Funding and Delivery of Syringe Services Programs in the United States, 2022

Abstract

Objectives. To describe the current financial health of syringe services programs (SSPs) in the United States and to assess the predictors of SSP budget levels and associations with delivery of public health interventions.

Methods. We surveyed all known SSPs operating in the United States from February to June 2022 (n = 456), of which 68% responded (n = 311). We used general estimating equations to assess factors influencing SSP budget size and estimated the effects of budget size on multiple measures of SSP services.

Results. The median SSP annual budget was $100 000 (interquartile range = $20 159‒$290 000). SSPs operating in urban counties and counties with higher levels of opioid overdose mortality had significantly higher budget levels, while SSPs located in counties with higher levels of Republican voting in 2020 had significantly lower budget levels. SSP budget levels were significantly and positively associated with syringe and naloxone distribution coverage.

Conclusions. Current SSP funding levels do not meet minimum benchmarks. Increased funding would help SSPs meet community health needs.

Public Health Implications. Federal, state, and local initiatives should prioritize sustained SSP funding to optimize their potential in addressing multiple public health crises. (Am J Public Health. 2024;114(4):435–443. https://doi.org/10.2105/AJPH.2024.307583)

Over the past decade, morbidity and mortality among people who inject drugs (PWID) has steadily increased throughout the United States. A total of 106 699 unintentional drug overdose deaths occurred in the United States in 2021,¹ with a 30% increase from 2019 to 2020² and another 14% increase from 2020 to 2021.¹ Viral infections such as HIV and hepatitis C³ as well as skin and soft tissue infections⁴ remain prevalent among PWID. Syringe services programs (SSPs) are an evidence-based, low-threshold public health intervention designed to reduce the risk of infections and fatal overdose for PWID, if supported adequately.⁵

SSPs were first implemented in the 1980s as a direct response to spread of viral hepatitis and HIV among PWID and, over time, have become recognized as one of the most cost-effective HIV prevention interventions.⁶ Over the years, many US-based SSPs have broadened their delivery to include provision of equipment for safer snorting or smoking of drugs, naloxone kits and training to identify and reverse opioid overdoses, drug checking services, wound care, and education on safer drug use and infection prevention. Many SSPs also provide basic medical care, infectious disease screening and treatment, and linkage to psychosocial care and support for basic needs such as food and housing. SSPs are the most common place for PWID to seek substance use disorder treatment or basic medical care.^7,8 During the COVID-19 pandemic, the US federal government loosened regulations to allow buprenorphine treatment inductions via telehealth, which many SSPs implemented.⁹

SSPs provide services at no cost to their participants and are not-for-profit entities. As such, SSPs require funding from individual donations, foundations, or governmental agencies to operate; many are set up as mutual aid collaboratives, and many are volunteer-based before formally receiving funding. When first implemented in the United States, SSPs were often illegal and depended solely on private funding.¹⁰ Some local health departments started funding SSPs in the 1990s as they gained legal status, and some state health departments started funding SSPs in the late 1990s.¹¹ US federal funding for SSPs was not available until the past decade, has been limited, and until the 2020s has included many restrictions.¹²

In 2018, even before the COVID-19 pandemic led to costly supply chain disruptions and severe staffing challenges, Teshale et al. at the Centers for Disease Control and Prevention (CDC) estimated the minimum costs of running a comprehensive SSP.¹³ They did this accounting for costs of personnel (including a program director, part-time accountant, peer navigators, part-time nurse, and counselors), operational costs (i.e., lease or rent, insurance, utilities, mail, and janitorial), prevention services (i.e., syringes, injection equipment, naloxone, sharps containers, and hazardous waste management), onsite medical or testing services (i.e., point-of care testing for HIV and hepatitis C, hepatitis A and B vaccination, wound care, and pregnancy tests), and a mobile van unit to serve people who cannot or will not come to a fixed site (i.e., van leasing, maintenance, registration, gas, storage, and insurance). The estimated costs for running a comprehensive SSP ranged from US $400 000 for a small rural SSP (serving 250 clients) to US $1.8 million for a large urban SSP (serving 2500 clients).

