The Ukrainian SORT IT Course Linking intravenous drug users to treatment through non-governmental organizations in Ukraine: how well is it working?

Introduction: Alliance for Public Health, the International Charitable Foundation, coordinates HIV prevention in Non-Governmental Organizations (NGO) working with people who inject drugs (PWID) in Ukraine. We aimed to describe the performance of the differential model of linking PWID to HIV care and treatment (Community Initiated Treatment Intervention – CITI). Methodology: A retrospective cohort study using routine program data was conducted among 8,927 PWID who were tested positive for the first time during January 2016 – June 2017. Study outcomes were enrollment into CITI and initiating antiretroviral treatment (ART). Factors associated with outcomes were estimated by logistic regressions with random effects. Results: Among the study participants, 54% enrolled into CITI and 23% initiated ART. CITI enrolment was associated with being married (adjusted odds ratio (AOR) = 1.17; 95%: 1.02-1.34); less than weekly compared to daily (AOR = 1.31; 95%: 1.13-1.52); less than 5 years of drug use compared to > 14 years (AOR = 1.73; 95%: 1.40-2.13), and having no criminal records (AOR = 1.30; 95%: 1.12-1.50). Factors of non-ART initiation were male gender (AOR = 1.33; 95%: 1.16-1.53); being single (AOR = 1.48; 95%: 1.21-1.82); drug use duration > 14 years compared to < 5 years (AOR = 1.38; 95%: 1.03-1.85), unemployment (AOR = 1.45; 95%: 1.15-1.83) and history of incarceration (AOR = 1.21; 95%: 1.003-1.45). Conclusion: Mobilizing the NGO community and PWID to engage in outreach HIV testing activity and harm reduction for key populations has succeeded in opening the gateway to prevention, care and ART for thousands of PWID in Ukraine.


Introduction
The Sustainable Development Goals (SDGs) aims to end the HIV/AIDS epidemic by 2030 [1]. A vital component of this strategy is the 90-90-90 targets [2]:90% of all people living with HIV know their status; 90% of those diagnosed with HIV will receive sustained antiretroviral therapy (ART) and 90% of those on ART have viral suppression [2]. As about 50% of new HIV infections occur in key populations, the first and second targets can only be achieved if there is an improved focus on people who are at increased risk of acquiring and transmitting HIV due to their high-risk behaviors including People Who Inject Drugs (PWID) [3].
Worldwide, 158 countries have reported injecting drug use, of which 78% have reported HIV among PWID [4,5]. The risk of HIV infection in PWID is on average22-50 times higher than in the general population [6]. Injecting drug use contributes to around 10% of HIV infections globally [7]. The overall proportion of HIV positive PWID ranges from 3% in Kazakhstan to 58% in Vietnam [8]. An estimated 20% of new HIV infections occur among PWID and their sexual partners (compared to 0.9% in the general population) in Ukraine [6].
To prevent HIV transmission and improve survival, PWID should know their HIV status and access ART. However, most countries have inadequate coverage and quality of HIV services for PWID [1]. Ways forward in addressing this issue are urgently needed.
The International Charitable Foundation "Alliance for Public Health" (APH), a Non-Governmental Organization (NGO) in Ukraine coordinates the HIV 96S prevention and care activities of about 100 implementing NGOs working with PWID in Ukraine. A unique aspect of their work is "differentiated service delivery model" which aims at simplifying and adopting HIV services to serve the needs of PWID better. One of the mentioned differentiated care models the Community Initiated Treatment Intervention (CITI), a case-management approach, which engages PWID, their peers and social workers in enhancing health-seeking behavior. CITI assists PWID to navigate the health system to overcome barriers to ART initiation ( Figure 1) and is recognized by the World Health Organization (WHO) as a good practice intervention [2]. Although previous studies have shown that involving social workers and peers is effective in linking PWID to HIV care, there are no operational research studies from Ukraine assessing large programs focused on the linkage between HIV testing and ART initiation [9][10][11][12][13]. Accessing this service delivery model would be useful both to evaluate possible performance gaps and its influence on the access to care for PWID.
We aimed to describe the differential model of linking HIV positive PWID to CITI and ART. The specific objectives were to determine a) the total numbers of PWID tested for HIV and found positive b) the socio-demographic characteristics of HIV positive individuals enrolled (and not enrolled) in CITI c) among those in CITI, the numbers initiated on ART within six months of HIV testing and d) risk factors for non-ART initiation within this period.

Study Design, General and Specific settings
A retrospective cohort study was conducted using routine program data in Ukraine which is the largest country of Eastern Europe having estimated population of 42 million and national HIV prevalence of 0.9% [14].
The study includes 16 out of 27 regions in Ukraine. In all 16 regions, HIV/AIDS services are provided by NGOs in collaboration with the public health system. There are 52 NGOs working at the selected study sites with key populations including PWID. The activities of NGOs are coordinated, supervised and monitored by the APH Injectable drugs include psychotropic substances (for example, opioids, methadone, amphetamine-type stimulants, hypno-sedatives, cocaine and hallucinogens) [1]. All NGOs offer a package of harm reduction services according to WHO guidelines [1] including the provision of sterile injecting equipment through needle and syringe programs.

