Prevalence and high risk behaviours associated with HCV testing among people who inject drugs: a systematic review and Meta-analysis

Background Hepatitis C diagnosis could be a gateway to behavioral change and subsequent decline in transmission among people who inject drugs (PWIDs). We assessed the association between the social determinants of PWID, their risk behaviors and hepatitis C testing. Methods We searched for studies in English published before May 1, 2020, on PubMed, Scopus, Cochrane, and Web of Science to identify primary studies on the factors associated with hepatitis C virus (HCV) testing among PWID. After reviewing for study duplicates, the full-text of selected articles were assessed for eligibility using Population, Intervention, Comparator, Outcomes (PICO) criteria. i) population: individuals who report injecting drugs; ii) intervention: HCV testing in the past year; iii) comparator: PWIDs who did not have an HCV test; iv) outcome: HCV testing among PWIDs and v) study type: cross-sectional, cohort, and case-control studies. Two independent reviewers (author BA and AB) chose the references in a two-phased monitoring process. The authors gathered data from selected papers, including the surname of the first author, publication date, participant demographic data (age, sex, and level of education) and other characteristics like previous HCV testing, past treatment attempts, duration of injecting drug use and condomless sex. We used fixed and random-effects meta-analysis models to estimate the pooled prevalence, pooled odds ratio (OR), and 95% confidence intervals. The data were analyzed using Stata 12.0 software. Results After a detailed assessment of over 12,000 articles, a total of 16 studies containing 38,952 participants met the eligibility criteria. Our findings showed a pooled prevalence rate of 61.01% (95% CI, 34.65–84.32%) for recent HCV testing among PWIDs. Being female (OR = 1.69, 95%CI = 1.13, 2.26), aged > 30 years, (OR = 2.61, 95%CI = 1.66–3.56) having past treatment attempt (OR = 2.24, 95%CI = 1.80–2.68), and reporting a previous test (OR = 2.03, 95%CI = 1.23–2.82). were significantly associated with having a recent HCV test.,,. Finding of present study was that unprotected sex had a negative association with HCV testing. Those PWIDs who had unprotected sex were 0.56 times less likely to have completed HCV testing during last year (OR = 0.56, 95%CI = 0.33–0.78). Conclusion Prevention programs that address age > 30 years, being female, past treatment attempt, previous testing of safe sexual practices, are strongly recommended to prioritize HCV risk reduction strategies.


Background
Hepatitis C virus (HCV) infection is a major public health problem, causing a preventable liver-related morbidity and mortality globally [1][2][3]. The World Health Organization (WHO) has defined targets for HCV treatment and diagnosis, representing a step towards the goal of elimination by 2030 [4]. Nevertheless, rates of HCV testing, treatment, and linkage to care are still low in many places around the world [5]. A number of social factors impact on treatment access including unstable housing, social stigma, health care providers' attitudes, criminalization of drug use, and gender [6][7][8][9]. Historically, HCV infection has been described as a silent epidemic, with estimations suggesting that half of those infected are not currently aware of their HCV status [10]. People who inject drugs (PWIDs) carry a higher HCV burden than other people, with HCV prevalence in this group being estimated at over 50% [11,12]. The criminalization of drug use and the political response means custodial settings have high rates of HCV prevalence [13][14][15].
The sharing of previously used injecting equipment means that PWID are the population most at risk for HCV infection in nearly all middle and high-income countries. Undiagnosed and untreated HCV infection can cause cirrhosis, hepatic decompensation, liver cancer, and death [16]. HCV diagnosis itself could be a helpful strategy [17] in behavioral change [18] and producing a decline of HCV transmission [19,20].
As part of the elimination strategy, WHO suggests regular testing of PWIDs (at least yearly) [21]. However, such regular HCV testing by active drug users relies on many factors, for example, heightened levels of risk awareness among PWID, availability of testing sites and drug-treatment workers who integrate testing as a normal part of their work [22]. Low rates of HCV treatment uptake among PWIDs have been attributed to barriers at the provider, patient and systems levels. Not least of which have previously been treatment complexity together with limited capacity, the side effects from interferon-based treatment, social stigma and discrimination which may influence the willingness of physicians to provide treatment for PWIDs [23][24][25].
In addition to financial barriers, other barriers that might prevent PWIDs from accessing HCV treatment and testing include the high levels of discrimination and stigma that may exist inside of conventional healthcare settings, which may influence on patient-provider relationships and the willingness on the part of physicians to treat PWIDs [22,25].
Despite evidence that the availability of direct acting antivirals has made HCV testing and treatment more accessible [26] PWIDs continue miss out. Making HCV (anti-HCV) and verifying chronic HCV infection by RNA testing are the essential first steps along the treatment cascade [27]. The aim of this investigation was to clarify the contributing factors for engaging PWIDs in HCV testing.

