According to a World Health Organization (WHO) report in 2018, there are approximately 38 million people living with HIV/AIDS in the world, the largest number being in Africa, with 25.7 million [1
]. The WHO’s goal is to identify 90% of HIV patients, treat 90% of those identified and virally suppress 90% of those treated by 2020 [2
]. Hepatitis C virus (HCV) and hepatitis B virus (HBV) have affected 71 million and 257 million people worldwide, which can lead to cirrhosis and liver cancer, respectively. Removing these viruses, which requires significant economic and sanitary capital, will prevent more than 1.2 million deaths worldwide annually [3
]. HCV has the highest prevalence in the Eastern Mediterranean region, followed by the European region, with a prevalence of 2.3% and 1.5%, respectively. It varies from 0.5% to 1% in other areas. HBV is most prevalent in the Western Pacific and African regions, with values of 6.2% and 6.1%, respectively. The lowest prevalence is reported in the US, with a value of 0.7% [5
]. Viral hepatitis elimination program has been adapted by WHO in 2016 [6
]. In line with this global program, several countries are working towards this elimination platform [7
]. In this program, engagement of high-risk groups and marginal populations has been highlighted [8
]. The efficient implementation of prevention and control measures incorporated in this program, requires detailed insight on the epidemiology of hepatitis viruses and other viral infectious agents circulating in these cohorts [9
People who inject drugs (PWID) and female sex workers (FSWs) are key populations for blood-borne viral infections (BBVI), including HIV, HCV and HBV. PWID are usually infected through shared needles, syringes and other infected injection equipment, as well as other high-risk behaviors [10
]. There are an estimated 15.6 million PWID aged 15–64 years worldwide. It is said that 23% of the new HCV cases and 33% of the annual HCV deaths are related to PWID [11
]. According to the WHO report in 2018, HIV prevalence in PWID is estimated to be from less than 1.8% to 13.5%, but viral hepatitis has not been reported in many countries. HBV is 0.7% in some of these countries, including Afghanistan, Germany and Nepal, and 7.3% in Azerbaijan, and HCV is less than 7% in some countries, including Germany, Afghanistan and Madagascar, and more than 60% in Kazakhstan [13
FSWs are also more exposed to high-risk behaviors, especially through sex. A study in a region with a high prevalence of HIV in India stated that about 77.5% of FSWs had drug injection history, and that they were at a higher risk of BBVI than those without drug injection. The prevalence of HIV in FSWs worldwide ranges from less than 1.4% to over 11%. The prevalence of viral hepatitis is not high in FSWs, but WHO reports that the prevalence of HBV in FSWs ranges from less than 0.3% in Brazil to over 3.6% in Peru, and hepatitis C prevalence ranges from less than 1.9% in Brazil to 6.2% in Kazakhstan in the same group. In fact, HIV is more prevalent among FSWs than HBV and HCV infections [14
]. The prevalence of HIV is 22 times higher among PWID and 21 times higher among sex workers (SWs) than the general population. In 2018, about 54% of new HIV cases occurred among key populations and their sexual partners [17
In patients co-infected with two or three HIV, HCV and HBV infections, HIV-induced immunodeficiency increases the likelihood of HBV and HCV persistence, and hepatotoxicity associated with anti-HIV treatment can worsen HBV-related liver diseases or HCV persistence. Evidence suggests that HIV infection increases the risk of HCV- and HBV-associated hepatocellular carcinoma. On the other hand, liver diseases associated with chronic HBV and HCV are leading to increased mortality and complications in HIV patients [18
]. Adverse drug reactions (ADRs) related to antiretrovirals (ARVs) are higher in HIV patients co-infected with HBV or HCV than in HIV-monoinfected patients [20
The purpose of this study was to conduct a systematic review and meta-analysis to estimate the prevalence of HIV (anti-HIV), HCV (anti-HCV), HBV (HBsAg) and their co-infections among PWID and FSWs, as separated by WHO geographical areas in the general population of these groups. The aim is to understand the current situation, in order to make better decisions regarding prevention, identification and treatment.
The protocol of this study had been registered in the International Prospective Register of Systematic Reviews (PROSPERO) (CRD42018115115).
