Introduction

Medications for opioid use disorder (MOUD) include naltrexone, buprenorphine and methadone. Only opioid agonists like methadone and buprenorphine, however, substantially reduce crime, incarceration, overdose, and mortality and increase employment and social functioning [1]. For individual and public health, however, initiating and remaining on MOUD with these opioid agonists is inherently prosocial. MOUD improves an individual’s health and functioning while also generating broad societal benefits by reducing overdose mortality, destabilizing cycles of withdrawal-driven injecting, and lowering risks of HIV and HCV transmission. Through these pathways, it can also reduce crime, improve employment and relationships, and strengthen community safety [2–4]. The combination of these benefits results in improved health-related quality of life and general social functioning [5]. Due to the inextricable link between non-medical opioid use and crime, criminal legal settings concentrate people with OUD, HIV and HCV [6–8] and serve as potential entry points for screening and treatment [8]. While most screening and treatment models in the criminal legal settings have focused on this strategy in prisons and pre-trial detention centers [9], community supervision settings like probation, where people live freely in the community, are likely to interact with large numbers of people with OUD [10]. These probation “touchpoints”, however, have not been adequately deployed as settings to align public safety and health. This is especially true as people on probation, in the absence of MOUD, have substantially heightened rates of overdose-related mortality and criminal activity [11,12].

As many settings align drug policy and public health [13], there has been a recent shift globally to reduce incarceration levels in prisons and jails, in part by shifting people with non-violent and less severe crimes to community-based supervision, like probation. This shift represents an intermediary step on the path to decarceration efforts, potentially resulting in increasing numbers of people with or at risk for OUD, HIV, HCV, and tuberculosis to be concentrated in probation settings where there is a mandate to ensure public safety, yet not for public health. Probation settings, however, share a common mandate by addressing OUD, which, when untreated, undermines public safety efforts. For example, when OUD is left untreated, it is associated with higher levels of criminal activity [14]. Moreover, as probation settings also lack infrastructure and staffing to independently align both public safety and public health, it is crucial to identify unique opportunities for integrating such activities.

The Eastern European and Central Asian (EECA) region includes countries with some of the highest prevalences globally of OUD, HIV, HCV, tuberculosis, as well as high incarceration rates [5,6,15–18]. This is especially true as HIV incidence and mortality in EECA continue to increase among people who inject drugs (PWID), with high levels of people with HIV (PWH) in this region [19] who are concentrated in criminal legal settings due to the criminalization of drug use [20]. The rising incidence of HIV among PWID, coupled with the increased concentration of PWID/PWH in criminal legal settings suggest the need for expanding opportunities to scale-up HIV prevention using MOUD to achieve epidemic control. MOUD is an evidence-based primary and secondary [21,22] HIV prevention strategy and is key to reducing transmission and mortality [23]. It also improves treatment outcomes for HCV [24–26], and tuberculosis [27]. Moreover, when scaled up, MOUD can significantly reduce HIV incidence and mortality [1,6,23], and is particularly effective at preventing HIV and transmission among criminal legal-involved individuals with OUD [27,28]. Most MOUD programs in EECA, aside from Ukraine, which has been scaling up substantially since 2014 [29,30], remain under-scaled and far below targets set by the World Health Organization [31].

Moldova is a lower middle-income country (LMIC) in EECA with a population of approximately three million people, with about a third of people living in rural regions [32]. Across the country, including in both urban and rural settings, the treatment cascade for OUD and HIV remains suboptimal. HIV incidence and mortality continue to increase in Moldova, unlike patterns globally, partly due to suboptimal scale-up of HIV prevention programs [33]. Among PWID, opioid injection remains common, and HIV prevalence is elevated, while MOUD coverage remains far below population need. Structural barriers, including stigma, criminalization, and administrative restrictions that shape daily functioning and employment, further constrain uptake and retention [34,35]. Together, these conditions create a compelling rationale to test whether probation can serve as a scalable entry point to identify OUD and facilitate linkage to MOUD as an HIV and HCV prevention strategy.

