Abstract
Background & Aims:
We evaluated the impact of integrating HCV testing/education into integrated care centers (ICCs) delivering HIV services to people who inject drugs (PWID) across India using a cluster-randomized trial.
Methods:
We compared ICCs with usual care (UC) in the PWID stratum (12 sites) of a 22-site cluster-randomized trial. In 6 sites, ICCs delivering HIV testing, harm reduction, other preventive services and linkage to HIV treatment were scaled from opioid agonist therapy centers and operated for two years. On-site rapid HCV antibody testing was integrated after one year. To assess impact, we conducted baseline and evaluation surveys using respondent-driven sampling (RDS) across the 12 sites (n=11,993 recruited at baseline; n=11,721 recruited at evaluation). The primary outcome was population-level self-reported HCV testing history.
Results:
At evaluation, HCV antibody prevalence ranged from 7.2%–76.6%. Across 6 ICCs, 5,263 ICC clients underwent HCV testing, of whom 2,278 were newly diagnosed. At evaluation, PWID in ICC clusters were four times more likely to report being tested for HCV than UC clusters adjusting for baseline testing (adjusted prevalence ratio [aPR]: 3.69; 95% confidence interval [CI]: 1.34, 10.2). PWID in ICC clusters were also 7 times more likely to be aware of their HCV status (aPR: 7.11; 95% CI: 1.14, 44.3) and significantly more likely to initiate treatment (aPR: 9.86; 95% CI: 1.52, 63.8).
Conclusions:
We provide among the first empirical support for integrating HCV testing with HIV/harm reduction services. To achieve elimination targets, programs will need to scale-up such venues to deliver comprehensive HCV services.
Keywords: HIV, hepatitis C virus (HCV), People who inject drugs (PWID), India, Cluster-randomised trial, integrated care
Lay summary:
Delivering HCV testing to people who inject drugs (PWID) in places where they also have access to HIV prevention and treatment services is an effective way to improve uptake of HCV testing among communities of PWID.
INTRODUCTION
An estimated 71 million people are chronically infected with HCV.[1] The availability of safe, short duration, curative therapies[2–4] prompted the World Health Organization (WHO) to release targets for HCV elimination – 80% reduction in incidence and 65% reduction in mortality.[5] Achieving these targets requires treating 80% of all persons with active infection and thus making major inroads among people who inject drugs (PWID) in low- and middle-income countries (LMICs). In these settings, awareness of HCV status is well below 10% and most have not even received the most basic HCV education.[6]
In 2015, it was estimated that there were 15.6 million PWID globally, of whom 8.2 million were exposed to HCV[7] and 6.1 million had active HCV infection requiring treatment.[8] In India, data from a 2013 cross-sectional serosurvey demonstrated one in three PWID were infected with HCV.[9] Despite this high burden, at the time fewer than six percent of HCV-infected PWID were aware of their status. Moreover, the majority had not been tested because they had never heard of HCV, highlighting a compelling need for testing and educational programs. Calls have been made for integrating HCV testing with existing services including harm reduction and HIV prevention and treatment particularly for drug using populations, but there are few empirical trials to date.[10–12]
We evaluated the impact of integrating HCV antibody testing and education into community-based centers delivering integrated HIV prevention (including harm reduction) and treatment services to PWID across India using a cluster-randomized trial.
MATERIALS AND METHODS
Study design of the parent trial
The National Collaboration on AIDS (NCA) study was a cluster-randomized trial (ClinicalTrials.gov identifier: NCT01686750) designed to evaluate the effectiveness of integrated HIV prevention and treatment service delivery on recent HIV testing among men who sex with men (MSM) and PWID across 22 cities in India. A detailed description of the trial[13] and primary outcomes[14] has been published. Briefly, the intervention, integrated care centers (ICCs), was nested between serial cross-sectional surveys (baseline and evaluation) conducted in 2013 and 2016–17 (Figure 1). The parent trial was powered to examine improvements in HIV testing among MSM and PWID. A stratified restricted randomization approach was used to assign sites to either intervention (ICC) or usual care (UC).[13] Masking was not feasible. ICCs were established in 2014 at 11 intervention clusters and operated for two years prior to evaluation. We present results from an add-on study at the 12 PWID sites, 6 of which were assigned to the ICC intervention and 6 to UC.
Fig. 1. Cluster-randomized trial design.
ICC: integrated care centers; NACO: National AIDS Control Organisation, India; PWID: people who inject drugs; RDS: respondent driven sampling; UC: usual care.
A detailed description of the ICC intervention has been published.[13] Briefly, for PWID, ICCs were stand-alone centers established to provide services for PWID in a single PWID-friendly venue either within a government facility or non-governmental organization. All ICCs provided the following: rapid HIV counseling and testing (HCT), antiretroviral therapy (ART) via a link model, information, education and counseling (IEC) services, screening for tuberculosis (TB) with referrals, condoms, syndromic sexually transmitted infection (STI) management, opioid agonist therapy (OAT) and field-level syringe service programs (SSP). In UC sites, all services available in the ICC sites were also available free-of-charge but at discrete venues that were not PWID-specific but rather catered to all populations.