Because governmental funding of SSPs has been sparse and varies greatly by geography, broader development of SSP organizational structures and staff is often not properly optimized. In turn, the public health benefits that SSPs can confer are geographically disparate, fluctuate over time, and do not typically deliver the full continuum of services that can improve the health and well-being of PWID.¹⁴

After decades of SSPs operating underground because of prohibitive laws and lack of public support, since 2020, the US government has officially supported SSPs as a critical public health intervention.⁵ In December 2021, the Substance Abuse and Mental Health Services Administration announced it would provide $30 million in grant awards to harm reduction programs through the American Rescue Plan,¹⁵ and, in 2022, the Biden administration awarded another $1.5 billion in funding for states and territories to address the opioid epidemic.¹⁶ These shifts have made it increasingly more common for state and local governments to provide funding for SSPs in their jurisdictions. Most of the states that fund SSPs have Democratic legislatures. Among states with Republican legislatures, existing laws often prevent operation of SSPs.¹⁷

To better understand the current funding environment of SSPs in the United States, we analyzed data from the 2022 National Survey of Syringe Services Programs (NSSSP) to (1) describe current funding levels, (2) assess the factors related to annual SSP budgets, and (3) understand how funding levels are associated with the amount and types of public health interventions that SSPs provide.

METHODS

As part of a larger effort supported by Arnold Ventures to understand the impact of state-level policy initiatives on service delivery from SSPs, RTI International conducted a study of all known SSPs operating in the United States from February to July 2022, which included an online, cross-sectional survey about the services the organization provided in 2021. The survey was administered using the Voxco platform (Voxco, Montreal, Canada). Referred to as the NSSSP, this survey has been repeated annually since 2019, following consistent procedures previously described.¹⁸

To recruit SSPs to participate, the North American Syringe Exchange Network (NASEN) emailed SSP contacts from a database of SSPs operating in the United States, continuously maintained for the last 30 years. To build this SSP database, we proactively contacted, searched, and followed up with SSPs from a variety of different sources, including NASEN’s online directory, NASEN’s Buyers Club, state and county public health department Web sites, social media platforms, regional and national networks of SSPs, webinars, and conferences. SSP organizational directors were e-mailed up to 3 times asking them to participate, and for those who did not respond, we conducted additional follow-up with individual programs via e-mail, phone calls, or both. SSPs were offered a $75 honorarium if they completed the survey. Our target population was the total number of known SSPs at the time of the 2022 NSSSP (n = 456), of which 311 (68%) completed our survey.

Measures

The NSSSP included questions about distribution of syringes, naloxone, fentanyl test strips, and buprenorphine treatment initiation, as well as other organizational characteristics, including annual budgets, funding sources, and organizational types. We identified 2 types of dependent variables. First was the SSP’s 2021 annual budget, treated as a continuous variable with budgets rounded to the nearest US dollar. Each SSP’s annual budget was ascertained with the following question: “What was your syringe services program’s annual budget last fiscal or calendar year? (Please estimate if records are not easily available.)” No specific instructions were provided regarding the inclusion or exclusion of in-kind costs. We also asked SSPs “What were your syringe services program’s sources of funding for the last fiscal or calendar year?” and SSPs could select funded or not from a list of different funding sources.

For a separate set of analyses, we examined dependent variables from a series of questions about the (1) quantity of equipment or services provided in 2021, including the number of syringes distributed, the number of participant contacts for syringe distribution, the number of naloxone doses distributed, and the number of participant contacts for naloxone distribution, and (2) whether the program offered fentanyl test strips or buprenorphine to their participants, either in person or via telehealth.

For independent variables, we classified SSPs as being operated by a city, county, or state department of public health (DPH-SSP) or as their own community-based organization (CBO-SSP), defined as a standalone community-based, nonprofit organization with or without a fiscal sponsor or a program within a larger community-based, nonprofit organization. Some CBO-SSPs received partial funding from public health departments but were still considered CBO-SSPs if they were independently managed and operated.

