Aiming for herd immunity just won’t do
In the last four months, countries that have won the battle against the coronavirus -- such as China, New Zealand, Vietnam, Taiwan, and Germany -- did so with strict and highly organized execution of country-wide policies and evidence-based control measures with long-term course of the pandemic in mind.
Notable were three simple strategies that were uncompromisingly enforced by administration and religiously followed by people: Social distancing, mask-wearing, and aggressive testing.
These strategies were not only critical to contain the disease and limit mortality, but also to enable the economy to open up sustainably without the worry of unbridled disease progression. As a nation, in the debate of life versus livelihood, we have clearly said it is not a “versus debate” but an “and discussion.” It has been a wise choice, because while we must fuel our economy, we must also ensure that we do not lose a single leader, frontline worker, or anybody vulnerable because not a single life -- young or old -- is dispensable.
How does Bangladesh fare in the three strategies?
Mask-wearing is being taken increasingly seriously by the government and hence, by the citizens also. In three circulars on May 28, June 30, and July 21, wearing a mask in public was declared mandatory, and specific parties responsible for enforcing it were assigned. The Penal Code provides the mandate for punitive legal action for non-adherence as well.
Social distancing in Bangladesh has been a mixed bag so far as observed during a government enforced lockdown from March 25 to May 30. While various circulars from multiple ministries specified the recommended behaviours in public spaces, compliance has been lacking because of appropriate monitoring and consequential measures for non-adherence..
According to Worldometer, we have done 6,400 tests per million population with RT-PCR (Real Time PCR) method. The test positivity rate in Bangladesh has been steady, at a high value of about 20% (one out of five people who are tested are detected Covid-positive) since mid-June whereas the World Health Organization considers 5% in at least the last two weeks to be the threshold (one out of 20 people at maximum) beyond which the disease is considered as not contained.
Currently, we are limiting our testing only to people with symptoms, and those who are exiting the country, although neither the severely ill nor the travellers represent the general population. It is safe to say that Bangladesh currently has very little knowledge about the asymptomatic Covid-positive portion of the population, who can spread the infection without getting detected by the system in place.
A case for why high-volume testing is the way forward
Among the most populous 10 countries, Bangladesh is at the very top in terms of density with a high 1,116 per sq-km. India, another country among 10 most populous countries, comes second with only 420 per sq-km, about a third of that of Bangladesh. This puts us in a unique position of tackling a pandemic, where chances of transmission are very high, resource constraints are dire, and countries to get lessons from are unavailable.
On top of that, unwaveringly, our need is that of balancing life and livelihood.
It is of utmost importance for a country with these criteria to have an accurate assessment of the size and proper control on the course of the pandemic to safely allow resuming economic activities. But, what does safely resuming economic activity really look like?
It means that a person can go out to buy groceries without feeling a sense of panic; police forces can work in the public spaces without feeling threatened by a possibility of getting infected; political leaders can attend public events without feeling uncertain about its consequences on their health and the health of the people in the audience; doctors can treat all patients without seriously risking their health; and patients can go to doctors/health care providers without fear of catching something more severe than what they have.
After all, the carriers of the disease may not show any visible indication of it, nor are all the carriers of the infection isolated. To lift the shroud of fear about coronavirus from the minds of the people, a large percentage of the population should be tested and the carriers of the disease isolated, with no contact with the rest of the population until recovery.
Only when society trusts that infected people are not out in the public or in zones of economic activities will they return to these spaces. This is where we must utilize the merit of continually testing a large number of people, regardless of whether they have any symptoms or not.
Current status of testing
Bangladesh has the lowest number of tests per million people in the world, followed by Mexico at second to last (see Table 1). When a very small fraction of the population is tested, the results do not tell the story of the entire population. Consequently, the epidemiologic indicators (reproduction rate and doubling time) calculated from this test data do not reflect the true spread of the disease. Inaccurate metrics impair the government’s ability to take the right actions.
One indicator of a pandemic spread is R(t) which tells how many people are subsequently infected by a single sick individual, and is thus a metric that assesses how easily the infection is spread. An R(t) below one means the spread is shrinking. Epidemiologists have various tools to estimate this number, but all of these methods require widespread testing of the population to be reliable.