This systematic review and meta-analysis examined the prevalence of HIV, HCV, HBV and their co-infections among PWID and FSWs worldwide, and the results were shown by different WHO regions. The prevalence of HIV in PWID and FSWs was 15% and 5%, respectively. That means one of seven PWID and one of 20 FSWs get infected with HIV, with the highest prevalence in PWID being in Africa (24%), South-East Asia (22%) and Europe (12%), and in FSWs being in Africa (19%) and South-East Asia (18%). The number of people living with HIV worldwide in 2017 was estimated to be 36.8 million [145
]. HIV is most likely transmitted through unprotected sex and syringes and needles used for injections. A total of 69.5% of HIV infection in the general population occurs through needle sharing and 10% through unprotected sex [146
]. In a meta-analysis study in 2017, the prevalence of HIV among PWID worldwide was reported to be 17.8%, and the largest population under study was from sub-Saharan Africa. A total of 95% of new HIV cases were among the key populations in the Middle East and North Africa (MENA) [13
]. In a 2019 study, the prevalence of HIV in PWID was 21% in Africa; in another study, HIV incidence in PWID in the MENA region was significant, with 75% of new cases occurring in PWID and their sexual partners [149
]. The meta-analysis by Leung et al., in 2019, also estimated the highest prevalence of HIV in Africa and Asia, which is consistent with the results of our study [151
]. The prevalence of HIV in Brazilian FSWs was 5%, and in part-time sex workers in Burkina Faso it was 6.5% [31
The prevalence of HCV in PWID and FSWs was 60% and 1%, respectively. That means that almost one in every two PWID has HCV, with the highest prevalence of HCV in PWID being in the Western Pacific, the Americas and the Eastern Mediterranean, with 75%, 64% and 60%, respectively, and the lowest being in Africa (38%); the highest prevalence in FSWs is in Africa (9%), and the lowest prevalence is in the Americas and Western Pacific (1% each). In a 2007 meta-analysis study, the prevalence of HCV in PWID worldwide was estimated to be 50% [153
]. In another study published by Nelson et al. in 2011, the prevalence of HCV in PWID was reported to be 60–80% in 25 countries and over 80% in 12 countries, with approximately 10 million PWID suffering from HCV, China and the US having the largest population [154
]. Another meta-analysis study estimated the lowest prevalence of HCV was in the Middle East, North Africa, East and South-East Asia [152
]. Worldwide, it is estimated that 14 million PWID are at risk for HCV exposure from contaminated blood. In 2011, the prevalence of hepatitis C in PWID worldwide was estimated to be 67%, the highest prevalence being in Eastern Europe (2.3 million) and East and South-East Asia (2.6 million) [154
]. Degenhardt et al. (2017) reported a 52.3% prevalence of HCV among PWID worldwide [11
]. In another European Union study, the prevalence of HCV among general population was estimated to be 0.54% to 1.50% in 2019, with the highest prevalence in PWID being in the range of 7.9–82% [155
]. In a study in Brazil, the prevalence of hepatitis C in FSWs was estimated to be 0.9% [31
]. The relatively high prevalence of needle/syringe sharing, low condom use, high levels of sex with sex workers, homosexual sex between men, and selling sex, indicate high-risk behaviors associated with HIV and HCV prevalence in different regions [156
]. Another parameter influencing the prevalence and natural history of HCV infection is the host immune response-related genetics such as IFNL3/4
polymorphisms which impact the spontaneous clearance of HCV, and it was also observed that the genotypes associated with favorable outcome has different prevalence in ethnic groups [157
In this study, the prevalence of HBV in PWID and FSWs was 6% and 3%, respectively, with the highest prevalence in PWID being in South-East Asia (9%) and Africa and the Eastern Mediterranean (5% each), and the lowest prevalence being in the Americas and Western Pacific (1% each). Moreover, the highest prevalence in FSWs was in Africa (5%) and South-East Asia (4%), and the lowest prevalence was in the Americas and Eastern Mediterranean (1% each). Nearly 3.6% of the world’s population (257 million people) have chronic hepatitis B, with a prevalence of 0.01–2% in the UK, the US, Canada, Western Europe, and Japan, and over 8% in most sub-Saharan areas in Africa and some countries in the Western Pacific region [5
