Trend in seroprevalence of SARS-CoV-2 (IgG antibody) among tribal-dominated population: Findings from Jharkhand, India

Till March 30, 2021, the COVID-19 pandemic had affected all the countries with 127 million confirmed cases and 2.78 million confirmed deaths1. India reported more than 12.14 million confirmed cases and more than 0.16 million deaths, of which Jharkhand reported more than 0.123 million cases and more than 1100 deaths2,3. India is the second most populous country, and with a rapid increase in the number of reported cases mainly from urban areas, and little information exists on the disease status in tribal dominated districts4. With a population of 37.5 million, Jharkhand State has carried out over 5.9 million COVID-19 testing with a case-positivity rate of 2.3 per cent as of April 5, 20215,6. The presence of SARS-CoV-2 specific antibodies in the collected blood samples is indicative of previous exposure or individuals vaccinated against COVID-197,8. However, both rounds of serosurveys were undertaken before COVID-19 vaccination in India gained momentum, which rules out the latter possibility. Repeated cross-sectional serosurveys in the same geographical location are useful to monitor the trends of seroprevalence over time and provide evidence for public health decision-making for a pandemic response plan. Infectious disease transmission or its reoccurrence depends upon the available susceptible individuals who can contract and transmit the infection. Therefore, the WHO had recommended conducting sequential serosurveys to monitor the trend of infections that could be used for planning an effective public health response9,10.

In India, Scheduled Tribes (STs) constitute around eight per cent of the population, and Jharkhand is unique amongst Indian states due to its large (23% ST population) and diverse (32 ST) tribal population11,12. The majority of these tribes inhabit hilly or forested areas, and depend on agriculture, forest resources and labour, which are primarily subsistence-based, without stratification or specialization13-15. Such indigenous communities are considered to be socially and economically marginalized, and their health is disproportionately at risk in times of public health emergencies14,15. According to earlier reports from India, the tribal districts are least affected by the pandemic and recovery rate is also very high compared to their counterparts16,17. The innate immunity and lifestyle of tribal people might have helped them to prevent the disease spread18-20. Therefore, to understand the trend in COVID-19 progression, age-specific prevalence of SARS-CoV-2 antibodies and status amongst different risk groups in the general population and healthcare workers, two rounds of COVID-19 serosurveys were carried out in the tribal dominated State of Jharkhand during June 2020 and February 2021, respectively.

Material & Methods

This study was carried out in ICMR-Regional Medical Research Center (RMRC), Bhubaneswar. The study was approved by the Institutional Human Ethics Committee and all the procedures were performed according to the ICMR-National Ethical guidelines21 for biomedical research involving human participants.

Study settings: Jharkhand is one of the eastern States of India which shares its border with the States of Bihar to the north, Uttar Pradesh to the north-west, Chhattisgarh to the west, Odisha to the south and West Bengal to the east. It is the 15^th largest Indian State by area and the 14^th largest by population. Population based, repeated COVID-19 serosurvey was carried out in 10 districts of Jharkhand during June 2020 and February 2021, respectively. The study population was randomly selected through a multistage random sampling technique to ensure state representation. The same 10 districts of the State were selected during both rounds of serosurveys. Individuals aged more than 18 yr were included in the survey after obtaining informed consent for data and blood sample collection. Bedridden patients, pregnant women, and individuals with recognizable cognitive impairment were excluded.

Sampling: For the first round, sample size was calculated with an assumption of one per cent SARS-CoV-2 seropositivity, 0.4 per cent absolute precision, 95 per cent confidence interval, design effect of 2-5 per cent non-response rate19. The estimated sample size was 4980 rounded off to 5000. Based on the findings from national serosurvey round 1 and 2, we were expecting a 10-fold increased seroprevalence in the second round of State serosurvey7,20. Thus, we assumed 5.4 per cent seroprevalence in the second round of State serosurvey in Jharkhand. Considering one per cent absolute precision (based on the findings from the first survey), 10 per cent non-response rate and a design effect of 2, the minimum sample size was estimated to be 4309.