As explained in the inset below, at the moment the insufficient testing levels in our country means that we cannot obtain a reliable estimate of R(t). In particular, the high -- and growing -- positivity rate of around 20% means that we cannot yet say that R(t) is below one.
The collection and testing of specimens are currently being rationed. Tests are priced at Tk200-500 in government hospitals and Tk4,000 at private hospitals.
Rationing the tests for specific sets of people will not give us unbiased estimates of disease spread as stated earlier. The paywall creates an inequality in access to the test kits. This systematically excludes certain population segments and thereby introduces bias in the data.
According to a study by JAMA internal medicine published on July 21, 2020, studies on the presence of the infection in the population show six to 24 times more infections than diagnosed coronavirus disease. Asymptomatic people who had mild or no symptoms and do not undergo testing still contribute to viral transmission in the population.
In the absence of general lockdown measures, the only way to reduce transmission is to test more, make tests more accessible, and identify cases including asymptomatic ones, and isolate infected people.
How do we test more?
When it comes to epidemic control, there is no one size fits all approach but there is no alternative to widespread testing. While RT-PCR is considered the gold standard for diagnosing Covid infections, many countries are unable to perform large numbers of RT-PCRs rapidly due to the high price (see Table 2 for comparison), inadequate infrastructure, and lack of human resources with the right expertise.
On the other hand, the significant delay from sample to result is counterproductive, as infected ones continue to infect others during the delay.
A cost-effective strategy is using rapid antigen tests for population level surveillance and as a first screening in clinical settings. Suspected individuals who test negative for Covid-19 by rapid antigen test should be tested subsequently by RT-PCR to rule out infection, whereas a positive test should be considered as a true positive and does not need reconfirmation by RT-PCR test.
Many have concerns regarding the lower sensitivity (ability of a test to detect cases) of antigen tests. However, simulation studies and recent work by Paul Romer (2018 economics Nobel Laureate) show that frequent testing with even a less sensitive test helps reduce transmission of the infection.
For example, if for 10 PCR positive cases there are 100 more in the community, a rapid antigen test with 50% sensitivity would enable us to isolate 50 additional infections and thereby reduce transmission.
The availability of rapid antigen tests in private hospitals and local pharmacies would lead many to self-isolate. Testing household members of known cases, and people who interact with many on a daily basis, such as physicians, police personnel, politicians, garment workers, and shop-keepers should be prioritized.
Testing strategies should prioritize accessibility, frequency, and collection-to-answer time; lower sensitivity should not be a barrier. India has recently approved rapid antigen tests (manufactured in South Korea) for both zone containment and clinical settings. In Bangladesh, rapid antigen tests can be administered by the Community Support Teams deployed by Brac and DGHS.
Another cost-effective way to scale up testing capacity is to pool samples for PCR testing in low incidence areas. In pooling, the idea is to collect samples from four to five people and combine them into one specimen, then test it as one sample rather than running each individual sample on its own test. If the pool is positive, it means that one or more of the individuals tested in that pool may be infected, so each of the samples in that pool are tested again individually.
Because the samples are pooled, it is expected that fewer testing supplies are used and more tests can be run at the same time, allowing patients to receive their results more quickly. For example, suppose in a population, 3% samples are positive.
We decide to pool the samples into a batch of five for every 100 samples (total 20 tests). Assuming 15 samples would be re-tested for the three positive pooled samples, we use 35 tests in total, saving 65 tests.
This testing strategy is most efficient in areas where most results are expected to be negative and this could in fact work in many parts of the country. The US has recently approved sample pooling, and Germany has used sample pooling for surveillance testing of health care workers. Automating sample processing in the labs can also increase testing capacity of RT-PCR, but the high cost is still prohibitive for mass testing.
Widespread testing would allow us to identify asymptomatic and pre-symptomatic cases and an accompanying transparent reporting system can help us regain the global confidence about our data.
Moreover, testing in large numbers ensures isolation, self-imposed or led by administration, of infected individuals without enforcing disruptive lockdown procedures; this is particularly important as we have started to allow economic activities.
Why alternatives -- herd immunity, lockdown -- will not work
Some propose that given our low mortality rates, we can aim to reach herd immunity. According to The New York Times “the world is still far from herd immunity for coronavirus,” many epidemiologists say herd immunity may be possible to achieve only when 60-80% of the population has developed immunity.
