The present review was registered with the International Prospective Register of Systematic Reviews (PROSPERO ID: CRD42022334938) [15]. We followed the Preferred Reporting Items for Systematic Review and Meta-Analysis (PRISMA) guidelines to perform and present this review (Supplementary Table 1) [16].

Observational studies that reported the data on community-based prevalence of Hepatitis B among tribal populations in India, including short reports, were included. Qualitative studies, reviews of any kind, meta-analyses, conference abstracts, and editorials were excluded.

Studies conducted in India that documented the prevalence of HBV infection among healthy subjects belonging to tribal communities and reported the prevalence of HBsAg (our main preference) and/or other markers were included in this study. Individuals with sexually transmitted infections were excluded, as we wanted to estimate the prevalence of HBV in healthy individuals.

Three electronic databases, Embase, Medline via the PubMed search engine, and CINAHL, were systematically searched. Additionally, to make the search more comprehensive, the first 10 pages of Google Scholar were searched, and the references of the included studies were also searched manually.

We used three major concepts: “Hepatitis B″, ‘Tribal’ and ‘Prevalence’. PubMed was used to create an exhaustive search syntax, which was later adopted for the other two databases. MeSH terms were employed for PubMed, whereas Emtree terms were used for Embase. The MeSH terms utilized were “hepatitis B”[MeSH] and “Indigenous Peoples”[MeSH] for the respective major concepts. The relevant keywords were identified from the literature to make the search more comprehensive. All three concepts were joined using the Boolean operator “AND.” We did not apply any other restrictions, such as language, but we considered studies published until April 2025. The detailed search strategy for each of the databases is provided in Supplementary Table 2.

Two reviewers (AS and GN) independently conducted the primary screening based on the titles and abstracts of the articles and categorized the studies as included, excluded, or unsure. Studies deemed irrelevant by both reviewers were excluded. Next, the full texts of the included articles were thoroughly screened by two reviewers (AS and GN) on the basis of pre-established inclusion and exclusion criteria. Any differences between the two reviewers were resolved by the entire team through consensus. Data, including the first author’s name, publication year, sociodemographic characteristics (such as age and sex), study location, name of tribes, study setting, study period, sampling method, testing method, markers to determine the prevalence of Hepatitis B (e.g., HBsAg, HBV DNA, Anti-HBc, Anti-HBs, and HBeAg), and the reported risk factors were extracted from included studies by two independent reviewers (AS and GN). Before entering the data into a pre-formatted data extraction sheet, two reviewers (PKS and SP) further verified the data.

The quality of the studies was independently assessed by two reviewers (AS and PKS) using the evaluation tool for cross-sectional studies (AXIS). This tool consists of 20 questions from five key domains, such as introduction, methods, results, discussion, and others [17]. The key domains address questions related to study design, sample size justification, target population, sampling frame, sample selection, and methods, among others. Each question was marked as “yes = 1” or “no/do not know = 0,” and studies obtaining 0%–50% were reported to have “high risk of bias,” 51%–80% as “medium risk of bias,” and 81%–100% as “low risk of bias.” Any disagreement about the risk of bias was resolved in consultation with PKS. The AXIS tool’s detailed questions are presented in Supplementary Table 3.

A qualitative and quantitative summary of the data was produced. The qualitative summary included the characteristics of the included studies along with their basic description. R Studio (version 4.1.3, 2022; The R Foundation for Statistical Computing, Vienna, Austria) was used to perform the meta-analysis. The “metaprop” package (Rdocumentation.org) was used to perform the meta-analysis of single proportions. Proportional variances between studies were stabilized using the Freeman-Tukey double arcsine transformation. The standard error (SE) for the proportions was computed based on the extracted data. The formula used for the calculation was: “SE = √ (P × (1–P)/n),” where P represents the proportion and n is the population size. The meta-analysis was conducted using a random-effects inverse-variance model, which included calculating a weighted average using standard errors (SEs) and supplying the data frame for the “metaprop” package in R Studio. We presented the pooled estimates as percentages with 95% confidence intervals (CIs), which were represented by a forest plot. I² statistics were used (I² statistics: 25% - mild, 25%–75% - moderate, and >75% - high) to measure heterogeneity across studies [18]. Additionally, a subgroup analysis was performed based on the administrative zones in India to account for geographic variability. To further address the issues of high heterogeneity between studies, a prediction interval was provided with each pooled estimate. A sensitivity analysis was performed using the Leave-one-out method, which enabled us to predict the variation in the pooled estimates and the stability of the model [19]. We constructed a funnel plot to analyze the publication bias, examining it through an Egger’s test, where a p-value of ≤0.05 was deemed to be statistically significant.

This review was conducted based on the published literature, thus eliminating privacy concerns.