There were no eligibility requirements to utilize the PWID ICCs. ICCs were open to all persons regardless of gender, sexual orientation or drug use history; however, almost all clients self-reported a history of drug use. All services at the ICC were offered free-of-charge and clients could use any service on a voluntary basis. Biometric data were used to track service utilization at ICCs. Clients could refuse to provide biometric data and still utilize services. Participants were not compensated for visiting ICCs.
HCV testing and educational intervention add-on
In September 2015, on-site rapid HCV antibody testing and counseling was integrated into the 6 established PWID ICCs. Existing and new clients visiting ICCs for any purpose (e.g., OAT, SSP, HIV testing, counseling) were offered one-time free HCV testing with pre- and post-test counseling and delivery of results. In addition, information on availability of HCV testing was disseminated to the broader PWID community through outreach workers and existing ICC clients with the goal of engaging new clients in the ICCs. Participants who tested positive for HCV antibodies were counseled on ways to prevent liver disease progression (e.g., reducing alcohol consumption) and were referred to HCV treatment centers where available (Ludhiana) or to the medical gastroenterology departments of government hospitals for further evaluation as is the standard of care in India. In addition to the individual-level testing, ICCs provided group HCV educational sessions which included information on the risk factors for HCV infection, prevention strategies, interpretation of diagnostic tests, clinical course, management and treatment.
In UC sites, HCV testing availability was variable. Across all sites, antibody testing was available through private labs for a nominal charge (INR 150 – 300 [USD 2 – 4.5]). In Punjab, HCV antibody testing was available free-of-charge through government district hospitals as part of an HCV elimination program. Free HCV testing was available via non-governmental organizations in Churachandpur; further, sporadic free HCV antibody testing was conducted in UC sites by pharmaceutical companies and other organizations. Across all six UC sites, free HCV testing is recommended for all persons infected with HIV who visit an ART center.[15]
Baseline and evaluation surveys
The community-level impact of the intervention was assessed via two independent cross-sectional surveys. The first survey (baseline) took place prior to when HCV testing was incorporated into the ICCs. The second survey (evaluation) took place approximately one year after the integration of HCV testing. Both used identical eligibility criteria and procedures and recruited participants using respondent driven sampling (RDS) – a chain-referral sampling method that is expected to provide unbiased estimates of the outcome in the community by accounting for recruitment bias.[16, 17] RDS was initiated by selecting “seeds” who are considered to be highly connected and influential members of the PWID community in each city. Each seed was given two hologram-labeled referral coupons to randomly recruit two members of the local PWID community who they knew. When recruits visited the study site and enrolled, they were in turn given two coupons to recruit two more members. Serial recruitment continued until the desired sample size of 1000 PWID was achieved per site. The theory of RDS is that the deeper the chains, and the more waves of recruitment achieved, the more the sample becomes independent of the initial seeds selection and the less subject to referral bias.[16, 17]
Inclusion/exclusion criteria were identical for both surveys. Participants had to: (1) be age ≥ 18 years; (2) self-report of illicit drug injection in the prior 2 years; (3) provide informed consent; (4) possess a valid referral coupon (except seeds). Participants of all genders (male, female and transgender) and sexual orientation (homosexual, heterosexual or bisexual) were eligible. Participants could only participate once in the baseline survey and once in the evaluation survey; however, the same participant could participate in both surveys. Biometric data was used to track participants across surveys and identify duplicate enrollments.
Procedures for each of the surveys were as follows. Following verbal consent, participants underwent biometric capture. No other identifying information was collected. Participants completed an interviewer-administered survey, which captured information on demographics, PWID network characteristics, risk behavior, history of HIV and HCV testing and treatment, and substance use including alcohol. Participants underwent rapid HIV testing with pre- and post-test counseling. Upon completion of study procedures for the baseline and evaluation surveys, all participants were compensated 250 Indian Rupees (INR) (3.5 US Dollars [USD]) and provided two referral coupons. Participants received an additional monetary incentive of INR 50 (0.7 USD) for each participant referred who completed study procedures.
Samples were shipped to the central laboratory for additional testing and repository storage. HCV antibody and HCV RNA testing were conducted using repository specimens from RDS surveys. As no contact information was collected in RDS surveys, we were not able to provide HCV results from surveys to the participants. The Genedia HCV ELISA 3.0 (Green Cross Medical Science, Chungbuk, Korea) and Murex anti-HCV ELISA 4.0 (Murex Biotech, South Africa) were used to assess exposure to HCV (anti-HCV antibody) in baseline and evaluation surveys, respectively. HCV RNA was quantified in all specimens testing positive for HCV antibodies in both cross-sectional surveys using the RealTime HCV assay (Abbott Molecular, Des Plaines, IL, USA) with a lower limit of quantification (LLOQ) of 30 IU/ml.