Multistage random sampling method was used where districts were stratified based on their reported cases per million population as low, medium and high burden. Three districts each from low- and medium-burden strata and four districts from high-burden strata were selected in consultation with the state health department of Jharkhand (Supplementary Tables I and II). From each of the selected districts, six clusters (villages in rural areas or wards in urban areas) were identified following the probability proportionate to size method. From each cluster, 40 households were selected through systematic random sampling, and one individual from each selected household was included in the serosurvey adhering to the participant selection matrix (Supplementary Table III). In addition, from each study cluster, 45 participants from high-risk groups (vulnerable to contract the infection and develop complications) were recruited (Supplementary Table IV). Hence, from each study district, 240 participants from the general group and 260 from the high-risk group (a total 500) were included.

Data collection: Data on the sociodemographic variables, history of exposure to a confirmed (and/or suspected) case of COVID-19, symptom profile over the last 30 days, clinical history and status of comorbidities, history of travel and testing were collected by trained field staff before blood sample collection. An open kit-based electronic data capture tool was employed for this purpose.

Sample collection and transport: Four millilitres of blood samples was collected through venepuncture and serum was separated. The collected samples were transported to the laboratory at ICMR-RMRC, Bhubaneswar, by maintaining proper cold chain. Samples were subjected to qualitative antibody detection (including IgG) against the nucleocapsid protein of SARS-CoV-2 in electro-chemiluminescence immunoassay-based platform (Roche Cobas e41-Roche Diagnostics International Ltd, Rotkreuz, Switzerland) using the Elecsys^® anti-SARS-CoV-2 kit (Roche Diagnostics GmbH, Mannhein, Germany) following manufacturer’s instruction. The value generated after testing was expressed in cut-off index (CoI), and a value of <1.0 was considered nonreactive and CoI ≥1.0 as reactive.

Statistical analyses: The prevalence of COVID-19 antibodies was estimated with a 95 per cent confidence interval (CI), and its distribution across different characteristics was assessed across both rounds. Pooled seroprevalence was calculated for each district and different strata. The results from first and second round were compared and analyzed. Statistical analyses were carried out using STATA 16.0 (Stata Corp. LLC, TX, USA) to estimate frequencies and percentages of the variables. Bivariate and multivariate logistic regression was carried out to calculate unadjusted odds ratio and adjusted odds ratio (AOR) respectively to estimate the odds of having SARS-CoV-2 antibody with regard to age, gender, ethnicity, residence, occupation, education, etc. at a significance level of P=0.05. The infection-to-case ratio (ICR) was calculated using the formula, ICR=Estimated number of infections (seroprevalence×population of the area)/reported cases (as reported in government database)22. Moreover, infection fatality ratio (IFR) was calculated as: IFR=Number of deaths from the disease×10,000/number of infected individuals. Geographical information system (GIS) maps analysis was carried out using an open-source software QGIS (ver. 3.10; https://www.qgis.org/en/site/).

Results

This study was carried out in 10 districts of Jharkhand in two rounds in which a total of 4761 participants during the first round and 3855 participants during the second round were enrolled. During the first round, out of 5157 participants approached, 4779 participated, and during round 2, out of 4290 participants approached, 3878 participated in the sero-survey. The non-response rate during the first round and second round was 7.3 and 9.6 per cent, respectively. After analysis of blood samples and matching the data collected, 4761 participants from the first round and 3855 participants from the second round were included for final analysis (

The mean age±Standard deviation of the study participants during round 1 and round 2 was 39.42±13.02 yr and 38.53±14.06 yr, respectively. The age- and gender-standardized seroprevalence for COVID-19 during round 1 was 0.54 per cent (95% CI: 0.36-0.80) which increased to 41.69 per cent (95% CI: 40.16-43.22) during round 2. Strata-wise (based on case burden) distribution of seroprevalence for both rounds are provided in Table I. The overall seroprevalence after adjusting for test performance for round 1 was 0.35 per cent (95% CI: 0.2-0.5) and round 2 was 41.58 per cent (95% CI: 40.2-43.0).