PWID in community settings rarely utilize conventional primary care and specialty services [36], in part due to high levels of stigma in such settings [37–39]. Instead, they are more likely to intermittently rely on other touchpoints for care like syringe exchange programs, emergency departments, and prisons or jails [40], rather than engage in long-term care. Criminalization of drug use, stigma, and discrimination, however, remain major barriers to care for HIV prevention services among PWID [41]. In Moldova, coverage of HIV prevention services among PWID remains suboptimal, and MOUD remains underutilized relative to need.

As part of a larger research study, we pilot-tested a similar protocol for both Ukraine and Moldova. The procedures for Ukraine have previously been described to assess the deployment of a modified screening, brief motivation and referral to treatment (SBIRT) introduced in probation settings [42]. The modification included rapid screening using a single-item screening question, evaluation for treatment, and treatment initiation (SET) by greenlighting a treatment pathway. This study was conducted in parallel with the Ukrainian one [43]. The primary goal of this study was to assess the feasibility of probation as a clinical opportunity to identify people with OUD (and HIV and HCV) from a syndemic perspective [44] and facilitate linkage to MOUD using an externally delivered, modified SBIRT strategy. In this study, feasibility was defined as the ability to: 1) systematically screen, evaluate and treat (SET) consecutive probation clients for OUD; 2) deliver a brief intervention in the probation setting without disrupting routine operations; and 3) successfully link interested individuals to MOUD, with verification of treatment initiation and retention. As a pilot study, it did not assess implementation outcomes such as adoption by probation staff, long-term sustainability, or costs. There is a dearth of literature on the use of modified SBIRT for substance use disorders, particularly in settings where MOUD uptake is suboptimal and HIV prevalence is high. We hypothesized that a modified SBIRT strategy would increase interest in MOUD among people on probation in the Moldovan setting, and thus MOUD uptake. Additional details on Moldova’s epidemiologic profile, incarceration and probation landscape, and the availability and financing of MOUD and syringe services programs are provided below.

Methods

Results

Participant disposition

Of the 900 consecutive entrants screened, 136 (15.1%) met criteria for OUD. Of these, 17 did not meet study eligibility because they did not live within 30 kilometers of an MOUD site, resulting in 119 (100%) who enrolled in the study, all of which completed the brief intervention (Fig 1). These findings demonstrate the feasibility of identifying individuals with OUD and linking them to treatment through probation-based screening, even in the absence of a formal healthcare mandate. Participant characteristics are shown in Table 1.

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Baseline participant on probation with opioid use disorder, stratified by initiation of OAT (N = 119).

https://doi.org/10.1371/journal.pgph.0006317.t001

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. Participant disposition.

https://doi.org/10.1371/journal.pgph.0006317.g001

Treatment outcomes

ROC analysis demonstrated excellent discrimination of baseline MOUD interest for predicting MOUD initiation (AUC = 0.86). An interest score of ≥8 maximized sensitivity and specificity and was therefore selected as the threshold for defining high MOUD interest in subsequent analyses. Using the cutoff of ≥8, 23 (16.9%) stated they were interested in initiating MOUD before the brief intervention, which increased to 27 (19.9%) afterwards. Using a lower cutoff of ≥4, 103 (75.7%) expressed interest in initiating MOUD and did not increase appreciably after the brief intervention (N = 104, 76.5%). Median interest (Table 2) in starting MOUD did not increase following the brief intervention (median [IQR] pre-intervention: 5 [3–7] vs post-intervention: 5 [3–7]; Wilcoxon signed-rank test, p < .001). Although the pre- and post-intervention summary medians were similar, paired analysis showed a statistically significant upward shift in MOUD interest scores after the brief intervention (Wilcoxon signed-rank test, p < 0.001), indicating a small but consistent within-person increase. Median participants’ attitudes towards and knowledge about treatment also remained unchanged after the brief intervention (Table 3). The perceived effectiveness of these treatments differed significantly between those who initiated MOUD and those who did not. Those who initiated MOUD perceived greater effectiveness of MOUD, family support, being in new environment, religion, detox (supervised withdrawal), inpatient rehabilitation, incarceration, employment, and other treatments with medication than those who did not initiate MOUD (p < 0.05) (Table 4). Reasons for not initiating MOUD among those who did not initiate MOUD are presented in Table 5. Reasons for not being interested in MOUD initiation among PWID who have never been on MOUD are presented in Table 6.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Changes in interest in, difficulty in accessing, and importance to initiate medications for opioid use disorder before and after a brief intervention among participants on probation with opioid use disorder (N = 117).

