Seroprevalence and Knowledge of Viral Hepatitis and HIV Among People Who Use Drugs Attending Methadone Therapy Clinic in Tanzania: A Cross-sectional Study.

Background: Methadone therapy clinics have been recently introduced in Tanzania, aiming at reducing risk behaviors and infection rates of viral hepatitis and HIV among people who use drugs. The objective of this study was to estimate the prevalence, associated factors and knowledge level of these conditions among people who use drugs attending a methadone clinic in Tanzania. Methods: We enrolled 253 People who using drugs receiving Methadone therapy. Clinical data was retrospectively collected from the medical records and face-to face interviews were conducted to determine the behavioral risk factors and respondents’ knowledge on viral hepatitis and HIV. Results: An overall seroprevalence of viral hepatitis (either hepatitis B surface antigen or anti-hepatitis C virus) was 6.3%, while that of hepatitis B virus mono infection was 3.5% and anti-hepatitis C virus mono infection was 3.5%. Seroprevalence of HIV was 12.6%. Viral hepatitis was strongly predicted by advanced age (>35 years) (p=0.02) and staying at Kirumba area (p=0.004), and HIV infection was predicted by increased age (>37 years) (p=0.04) and female sex (p<0.001). Regarding the knowledge of viral hepatitis, majority of the respondents were unaware of the transmission methods and availability of hepatitis B virus vaccines and only 17% were classied as well informed (provided ≥ 4 correct answers out of 7 questions). Good knowledge was highly predicted by higher education level of the individual (p=0.001). Conclusions: Despite the efforts to curb viral hepatitis and HIV infections through Methadone clinics, infection rates among people who use drugs are still high and the general knowledge on preventive measures is inadequate. A specic questionnaire was developed for this study where face-to-face interviews were conducted and data on the types of drug used, route, duration, behavioral risk factors for viral hepatitis, and HBV vaccination status, were captured. A questionnaire also included a series of seven closed-ended questions regarding the respondents’ knowledge on the transmission of viral hepatitis and availability of HBV vaccine to PWID in the public health facilities. Primary outcomes were the prevalence’s’ of VH and HIV, and the knowledge level regarding HBV. Individuals were considered to have good knowledge when they answered more than 3 correct answers of seven questions. comparisons baseline done in set The sensitivity and specicity of independent factors in the nal assessed to determine their discriminative ability. The Receiver Operating Characteristic (ROC) curves were used according to Hanley and McNeil's method to determine the cut points with the best sensitivity and specicity for continuous variables which were reported as proportions with 95% Condence interval (CI).


Background
Viral hepatitis (VH) is an increasing globally public health problem and various hepatitis viruses (A, B, C, D, and E viruses) have been implicated. While Hepatitis A (HAV) and hepatitis E (HEV) are frequent causes of acute sporadic infections and outbreaks, Hepatitis B (HBV) and hepatitis C (HCV), usually leads to chronic infections that can complicate to liver cirrhosis or hepatocellular carcinoma if un-intervened. Globally, about 257 million people are living with chronic hepatitis B (CHB) and 71 million people with hepatitis C. In sub-Saharan Africa (SSA), around 60 million people are estimated to have CHB and 10 million more to have chronic HCV infection (1). Additionally, Africa is an epicenter for Human immunode ciency virus (HIV), where nearly 25 million people are currently living with the disease (2).
In Tanzania, the overall seroprevalence of hepatitis B surface antigen (HBsAg), hepatitis C antibody (anti-HCV), and HIV infection is estimated to be 4.1%, 2.0%, and 4.7% respectively in the general community (3). While the rate of HIV infection in the country has decreased from previous estimates (4), studies among different sub-populations suggest an escalating rate of viral hepatitis. For instance, among people who inject drugs (PWID), HBsAg seroprevalence of 7.8% has been reported recently (5), which is over 3-fold of the earlier reports in 2006 (6). The seroprevalence of anti-HCV in this subgroup of people varies from 16.2%-50.2% (5,7,8) People who inject drugs are at increased risk of acquiring HIV, HBV, and HCV infections due to associated risk-related injecting behaviors and risky sexual practices. Following a surge of HIV and other blood-borne infections among PWID, the rst Methadone therapy clinic (MTC) was established in Tanzania in 2011, and currently there are 6 functional clinics (three in Dar es Salaam, one in Mbeya, one in Dodoma and one in Mwanza). Mwanza clinic, where this study was done, was rst opened in 2018. At present in these clinics, Methadone, which is Opioid receptor agonist, is the only used drug for the management of Opioid use disorders. However, the plans are underway to include other available approaches in the future like use of new adrenergic agonists such as Buprenorphine, use of Opioid receptor antagonists such as Naltrexone, implementation of harm reduction programs, and comprehensive HIV, Tuberculosis& hepatitis services. Around 300,000 people are estimated to be abusing narcotic drugs in the country, and among them, 30,000 use injectable drugs. Only 20% of them have been enrolled in MTCs (9). Mwanza city, which is the second largest in the country is estimated to constitute 300 PWID (10).
The majority of studies regarding the status of this population have been based in Dar es Salaam. The aims of this study, therefore, were to estimate the seroprevalence of HBsAg, anti-HCV, and HIV and to assess the knowledge regarding viral hepatitis and HIV among PWID attending MTC at Sekou Toure Regional Referral hospital (STRRH) in Mwanza.