Study Outcomes
Community-level outcomes were assessed in the evaluation RDS survey. The primary outcome of interest was self-report of prior (ever) HCV testing. Secondary outcomes included recent HCV testing (in the prior 12 months, excluding individuals who report being diagnosed with HCV more than a year prior to the survey) and other HCV care continuum outcomes including awareness of status, linkage to care, treatment initiation and sustained virologic response. Awareness of status was defined as the proportion of HCV-antibody positive persons who self-reported that they were HCV positive. Linkage to care was defined as the proportion of HCV RNA positive participants (or HCV RNA negative but self-reported that they had been treated for HCV) who reported that they had consulted a clinician for HCV care. Treatment initiation was defined as the proportion of HCV RNA positive persons (or HCV RNA negative but self-reported that they had been treated for HCV) who self-reported that they had initiated HCV treatment. Sustained virologic response was defined as the proportion of HCV RNA positive participants (or HCV RNA negative but self-reported that they had been treated for HCV) whose HCV RNA was less than the lower limit of quantification (LLOQ) at the time of survey.
Statistical Analyses
All statistical analyses excluded the RDS “seeds” and incorporated RDS-2 weights as is the standard in RDS analyses.[18] At least one author had access to all of the data and can vouch for the integrity of the data analyses. The primary analyses incorporated an intent-to-treat (ITT) approach and estimated community-level prevalence ratios (PR) for ICC vs. UC of each study outcome at the evaluation assessment adjusting for the baseline prevalence of the outcome. Analysis was performed using linear regression, using log-normal transformations of the community prevalence estimates. For some sites, no individuals reported linkage to HCV care or initiation of treatment – therefore, a small prevalence was imputed (0.05%) so that the estimate could be transformed and included in the regression model. Even though only five of six ICC sites incorporated HCV testing within the ICCs, all six sites were used in the ITT analyses. The study outcomes were also compared between ICC vs. UC clusters in an as-treated analysis that excluded the one site (Chandigarh) without HCV testing. Additional sensitivity analyses were conducted: (1) we repeated the primary and secondary outcome analyses using unweighted community-level outcomes; (2) we used HCV RNA detection to define the denominator for the testing, awareness and linkage outcomes in both RDS-2 weighted and unweighted analyses; (3) we used an individual-level approach with multi-level random effects log poisson regression models with a random intercept for each site to account for the dependence of individual-level responses within site; (4) we restricted analyses to the ICC sites and compared study outcomes among those who had vs. had not visited the ICC, verified by biometric match; and (5) we compared people in the ICC sites who did and did not report HCV testing in the prior year, excluding those who had been diagnosed more than one year prior. All statistical analyses were performed using Stata Version 15 (StataCorp, College Station, TX).
Ethical clearance
Ethical oversight of the trial was provided by the Johns Hopkins Medicine and the YR Gaitonde Centre for AIDS Research and Education institutional review boards and an independent data safety monitoring board (DSMB).
RESULTS
Participant and cluster characteristics
Across 12 PWID sites, we recruited 11,993 and 11,721 participants at baseline (2013) and evaluation (2016–17), respectively (Table 1). A total 1,631 (14%) participated in both the baseline and evaluation surveys verified via biometric match, and 2,375 (39.5%) intervention site participants surveyed in the evaluation assessment had visited an ICC. One site (Mumbai) did not meet the enrollment target of 1000 in the evaluation assessment; recruitment was stopped at 724 participants due to slower than expected enrollment. In the evaluation surveys, the median number (range) of RDS recruitment waves was 16 (9, 27) and the median time (range) for recruitment was 149 days (95, 269).
Table 1.
Demographics and risk behaviors stratified by study arm
| Usual Care | ICC intervention | |||
|---|---|---|---|---|
| Baseline | Evaluation | Baseline | Evaluation | |
| Participant-level characteristics* | ||||
| Participants, n | 5997 | 5721 | 5996 | 6000 |