https://doi.org/10.1371/journal.pgph.0006317.t002

Download:

• PNG
  larger image
• TIFF
  original image

Table 3. Comparison of mean MOUD attitude and knowledge scores* between baseline and 1 month among participants on probation with opioid use disorder (n = 118).

https://doi.org/10.1371/journal.pgph.0006317.t003

Download:

• PNG
  larger image
• TIFF
  original image

Table 4. Perceived treatment effectiveness of various treatment strategies among participants on probation with opioid use disorder.

https://doi.org/10.1371/journal.pgph.0006317.t004

Download:

• PNG
  larger image
• TIFF
  original image

Table 5. Reason for initiating MOUD among people who initiated MOUD.

https://doi.org/10.1371/journal.pgph.0006317.t005

Download:

• PNG
  larger image
• TIFF
  original image

Table 6. Reasons for not initiating OAT among those declining OAT (N = 86).

https://doi.org/10.1371/journal.pgph.0006317.t006

Overall, 33 (27.7%) participants with OUD initiated MOUD. Nine started within 30 days after the brief intervention and 24 started after 30 days. Among participants who initiated MOUD, 72.7% had not previously attempted MOUD. In the multivariable analysis, covariates included in the logistic regression model are age, sex, HIV status, HCV status, prior MOUD attempt, registration in the narcology registry, and post-intervention importance score for MOUD. Analysis showed a higher level (score ≥8) of perceived importance for MOUD as an effective treatment (post-intervention) for OUD, having HCV (aOR = 4.96, CI = 1.57-18.25), and having HIV (aOR 4.02, CI = 1.11-16.02,) predicted MOUD treatment entry (Table 7). Of the 33 that initiated MOUD, 32 (97%) remained on treatment through 6 months (Fig 2).

Download:

• PNG
  larger image
• TIFF
  original image

Table 7. Predictors of initiating OAT among people with opioid use disorder (N = 119).

https://doi.org/10.1371/journal.pgph.0006317.t007

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Opioid use disorder continuum of care cascade for people in probation in Moldova.

https://doi.org/10.1371/journal.pgph.0006317.g002

Of the 33 participated who started OAT during the study, 1 (3%) stopped OAT before their month 6 follow up interview. The remaining 32 participants were still on OAT at the time of their 6 month follow up.

Discussion

Consistent with implementation science frameworks, this study focuses on early feasibility indicators rather than downstream implementation outcomes [69]. The study highlights the high prevalence of OUD (15.1%) and HIV (22.7%) among individuals on probation in Moldova, emphasizing the critical role of probation settings as a touchpoint to identify people eligible for treatment (and prevention). By employing a modified SBIRT (SET: screen-evaluate-treat) strategy, the intervention successfully identified a high prevalence of individuals with OUD and HIV as well as interest in initiating MOUD with methadone or buprenorphine. This modified SBIRT strategy was effective at identifying individuals with OUD in probation settings and increasing their interest in MOUD. A proportion of participants subsequently initiated MOUD following the intervention, suggesting potential to improve treatment engagement, which warrants evaluation in controlled studies.

Key among the findings is that over a quarter of the 119 participants with OUD initiated or remained on MOUD, with 97% being retained on treatment after six months. Additionally, the integrated screening identified substantial undiagnosed HIV cases, further underscoring the potential for probation to serve as an impactful entry point for syndemic screening and linking individuals to comprehensive treatment for chronic conditions. In the case of HIV, undiagnosed PWH substantially contribute to HIV transmission [70]. These findings, however, suggest that aligning probation’s public safety mandate with public health goals will require structural and procedural adaptations so that the SBIRT can be implemented by probation staff rather than trained researchers.