]. In high- and middle-income countries, HBV transmission is more perinatal and horizontal. In low-income countries, however, transmission occurs through drug injection and high-risk sexual behaviors [160
]. Asia and Africa have the highest HBV endemicity, but highly effective vaccination programs in some countries have pushed the pattern towards moderate or low endemicity. Therefore, China is currently the only country in Asia where HBV is of paramount importance. Countries with moderate endemicity include India, Korea, the Philippines, Taiwan and Thailand. Countries with low endemicity include Japan, Pakistan, Bangladesh, Singapore, Sri Lanka and Malaysia. Most countries in Africa have high endemicity, with the exception of Tunisia and Morocco, which have moderate endemicity [161
]. HBV vaccination is effective in reducing and eliminating HBV by 2030. According to the WHO, in 2017, 97% of blood donors were screened for HBV, but there are gaps in the program and strategies have been suggested to resolve the problems, including reducing insecure injections (it has reduced from 39% in 2000 to 5% in 2010), and safer sex practices, such as minimizing the number of sexual partners and using protection (condoms). On the other hand, according to WHO, 80% of people with hepatitis live without prevention, testing and treatment [5
]. In a study from Germany, the coverage of three HBV vaccines was 58% for drug users, one of the influencing factors being injection drug use [162
]. The prevalence of HBsAg in PWID in the study by Nelson et al is estimated to be 5–10% in 21 countries and over 10% in 10 other countries with a population of 1.2 million [153
]. In a 2017 study, the prevalence of HBV among PWID worldwide was 9.1%. In another study, its prevalence was reported to be 5% in Africa [149
The prevalence of HIV/HCV in PWID and FSWs was 13% and 3%, respectively, with the highest prevalence in PWID being in South-East Asia (17%), Africa and Europe (16% each), and the lowest in Eastern Mediterranean (8%); and the highest prevalence in FSWs being in Africa, and the lowest being in the Americas and South-East Asia (23%, 0% and 0%, respectively). The prevalence of HIV/HBV in PWID and FSWs was 2% and 1%, respectively, and the prevalence of HCV/HBV and HIV/HCV/HBV in PWID was 3% and 2%, respectively. The study by Larney et al. found a positive association between the high prevalence of anti-HCV and the prevalence of HIV in PWID worldwide [163
]. Co-infection with viral hepatitis in HIV-positive patients may worsen the prognosis [164
]. HBV reactivation is observed in patients with HBV/HCV co-infection during HCV treatment (direct acting antivirals (DAA)) or afterwards [165
Strength and Limitations
Previous systematic review studies have investigated the prevalence of HIV, HCV and HBV in PWID worldwide, but we have investigated the prevalence of these infections and their co-infections in the high-risk groups PWID and FSWs as distinguished by geographical areas. One limitation of this study is the changing sensitivity of HIV, HCV and HBV diagnostic tests over time, therefore, the results of the 2018 surveys may be different from 2008. In this study, all the reviewed articles had used anti-HCV serology test to detect HCV, which does not differ between past and present infections. The articles reporting HCV-RNA were scarce and excluded. Another limitation of our study is the high heterogeneity in studies. As the analysis was performed in different geographical areas, the heterogeneity may be due to different inclusion and exclusion criteria (e.g., type of drug, minimum duration of injectable drug use, sampling location (prison, the Behavioral Disease Counseling Center, MMT Center, homeless people), frequency of sex during a specific time, number of sexual partners and condom use).
4. Materials and Methods
4.1. Search Strategy
This systematic review and meta-analysis was designed based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyzes (PRISMA). Information sources from January 2008 to October 2018 were searched. Databases including PubMed, Scopus, Web of Science, Embase, Ovid, WHO and Google Scholar were searched, to find out the prevalence of HIV, HCV, HBV and their co-infections. The databases were searched using MESH keywords and the Boolean logic (AND, OR and NOT). The keywords included “Human Immunodeficiency Virus”, “HIV”, “Hepatitis C”, “HCV” “Hepatitis B”, “HBV”, “PWID”, “IDU”, “IVDU”, “FSWs” and “Co-infection”.