| Women, n (site % range) | 341 (0 – 22.7) | 272 (0.5 – 22.0) | 346 (0.2 – 18.7) | 221 (0.6 – 15.4) |
| Age (years), site median range | 24–34 | 25–35 | 26–34 | 26–35 |
| At least secondary school education, n (site % range) | 3331 (30.7 – 95.1) | 3451 (35.6 – 97.6) | 4341 (66.1 – 93.8) | 4652 (60.7 – 94.3) |
| Married or living with partner, n (site % range) | 2331 (26.3 – 48.1) | 1993 (26.5 – 48.9) | 2999 (33.4 – 63.9) | 2674 (27.7 – 62.5) |
| Monthly household income (USD), site median range | 92 – 306 | 138 – 459 | 107 – 389 | 153 – 459 |
| Recent incarceration, n (site % range) | 648 (4.2 – 27.7) | 512 (4.4 – 24.3) | 827 (0.5 – 30.1) | 572 (2.5 –12.3) |
| Lifetime number of sexual partners, site median range | 2–4 | 2–5 | 1–5 | 1 – 5 |
| Unprotected sex in prior 6 mo., n (site % range) | 2188 (24.4 – 51.8) | 2522 (23.2 – 70.3) | 2894 (37.4 – 62.9) | 2779 (37.6 – 61.6) |
| Hazardous alcohol use1, n (site % range) | 2127 (11.5 – 49.3) | 2253 (29.5 – 51.5) | 2721 (22.2 –66.8) | 1926 (8.3 – 49.2) |
| Any recent substance use2, n (site % range) | 5900 (95.8 – 100) | 5565 (85.8 – 100) | 5752 (88.7 – 99.9) | 5432 (81.8 – 98.1) |
| Ever tested for HIV, n (site % range) | 2728 (7.9 – 61.5) | 2940 (16.6 – 68.0) | 3055 (24.3 – 71.4) | 3702 (44.3 – 70.2) |
| Ever tested for HCV, n (site % range) | 478 (0.2 – 11.8) | 783 (0.9 – 28.0) | 632 (0.7 – 27.2) | 1902 (10.9 – 52.7) |
| HIV positive3, n (site % range) | 1220 (8.6 –30.8) | 1301 (8.9 – 38.0) | 1324 (8.9 – 31.1) | 1216 (5.6 – 32.2) |
| HCV Ab+ positive, n (site % range) | 2753 (15.3 – 63.6) | 3099 (16.5 – 66.8) | 2691 (9.0 –64.9) | 3389 (7.2 – 76.6) |
| HCV active infection4, n (site % range) | 2242 (13.2 – 52.8) | 2420 (10.2 – 53.9) | 2225 (7.9 – 54.9) | 2698 (4.1 – 66.4) |
| Injection in past 6 mo., n (site % range) | 5611 (87.3 – 99.1) | 4975 (57.7 – 98.3) | 5211 (68.6 – 98.2) | 4373 (27.1 – 89.6) |
| Age at first injection (years), site median range | 18–26 | 18–25 | 18–21 | 20–22 |
| Ever shared needle/syringe, n (site % range) | 3327 (37.0 – 72.1) | 3113 (24.8 – 66.7) | 2944 (19.9 – 80.4) | 3221 (25.7 – 82.4) |
| Drugs injected in prior 6 mo., n (site % range) | ||||
| Heroin | 2875 (2.5 – 97.0) | 2967 (7.4 – 96.0) | 1958 (0.6 – 98.2) | 2699 (0.04 – 89.5) |
| Buprenorphine | 2134 (0.3 – 75.7) | 2066 (0 – 90.9) | 2289 (0.1 – 89.9) | 1640 (0 – 77.2) |
| Other pharmaceutical opioids | 1634 (0.7 – 88.7) | 844 (1.2 – 26.2) | 1396 (0.9 – 62.9) | 368 (0 – 23.6) |
| Cocaine or other stimulant | 54 (0 – 2.6) | 79 (0.1 – 3.8) | 50 (0.2 – 1.5) | 44 (0 –1.3) |
| Sedative/antianxiety | 283 (0.1 – 20.3) | 311 (0.1 – 27.1) | 627 (0.4 – 23.6) | 66 (0 – 2.3) |
| Antihistamines | 2125 (0.01 – 90.1) | 2117 (0 – 96.2) | 1992 (0.04 –73.0) | 1418 (0 – 70.4) |
| Ever use needle exchange program, n (site % range) | 2703 (6.8 – 73.8) | 2520 (8.3 – 63.2) | 2279 (7.7 – 59.2) | 2083 (6.1 – 61.1) |
| Ever use opioid agonist therapy, n (site % range) | 1627 (6.4 – 48.7) | 1975 (4.7 – 56.4) | 1296 (0 – 33.0) | 2486 (10.7 – 54.2) |
| Site recruitment characteristics | ||||
| Sites, n | 6 | 6 | 6 | 6 |
| Sample size per site, median (range) | 1000 (999 – 1000) | 1000 (722 – 1000) | 1000 (996 – 1000) | 1000 (1000 – 1000) |
| Number of seeds used per site, median (range) | 2(2–2) | 2(2–2) | 2(2–2) | 2(2–2) |
| Time to recruit sample (days), median (range) | 139 (52 – 190) | 171 (101 – 269) | 136 (89 – 200) | 139 (96 – 203) |
| Median network size, site median5 range | 16 (10 – 20) | 12 (7 – 30) | 13 (7 – 20) | 11 (4 – 20) |
| Number of recruitment waves reached, median (range) | 23 (14 – 31) | 18 (11 – 27) | 32 (14 – 50) | 16 (9 – 24) |
| Biometric overlap between evaluation and baseline RDS (%), n (site %5 range) | - | 848 (7.5 – 23.5) | - | 765 (5.8 – 18.8) |
| Biometric overlap between evaluation RDS and ICC register (%), n (site %5 range) | - | - | - | 2375 (16.7 – 55.5) |
Weighted using RDS-II weights and PWID size in each city;
Measured using AUDIT, score ≥8;
Any alcohol use or illicit drug use;
By rapid HIV testing;
HCV RNA+ or reported initiating treatment for HCV;
Unweighted; hepatitis C virus (HCV), antibody (ab), respondent-driven sampling (RDS), Integrated care center (ICC)
At the evaluation survey, the median age ranged from 26 to 36 years and 493 women were recruited. Forty percent (n=4667) of participants at evaluation were married or living with a partner and 69% (n=8103) had at least secondary schooling. The median site-level monthly household income ranged from USD 138 to 459. Unweighted data are presented in Supplementary Table 1. Overall, 9348 (80%) participants reported injecting in the prior 6 months in the evaluation assessment.