Study design and setting
This was a cross-sectional study among PWID attending MTC in Mwanza, Tanzania from September 2019 to May 2020. A convenient sampling was done, and all 256 registered participants at Sekou Toure MTC were included.

Data collection
The included patients had attended MTC for at least 2 weeks. The patients' les were reviewed where socio-demographic information, latest status of HBsAg, HCV antibodies, and HIV were recorded. These tests are usually done 6 monthly according to the guiding standard operating procedures (11). Hepatitis B test was done by rapid HBsAg (Meriscreen HBsAg, Gujarat India) and hepatitis C was tested using rapid test for HCV antibody (Camp medica group, Bucharest Romania) according to the manufacturer's instructions. HIV tests were performed according to the Tanzanian national guidelines for HIV testing (12).
A speci c questionnaire was developed for this study where face-to-face interviews were conducted and data on the types of drug used, route, duration, behavioral risk factors for viral hepatitis, and HBV vaccination status, were captured. A questionnaire also included a series of seven closed-ended questions regarding the respondents' knowledge on the transmission of viral hepatitis and availability of HBV vaccine to PWID in the public health facilities. Primary outcomes were the prevalence's' of VH and HIV, and the knowledge level regarding HBV. Individuals were considered to have good knowledge when they answered more than 3 correct answers of seven questions.

Data analysis
Data were computerized using Epi data version 3.1 and STATA 13 (Stata Corp LP, college station, TX) was used for analysis. Continuous variables were summarized as medians with interquartile range (IQR) and categorical variables as proportions with percentages. The comparisons between the baseline characteristics of PWID in relation to the presence of HBV, HCV or HIV were done using the non-parametric Mann Whitney U test. To determine the factors that were independently associated with HBV, HCV, and HIV infections, the simple binary logistic regression was rst performed, and the variables with p < 0.2 were included in the multivariate model. The level of signi cance was set at p < 0.05. The goodness of t for the nal model was assessed subsequently. The sensitivity and speci city of independent factors in the nal model were also assessed to determine their discriminative ability. The Receiver Operating Characteristic (ROC) curves were used according to Hanley and McNeil's method to determine the cut points with the best sensitivity and speci city for continuous variables which were reported as proportions with 95% Con dence interval (CI).

Ethical Clearance
The permission to conduct and publish the ndings from this study was sought from the Catholic University of Health And Allied Sciences (CUHAS) /Bugando Medical Centre (BMC) joint ethical committee with an ethical clearance certi cate number CREC/1399/2019. The patients' information was handled by the researcher alone and their identi ers including names and registration numbers were not included in the nal analysis to further conserve con dentiality.

