To quickly detect Covid-19 positive cases and slow the virus’ spread in case of a possible surge due to the Omicron variant, India must increase its use of rapid antigen tests (which take 30 minutes or less), in addition to the “gold standard” RT-PCR tests, research shows.
The rapid antigen tests can help improve access to testing, without ramping up RT-PCR capacity, help identify more positive cases, at a lower cost and help better handle a possible third wave, shows a project by teams from the Max Institute of Healthcare Management at the Indian School of Business and health nonprofit PATH, funded by the Rockefeller Foundation.
The results are from a model-based analysis that uses data from news reports and government portals. The study also includes learnings from pilot projects in Maharashtra and Punjab between April and September.
The research compared several different combinations of RT-PCR and rapid antigen tests and found that it was cheaper, faster and effective if all Covid-19 suspects were first given the rapid antigen tests and only those who were symptomatic and still tested negative on the rapid test were given the RT-PCR. Using this strategy when testing capacity is overwhelmed across the country or in localised hotspots, and in rural areas where RT-PCR testing capacity is lower, can help India control a surge quickly.
In the second Covid-19 wave, India faced challenges in testing. As of June, towards the end of the second wave, India had conducted 294 tests per 1,000 population, compared to 1,416 tests per 1,000 population in the United States and 2,800 tests per 1,000 population in the United Kingdom.
In India, some people with symptoms of Covid-19 were unable to get tested in the second wave because of a shortage of tests and staff to carry out the tests, we had reported in April. In rural areas, infrastructure to test did not exist, we had reported in August.
In a potential third wave, ramping up test capacity would present similar challenges, addressing which would need development of new labs or installation of additional machines, which is difficult due to resource constraints. Even if that were possible, the turnaround times for tests would continue to stay high. The rapid antigen tests could be used in such a case to quickly detect positive cases, as research shows.
Rapid tests’ accuracy
Like the RT-PCR test, antigen testing looks for the presence of the SARS-CoV-2 virus in the body. In an antigen test, a swab is taken from a person’s nasal cavity and it is tested to detect fragments of proteins that are found on or within the SARS-CoV-2 virus. The RT-PCR tests look for the genetic material of the virus, we had reported in August 2020.
To know how good tests are, researchers look at two characteristics: One is the sensitivity of the test or the likelihood that the test will pick up a positive sample. The second is the specificity of the test or the likelihood that a negative sample will be classified as such.
RT-PCR tests have high sensitivity (around 95%) and specificity (almost 100%) with a processing time ranging from three to six hours, and they need additional time and resources for sample transportation and reporting test results.
Rapid tests typically detect the presence of viral particles in the sample within 30 minutes with high specificity (nearly 100%) but low sensitivity (50%–90%), depending on the antigen test kit used. So rapid tests would miss out on more positive cases than RT-PCR tests.
Combining tests
High-frequency mass testing using low-sensitivity tests (such as rapid antigen tests) and reducing turnaround time even by a day or two can improve the epidemiological impact of testing, show studies from India, France, the United States and Italy.
Going forward, India could rely on a combination of both tests, based on the scenario of disease spread in the country, the research team from the Indian School of Business and PATH concluded. For instance, if there are very few cases and the spread is slow, and the aim is to identify every positive case, using RT-PCR tests would suffice. But in case of a surge, such as during the second wave, a combination of both testing techniques would help identify more cases.
Similarly, a combination would work better for localised hotspots or sudden surges. Rapid tests would also help in testing in rural and remote locations where RT-PCR capacity is low.
“We are not using Rapid Antigen Tests enough,” said Gautam Menon, a professor of physics and biology at Ashoka University. “They detect Covid-19 at the point when you are most likely to infect others,” he said, adding that RT-PCRs might sometimes be too effective and give positives even when a person is past being infectious.
The researchers from the Indian School of Business and PATH also conducted diagnostic demonstration studies or pilots in Ahmednagar in Maharashtra, and in Mohali in Punjab. For this, the testing ecosystem was created by the public health department, which provided the space and laboratory technicians for testing.
PATH trained these technicians and gave technical assistance, test kits and confirmatory diagnostic technologies. The demonstration studies used existing RT-PCR capacity – roughly 600 tests per day per lab.
They compared four different combinations of rapid tests and RT-PCRs.
Twelve times the number of people can be tested by the rapid antigen+RT-PCR, as compared to using RT-PCR in 70% of tests, and six times the baseline, the researchers found. It can also prevent excessive burdening of the labs by selectively utilising the limited RT-PCR capacity only for those individuals who are symptomatic and test negative on the rapid tests and not for all suspects.