The weighted prevalence of active HCV infection (HCV RNA positive) ranged from 4.1% to 66.4% across the 12 sites in the evaluation assessment. Unweighted estimates are in Supplementary Table 1.
Uptake of HCV testing services in intervention sites.
ICCs were operational in 6 sites for a median 26.3 (range: 25.6 – 28.7) months prior to the evaluation survey. Across six sites, a total 10,757 unique clients recorded 706,978 ICC visits during the intervention period. HCV antibody testing and counseling was available in 5 of the 6 ICCs for a median 11.4 (range: 11.3 – 11.5) months prior to the evaluation. Across five sites where HCV antibody testing was incorporated, 8524 visited the ICC after testing became available, of whom 5,263 clients (62%) voluntarily underwent HCV antibody testing with appropriate pre- and post-test counseling and delivery of results, and 2,278 (43.3%) clients were newly diagnosed as HCV antibody positive.
Site-level study outcomes
At evaluation, the weighted mean percentage self-reporting ever having been tested for HCV were 28.6% and 11.6% in the ICC and UC clusters, respectively (Supplementary Table 2). Weighted mean percentages tested in the prior 12 months were 22.0% and 7.4%, respectively (Supplementary Table 3). In the community-level analysis, compared to PWID in the UC clusters, PWID in ICC clusters were 3.69 (95% confidence interval [CI]: 1.34, 10.2) times more likely to have ever been tested for HCV and 5.55 (95% CI: 1.48, 20.8) times more likely to have been tested in the prior 12 months adjusting for the baseline levels of HCV testing in the clusters. The weighted absolute mean change in ever having been tested in the ICC and UC clusters at the evaluation assessment compared to baseline were 19.6% and 5.1%, respectively – the mean difference in absolute percentage point change in the ICC vs. UC clusters was 14.5% (95% CI: 1.5, 27.5; Figure 2). These changes were 18.8% and 3.7% respectively for testing in the prior 12 months in the ICC and UC clusters at the evaluation assessment compared to baseline - the mean difference in absolute percentage point change was 15.1% (95% CI: 0.9, 29.2).
Fig. 2. Indian map depicting the absolute percentage point change in ever being tested for HCV between evaluation and baseline assessment by study arm.
Height of triangle is proportionate to the absolute percentage point change at each location. Gray triangle pointing upwards represents a positive change and black triangle pointing downwards represents a negative change.
The weighted site-median and range for HCV care continuum outcomes in the baseline and evaluation assessment are listed in Table 2. Individual site outcomes are in Supplementary Table 4. Adjusting for baseline levels of awareness of HCV status, PWID in the ICC clusters were 7 times (aPR: 7.11; 95% CI: 1.14, 44.3) more likely to be aware of their positive HCV antibody status compared to the UC clusters. PWID in the ICC clusters were also more likely to have linked to care, initiated treatment, and achieved cure compared to PWID in the UC clusters after adjusting for baseline levels of each outcome though linkage to care and cure failed to achieve statistical significance (Figure 3). Inferences remain unchanged in as-treated (excluding Chandigarh) and other sensitivity analyses (Supplementary Table 5).
Table 2.
Weighted site-level HCV care continuum outcomes in the integrated care centers (ICC) vs. usual care (UC) clusters*
| Usual Care | ICC intervention | |||
|---|---|---|---|---|
| Baseline | Evaluation | Baseline | Evaluation | |
| Aware of HCV Ab+ status, n (site % range) | 145 (0.6 – 7.3) | 252 (0.03 – 14.4) | 264 (0.08 – 17.4) | 642 (1.0 – 37.7) |
| Linked to care among HCV RNA+, n (site % range) | 60 (0.05 – 5·7) | 108 (0 – 6.0) | 122 (0 – 10.6) | 201 (0.4 – 13.6) |
| Initiated HCV treatment, n (site % range) | 30(0 – 6.7) | 52 (0 – 4.9) | 42 (0 – 6.0) | 119 (0.5 – 9.0) |
| Achieved SVR, n (site % range) | 8 (0 – 1.9) | 25 (0 – 2.8) | 9 (0 – 1.8) | 47 (0.1 – 3.2) |
Weighted using RDS-II weights; hepatitis C virus (HCV), antibody (Ab), sustained virologic response (SVR)
Linkage, treatment initiation, and SVR outcomes among HCV RNA+
Fig. 3. Community-level impact of integrated care centers vs. usual care on HCV care continuum outcomes among PWID in India.
Adjusted prevalence ratios compare the relative likelihood of outcome in the ICC vs. UC clusters. All data presented were estimated using an intent-to-treat (ITT) approach incorporating RDS-2 weights and adjusted for baseline prevalence of outcome.