During the first round, the seropositivity was highest (0.84%) among the middle age group (45-59 yr) and lowest (0.42%) among age group 60 yr and above. In the second round, 46.35 per cent were seropositive in the middle age group and 36 per cent in the age group 60 yr and above. The seropositivity among male and female participants were 0.73 and 0.45 per cent, respectively in the first round and 51.35 and 33.70 per cent in the second round. During the first round, 17.37 per cent of the participants were tribal with seropositivity of 0.24 per cent (95% CI: 0.02-0.87), while during the second round, 21.14 per cent of the participants were tribal when their seropositivity increased to 39.14 per cent (95% CI: 35.77-42.59). The detailed sociodemographic characteristics of the study participants and distribution of seroprevalence are provided in Table II.

Table III compares the sociodemographic characteristics and COVID-19 seropositivity between tribal and non-tribal participants in both rounds. During the first round, compared to participants from general group, participants from high-risk group had lower COVID-19 seroprevalence with an AOR of 0.36 (95% CI: 0.13-0.98) which was very different in the second round with an AOR of 2.44 (95% CI: 1.99-2.97) amongst high-risk group (Supplementary Table V). In the second round, the AOR for antibody positivity in the age group of 18-44 yr was 1.31 (95% CI: 1.02-1.67) and in the age group of 45-59 yr it was 1.37 (95% CI: 1.05-1.79) compared to that in individuals aged 60 yr and above. Males were more likely to be seropositive with an AOR of 1.61 (95% CI: 1.35-1.92) compared to females. Compared to the tribal group, non-tribal participants had an AOR for SARS-CoV-2 antibody positivity was 1.24 (95% CI: 1.04-1.48). Urban people had higher seropositivity compared to the rural ones with AOR of 1.56 (95% CI: 1.35-1.81).

The infection to case ratio (ICR) in Jharkhand was 2.9 (95% CI: 0.9-6.5) in round 1 and 1.4 (95% CI: 0.9-4.2) in round 2. In the first round, the highest ICR was found in the districts of Dumka 25 (95% CI: 15.1-36.3) and Khunti 25 (95% CI: 18.2-39.6). In the second round, the highest ICR was in Dumka 13.5 (95% CI: 8.1-23.6) district (Table IV). The IFR was also increased in Jharkhand from 0.06 (95% CI: 0.02-0.17) in the first round to 0.63 (95% CI: 0.47-0.92) in the second round. Heat maps across the districts for both the rounds are provided in Figures 1 and 2, respectively.

Close

Fig. 1

Jharkhand seroprevalence in round 1 (Source: QGIS v3.10, https://www.qgis.org/en/site/).

Export to PPT

Close

Fig. 2

Jharkhand seroprevalence in round 2 (Source: QGIS v3.10, https://www.qgis.org/en/site/).

Export to PPT

Discussion

In Jharkhand, the first case of COVID-19 was reported on April 1, 2020, from the capital city of Ranchi. In the first round of serosurvey in the State, seroprevalence was found to be 0.54 per cent, which was much lower than the seroprevalence detected during the National Serosurvey conducted by ICMR during May-June 20207. It was also lower than the seroprevalence detected in its neighbouring State of Odisha, which is a tribal dominated State of India. In Odisha, the seroprevalence was 1.55 per cent in Bhubaneswar in July 2020, 24.59 per cent in Rourkela (August, 2020) and 31.14 per cent in Berhampur (August, 2020)23,24. The low seroprevalence in Jharkhand could be due to the strict implementation of a State-wide lockdown since March 22, 2020. Jharkhand was one of the first five States to adopt and implement State-wide lockdown in India25. A community-based study from British Columbia in Canada, involving serial cross-sectional sampling, reported a seroprevalence of only 0.28 per cent in March 2020 and 0.55 per cent in May 202026.