The percentage of the population with antibodies is lower than 25% in the countries with this information. In New York, which has seen the hardest hit in the US, 19.9% people have antibodies. Sweden and Britain, which have relaxed lockdown, have only seen 7 to 17% of the population to have antibodies.
A survey by the Indian government said 23.48% of the people whose blood samples were tested had Covid-19 antibodies. The plausibility of achieving herd immunity is still quite low.
The 1918-19 flu pandemic lasted two years in three waves with a second wave by a mutated, deadlier virus. In the Indian sub-continent, deaths were packed in the three cruel months of fall 1918 killing at least 12 million people, about 4% of India’s total population at the time.
We of course live in a better world now, with advanced therapeutic medicine and improved knowledge on hygiene measures. Given what we have seen so far with the Covid-19 pandemic and what we know happens with a new virus, we should not assume that the pandemic will disappear very soon. It is unlikely that we will see devastation at the scale of the Spanish flu, but we need to remember the past as we plan for the future.
On developing a collective response
Managing a pandemic of a new infection will require a coordinated effort. Bangladesh Preparedness and Response Plan (BPRP) for Covid-19 published in July by the Health Services Division and a corresponding UN plan published in April underscore the importance of such a collaborative arrangement across government agencies, private sector, civil society organizations, academia, UN agencies, various development partners, and of course, the media. Both frameworks, which are in alignment, stress “multi-sectoral” strategy.
We are already seeing great examples where RT-PCR labs set up in both public, private, and academic institutions took daily testing from only a few hundred in April to nearly 15,000 in the June-July timeframe. Brac’s sample collection booths increased access to tests for common citizens.
The four mobile companies, in an unprecedented collaborative arrangement, have joined hands with the government to develop a highly sophisticated, artificial intelligence-supported syndromic surveillance system for early detection of cases.
The Community Support Teams (CST), jointly conducted by Brac and DGHS, have laid the foundation for community-based syndromic surveillance. The CSTs can be trained to become proficient rapid antigen testers.
Decision-making coalitions with public health and infectious disease experts, as well as economists, teachers. and community members can be formed to bring their unique expertise to this issue.
Target setting is key
As a nation, we have done well in setting targets to achieve the Millennium Development Goals, LDC graduation, immunization, women’s empowerment and now are gunning to be a poster child nation for achieving the Sustainable Development Goals.
Similarly, we can and should take the challenge of swinging from a laggard to a champion of aggressive testing in a matter of months. As Covid-19 will not disappear any time soon, we should prepare comprehensively and aggressively, and for the long term until we get everyone in the country immunized against the coronavirus.
The current high humidity and temperature is a fortuitous opportunity for us to prepare and devise a plan for the upcoming effect of the pandemic; however, this turnaround will require us to set targets. Without losing any more time, we must set aggressive but achievable targets for disease containment, curbing mortality, and testing coverage at all administrative levels from national to local.
The important thing to realize is that this virus is faster than our typical decision-making process. Our processes must be highly nimble and responsive to new learnings. The expert advisor groups within the government, private sector, and development communities must possess that capability to rapidly learn, assimilate, and recommend.
More than a headcount
By now, for most of us the disease has reached very close to home. It’s no longer affecting people from another city, another household, or another family. It is now among us and in our homes. Even with the low mortality rate, the ones we risk losing are our parents, grandparents, teachers, and highly respected elders.
If we zoom out for a minute, we can see that they are the most senior members of our society with the wealth of experience and knowledge, the ones who built the country for us, and the ones who are our emotional and intellectual caregivers.
The loss of the senior members and leaders of the society is much larger than a headcount. It is not ideal for a country to aim for such a future. It is not desirable for any of us to allow such an episode in the history of Bangladesh.
Ayesha Sania is Research Scientist (Epidemiologist), Columbia University. Shams El Arifeen is Senior Director and Senior Scientist, Maternal and Child Health Division (MCHD), International Centre for Diarrhoeal Disease Research, Bangladesh. Anir Chowdhury is Policy Advisor, a2i, ICT Division/Cabinet Division/UNDP Bangladesh. Tamanna Urmi is Data Analyst in Fraud detection, Gojek, Indonesia.