Individual-level analysis
In an analysis restricted to the 6 intervention sites, we further compared persons who did and did report receiving HCV testing in the prior year. In a multivariable age-adjusted model, compared to those who had not been tested in the prior year, those who reported recent testing were significantly more likely to have visited the ICC (aPR: 1.74; 95% CI: 1.44, 2.10) accessed OAT in the prior six months (aPR: 1.72; 95% CI: 1.22, 2.43) and be recently tested for HIV (aPR for HIV negative and tested in the prior year vs. HIV negative and not tested in the prior year: 5.23; 95% CI: 3.10, 8.90) or aware of their HIV positive status (aPR: 4.35; 95% CI: 2.51, 7.51).
CONCLUSIONS
In “Test.Treat.Hepatitis”, the WHO recognized the urgent need to improve access to HCV testing and diagnosis if ambitious elimination targets for HCV are to be met.[19] In this trial, we demonstrated a statistically significant population-level impact of integrating HCV testing with HIV/harm reduction services on critical HCV care continuum outcomes including HCV testing, awareness of status and treatment initiation. However, the overall proportion of the target population who were aware of their status after this intervention remained low, and we did not observe statistically significant improvements in SVR. Nonetheless, the simplification of HCV treatment and availability of generic DAAs in over 100 countries further supports task-shifting from tertiary specialty centers to community-based centers such as the ICCs described in this trial to potentially further improve HCV testing and downstream outcomes of SVR and consequently, reduce HCV incidence.
HCV testing and counseling was incorporated into integrated HIV/harm reduction centers one year prior to the evaluation assessment. Yet, in this brief period across five cities, over 5000 clients were tested of whom nearly half were newly diagnosed with HCV. Moreover, acceptability of HCV testing was high and comparable among persons on and off OAT. The benefits of integrating HCV testing with OAT has been previously demonstrated[20] including some limited data from LMICs. For example, integration of HCV testing into an OAT program in Tanzania resulted in all 1350 PWID on OAT undergoing testing.[21] A study from Ukraine further demonstrated a correlation between being engaged in OAT and awareness of HCV status.[22] However, the impact of integration of HCV testing with HIV/harm reduction services at the community-level (regardless of whether the clients were enrolled in OAT or HIV treatment) has not previously been reported to our knowledge. Indeed, recent systematic reviews of HCV interventions[23, 24] which have highlight successful interventions for PWID including service integration, clearly point to the lack of randomized trials and a dearth of evidence on successful interventions from LMICs, a gap which this trial sought to fill.
In this study, at the community-level, PWID in the ICC cities were more than 3 times more likely to be tested for HCV compared to PWID in UC communities. Moreover, this model reached diverse clients including those not engaged in harm reduction or other HIV prevention/treatment strategies who may be most at risk of re-infection.[25] PWID in ICC cities were also seven times more likely to be aware of their HCV antibody status. Nonetheless, it is important to note that only 642 out of the 3,389 (19%) HCV antibody positive PWID in the ICC cities were aware of their status. To achieve elimination, awareness levels need to be substantially higher. Moreover, there was important variability in awareness across sites, ranging from 1% in Bilaspur to 37.7% in Imphal. Allowing ICCs to function longer and consequently reach a larger proportion of the population might have further improved community-level awareness. However, our experience with HIV service integration suggests that additional “demand” creation strategies[14] might be needed to improve population-level HCV awareness to meet elimination targets. We have previously demonstrated the potential role of network-based recruitment in improving case detection rates of HCV among PWID[26]; incorporating network referral strategies could further improve community-level uptake of HCV testing.[26] Additionally, incorporating free HCV treatment into centers like ICCs could further enhance uptake of HCV testing analogous to what was observed with HIV testing when free antiretroviral therapy programs were introduced. OAT centers represent ideal venues for HCV treatment as many have clinicians on-site who could be trained to deliver treatment and SVR among OAT clients has been comparable to non-PWID.[20]
The findings presented need to be interpreted with caution. The parent trial was designed in 2012 to evaluate the impact of an integrated care model on HIV testing among MSM and PWID. HCV testing and care continuum outcomes were not originally part of the trial. However, over the course of the study, two major developments emerged. First, sofosbuvir was licensed for use in the US in 2013 and negotiations began for generic licensing of sofosbuvir in India. Second, during pre-trial ethnography, HCV testing at the ICCs was raised as a need by PWID in the Northeast. These findings prompted us to seek independent funding to incorporate HCV testing at the ICCs and testing of stored specimens to evaluate population-level impact.
HCV testing increased substantially in two UC cities, both of which had testing available free-of-charge at least some time during the study period through either governmental or nongovernmental program, so we cannot rule out the possibility that free rather than integrated testing at least in part drove results. Several of the study outcomes were based on self-report. However, it appears unlikely that there would be differences in social desirability between ICC and UC clusters. Baseline and evaluation assessment samples were accrued using RDS and RDS weights were utilized to arrive at population-level estimates. There has been criticism of the use of RDS[27, 28] but given the inability to randomly sample PWID, RDS is the most widely used method to sample “hidden” populations. Further, all samples satisfied RDS process measures. Finally, HCV testing was integrated only in the PWID sites which nearly halved the number of clusters used resulting in large confidence intervals and limited power.