Characteristics and demographics of the study participants
We recruited a total of 256 PWUD attending MTC at STRRH from September 2019 to May 2020. Three clients were excluded from the study; two did not consent and one had missing data, leaving 253 for analysis. Out of 253 PWUD, the seroprevalence of viral hepatitis (either HBsAg or anti-HCV) was 6.3% (95%CI: 3.7-10.1), while that of HBV mono infection was 3.5% (95%CI: 1.2-6.6)and of anti-HCV was 3.5%(95%CI: 1.2-6.6). HBsAg and anti-HCV coexistence was detected in 0.8% (95%CI: 0.09-2.8). The seroprevalence of HIV was 12.6% (95%CI: 8.8-17.4). Coexistence of HIV and anti-HCV was detected in 0.8% (95%CI: 0.09-2.8). None of the clients was found to have HIV-HBV co infection. Demographics and characteristics of the patients with HBV infection, anti-HCV positivity, and HIV infection have been illustrated in Table 1.
People who injected drugs were predominantly males with a female: male ratio of 1:10. None of the females were found to be either positive for HBsAg or anti-HCV. Regarding the HIV infection, the females were more likely to be positive compared to their males counterparts (37.5% vs. 4.9%, p < 0.001). The age group of 25 to 30 years comprised the majority (37.9%) of the study population, while the elderly group (> 50 years) embraced the minority (0.8%). Signi cantly higher frequency of both, HBV infection and anti-HCV occurred in the age group of 36 to 40 years (p = 0.01 for both). Also, there were increased odds of higher HIV infection rate towards advanced age (41-45 years), though it was not statistically signi cant (p = 0.07). In the younger population (< 30 years), occurrences of HBV infection and anti-HCV were insigni cant.
HIV infection was signi cantly higher among the married or cohabiting PWUD (62.5% vs. 40.7%, p = 0.02). No statistical difference was observed between the marital status and HBV infection and anti-HCV seropositivity. The vast majority (92.9%) of the study subjects were not formerly employed. However, subjects that had formal employment comprised signi cantly higher rates of anti-HCV positivity (p = 0.002). This was not signi cant for HBV and HIV infections.
Regarding Opioid use, nearly 30% reported being using additional drugs. More than one-half of the study participants were using the injection method, and out of these, 124 (79.5%) were both injecting and snorting and 32(20.5%) were injecting only. About 30% of all drug users had used the Opioid for more than 10 years, and only 39(15.4%) had used for < 5 years. The types of drugs used, route and duration were all insigni cant regarding the occurrence of HBV, HCV, or HIV. Surprisingly, none of the client have had received vaccine for HBV infection (Table 1).  Table 2 and Table 3 respectively illustrate the predictors of viral hepatitis (HBsAg or anti-HCV) and HIV infection by multivariable logistic regression. Age above 35 years (p = 0.02) and Staying in Kirumba area (p = 0.004) were the only factors that were independently associated with viral hepatitis while female sex (p < 0.001), and age above 37 years (p = 0.04) strongly predicted HIV positivity. Hosmer-Lemeshow tests for goodness of t did not indicate evidence of gross lack of t with the area under the ROC curve of 0.83 for viral hepatitis (Fig. 1) and 0.80 for HIV (Fig. 2).