A total of 611 articles were identified by searching the three databases indicated above and by conducting manual searches (cross-referencing). We removed 173 duplicate studies (Figure 1). Furthermore, the full texts of 37 articles were reviewed after primary screening based on titles and abstracts. A total of 24 studies finally fulfilled the inclusion criteria and were selected for the study [20–44].

Of the 24 included studies, seven were conducted in the Andaman and Nicobar Islands, making it the region with the highest number of studies on HBV prevalence among tribal populations. [34–37]^, The age of the participants ranged between 0.04 years of age and 90 years of age. Reporting of participant sex was limited to 18 studies; the other studies did not present corresponding data (Table 1) [14, 20–42].

The HBsAg status was reported in all studies.1 [14, 20–42] Along with HBsAg, the HBV DNA status, which was analyzed using polymerase chain reaction (PCR), was reported by seven studies. The prevalence of HBeAg was reported in seven studies, while anti-HBs and anti-HBc were detected in 15 and 14 studies, respectively, with 11 studies containing data from both (Table 2). Regarding the testing method, the enzyme-linked immunosorbent assay (ELISA) was performed in 22 studies, while one of the studies used the Enzyme Linked Immuno-Filtration Assay (ELIFA) and one used an enzyme immunoassay (EIA) (Supplementary Table 4).

The HBV prevalence among the Indian tribal population was found to range from 0% in the Kurichiya tribes of Kerala to 65% in the Jarawa tribes of the Andaman and Nicobar islands. A very high level of HBsAg positivity was also reported among the tribes of the Andaman and Nicobar islands (ranging from 23.3% in the Nicobarese tribe to 37.8% in the Shompen tribe). The point prevalence of HBsAg was found to be 21.2% among the Idu Mishmi tribes of Arunachal Pradesh. However, the most recent HBV status report found the prevalence of HBV to be 3.6% (HBsAg positivity) in the tribes of Arunachal Pradesh. Among the Tripura tribal populations, the highest and lowest prevalence rates were observed among the Chakma (11.41%) and Lusai (2.7%) tribes, respectively. Low levels of HBsAg prevalence were observed in the primitive tribes of Odisha, ranging from 0.8% in the Lodha tribe to 3.8% in the Mankidia tribe. Among the Spiti tribes of Himachal Pradesh, HBsAg positivity was reported at 17.2%–22.05%, whereas it was observed to be 2% among the Paharias of Jharkhand and at 2.9%–4.4% in Madhya Pradesh, where the specific tribal groups were not reported. The lowest prevalence levels were observed in the Didayi and Chuktia Bhunjia tribes (1.06% each) and among the Mankidia (1.76%) and Kharia (3.13%) tribes. The Kutia Khond tribe showed the highest HBsAg positivity at 14.18%, followed by the Paudi Bhuyan tribe (6.06%) and the Gond tribe (6.00%). Other tribes, such as the Khond (3.60%), Bhuyan (3.13%), and Kol (4.26%), were found to have moderate HBsAg positivity levels. A more recent assessment by Sahoo et al. among an unspecified tribal population in Odisha showed a relatively low HBsAg positivity rate of 1.14% (Table 2). The presence of HBeAg in the serum ranged from 7.14% in the Attapady tribal population of Kerala to 36.6% in the Idu Mishmi tribe of Arunachal Pradesh. In total, 85.5% of the vaccinated Nicobarese individuals were anti-Hbs positive, indicating protection against HBV. A total of 92.3% of the Idu Mishmi tribe were anti-HBc positive, indicating a past or current Hepatitis B infection. Cases of Occult Hepatitis B infection (OBI) were also found in 11.1% of the Nicobarese tribe tested for HBV DNA (Table 2).

The HBV DNA was reported on an individual basis, not a tribal basis. The nationwide pooled prevalence of HBV (as measured by hepatitis B surface antigen positivity) was estimated to be 9.99% (95% confidence interval (CI) 6.07–14.75, I² = 98.7%, p < 0.01) with a prediction interval ranging from 0.00 to 44.13 (Figure 2). To account for the high heterogeneity, subgroup analysis was performed based on disaggregated estimates of HBV prevalence in different administrative zones in India. The highest HBV prevalence was noted in the northern zone of the country (19.60%, 95% CI 15.09–24.54, I² = 84.4%, p < 0.01), followed by the northeastern zone (13.43%, 95% CI 6.09–23.08, I² = 98.4%, p < 0.01) and the southern zone (10.44%, 95% CI 4.75–18.01, I² = 98.9%, p < 0.01). Residual heterogeneity was found to persist within subgroups after disaggregated analysis, although the test of subgroup difference was statistically significant (p < 0.01).