The rapid antigen+RT-PCR combination has a sensitivity of 77.8%. That is, it detects 77.8% of positive cases, compared to 81.9% in the baseline. Despite this reduction in sensitivity, this algorithm can find more positive people – close to eleven times – compared to using RT-PCRs 70% of the time, and more than five times compared to the baseline. Test sensitivity improves to 81.9% by using newer generation rapid antigen test kits, the research found.
“You can compensate for the lower sensitivity of rapid antigen tests by testing more,” Menon said.
A higher proportion of people in the rapid antigen-RT-PCR combination receive a confirmed diagnosis within one hour – 94.4% – instead of 30% in the 70% RT-PCR algorithm and 66% at baseline.
“In a hotspot, rapid antigen tests help you eliminate those who don’t have the virus and quickly detect those who do,” said Menon. This is important to control a surge, he explained, in contrast to what had happened in Delhi, for instance, during the second wave, when RT-PCR results would take three to four days.
RT-PCR tests and rapid antigen tests cost Rs 500 and Rs 150 each, respectively (according to previous studies and field experts who conducted the pilots). The cost per person tested reduces to Rs 178 in the rapid antigen+RT-PCR algorithm, compared to Rs 395 in the 70% RT-PCR algorithm and Rs 268 at baseline.
Some variations of the rapid antigen test+RT-PCR algorithm, such as conducting the follow-up RT-PCR test for all RAT-negatives instead of just symptomatic RAT-negatives, and directly testing everyone using RT-PCR, result in worse performance – they found a lower number of positives and reduced access and affordability.
India’s testing strategy
As of August 2020, 85% of the tests in India were reportedly by RT-PCR. The Indian government does not provide updated disaggregated data for testing by different methods. A press release from November 2020 had said that 46% of tests were conducted using RT-PCR while 49% used rapid antigen tests. We have reached out to the Ministry of Health and Family Welfare and the Indian Council of Medical Research for disaggregated data on testing and will update the story when they respond.
In March, the central government urged states to conduct at least 70% of overall testing by RT-PCR. Then, in May, the Indian Council of Medical Research suggested that rapid antigen tests be used to support limited testing capacity overburdened by the second wave and be ramped up across health facilities and through testing booths at community centres and offices. The short turn-around time of 15 minutes-30 minutes “offers a huge advantage of quick detection of cases and opportunity to isolate and treat them early for curbing transmission”, the ICMR said.
The ICMR guidelines from September 2020 asked that rapid tests be preferred for routine surveillance in hotspots and those who are symptomatic but negative on the rapid tests, be re-tested with the RT-PCR.
We have reached out to the ICMR and the Union health ministry to ask for their view on ramping up rapid testing in times of a surge and will update the story when we receive their response.
There is no standardised testing strategy across states and states can decide their own strategy, and different districts, even in the same state, may have different rules for how these tests are used. For example, in one of the demonstration exercises in the research project, the testing facility was dependent on rapid antigen tests because they did not have enough RT-PCR tests.
In general, the availability of resources and prevailing norms within the state was determining what was followed in practice, the research found. Another factor was the preference of beneficiaries –those being tested might prefer the RT-PCR over a rapid antigen test.
Challenges in testing
Through a combination of both kinds of tests would be more effective than one kind alone, implementing such a combination would be challenging. For instance, health facilities would have to ensure that all suspects receive a rapid antigen test, and follow up on those who are symptomatic but get a negative result for a rapid antigen test.
More personnel (laboratory technicians and data entry operators) who are well-trained in infection prevention practices would be required at rapid antigen test centres to ensure coordination with the RT-PCR labs. For this, a dual swab collection strategy where two samples are collected – first sample for rapid antigen test and second for RT-PCR – from the same suspect may help, the researchers observed based on the pilot studies.
Additionally, all health facilities should have a high quantity of rapid antigen tests to ensure that all suspects are able to get screened through rapid antigen tests. Moreover, labs would require the newer rapid antigen test kits that have higher sensitivity, which may take longer to procure.
Also, this algorithm might not work for all situations. For example, if the focus of a state is on avoiding missing any positive cases, rather than simply finding more positive cases, implementing an alternative algorithm having an almost perfect sensitivity would be a better choice for that state. The World Health Organization does not recommend rapid antigen tests when Covid-19 transmission is low.
This article first appeared on IndiaSpend, a data-driven and public-interest journalism non-profit.