Discussion
This cross-sectional study was conducted to establish the magnitude of viral hepatitis and HIV, and to determine the knowledge level among PWUD. We have reported a high seroprevalence of viral hepatitis (6.3%) and HIV (12.6%) in this population. We have also demonstrated that the occurrence of viral hepatitis is predicted by age > 35 and injection route of drug use, while that of HIV is highly predicted by age > 37 years and female sex. The general knowledge regarding to viral hepatitis is modest.
Seroprevalence of anti-HCV (3.5%) and HIV (12.6%) in this study were found to be higher than the general population in the country which is estimated to be 2% and 4.7% for HCV and HIV respectively (3,4).The same trend of results has been previously reported in Dar es Salaam, Tanzania where signi cantly higher seroprevalence of anti-HCV (57%) and HIV (40%) compared to the general population among PWUD attending MTC were reported (13). These results support the notion that PWUD forms an important key population for HCV and HIV infections. The vast difference of rates between the later study and ours might be explained by the reasons that the later study was conducted from 2011-2013, immediately after commencement of MTC and harm reduction program in the country; therefore many clients were unaware of infection protection methods. Also, different study sites have different infection rates due to the urbanization effect which is a major factor for viral transmission (14)(15)(16).This effect has further been observed in the current study where the Kirumba area, which is in the Mwanza city center, comprised signi cantly more cases of viral hepatitis. Regarding HBV infection, the prevalence obtained from this study (3.5%) is lower than that of the general population in the country, which is estimated to be 4.1% (3). To date, there is no published data on the prevalence of HBV infection among PWUD enrolled in MTC. However, the magnitude of HBV infection among PWUD who are not enrolled in MTC is higher than that of the general population in the country (5,8)and globally (17). Therefore, our ndings suggest the positive impact of MTC in reducing VH in the country. Methadone replacement therapy in PWUD has been found to be effective intervention to prevent transmission of VH (18) and HIV (19), to reduce injecting-related risk behaviors (20), and to improve health-related quality of life (21), healthcare utility, utilization and expenditure (19). Nevertheless, this promising upshot should not enervate the efforts to scale up other available modalities of management of Opioid use disorders like Opioid receptors antagonists and harm reduction that are currently inexplicit in the country, as combined prevention measures have been shown to be more effective that any measure alone (22) We found a clear variation of age and gender pro les of PWUD and their association with VH and HIV infections. More than 90% of the studied population was males and one-half were below 30 years old. This is in line with a recent large meta-analysis review where it was reported that in low and mid-income countries PWUD were exclusively males and young adults while more aging with increasing women population was evidenced in highincome countries (17).Furthermore, it has been previously reported that the occurrence of VH among PWUD increases with age, with no effect on gender despite the males' preponderances (5,8,23), while that of HIV infection is strongly predicted by female sex (24,25). Similar ndings have been observed in this study. Tanzania is an endemic country for HBV infection and the commonest (> 90%) mode of transmission is perinatally through mother-to-child, with the trivial horizontal transmission in the general population. Most of the people born before the year 2002 when the vaccine against HBV was introduced in the country, remain unvaccinated to date (3), and due to unavailability of HBV Immunoglobulin, children born to HBsAg positive mothers, are likely to be habitually infected. Thus, by having insigni cant cases of HBV infection among the young adults that are mostly unvaccinated indicates that transmission is conspicuously horizontal among PWUD and likely driven by riskier substance use behaviors in this group of people.
To the best of our knowledge, this is the rst study to describe knowledge of VH among PWUD that are enrolled in MTC in the country. The present study demonstrates signi cant de cits in knowledge about viral hepatitis transmission and the availability of HBV vaccine in the public health service facilities. Only 17% of the respondents were classi ed as well informed, and < 1% responded correctly to all questions. Furthermore, none of the clients had ever received the HBV vaccine. These ndings are similar to previously reported data where lesser-educated responders and those who were negative for VH were likely to score low on VH knowledge (26,27). Some education (above primary school) strongly predicted good knowledge of VH in the index study while more than two-fold of respondents with good knowledge were observed among VH seropositive compared to their seronegatives counterparts. In the recent National Guidelines for Comprehensive Management of Opioid use disorder(28), it has been recommended that all PWUD attending MTC should receive a comprehensive health education on prevention of VH and HIV. For those who have been infected, should be linked immediately to the relevant care services while those who have not been infected by HBV, should be vaccinated. These ndings underscore the importance of intensifying heath education service, linking protocols and HBV vaccination uptake in our MAT clinics. One major limitation for this study is inability to validate the information given by the clients.

Conclusions
The current study has conclusively demonstrated that VH and HIV have substantial magnitude among PWUD attending MTC in Mwanza, Tanzania, and the rate increases signi cantly with age and geographical location for VH, and female gender for HIV. We have also revealed insubstantial knowledge of VH in this population with negligible uptake of the HBV vaccine. These ndings which are rare in the country are crucial as they do not only provide important statistics among PWUD but also feature important policy inference on tailoring comprehensive prevention programs and early intervention strategies to the risky-youth population. Moreover, our ndings inform the government and other stakeholders the discrepancies in the service provision systems that commenced nearly ten years ago in the country. Finally, these ndings might not be generalized hence they warrant a countrywide survey to all established MTCs for their e ciency. We recommend a more focused approach of different community programs dealing with PWUD to high-risk groups and close monitoring and enriching available MTCs in the country to meet the desired goals. the study was obtained from STRRH authorities. Written consent was obtained from each participant after explaining the study aims.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Not applicable
Authors' contributions SBK, DWG and DCM participated in planning, study design, data analysis, and preparation of manuscript. HJ, PMM and FK participated in data analysis and manuscript writing, ERS and FB participated in manuscript writing, GM and AS participated in collecting clinical data, interviewing the clients and manuscript writing. All authors have read and approved the nal manuscript.
28. Ministry of Health, Community Development, Gender, Elderly and Children. National guidelines for comprehensive management of Opioid use disorder. 2019.https://www.moh.go.tz/guidelines for comprehensive management of opioid use disorder. Accessed on 02 Aug 2020. Figure 1 Test for goodness of t for multivariate logistic model for viral hepatitis Figure 2 Test for goodness of t for multivariate logistic model for HIV