We also constructed a 3-tier, county-level measure of urbanicity from the National Center for Health Statistics (NCHS) Urban-Rural Classification Scheme, following guidance from the Pew Research Center.¹⁹ Urban counties are located in 53 metropolitan areas with at least a million people. In the NCHS classification system, they are called “large central metro” counties, where about 31% of Americans live. Suburban and smaller metropolitan counties, where about half of Americans (55%) live, include those outside the core cities of the largest metro areas, as well as the entirety of other metropolitan areas. This group includes “large fringe metro,” “medium metro,” and “small metro” counties in the NCHS classification system. Rural counties are located in nonmetropolitan areas. With a median population size of 16 535, only 14% of Americans live in rural counties.¹⁹

Next, we assessed opioid overdose mortality rates per 100 000 population as a standardized continuous measure with a mean = 0 and standard deviation = 1. We used opioid mortality rates from 2020,²⁰ the year before NSSSP data, as a proxy for the level of need in the community at the time funding was allocated in 2021. We obtained these data through the National Vital Statistics System, following details described previously.²¹

Finally, county-level data on percentage that voted for the Republican presidential candidate in the 2020 presidential election was obtained from the Massachusetts Institute of Technology Election Data and Science Lab County Presidential Election Return 2000–2020²² and represented as a continuous variable.

Data Analysis

We first summarized the data with descriptive statistics, with missing responses handled using listwise deletion. We then assessed how the annual budget of SSPs compared with minimum benchmarks established by Teshale et al. for small SSPs of 250 clients per year, as this was the most conservative (lowest-cost) benchmark against which to compare.¹³ We omitted the start-up costs used by Teshale et al. to compare annual operating costs of SSPs that completed the NSSSP with budget benchmarks for annual operating costs of small SSPs. In addition, we ranked SSP funding sources from most common to least common. To better understand the characteristics associated with SSP budget size, we then used generalized estimating equations with SSPs nested within counties, and an exchangeable correlation structure.²³ The outcome variable was SSP annual budget in 2021. Independent variables were the smoothed opioid-related mortality rate per 100 000 population for that SSP’s county in 2020, urbanicity, SSP type, and the percentage of voters who voted Republican in the 2020 election.

We then examined the effect of SSP annual budget (scaled per $100 000), smoothed opioid-related mortality in 2020, urbanicity, SSP type, percentage of voters who voted Republican in the 2020 election, and population obtained from the American Community Survey 5-year estimates (2016‒2020, scaled to 100 000)²⁴ on the number of participant encounters where syringes were distributed, number of syringes distributed, number of participant encounters where naloxone was distributed, and number of naloxone doses distributed, all scaled per 1000, using a negative binomial generalized estimating equation with SSPs nested within counties and an exchangeable correlation structure. We then used a logit generalized estimating equation in a similar way to look at the effect of the same independent variables on whether the SSP offered fentanyl test strips or offered buprenorphine either in person or via telehealth. We conducted sensitivity analyses to assess the impact of nesting SSPs within states rather than counties; we observed no substantive differences.

In the regression analyses, we considered all variables with P < .05 statistically significant. We conducted data preparation and analyses in SAS Enterprise Guide version 7.15 (SAS Institute Inc, Cary, NC).

RESULTS

The median SSP annual budget was $100 000 (interquartile range [IQR] = $20 159‒$290 000). One hundred thirteen responding organizations (36.3%) were DPH-SSPs, either local or state, and 198 (63.7%) were CBO-SSPs (Table 1). The median budget for DPH-SSPs was $50 000 (IQR = $9336‒ $150 000) and for CBO-SSPs was $149 000 (IQR = $42 000‒$359 890). Data were missing for key variables (annual budget, % voted Republican in the 2020 presidential election) for 32 SSPs, creating an analytic sample of n = 279 SSPs. Sensitivity analyses found that variables in Table 1 were not meaningfully different between the full (n = 311) and analytic sample (n = 279).