The non-significant Egger’s p-value (0.56) and visual inspection of the funnel plot indicated that publication bias is unlikely (Supplementary Figure 1). To assess the robustness of our HBsAG national prevalence estimate, a leave-one-out analysis was performed. The pooled estimates, when each study was omitted one at a time, ranged from 10.70% to 8.74%, while the overall study estimate was 9.99% (Figure 3).

We observed that 21 studies had a moderate risk of bias [14, 20–23, 25, 26, 28, 30–42], while three studies [24, 27, 29] had a high risk of bias (Supplementary Table 5).

We observed a significant pooled prevalence of Hepatitis B among tribal populations. The majority of the included studies used ELISA to test for Hepatitis B. The majority of the studies measured HBsAg to report the burden of Hepatitis B, while ELISA was the most commonly conducted serological test.

We observed wide diversity in HBV prevalence among different tribes, ranging from 0% (Kurichiya tribes of Kerala) to 65% (Jarawa tribes of the Andaman and Nicobar islands). However, a recent systematic review reported that the prevalence of HBV infection in the general population ranged from 0.87% to 21.4%, which is lower than the findings of the present study [44]. This further strengthens the idea that tribal communities are a vulnerable group and should be the focus of HBV-related interventions. We reported the pooled prevalence of HBV infection to be approximately 11.09%, which is lower than the 15.9% HBV prevalence reported in a meta-analysis conducted in 2007 among tribal populations in India [4, 6]. Additionally, our recent study conducted among under-five children from tribal communities also observed the prevalence of HBV to be low (1.14%), which further supports the notion that the introduction of a vaccine against HBV in routine immunization programs has resulted in a steady decrease in HBV burden [45]. Nonetheless, tribal populations remain vulnerable to this infection, which highlights the need for continued vaccination strategies along with efforts to educate these communities about the infection and its transmission routes so that Hepatitis B can be prevented.

The present review found ELISA to be the most commonly used serological test for the detection of HBV, which is consistent with the findings of a systematic review published in 2015 that also reported ELISA to be the most common HBV diagnostic method [46]. A probable reason for this could be the popularity and the cost-effectiveness of ELISA. Nonetheless, other novel and highly sensitive techniques, such as flow cytometric quantification and covalently closed circular DNA (cccDNA), may be used for diagnosing the etiology of chronic hepatitis and predicting the recurrence of disease, respectively [47].

There are approximately 550 different tribes residing in different regions and spread throughout India [47]. Their subsistence largely depends on farming. This population has distinct sociocultural practices such as tattooing, pricking, and piercing of body parts, which are important risk factors for HBV infection. These sociocultural practices, along with their living conditions, sexual promiscuity, higher rates of poverty, and limited literacy, make this group vulnerable to HBV [1]. This highlights the need to raise awareness among the general public about the possible routes of HBV transmission, which include exposure through the mucosal membrane, body fluids, and perinatal transmission. Moreover, HBV infection may lead to chronic liver diseases such as cirrhosis and hepatocellular carcinoma. Furthermore, individuals who have ever been infected with HBV may be at risk for viral reactivation, which occurs when host immunity interplays with HBV replication, necessitating complete treatment of infected individuals. However, limited data did not allow us to assess the proportion of subjects in need of treatment, which should be assessed in future studies. Nonetheless, a systematic review reported that there may be 12%–25% of patients requiring treatment in various settings, i.e., community and clinical.

To eliminate viral hepatitis as a public health threat by 2030, one of the critical actions is to identify and prioritize high-burden and vulnerable populations for targeted interventions. Given the considerable and disproportionate burden of Hepatitis B observed among tribal populations in this review, this group clearly falls into that high-priority category. This can directly inform national and subnational immunization strategies and can guide the equitable allocation of resources under the broader umbrella of India’s commitment to the GHSS targets. The integration of tribal-specific vaccination plans, supported by decentralized health governance and resource deployment, is essential. Moreover, the identification of traditional practices as risk factors underscores the need for culturally sensitive programming. Additionally, Information, Education, and Communication (IEC), along with Behavior Change Communication (BCC) strategies, are required. Culturally appropriate and linguistically tailored messages should be provided to tribal communities so that they can understand and adopt safe practices, especially with regard to the use of injections. IEC materials may include ‘fotonovelas’ or books containing photos showing how HBV infection spreads. Nonetheless, vaccination remains a key to controlling the infection, and it should reach all segments of society. Although India has high vaccination coverage, these populations should still remain the focus.

We conducted a thorough literature search in three databases. Additionally, we followed a prospectively registered protocol to conduct this review. Two independent reviewers assessed screening, data extraction, and risk of bias, which gave additional methodological rigor to the review. However, the paucity of data on vaccinated and non-vaccinated individuals did not allow us to synthesize and compare data based on vaccination status, which could have been useful to inform new programs and policies.