Although the median interest in MOUD score and interquartile range were unchanged from pre- to post- brief intervention assessments (median 5 [IQR 3–7] at both time points), the post-intervention scores were shifted toward higher values overall (Wilcoxon signed-rank test, p < 0.001) after the brief intervention and 33 of 119 participants initiated MOUD and 32 were retained on treatment over 6-months. Predictors of initiating MOUD were perceiving MOUD to be an important treatment, having HCV and having HIV. Twenty-seven participants tested positive for HIV, with seven being newly diagnosed during screening procedures. Particularly in LMICs like Moldova where MOUD coverage is low and decarceration efforts have yet to be fully realized, probation has been an under-utilized opportunity to intervene with PWID to support access to prevention and treatment services. Interventions that target the probation period among PWID can be especially impactful as they can address both OUD and HIV among criminal legal-involved individuals. Notably, nearly 40% of participants reported experiencing an overdose within the past six months, a critical indicator of elevated risk for morbidity and mortality within this supervised population. Research indicates that individuals under probation face overdose rates up to 15 times higher than the general population, and overdose fatalities remain strikingly high during periods of community supervision [11,71]. This alarming prevalence underscores the urgent need for enhanced surveillance of OUD and rapid linkage to OAT like methadone, buprenorphine, [72] which are proven to substantially reduce overdose mortality [73] and help reduce the transmission of blood-borne infections like HIV and HCV [2–4]. This finding further highlights an important opportunity for probation officers to play a pivotal role in harm reduction efforts through proactive screening for substance use, facilitating access to OAT, and distributing naloxone to protect both individuals and the wider community. Findings here suggest that probation can and should serve as a key entry point into OUD treatment for PWID, which is similar to findings in nearby Ukraine [43] and points to the need for more broadly intervening in probation.

To reduce the burden on prisons, which have a mandate to provide healthcare that is equivalent to that in the community, the criminal legal system has transitioned many people who would otherwise be in prison to probation. A key tenet of probation, however, is to ensure that crime is reduced (i.e., either illicit drug use and/or the crimes committed to procure drugs). As MOUD is associated with substantial reductions in criminal activity, screening for OUD and supporting clients to gain treatment could be a critical strategy to align public safety and public health. As probation services are mostly designed to promote public safety, they have not yet has been leveraged to simultaneously promote public health. For probation to fully align their public safety mission with public health, it will be crucial to be able to implement SBIRT in such settings by the staff who operate probation as our findings were observed when people external to probation implemented the screening. Though not explicitly understood by probation officers, supporting their clients in accessing MOUD (a public health benefit) also promotes public safety by reducing or eliminating substance use, reducing other crime and supports overall recovery [7,74]. In real-world settings, developing strategies that allow probation officers to modify SBIRT procedures requires a clear understanding of how MOUD does promote public safety (through its reduced drugs use and criminal activity). Inquiring about substance use by probation officers, however, may require an alternative delivery strategy, perhaps by reducing opportunities to stigmatize and discriminate against those using drugs who are “breaking the law”.

Redesigning or modifying SBIRT in probation might include deployment behavioral design interventions (BDI), which uses choice architecture, framing and nudging. Choice architecture focuses on the different way choices can be presented and how this presentation may impact decision-making [75]. For example, do clients self-screen by having a self-screening instrument (e.g., using a QR code, pen and paper or online decision tool) that is anonymous and provides an actionable output that asks probation officers to assist in supporting their recovery (without directly discussing illicit opioid use). Framing MOUD may also be impactful in whether people with OUD decide to initiate MOUD [76]. For example, it can be framed as a tool to manage one’s addiction or as a strategy to remain in the community and not go to prison. Other types of framing may derive from the officer directly by stating upfront that their goal is to keep their clients in the community and one way of doing this for clients to engage in treatments that support their recovery. Framing can also be supported by other implementation tools like decision aids that allow patients to make preferences about treatment that are aligned with their values [77,78]. Last, nudging is a small push without forcing people to make decisions [79]. People with OUD may be “nudged” to make a healthier decision by including facts about rates of recidivism and contraction of infectious diseases among those who continue to inject opioids. Alternatively, probation officers themselves might be nudged to reduce their likelihood of sending people to prison by setting rewards for successful retention in community treatment. The aforementioned strategies taken together may be synergistic and promote the uptake of OUD treatment if implemented correctly.