TABLE 1—

Descriptive Characteristics of the 2022 National Survey of Syringe Services Programs: United States

Characteristic	No. (%) or Median (IQR)
Urbanicity
Rural	74 (24)
Suburban	154 (50)
Urban	81 (26)
Syringe services program (SSP) type
Department of public health run	113 (36)
Community-based organization run	198 (64)
Supplemental services offered at SSP
Fentanyl test strips	201 (65)
Buprenorphine	107 (34)
Annual budget, $	100 000 (24 000‒296 583)
No. of syringe contacts	1 432 (306–5 000)
No. of syringes distributed	125 000 (32 741–457 963)
No. of naloxone contacts	500 (150–1 873)
No. of naloxone kits distributed	1 500 (396–4 585)
Opioid-related mortality (per 100 000 population; smoothed), 2020	21 (13‒32)
% of voters in the SSP catchment area who voted Republican during the presidential election, 2020	40 (25‒51)

Note. IQR = interquartile range; SSP = syringe services program. The sample size was n = 311.

We observed the following ranking of most-to-least-common funding sources: state health department (63%); fundraising or donations from individuals (42%); private foundations (29%); city or county health department (25%); other sources, including national networks and coalitions (20%); out of pocket, including staff donations (19%); and federal government (11%). Funding levels for SSPs also varied by urbanicity, yet most SSPs operating in rural, suburban, and urban environments had an annual budget that met 5%, 23%, and 46%, respectively, of the minimum benchmarks established by Teshale et al. for a small-scale program¹³ (Figure 1).

FIGURE 1—

Minimum Benchmark and Median Actual Annual Budgets in 2021 for Syringe Services Programs in United States, by Urbanicity

^aBenchmarks were taken from Teshale et al.¹³ omitting start-up costs and based on a small (250 clients per year) syringe services program using 2016 US dollars.

SSPs operating in areas with higher levels of opioid-related overdose mortality in the previous year had higher budget levels (adjusted mean difference [AMD] = $77 949; 95% confidence interval [CI] = $7216, $146 681; Table 2). SSPs in urban environments also reported significantly higher budget levels compared with SSPs operating in rural environments (AMD = $241 965; 95% CI = $89 664, $394 265). SSPs operating in suburban environments had similar budget levels to those operating in rural environments. SSPs operating in counties with a higher percentage of voters who voted Republican in the 2020 presidential election reported significantly smaller SSP budget levels (AMD = ‒$80 890; 95% CI = ‒$153 821, $7958). While, on average, CBO-SSPs had higher budget levels than DPH-SSPs, these differences were not statistically significant.

TABLE 2—

Association of Community and Organizational Characteristics With Syringe Services Program (SSP) Budget Size in the United States, 2021

	AMD, $ (95% CI)	P
Opioid-related mortality per 100 000 population (smoothed), 2020	77 949a (7 216, 148 681)	.031
Urbanicity (Ref = rural)
Suburban	10 856 (−75 694, 97 406)	.81
Urban	241 965 (89 664, 394 265)	.002
CBO-SSP (Ref = DPH-SSP)	64 146 (−10 136, 138 428)	.09
% of voters in the SSP catchment area who voted Republican during the presidential election, 2020	−80 890a (−153 821, 7 958)	.03

Note. AMD = adjusted mean difference; CBO-SSP = community-based organization‒run SSP; CI = confidence interval; DPH-SSP = department of public health‒run SSP.

^a In the case of continuous variables (which were standardized with mean = 0 and SD = 1), the AMD is presented for a 1-SD increase.

When adjusting for the opioid-related mortality rate per 100 000 population in 2020, urbanicity, total population (per 100 000), and percentage of voters who voted Republican in 2020, the total SSP budget had a statistically significant association with the number of participant encounters where syringes were distributed (incidence rate ratio [IRR] = 1.16 per $100 000 budget increase; 95% CI = 1.13, 1.20), the number of syringes distributed (IRR = 1.21; 95% CI = 1.18, 1.24), the number of participant encounters where naloxone was distributed (IRR = 1.09; 95% CI = 1.06, 1.12), and the number of naloxone doses distributed (IRR = 1.09; 95% CI = 1.06, 1.12; Figure 2). SSP budget also had a positive association with the offering of fentanyl test strips and buprenorphine treatment, though those adjusted odds ratios (AORs) were not statistically significant (AOR = 1.27; 95% CI = 0.85, 1.92 and AOR = 1.13; 95% CI = 1.00, 1.29, respectively).