4.2. Study Selection and Data Extraction
Cross-sectional studies published in English that assessed anti-HIV, anti-HCV and HBsAg were selected. The studies that did not have a specific serology test to detect infections, those in which the prevalence rate was based on self-reports, the ones that used HBsAb, HBcAb, HBeAg and HBeAb tests to detect HBV and used HCV-RNA to detect HCV, and non-cross-sectional studies (case reports, reviews, case-control and cohort studies) were excluded.
After consulting with experts, two authors (RR and YM) extracted the data from the original articles. Extracted data include: (1) name of first author,(2) date of publication,(3) date of study,(4) country,(5) study subjects,(6) age of patients,(7) sample size,(8) method of sampling,(9) prevalence of HIV among PWID and FSWs,(10) prevalence of HCV among PWID and FSWs,(11) prevalence of HBV among PWID and FSWs,(12) prevalence of HIV/HCV among PWID and FSWs,(13) prevalence of HIV/HBV among PWID and FSWs,(14) prevalence of HCV/HBV among PWID and FSWs,(15) prevalence of HIV/HCV/HBV among PWID and FSWs.
All the steps ranging from searching to extracting data were independently performed by two researchers. In case of disagreement between the two, the problem was solved by referring to the article, discussing the problem and, if necessary, seeking help from a third reviewer (Kappa statistic for agreement for quality assessment; 0.75).
4.3. Quality Assessment
In this study, the Newcastle-Ottawa Quality Assessment Scale (NOS) checklist for cross-sectional studies was used to evaluate possible bias and quality of studies. This checklist was completed by two people. The quality of studies were judged based on such aspects as selection, comparability and outcome using the “star” rating system. Scores ranged from 0 stars (worst case) to 9 stars (best case). Studies with a score of 0–4 were categorized as low quality, 5–7 as moderate and more than 7 as high quality.
4.4. Statistical Analysis
Meta-analysis was performed on the eligible data to determine the prevalence of HIV/HCV, HIV/HBV, HCV/HBV, HIV/HCV/HBV, HIV, HCV and HBV in PWID and FSWs. A chi-square test was used to investigate the heterogeneity of the studies. The heterogeneity results from random-effects model were used to analyze the data at 95% confidence level. The MetaProp command was used to estimate the prevalence.
Egger’s statistical test and funnel plot were used to evaluate publication bias. Subgroup analysis was performed to investigate any qualitative confounding factors that may influence the prevalence of the disease. Subgroup analysis was conducted for the two high risk groups of PWID and FSWs in WHO geographical areas. Meta-regression was performed for mean age and sample size. All two-way statistical tests were considered with α = 0.05. Meta-analysis was performed in STATA 16.
The results show that the prevalence of HCV and HIV infections in PWID, and the prevalence of HIV in FSWs are higher than in the general population. The results indicate that the coverage of interventions for HIV and HCV prevention in PWID appear to be poor, and may not be sufficient to effectively prevent HIV and HCV transmission. Additionally, the lack of political commitment and, as a result, inadequate investment, reluctance to address sensitive issues related to young people’s sexual and reproductive needs and rights, and issues related to key populations and harm reduction, and a lack of systematic prevention implementation, even with regard to policy, are three interconnected reasons that seem to underpin the failure to implement effective programs at scale. Increasing the interventions for PWID and FSWs, such as HBV vaccination for the prevention of HBV, and the use of harm reduction programs, such as reducing the number of sexual partners per person, condom distribution, the use of clean needles and syringes, opiate substitution therapy (e.g., methadone) and the treatment of people living with HIV to reduce viral load and prevent onward transmission of HCV, are still a top priority in stopping the HIV and HCV epidemics. For HCV patients, education and counselling on options for care and treatment; immunization with the hepatitis A and B vaccines to prevent co-infection from these hepatitis viruses, and to protect their liver; early and appropriate medical management, including antiviral therapy; and regular monitoring for early diagnosis of chronic liver disease are necessary. Key population should be regularly monitored and screened for these infections and their associated infections.