Differentiated HIV/AIDS care models for linking PWID to ART
A brief description of the models that were reviewed in the study and implemented by the APH in Ukraine is presented below.
Directly assisted HIV self-testing NGO workers provide services at the places convenient for PWIDboth at outreach and community centers. HIV testing is provided using rapid blood tests [1]. From 2007 until 2014, HIV testing was performed at NGO sites by trained medical teams. To improve access to HIV testing, a new approach called directly assisted HIV self-testing (DAST) was introduced in 2015. DAST is performed with the help of trained peer outreach workers. These workers carry rapid HIV tests when they visit clients, or when clients visit community centers for syringe or condom distribution. They provide pre-and post-test counselling and ensure HIVtesting and follow up. HIV testing is provided at any convenient place for the client.

Optimized Case Finding (OCF)
This approach was introduced in 2016 as an additional strategy to DAST and aimed at improving HIV case-finding. This is a two-step chain referral process using an HIV positive index case to recruit peers from the extended risk network (anyone known by the client who might be at risk) for assisted testing by trained case finder [15].
Linkage to treatment services There are 2 major community intervention models for PWID available in Ukraine -basic nation-wide accepted set of standards of care and its extension, CITI  (Table 1). Essentially, CITI provides case management for HIV positive individuals by peer outreach workers so called case-managers. Case-managers conduct motivational counselling, initiate dialogue with doctors, and manage formal arrangements for clients' initiating their ART. CITI is effectively "peer navigation service" which helps eligible PWID initiate and adhere to HIV treatment. The case-managers are highly trusted by the PWID community. If a PWID enrolls in CITI and the case manager is unable to have the person start ART within 6 months thereafter, the file is closed with a possibility of reopening per clients need.

Antiretroviral treatment
ART is offered according to the WHO [1] and National guidelines and initiated at HIV/AIDS centers, which are public health facilities. Drug refills are provided at the same centers.

Study population and period
All PWID aged over 14-years who were HIV tested and found positive between January 1 st 2016 to June 30 th 2017 were included in the study.

Data and statistical analysis
Data on HIV testing and linkage to CITI and ART was collected through the Harm Reduction Program with paper based questionaries' and further entering into a dedicated database for key populations (SyrEx 2+) [16]. All data entry clerks and program staff were trained in data entry and were supervised by a data manager.
Primary outcomes included CITI enrollment and ART initiation. CITI enrollment was defined as enrolled in CITI during the period of 01.01.2016 until 30.06.2017. The first positive HIV test was used as the entry point of any given PWID for the study. ART initiation was assessed up to six months from HIV testing and was censored on 31.12.2017. ART initiation later than 6 months after HIV testing or non-initiation of ART was clumped as an unsuccessful outcome. Independent variables were baseline social and demographic characteristics. We included gender, age in years and categorized by groups, marital status, frequency of drug use, duration of injecting drug use in years and categorized by groups, drug type, employment and history incarceration. In categorical   98S variable, all missing data were included as separate category ("unrecorded"). SyrEx data were imported, validated and analyzed in R, version 3.5.1. Risk measures were estimated using crude and adjusted odds ratios (OR) and their 95% confidence intervals (CI) Logistic regression models with random effects were used for calculating odds ratios.

Ethics
Permission to conduct the study was secured from the Senior Management of the Alliance for Public Health in Ukraine and ethics approval was obtained from the Institutional Review Board of the Ukrainian Institute on Public Health Policy, Kyiv, Ukraine.

HIV testing and the cascade to ART initiation
There was a total of 194,983 PWID tested for HIV, of whom 8,927 (5%) were found HIV positive. Figure  2 shows a progressive drop from being HIV positive to CITI enrollment (54%) and eventual ART initiation (23%). HIV (human immunodeficiency virus), COR (crude odds ratio), AOR (adjusted odds ratio), SD (standard deviation), IQR (interquartile range).

Socio-demographic characteristics of individuals enrolled (and not enrolled) in CITI
A total of 4,961 (56%) of the entire cohort of 8,927 HIV positive PWID were on opioids and stimulants, 2,639 (30%) were daily drug injectors and 2,134 (24%) had been injecting drugs for over 14 years. Table  2 shows the socio-demographic characteristics of 4,783 (54%) individuals enrolled and not enrolled (4,114; 46%) in CITI. The median time between the first positive test and CITI enrollment was one day (Inter Quartile Range, IQR: 2, range: 1-640 days).
After adjustment, CITI enrolment was significantly associated with being married, not using drugs daily, using drugs for periods less than 14 years and having no criminal records. Several variables including duration of drug use, drug type, employment status and history of incarceration had unrecorded data (ranging from 42-51%). Some unrecorded data showed significant statistical associations.