FIGURE 2—

Relationship of Syringe Services Program (SSP) Budget Level and Service Scale, by (a) Continuous Outcomes and (b) Dichotomous Outcome: United States, 2021

Note. AOR = adjusted odds ratio; CI = confidence interval; IRR = incidence rate ratio; LCL = lower confidence limit; UCL = upper confidence limit. The vertical, blue dotted line represents 1.0, or the null. Outcomes are shaded in green. IRR and AOR point estimates for each exposure’s association with the outcome above it are represented with black dots, and the 95% CI is demonstrated by the horizontal line. All models were adjusted for SSP annual budget (scaled per $100 000), smoothed opioid-related mortality in 2020, urbanicity, SSP type, percentage of voters who voted Republican in the 2020 election, and population (scaled per 100 000).

DISCUSSION

One of our most striking findings was that the majority of SSPs’ actual annual budget levels were far below minimum benchmarks for a small-scale SSP. This funding gap was even more evident with decreasing levels of urbanicity. This is particularly alarming given the potential that SSPs hold for reaching PWID and delivering evidence-based public health interventions that can prevent infectious diseases and overdose deaths. The unrealized potential of SSPs is substantial; even though the federal government has recently begun funding these organizations, SSPs require substantially greater sustained resources from local, state, and federal sources to be optimally effective.

In our analysis, we found that most SSPs had diverse funding sources, with fundraising or donations from individuals the second-most-common source of funding. Individual donations, while beneficial, are typically not in amounts comparable to large government contracts and are typically inconsistent over time, creating an uncertain climate in which to build staffing and broader organizational structures. Previous research has shown that limited, fragmented funding can be detrimental to implementation of evidence-based interventions.²⁵ Many SSPs with small budgets rely heavily on volunteer labor, especially CBO-SSPs. While volunteer support is welcomed, it increases instability for these underresourced programs, and civil society should not rely on volunteerism to provide essential public health interventions.

We also found that higher levels of SSP funding led to greater distribution of a variety of SSP services, regardless of underlying community or organizational characteristics. SSPs have repeatedly been shown to be cost-effective, life-saving public health interventions,^26–28 and public health departments have a responsibility to make data-informed funding decisions. Throughout the United States, we need more institutional structures for people with substance use disorders to receive culturally sensitive treatment services that aid recovery from problematic use.^29,30 As SSPs are a trusted, culturally sensitive source of care for many PWID, they are ideal settings for building infrastructure to improve access to treatment. Without increasing funding for SSPs, it will be difficult for public health institutions to meet the emerging and changing needs of PWID.¹⁴

We found many variables that were significantly associated with SSP budget size. SSP budget size was significantly greater in counties with higher opioid-related mortality in the previous year, and, while encouraging, the level of funding remains inadequate. Urban SSPs had higher budgets than suburban and rural SSPs, even after accounting for opioid-related mortality rates and voting history; as more than two thirds of the US population live in suburban and rural areas,¹⁹ there is a clear need to improve SSP funding in less densely populated areas.

Finally, while we found significantly lower budget levels for SSPs in areas where a higher percentage of voters voted Republican in the 2020 presidential election, it remains unclear whether the driving factors for less-resourced programs in Republican districts are unique to SSPs. For example, it could also be that similar underlying factors are driving findings from other recent research showing Republican-dominated districts in the United States have had the lowest COVID-19 vaccine uptake³¹ and the highest COVID-19 mortality rates.³² As long as the United States remains highly politically polarized, funding for SSPs among states with Republican-dominant governments may have little chance of increased support despite the evidence supporting implementation. Further work is needed to untangle and address partisan differences with regard to individual autonomy, use of public resources, science-based approaches, and morality-based decision-making specific to SSP support.