The seroprevalence during the second round in the month of February, 2021 was 41.69 per cent (95% CI: 40.16-43.22), whereas the percentage increase of 41.11 per cent took place over a period of seven months. This was also lower than the seroprevalence detected in Bhubaneswar (54%) and Agartala (55.65%), the capital city of the north-eastern State of Tripura, conducted during March 202127. The third national serosurvey conducted during December and January 2021, by the ICMR demonstrated an overall seroprevalence of 24.1 per cent in the country28. This could be due to heterogeneous spread of SARS-CoV-2 infection in India and different study timings.

In both the rounds of serosurveys, higher seropositivity was found among the middle-aged group (45-59 yr), which might be due to their higher mobility, resulting in increased exposure. This age group represents working population, making them more susceptible to exposure to infection while engaging in different occupations. There was a gender-wise difference in seropositivity with higher seropositivity amongst males, and this was significant in the second round (51.35% in males vs. 33.7% in females). Among the tribal group, although the gender difference in seroprevalence was not significant during round 1 (0.26% in males vs. 0.21% in females), it was significantly higher in males during round 2 (50.57% in males vs. 27.35% in females). Noticeably, the role of gender towards vulnerability to infectious diseases has been reported by Díaz et al29. The susceptibility to exposure or infection is influenced by age and gender and varies between cultures and tribes30.

The results indicated that about 0.81 per cent of the population in the community and 0.41 per cent in the high-risk group had developed antibodies in round 1, whereas it was 36.92 and 56.69 per cent in round 2, respectively. The first round findings were similar to the first nationwide serosurvey conducted by the ICMR in May-June 2020, in which overall unweighted seroprevalence was 0.56 per cent7. Similarly, the ICMR had conducted a third round serosurvey in the month of December 2020 where seroprevalence amongst adults was detected to be 24.3 per cent and in healthcare workers it was 25.6 per cent28. The second round of our study survey showed higher seroprevalence which might be due to the study taking place after two months from ICMR third round serosurvey giving ample time for the spread of infection The majority of healthcare workers received the first dose of vaccine by that time as vaccination started in the month of January 2021.

This study was conducted in ten different districts of Jharkhand, which provided a true picture of seropositivity in the State. Second, the study shows the possibility of herd immunity through natural infection, as during round 2, only healthcare workers were eligible for COVID-19 vaccination. This is also the first study in India that studied the seropositivity among tribal population and compared the situation between tribal and non-tribal populations. Our study had a few limitations. The possibility of selection bias cannot be ruled out. The sample size was estimated for the whole community considering 23 per cent of the study population as tribal11. While 20 per cent of our study participants belonged to the ST community, this might have affected the group comparison. Finally, we might have overestimated the ICR by using COVID-19 cases reported one and two week before the median date of all survey districts.

The low seroprevalence in rural areas compared to urban settings in both the rounds is as expected as the people in rural areas live in sparsely populated villages, in forest areas with less pollution and well ventilated houses. Round 1 serosurvey was conducted in the lockdown period, whereas round 2 serosurvey was conducted after complete unlocking of the State. This could be the major reason for high seropositivity in round 2, as during round 1, several travel restrictions were imposed and people were advised to keep themselves isolated in their homes. The government-imposed lockdown possibly was also a key factor that restricted the travel movement towards cities, and due to the self-reliance of villagers, the spread of infection in rural areas was limited.

Overall, our study identified a rapid rise in SARS-CoV-2 seroprevalence in Jharkhand over a period of eight months (June 2020-February 2021), which might have been due to unlocking of different districts. Both the serosurveys were carried out during the lockdown phase, but there was a period in between when lockdown was not in place. The prevalence during the first lockdown was attributed to lifestyle, travel restrictions, geography of the state and low work-related migration. Strict adherence to COVID-19 protocols during the initial phase such as social distancing and wearing of masks could be the factors leading to low prevalence. Our findings generated key insight into the dynamics of COVID-19 across the waved in a tribal-dominated state and this could inform in designing and implementing effective public health strategies to mitigate the burden of COVID-19 in the State. The study findings also suggest the need for periodic serosurveys in these tribal dominated regions of the country to inform appropriate intervention strategies.