A whopping 44.5% of Karnataka’s population was infected with Covid-19 by the end of August, said a team of Indian and American researchers who completed a randomised Covid-19 testing study.
This estimate stands in sharp contrast to the state’s official sero-survey which determined that 27.3% of the population was exposed to the virus. The diverging studies appear to suggest that this is because the state’s sero-survey did not conduct enough random tests.
The survey, titled “Karnataka Seroprevalence Study” (KSS), began on June 15 and ended on August 29. It saw researchers from Duke University in North Carolina, the University of Chicago, the Centre for Monitoring Indian Economy (CMIE) and Mapmygenome, supported by the State Covid Task Force, reach out to 1,907 random households across 20 districts of Karnataka.
The homes were identified from an existing and representative sample of 9,717 households, drawn from the Pyramids Household Survey (CPHS), a longitudinal household panel data set assembled by the Mumbai-based Centre for Monitoring Indian Economy (CMIE).
The survey has now estimated that a shocking 3.15 crore residents had been infected by the virus by the end of August 2020, which is higher than the figure of 1.93 crore persons currently infected or recovered from Covid according to the government survey.
Another key finding is that rural areas have been hit equally badly as urban areas because of returning migrants and because there were fewer lockdowns there due to ongoing agricultural activities.
Commissioner of Health Pankaj Kumar Pandey said the disparity in findings could be due to KSS’ small sample size.
“A small sample size can skew estimates. Plus the KSS’ use of households sourced from an economic survey can create a distorted picture. The state’s sero-survey results, meantime, have tallied largely with nationwide sero-survey results,” he said.
One of the authors of the KSS study, Associate Professor Manoj Mohanan of the Sanford School of Public Policy at Duke, told DH that one reason for the disparity is the different strategies used.
“We aimed to get at the population-representative sample while the state’s study focused on target-risk groups. This is likely to cause some variation, especially because consent rates are likely to differ,” he said.
That the studies used different antibody tests is also a factor, he added. “We relied on the Elisa test for the RBD spike protein that was developed by THSTI while the other study relied on the Covid Kavach Elisa,” he said.
Scientists of the DBT India Consortium for Covid-19 Research had published a paper in August showing that the ICMR-designed Covid Kavach Elisa test has a specificity of 99.5% and a sensitivity of 75.7%. In contrast, THSTI’s Elisa test was shown to have a specificity of 100% and a sensitivity of 84.7%.
“Sensitivity” is the ability to identify those with antibodies to Sars-Cov-2 (true positive rate), and “specificity” is the ability to identify those without antibodies to Sars-CoV-2 (true negative rate).
Professor Mohanan also pointed out that the RT-PCR positive rates in the state’s sero-survey (12.7%) are “slightly higher than the high-end of our estimates for current infection”.
Deccan Herald is on WhatsApp Channels| Join now for Breaking News & Editor's Picks