and see clients face-to-face during the UK national lockdown, and as a result, NS provision declined by almost one third in England, and by around 26% in Wales (Public Health Wales, 2021;Whitfield et al., 2020).The effects of these barriers are compounded by other effects of the pandemic, including increased rates of homelessness and unstable housing, and reduced access to healthcare services (Public Health England, 2020d).
There are a number of limitations to our study.Participants in the UAM Survey are, by recruitment design, engaged in the range services offered to people who use drugs; this survey does not capture people who are not currently in contact with any services and may not have any access to injecting equipment.Moreover, the calculation of NS coverage used here does not account for the number of needles picked up for others, received from others via secondary distribution, or the extent of stockpiling equipment, which has been shown to increase the proportion of PWID with adequate NS by up to 8% (McCormack et al., 2016).Since 2020, the UAM Survey has also asked participants where they most often source their injecting equipment.It will be useful in future analyses to determine whether PWID collecting fewer needles from services are more likely to source them elsewhere, and where from.Additionally, it was not possible to investigate the relationship between current infection by individual BBV and inadequate NS coverage due to the sample size.
The data presented in this study must be caveated by acknowledging that our approach for estimating NS coverage was relatively simple, based on participants' answers to 'On the last full day that you injected, how many times did you inject drugs?' multiplied by the number of days they reported injecting in the last month.This assumed that the last day injected was representative of an average day.Calculation of NS coverage including unsuccessful injection attempts was based on answers to the question 'Last time you injected how many times did you insert a needle before getting a 'hit'?'; this also involved an assumption that the last day injected was representative of an average day and a participant taking more than four injection attempts could not be captured based on the available answers.This measure, although not perfect, utilises the data available to provide an 'upper estimate' to the potential extent of inadequate provision.PWID can often have difficulty accessing veins due to vascular damage and poor injection technique (Rhodes et al., 2007) and each injection attempt is associated with risks of injection site infections (Hope et al., 2016).Some potentially important structural factors were not collected as part of this study.For example, PWID living in rural locations may face more barriers to receiving adequate NS coverage than PWID in urban areas, as local services may be harder to access, requiring long-distance travel (Harm Reduction International, 2020).While we were able to separate and compare the questionnaire data from PWID across regions, smaller geographical breakdowns, necessary to understand rural versus urban coverage, were not possible due to small sample sizes.
There is significant variation in the provision of drug services by local authorities across the UK, with facilities provided by a network of different charities and organisations (Department of Health and Social Care, 2021).Funding for drug services has decreased by 26% in the last five years and, due to this, there has been a shift away from specialist services and towards pharmacy-based distribution of injecting equipment.This may improve accessibility to NS for some PWID, but will reduce access to the plethora of other harm reduction interventions that specialist drug and alcohol services can provide (Harm Reduction International, 2020).Among specialist services, there is support for an increasingly peer-led approach, to capitalise on and appropriately value the experience of PWID, to help minimise some of the stresses associated with using services, such as stigma and mistrust (Henderson et al., 2017).The COVID-19 pandemic has also driven a push towards online procurement systems for injecting equipment, which can now be ordered by service users or peers for free postal delivery (Exchange Supplies, 2021).There are concerns regarding the accessibility of these services to PWID who may be less able to access the internet or have limited literacy, but online services may hold promise for improving reach to younger PWID (Genz et al., 2015).NS should be available from a variety of sources, including outreach, traditional NSP services, pharmacy and online distribution, to cater to a varied population of PWID (Craine et al., 2010).
In 2021, following an independent review of drug use in the UK by Professor Dame Carol Black (Department of Health and Social Care, 2021), the government announced its 10-year drug strategy to tackle harmful drug use (UK Government, 2021).Promisingly, this included substantial investment in drug and alcohol services; however, identifying those most at risk for targeted intervention will be vital to ensuring these resources are utilised to their fullest potential (O'Keefe et al., 2019).The UAM Survey provides a sentinel approach by monitoring individual-level NS coverage among the PWID attending the network of drug services which participate, but there is also a need for a national population-level NS monitoring system to gain a more complete understanding of the extent of NS provision across the UK.Provision of NS should be tailored in response to local needs and regularly reviewed to optimise accessibility, and form part of a more holistic and interdisciplinary approach to the complex and varied needs of PWID, such as that laid out by the partnership between NHS Wales, Public Health Wales, and the South Wales Police and Crime Commissioner (Smith & Metcalfe, 2020;Welsh Government, 2011).Future research should focus on effective ways to reduce barriers to accessing adequate NS among high-risk PWID.
In conclusion, this study demonstrates the need for a targeted approach to increasing NS coverage among those particularly vulnerable, characterised here by younger age and recent initiation to injecting, to reduce harms associated with injecting psychoactive drugs.PWID represent an often marginalised population, disproportionately affected by infectious diseases and the challenges of service access.Harms have been exacerbated by the COVID-19 pandemic, which, since this study, has likely reduced access to NS even further.Given recent disease outbreaks among PWID, in combination with the significant proportion of PWID reporting inadequate NS coverage described here, it seems inevitable that further outbreaks are imminent.It is therefore critical to improve NS coverage monitoring nationally, increase access to safe injecting equipment to those in high-risk groups, and to capitalise on opportunities for harm reduction and interventions.
monitoring can be used to complement one another, as incorporating individual-level coverage monitoring allows identification of factors that put groups at higher risk of inadequate NS, and the more focused monitoring of requirement and risk among these groups (O'Keefe et al., 2019).There remains no national-level, high-quality population-level monitoring system for injecting equipment provision or coverage across the UK; although systems are in place to monitor NS provision in Scotland, Wales and a small number of locally driven initiatives in England (Public Health Scotland, 2020; Public Health Wales, 2021; Whitfield & Reed, 2021).

Figure 1 .
Figure 1.Map of inadequate needle and syringe coverage among people currently injecting drugs by region: England and Wales, 2017-2019.

Table 1 .
ns coverage by year of uaM survey: England and Wales, 2017-2019.

Table 2 .
summary of ns coverage among people currently injecting drugs by demographic factors, social determinants and risk behaviours:England and  Wales, 2017-2019 (N = 2,442).
People who began injecting in the last three years.b Drug(s) other than heroin, crack cocaine, powder cocaine or amphetamine, not specified.c sharing of needles/syringes, mixing containers or filters among those who had last injected in the four weeks preceding survey participation.d reported visit to general practice (gP), pharmacy, accident and emergency (a&E) or sexual health services in the last year, offered a Hiv or Hcv test this year or last year, or currently prescribed opioid substitution therapy (ost).e variable combining all current Hiv, Hcv, HBv results. a