Limitations notwithstanding, these data provide among the first empirical support of the benefits of integrating HCV testing with HIV prevention and treatment services for PWID. Over a short duration, we observed significant impact on community-level HCV testing and awareness of HCV status among PWID. While additional strategies might be required to improve population awareness levels, integration of HCV testing with HIV programs for PWID particularly given the high burden of HIV/HCV co-infection represents a critical first step. To achieve elimination targets, programs will also need to consider delivering HCV treatment from such venues.
Supplementary Material
ACKNOWLEDGEMENTS
This research study was supported by the National Institutes of Health (NIH) (R01DA032059, R01MH089266, K24DA035684, R01DA041034, DP2DA040244, F31DA044046, T32AI102623) and the Johns Hopkins University Center for AIDS Research (P30AI094189). Additional support was provided by the Office of AIDS Research, Division of Intramural Research, National Institute of Allergy and Infectious Diseases, NIH, Intramural-to-India program, and the Elton John AIDS Foundation. We thank the National AIDS Control Organisation and the State AIDS Control Societies for their support of this study.
Conflict of Interest Statement: SSS and SHM received speaker fees from Gilead Sciences for work outside the study.
Financial Support: National Institutes of Health and the Elton John AIDS Foundation.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ClinicalTrials.gov identifie: NCT01686750.
REFERENCES
- [1].Polaris Observatory HC. Global prevalence and genotype distribution of hepatitis C virus infection in 2015: a modelling study. Lancet Gastroenterol Hepatol 2017;2:161–176. [DOI] [PubMed] [Google Scholar]
- [2].Feld JJ, Jacobson IM, Hezode C, Asselah T, Ruane PJ, Gruener N, et al. Sofosbuvir and Velpatasvir for HCV Genotype 1, 2, 4, 5, and 6 Infection. N Engl J Med 2015;373:2599–2607. [DOI] [PubMed] [Google Scholar]
- [3].Foster GR, Afdhal N, Roberts SK, Brau N, Gane EJ, Pianko S, et al. Sofosbuvir and Velpatasvir for HCV Genotype 2 and 3 Infection. N Engl J Med 2015;373:2608–2617. [DOI] [PubMed] [Google Scholar]
- [4].Kwo PY, Poordad F, Asatryan A, Wang S, Wyles DL, Hassanein T, et al. Glecaprevir and pibrentasvir yield high response rates in patients with HCV genotype 1–6 without cirrhosis. J Hepatol 2017; 67(2):263–271. [DOI] [PubMed] [Google Scholar]
- [5].World Health Organization. Global Hepatitis Report, 2017. [cited 2018 October 23]; Available from: http://apps.who.int/iris/bitstream/handle/10665/255016/9789241565455-eng.pdf?sequence=1
- [6].World Health Organization. Guidelines for the Care and Treatment of Persons Diagnosed with Chronic Hepatitis C Virus Infection. 2018. [cited 2018 October 23]; Available from: http://apps.who.int/iris/bitstream/handle/10665/273174/9789241550345-eng.pdf?ua=1 [PubMed]
- [7].Degenhardt L, Peacock A, Colledge S, Leung J, Grebely J, Vickerman P, et al. Global prevalence of injecting drug use and sociodemographic characteristics and prevalence of HIV, HBV, and HCV in people who inject drugs: a multistage systematic review. Lancet Glob Health 2017;5:e1192–e1207. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [8].Grebely J, Larney S, Peacock A, Colledge S, Leung J, Hickman M, et al. Global, regional, and country-level estimates of hepatitis C infection among people who have recently injected drugs. Addiction 2019; 114(1):150–166. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [9].Solomon SS, Mehta SH, Srikrishnan AK, Solomon S, McFall AM, Laeyendecker O, et al. Burden of hepatitis C virus disease and access to hepatitis C virus services in people who inject drugs in India: a cross-sectional study. Lancet Infect Dis 2015;15:36–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [10].Reece R, Dugdale C, Touzard-Romo F, Noska A, Flanigan T, Rich JD. Care at the Crossroads: Navigating the HIV, HCV, and Substance Abuse Syndemic. Fed Pract 2014;31:37S–40S. [PMC free article] [PubMed] [Google Scholar]
- [11].Kresina TF, Bruce RD, Cargill VA, Cheever LW. Integrating care for hepatitis C virus (HCV) and primary care for HIV for injection drug users coinfected with HIV and HCV. Clin Infect Dis 2005;41 Suppl 1:S83–88. [DOI] [PubMed] [Google Scholar]
- [12].Birkhead GS, Klein SJ, Candelas AR, O’Connell DA, Rothman JR, Feldman IS, et al. Integrating multiple programme and policy approaches to hepatitis C prevention and care for injection drug users: a comprehensive approach. Int J Drug Policy 2007;18:417–425. [DOI] [PubMed] [Google Scholar]