Important in this pilot study was the integration of screening for comorbid conditions like OUD, HIV and HCV, which are syndemic [44] and well-suited to advance multiple public health goals [42]. These integrated programs could be even more impactful in high-prevalence settings or among key populations such as PWID in Moldova, where it is estimated that a third of all people in prisons use drugs and only 12% are prescribed MOUD [6]. MOUD remains one of the most effective and cost-effective strategies to prevent and treat HIV [23,80], thus making it an intervention to treat OUD and address the HIV epidemic. The secondary outcome of HIV identification was therefore a critical element of this intervention. Evidently, there remains a need to improve the uptake of MOUD among PWID and improve HIV case detection.

While 27.7% of our sample initiated MOUD as part of the brief intervention, the brief intervention component only modestly increased interest in starting MOUD and unfortunately, did not change attitudes towards or knowledge about MOUD. Although evidence for the effectiveness of brief interventions in reducing drug use is mixed, they do appear to have benefits within SBIRT protocols for alcohol and OUD. In this study, the findings align with studies showing modest improvements in motivation but limited effects on behavioral outcomes [81]. This may be due to persistent negative attitudes and misinformation about MOUD within prisons in Moldova [82]. Similar attitudes and beliefs have been found in other countries in EECA, including Kyrgyzstan [83] and Ukraine [56,41,84], and in the case of Kyrgyzstan, did not increase methadone scale-up [28,85]. More broadly across EECA, punitive “Russian narcology” legacies, reinforced by institutional surveillance and abstinence-oriented norms and compounded in many settings by weakened support for evidence-based harm reduction, can entrench stigma and misinformation about MOUD in ways that blunt the impact of brief, individually focused interventions [86]. One potential explanation is that the brief intervention was not effective because of content, while another explanation might be delivery where deliverers had their own implicit biases or presented findings not congruent with context or not aligned with patient preferences [87,88]. One alternative delivery strategy could be using either informed- or shared-decision aides which improve knowledge, reduce decisional conflict and stimulate individuals to be more active in decisions regarding their health [89]. Potential future options regarding choice architecture in BDIs may include embedding a modified SBIRT process initially within an informed (or shared) decision aid for MOUD and the referral for treatment could derive from the client providing a decision-aid output providing recommendations for the client.

Numerous studies have examined the unique social dynamics in prison settings that may dissuade engagement with OUD services [82,84,90]. The extent to which they influence initiating MOUD in probation, however, is not known. In some cases, criminal subcultures, or informal systems of governance, in prisons forbid the use of drugs, and understandings of methadone vary. Whether a particular prison’s subculture deems methadone a drug or not can influence the uptake of methadone. Criminal subculture is deeply embedded in individuals leaving prisons as a social and cultural force which may supersede any individual-level intervention like an SBIRT, hence the lack of increase in MOUD post-intervention [28]. Additionally, because probation in community settings can be an extremely stressful period for individuals with low social support, this may suggest the need to supplement support systems that drive subcultural norms, including social, physical and economic levels [91]. This is especially true for those who are never detained or who transition from pre-trial detention for their probation period. While little is known about how dynamics brought on by criminal subculture persist outside of the prison setting and into the probationary period, we can hypothesize that criminal subcultural influence is lower in probation than in prisons, but we do know whether these norms continue into the community and may impact probation. One study from Kyrgyzstan documented low levels of MOUD uptake in prison using SBIRT [28] while a similar study from Moldova showed that one quarter of those screened in prison with OUD initiated methadone, but most did so soon after release when the criminal subculture may have had less influence on daily activities [92]. Sociologically rigorous qualitative work is, therefore, necessary to elucidate the underlying social, structural and cultural forces that shape decision making about MOUD uptake outside the prisons setting. Regardless, initiation of MOUD was relatively high among PWID in this study.