Methods
The present study was conducted based on the instructions in Protocols of Systematic Reviews and Meta-Analyses (PRISMA) [28,29].

Search strategy and study selection
The process of our study selection is presented in Fig. 1.
We selected 8348 papers from across 4 databases, and included additional manual searches of the reference lists for papers published before May 1, 2020. To identify and select studies two independent researchers (A.B. and B.A.) reviewed the electronic databases of PubMed, Scopus, Web of Science, and Cochrane independently (Table 1).
We included English language papers in the study, and also, we considered limitations such as time and geographic items. We reviewed papers twice based on the abstract and on the relevancy to the subject.
Inclusion criteria based on PICOS: Population: Individuals who report injecting drugs Intervention: HCV testing in the past year The Comparison Group: PWIDs who did not conduct HCV testing Outcomes: HCV testing among PWIDs. Study design: We included cross-sectional, cohort, and case-control studies.
We excluded qualitative studies, secondary studies, systematic reviews and, meta-analysis studies, and also, we excluded non-English language papers from our study.

Data extraction and study quality assessment
Two independent researchers (AB and BA) reviewed and assessed the papers applying a standardized data collection form. Any contradictions between the two researchers were eliminated by discussing with two other members of the research team (EA and YF). We used Microsoft Excel software for data extraction and management. Two independent reviewers (author BA and AB) chose the references in a two-phased monitoring process. First, we eliminated duplicated titles/abstracts (89% agreement) according to the criteria one through three mentioned below. Second, titles/abstracts that met these initial criteria were selected for full-text review based on the inclusion criteria (96% agreement).
The authors gathered data from selected papers, including the surname of the first author, publication date, participant demographic data (age, sex, and level of    education) and other characteristics like previous HCV testing, past treatment attempts, duration of injecting drug use and unprotected sex. The Newcastle-Ottawa Scale (NOS) [30] proposed by the Cochrane Collaboration [31] was used to evaluate the quality of the reviewed papers (Table 2).
To evaluate the quality of papers included in the meta-analysis, we applied a modified version of NOS. This was used to evaluate statistical quality, sample representativeness, sample size, and comparability between people were the domains used for the NOS when evaluating the quality of individual studies. We applied agreement beyond chance (unweighted kappa) for evaluating agreement between the two authors (BA and AB) during the quality assessment. The levels of poor, slight, fair, moderate, substantial, and almost perfect levels of agreement were showed by the values 0, 01-0.02, 0.021-0.04, 0.041-0.06, 0.061-0.08, and 0.081-1.00, respectively [48].

Data synthesis and statistical analysis
The present meta-analysis was conducted by generating pooled odds ratios (OR) and the 95% confidence intervals on recognizing factors associated with HCV testing among PWID. We computed the OR applying a 2*2 table, and we considered OR < 1 as a positive association between HCV testing and the target variable. An OR >1 (as the statistical threshold for assessing the relationship between outcome variables and expositive variables) suggests a protective relationship between variables and vice versa. To evaluate the lack of correlation between studies, we applied Q test with a P value < 0.05 and I 2 statistics with a cutoff of ≥50%. We assessed 95% confidence intervals for I2 where we considered negative values as zero. We applied the random-effects model to calculate pooled estimation, considering the different sampling methods of the studies. To identify any publication bias, we used Begg's and Egger's publication bias approach both in graphical and statistical manners [49,50]. Pvalues of less than 0.05 were considered significant. We demonstrated the association between social and demographical determinants by an OR and 95% CI, and showed the results in forest plots. For data analysis, we conducted the meta and metabias commands in STATA version 13.0 (STATA Corporation, College Station, TX.