ART initiation and risk factors for non-initiation
The characteristics of 2,050 (23%) HIV positive individuals who initiated ART within six months of HIV testing is shown in Table 3. Median time to ART initiation was 42 days (IQR: 50; range. 1-182 days) from the first HIV positive test. After adjustment, significant risk factors associated with non-ART initiation included being male, single, divorced or widowed, being unemployed and having a history of incarceration. Using drugs other than stimulants and opioids had a protective effect. Unrecorded data was again observed for variables mentioned above (32-56%), some of which were statistically significant.

Discussion
This study is one of the first studies from the EECA region conducted under the operational conditions and assessing the performance of the cascade between HIVtesting and ART among PWID.
The study shows that close to 200,000 PWID were tested for HIV thereby opening the gateway to harm reduction and care services. About half of the HIV positive cohort were enrolled into CITI and about twoin-ten eventually initiated ART. CITI enrolment was associated with a number of factors including drug use frequency and period as well as criminal records. ART initiation showed similar associations.
The study highlights the important role NGOs can play as a health system strategy of "reaching-out" to PWID and other key populations. This is important to the SDG goal of achieving Universal Health Coverage, ending the HIV epidemic by 2030 and "leaving none behind" [17].
Considering that PWID are a difficult-to-reach population, commendable numbers were enrolled in CITI and ART. However, the drop-outs in the journey to ART initiation needs focused attention. This is needed if the test-and-treat moto that maximizes HIV prevention through ART is to be achieved [18].
The strengths of the study were that it involved 16 of 24 regions in Ukraine with over 50 implementing partners and thus likely to be representative of the ground reality. "Cohort analysis" allowing an assessment of performance of the cascade towards ART initiation was useful to identify performance gaps and this is an identified national operational research priority. This is the first study using the cohort approach and assessing the journey from HIV testing to ART. Finally, we followed the STROBE guidelines for reporting of observational research [19].
The main study limitation was missing data on variables related to drug use and drug types, employment status and history of incarceration. This lapse may be linked to practical difficulties in gathering self-reports from people who may have an altered state of mind while on injectable opioids and sedatives. Alternatively, it may reflect the attitude of PWID who are simply hesitant to provide information due to fear or legal consequences. We cannot also exclude shortcomings in actual data recording which needs increased vigilance during supervised visits. In any case, findings ways to build trust of PWID and ensure data confidentiality would be important if the completeness of self-reported data on PWID is to be improved. This is all the more relevant since unrecorded data showed significant associations with CITI and ART initiation.
A number of initiatives to extend a default standard of care for involving PWID are known. Authors [20] assessed whether a strengths-based case management intervention in California, USA had helped smokers of crack cocaine and regular PWID to achieve the viral load suppression. The study claims that strengths-based case management may help highly vulnerable group to achieve undetectable HIV viral load over time. Another USA-based study [21] focused at the local intervention program in Maryland showed that IDUs and those who are not on methadone treatment are less likely to initiate ART. Another study has shown that engagement in methadone promoted ART initiation and fact of incarceration has negative association with ART initiation [22].

100S
The findings from this study have a policy and practice implications. First, although considerable numbers of PWID were enrolled into CITI and on ART, only 23% accessed ART. However, the latter is almost twice what has been reported in the Russian Federation [23,24]. ART enrollment in our study reflects the endpoint of cohort analysis approach which has not been the case with other published studies reported on ART uptake in Ukraine and other parts of the world [25,26]. Studies which do not use the cohort approach will tend to exaggerate ART uptake figures [27]. ART (antiretroviral therapy), HIV (human immunodeficiency virus), COR (crude odds ratio), AOR (adjusted odds ratio), IQR (interquartile range), SD (standard deviation).

101S
However, there is no room for complacency and APH teams need to find innovative ways to increase the proportions of initiating ART. We do not know the real reasons why individuals did not enroll in CITI and on ART. It might be related to cumbersome pathways and barriers in the journey to accessing ART. The mean time to ART initiation was a long 42 days which is suggestive. Patient-related factors may also be responsible for not initiating ART.
In any case, this calls for exploring new and innovative ways of further "differentiating" HIV/AIDS care for PWID. Qualitative research would help identify the real reasons why PWID do not enroll in CITI and eventually on ART. This is merited.
Second, on adjustment, CITI enrolment and ART initiation were significantly associated with being married, avoiding frequent and long duration of drug use and having no criminal records. This may be due to married individuals being exposed to a more supportive environment. Frequent and longer durations of injectable drug use particularly (opioids and stimulants) may affect the state of mind of PWID and have negative consequences on CITI and ART uptake. The fact that over half of the entire cohort were on opioids and stimulants and about a third of them were daily drug injectors is a pointer towards practical difficulties in offering counselling and dialoguing with these individuals.

Conclusion
In conclusion, mobilizing the NGO community and PWID to engage in outreach HIV testing activity and harm reduction for key populations has succeeded in opening the gateway to prevention, care and ART for thousands of PWID in Ukraine. Further steps are needed to increase ART initiation and assess if further differentiation of care for PWID subgroups would allow further gains.