Our findings, in part, are similar to findings in probation in Ukraine using a similar methodology where 24.6% of participants initiated MOUD, closely mirroring the 27.7% who did so in Moldova. Our findings differ, however, in the greater differences in perceived effectiveness of various treatments between those who did and did not initiate MOUD. Both studies found a modest but significant increase in interest in starting MOUD after the brief intervention [43]. In contrast to both these studies that demonstrated the utility of SBIRT for improving MOUD initiation, one systematic review and one meta-analysis found that brief screening interventions are unimpactful in improving the uptake of substance use treatment [93,94]; however, treatment services for OUD were poorly represented in these studies, and neither included interventions delivered in the probational setting [95], thus limiting the generalizability of their findings to our study setting and context. Given the little change in interest in MOUD after the intervention, modifying SBIRT to eliminate the brief intervention and perhaps creating a “green light” pathway to treatment for those interested might be considered.

A critical aspect of this intervention is the bundling of screening for comorbid conditions that occurred here, as 22.7% were found to be PWH, which was higher than the reported 11% of PWID with HIV in Moldova [96]. This may suggest that people on probation who also meet criteria for OUD are at a particularly high risk for HIV. Although this study was not primarily designed to identify new HIV cases, 25.9% of people with HIV in this study were newly identified and referred to care, an important tool in meeting UNAIDS 95-95-95 goals to reduce HIV incidence. If one extrapolates to the 11,970 on probation this would minimally translate to 1807 people identified with OUD, of which 486 could be initiated and retained on methadone and 106 new HIV infections identified.

Given nearly one third of people in Moldova do not know they have HIV, modified SBIRT screening that bundles screening for commonly co-occurring conditions like OUD, HIV and HCV can amplify the impact of such strategies [42,44]. Future studies should consider other medical conditions as a part of SBIRT screening including depression and other psychiatric disorders that often disproportionately afflict those with substance use disorders involved in criminal legal settings [97].

Despite the feasibility and important findings from our study, it does have limitations. First, this study did not assess implementation feasibility from the perspective of probation staff, including adoption, fidelity, or sustainability of SET delivered by probation officers themselves. These outcomes will be essential for future implementation trials. Second, our sample size of 119 is small and limits generalizability. This is, however, mitigated by screening a substantially larger number of 900 people on probation. Additionally, 17 of our screened participants were ineligible due to distance from a participating MOUD center. This exclusion criteria could introduce some bias as it likely would have been more difficult for those living further away from an MOUD site to initiate MOUD and be retained in treatment. Third, social desirability bias may have also influenced our findings. Although participants self-administered the survey on REDCap, there may have been social desirability bias introduced as they may have wanted to please motivational interviewers (research assistants). COVID-19 did not disrupt study procedures although it lengthened the recruitment phase. Fourth, though not included in our study was the influence of the national registry for people with OUD (i.e., restrictions on employment and driving and acts as a deterrent to MOUD initiation), it may partially explain modest uptake despite interest. This may, in part, have explained why a larger number of people did not initiate MOUD. The pre/post design without a control group limits causal inference regarding intervention effects on MOUD interest and initiation compared with randomized or controlled designs. Finally, it is likely that some individuals with OUD did not want to report their drug use for fear it would violate the rules of their probation. OUD prevalence, therefore, was likely significantly higher among the 900 we screened than we were able to capture. It is important to note, however, among those who did disclose their drug use, there were no instances or complaints of breaches of confidentiality.

Notwithstanding these limitations, this modified SBIRT strategy deployed in probation by trained interventionists was effective in identifying individuals with OUD and increasing their interest in MOUD. Among participants with OUD who were not receiving MOUD at enrollment, a substantial subset initiated MOUD during follow-up after exposure to the intervention. Given there are greater numbers of people on probation than in prisons, the probation period could be an optimal time point to encourage PWID with OUD to initiate MOUD. Further research should explore other brief, rapidly introduced interventions that can address specific barriers to MOUD uptake, and qualitative studies are needed to explain why the brief intervention does not change attitudes nor interest – perhaps other decision support tools might. Policy reform may also be necessary so that people on probation can disclose their drug use without violating their probation.