Results of the meta-analysis
In Table 2, the key characteristics of the included studies for factors associated with HCV testing uptake among PWIDs are presented. The impact can be inferred through plots 2-9. Figure 2 indicates the association between being age > 30 years and HCV testing among PWIDs. Our findings show PWIDs aged > 30 years were 2.61 times more likely to report having had an HCV test during last year (OR = 2.61, 95%CI = 1.66-3.56).
As illustrated in Figs. 3, and 4, there were no significant associations found with HCV testing among PWIDs and their duration of injecting (OR = 1.92, 95%CI = 0.50-3.33), and high school education (OR = 1.55, 95%CI = 0.32-2.78). Figure 5 shows the positive impact of being female and HCV testing among PWIDs, and the overall heterogeneity was about 0%. The pooled effect size has a respective positive impact and the lower boundary is about 1.13, and the higher is 2.26. The OR indicated a positive and effective role that being female has on HCV testing among PWIDs. Participants who were female were 1.69      Fig. 6 indicates a significant correlation between past treatment attempts and HCV testing among PWID. PWIDs in this situation were 2.24 times more likely to have had an HCV test in the previous 12 months than PWIDs who did not report any past treatment attempts (OR = 2.24, 95%CI = 1.80-2.68).
Another finding from our data was previous testing had a positive association with HCV testing. Those PWIDs who had a previous test were 2.03 times more likely to have completed a recent HCV test (last 12 months) (OR = 2.03, 95%CI = 1.23-2.82) (Fig. 7).
Finding of present study was that unprotected sex had a negative association with HCV testing. Those PWIDs who had unprotected sex were 0.56 times less likely to have completed HCV testing during last year (OR = 0.56, 95%CI = 0.33-0.78) (Fig. 8). Figure 9 shows a significant pooled prevalence of 61.01% (95% CI, %34.65-%84.32) for the prevalence of HCV testing among PWIDs.
To identify the probable publication bias, the Egger's test and the graph were performed. Considering the symmetry assumption, there was no significant publication bias in the reviewed studies selected for inclusion.
As regards the funnel plot, the distribution of the articles was not oriented and for most of them, it was identical confirming no publication biases observed in our study. The publication bias test indicates considerable bias based on Eggers test (coefficient = 3.66, P value < 0.001) (Fig. 10).