- [13].Solomon SS, Lucas GM, Celentano DD, McFall AM, Ogburn E, Moulton LH, et al. Design of the Indian NCA study (Indian national collaboration on AIDS): a cluster randomized trial to evaluate the effectiveness of integrated care centers to improve HIV outcomes among men who have sex with men and persons who inject drugs in India. BMC Health Serv Res 2016;16:652. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Solomon SS, Solomon S, McFall AM, Srikrishnan AK, Anand S, Verma V, et al. Integrated HIV testing, prevention, and treatment intervention for key populations in India: a cluster-randomised trial. Lancet HIV 2019;6:e283–e296. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [15].National AIDS Control Organization (NACO) I. National Technical Guidelines on Antiretroviral Treatment. 2018. [cited 2019 July 16]; Available from: http://naco.gov.in/sites/default/files/NACO%20-%20National%20Technical%20Guidelines%20on%20ART_October%202018%20%281%29.pdf
- [16].Heckathorn DD. Respondent-driven sampling: a new approach to the study of hidden populations. Social Problems 1997;44:174–199. [Google Scholar]
- [17].Heckathorn DD. Respondent Driven Sampling II: deriving valid population estimates from chain-referral samples of hidden populations. Social Problems 2002;49:11–34. [Google Scholar]
- [18].Volz EM, Heckathorn DD. Probability based estimation theory for respondent driven sampling. Journal of Official Statistics 2008;24:79–97. [Google Scholar]
- [19].World Health Organization. Urgent increase in hepatitis testing and treatment needed. 2018. [cited 2019 March 15]; Available from: http://www.who.int/hepatitis/news-events/hepatitis-testing-treatment-increase/en/
- [20].Hajarizadeh B, Cunningham EB, Reid H, Law M, Dore GJ, Grebely J. Direct-acting antiviral treatment for hepatitis C among people who use or inject drugs: a systematic review and meta-analysis. Lancet Gastroenterol Hepatol 2018;3:754–767. [DOI] [PubMed] [Google Scholar]
- [21].Mohamed Z, Rwegasha J, Kim JU, Shimakawa Y, Poiteau L, Chevaliez S, et al. The hepatitis C cascade of care in people who inject drugs in Dar es Salaam, Tanzania. J Viral Hepat 2018; 25(12):1438–1445. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Iakunchykova O, Meteliuk A, Zelenev A, Mazhnaya A, Tracy M, Altice FL. Hepatitis C virus status awareness and test results confirmation among people who inject drugs in Ukraine. Int J Drug Policy 2018;57:11–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Bajis S, Dore GJ, Hajarizadeh B, Cunningham EB, Maher L, Grebely J. Interventions to enhance testing, linkage to care and treatment uptake for hepatitis C virus infection among people who inject drugs: A systematic review. Int J Drug Policy 2017;47:34–46. [DOI] [PubMed] [Google Scholar]
- [24].Meyer JP, Moghimi Y, Marcus R, Lim JK, Litwin AH, Altice FL. Evidence-based interventions to enhance assessment, treatment, and adherence in the chronic Hepatitis C care continuum. Int J Drug Policy 2015;26:922–935. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [25].Martin NK, Vickerman P, Miners A, Foster GR, Hutchinson SJ, Goldberg DJ, et al. Cost-effectiveness of hepatitis C virus antiviral treatment for injection drug user populations. Hepatology 2012;55:49–57. [DOI] [PubMed] [Google Scholar]
- [26].Solomon SS, McFall AM, Lucas GM, Srikrishnan AK, Kumar MS, Anand S, et al. Respondent-driven sampling for identification of HIV- and HCV-infected people who inject drugs and men who have sex with men in India: A cross-sectional, community-based analysis. PLoS Med 2017;14:e1002460. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [27].Gile KJ, Handcock MS. Respondent-Driven Sampling: An Assessment of Current Methodology. Sociol Methodol 2010;40:285–327. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [28].Goel S, Salganik MJ. Assessing respondent-driven sampling. Proceedings of the National Academy of Sciences of the United States of America 2010;107:6743–6747. [DOI] [PMC free article] [PubMed] [Google Scholar]
REFERENCES
- i.Hopewell S, Clarke M, Moher D, Wager E, Middleton P, Altman DG, et al. CONSORT for reporting randomised trials in journal and conference abstracts. Lancet 2008, 371:281–283 [DOI] [PubMed] [Google Scholar]
- ii.Hopewell S, Clarke M, Moher D, Wager E, Middleton P, Altman DG et al. (2008) CONSORT for reporting randomized controlled trials in journal and conference abstracts: explanation and elaboration. PLoS Med 5(1): e20. [DOI] [PMC free article] [PubMed] [Google Scholar]
- iii.Ioannidis JP, Evans SJ, Gotzsche PC, O’Neill RT, Altman DG, Schulz K, Moher D. Better reporting of harms in randomized trials: an extension of the CONSORT statement. Ann Intern Med 2004; 141(10):781–788. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.