One case study that highlights the importance of adequate funding is the California Harm Reduction Initiative, or CHRI.³³ CHRI was a state government‒funded initiative that began in August 2020 and provided more than $15 million to SSPs over 3 years through direct funding and technical assistance from the National Harm Reduction Coalition. CHRI-supported SSPs in our study had a median total annual budget of $245 000, which was $151 080 more than the median annual budget across non-CHRI SSPs.³³ Though CHRI represents the largest single investment in harm reduction ever made by the State of California before 2023, the median annual budget provided to California SSPs through CHRI was $112 500, which still remains far below benchmarks set by Teshale et al. SSPs that were funded by CHRI had significantly more participant encounters than non‒CHRI-funded SSPs, and provided significantly more syringes, naloxone, fentanyl test strips, and buprenorphine treatment than those not funded through this initiative. Moving forward, this type of evidence-based investment in harm reduction service provision is one that other states can look to as a model to increase funding levels for their programs. Future work should investigate the impact of state-level funding initiatives on SSP operations in states other than California.

Limitations

There are a number of potential limitations to this analysis. First, we had a survey response rate of 68%, which may have contributed to some selection bias in the sample. It remains possible that other SSPs exist that are unknown to us and that our findings do not represent those SSPs that did not respond. Previous surveys have suggested that SSPs that do not respond to surveys like this one tend to be small programs.³⁴ This is unsurprising, as smaller programs would likely have less staff time available to fill out a survey. If this were true for the 2022 NSSSP, our study could be overestimating actual budget levels for SSPs in the United States.

Second, program budgets can be complex, and while we estimated the association of variables with budget size, other unmeasured factors driving annual SSP budgets in various communities could exist. This could vary by region, which we did not attempt to address in this analysis. Furthermore, while this analysis used the number of syringes distributed as 1 outcome, many jurisdictions face legal barriers to syringe distribution that limit distribution to a 1:1 ratio (only 1 sterile syringe can be provided in exchange for every used syringe returned). These and similar legal barriers were likely unmeasured confounders on the association between SSP budget and number of syringes distributed, and do not reflect CDC recommendations.³⁵

Third, we used county-level, smoothed opioid mortality rates in the previous year as a proxy for underlying need in a community. Smoothed estimates may over- or underestimate actual opioid overdose death rates, may not match drug overdose death rates obtained from NCHS, and do not fully represent community need, which would ideally also include measures of substance use disorder treatment accessibility and viral infection rates.

Fourth, this analysis was not designed for a detailed budget comparison of actual budget levels for specific services versus benchmarks for that service as put forward by Teshale et al. No data were available to differentiate budget and service outcomes for SSPs embedded within organizations that provide clinical services and those that are nonclinical. In addition, we did not capture information about SSP affiliations with other organizations whose mission includes delivery of clinical or other types of billable services, nor did we capture information about in-kind contributions. Future work should assess allocation of SSP budget levels to different services within SSPs.

Public Health Implications

SSPs are well positioned to have an impact on the health of PWID in ways beyond prevention of HIV and hepatitis C if they are funded and supported to distribute naloxone, integrate fentanyl test strips, and offer buprenorphine to those who are interested. Higher funding levels could yield greater staffing and supplies, better partnerships and integration throughout the community, improved workplace culture, and ability for SSPs to reach those who are most vulnerable to drug-related (and other) harms. Increased local, state, and federal funding must come with low administrative burden to ensure that SSPs retain organizational autonomy to adapt as needed. Otherwise, SSPs will need to carefully consider the trade-offs of accepting funding from specific sources.³⁶

With financial and political support, SSPs will be able to provide more services to a population at high risk of morbidity and mortality. Federal, state, and local health departments must issue more funding for SSPs—to at least benchmarks described by Teshale et al. and CDC colleagues—and provide political support for them to do this life-saving work.