Discussion
This meta-analysis study aimed to explore factors associated with recent HCV testing among PWIDs. The prevalence of recent HCV testing uptake was high. Past 12 months testing for HCV was associated with being aged > 30 years, being female, having a past treatment attempt and reporting previous HCV testing. HCV testing was reported by 61.01% of the study participants which could indicate enhanced access to harm reduction interventions. Many drug treatment services require initial HCV screening for PWIDs and such services are in the perfect position to conduct long-term follow-up and monitoring of HCV testing in this population.
Moreover, a strong predictor for a recent HCV test was being over the age of30 years. Given the correlation between a longer duration of drug injecting and the increased odds of high-risk injecting. This may reflect an association between increased HCV test-taking and Fig. 6 The association between past treatment attempt with HCV testing uptake among PWID Fig. 7 The association between previous HCV testing with HCV testing uptake among PWID longer injecting history. However, studies also show that HCV infection can also occur in the early stages of people's injecting careers [51]. While the risks are high in this group evidence suggests they ae also more likely to be conscious of the advantages in HCV testing [52,53]. In addition, while recent HCV testing and female gender were independently correlated the risks are more challenging to manage. Sex work and injection-assistance [54] have been shown to increase vulnerability to HCV infection [38]. Addressing injecting couple's relationships in terms of developing and conducting HCV testing and considering gender-specific needs are necessary if increased HCV testing is to be achieved [55]. Our study suggests that accessing drug treatment by PWIDs is a determining factor in recent HCV test-taking. In line with this result, other studies have highlighted the advantages of drug treatment in facilitating access to healthcare services for PWIDs [56,57] especially HIV and HCV treatment programs [58]. However, when excluding treatment from the variable definition, drug treatment and recent HCV testing were not associated and this may be a result of inadequate staff training on the risks and transmission methods of HCV.
There is evidence that providing education programs for drug treatment staff aimed at improving the HCV awareness i is beneficial in promoting HCV services among PWID [59]. HIV service integration highlights the need for developing strategies due to increased demands [9,60]. Such plans could help to improve the HCV-related knowledge of the general population, which in turn results in eliminating targets. Networkbased recruitment and the use of PWID peer facilitators might enhance the detection rates of HCV cases [61][62][63]. Accordingly, network referral strategies could be combined to increase HCV testing and treatment at the community level.
Furthermore, there was an independent association between confirmatory PCR testing and undergoing HCV testing within the past 12 months. Enhanced awareness of the need for confirmatory testing in healthcare providers and PWIDs is essential if WHO elimination goals are to be achieved. Approximately 33% of PWIDs with histories of active HCV infection had received HCV professional surveys (e.g., liver biopsy or FibroScan1) before being introduced to interferon-free regimens. However, there is a significant gap between healthcare engagement and the receipt of HCV treatment in the era of interferon-based therapy. It is reported that 1-2% of PWIDs received treatment per year. Additionally, a cumulative therapy of 10% at the end of 2011 are extensively in line with these data [39]. Consistent with prior research [33,39] we found that having unprotected sex was negatively associated with having been tested for HCV. There is a data discrepancy on HIV-related counseling programs and undertaking the relevant test in terms of high-risk sexual behaviors. A tremendous body of literature suggested a major decline in claiming multiple partnerships and unprotected sexual relationships. On the other hand, some investigators [64][65][66], documented either adverse or poor effects of test-taking on behavioral modifications [67,68]. Sex-related high-risk behaviors, such as the lack of using sexual protection tools and fisting, which in turn, could be mucosal traumatic and result in bleeding. These factors have also been suggested to be related to contracting HCV risk [69][70][71]. It is still debatable whether bleeding plays an essential role in the transmission of HCV. Some investigations have detected HCV in seminal and rectal fluids in males who were infected with HIV and supported that such fluids could interfere with the transmission of HCV [72,73].

Limitations and strengths of the study
The study limitations include first, relying on HCV selfreports (study samples might have failed to completely percept the HCV diagnosis approach, i.e., comprising 2 test forms, the reported data may be biased by overestimating). Although not commonly implemented such limitations could be eliminated by achieving confirmation of prior HCV testing. Second, most of the included studies were cross-sectional meaning causal and temporal relationships between risk behavior and HCV testing are not possible. However, this meta-analyses may enhance the statistical inference of analyses and incresase the reliability of the evidence. Third, few studies investigated the association between risk behavior and HCV testing, emphasizing this gap in the literature. Also, since we did not interfere with the setting of independent and dependent variables, we had to report only the data that were published in the articles. A.
The strengths of our study include the number of high-quality studies reviewed, the large representative sample and the multivariate analysis which regulaed for potential confounders. These factors provided greater statistical power and strengthened the results of the reviewed studies and enhanced the chance of recognizing a true effect of exposure [74]. Longitudinally design research aiming at chronic HCV infection detection as well as HCV treatment enrollment, are warranted in PWIDs.

Conclusion
The findings of the present study illustrate the important factors that may be effective to increase HCV testing rates among PWID, which may improve prevention and reduce transmission. This study concludes that better understand the social determinants of injecting risk