HOW DID PAKISTAN AVERT DISASTER?

Is Pakistan's Covid-19 fight over? Have we won?
Published August 30, 2020

All the predictions pointed towards doom and gloom with respect to Covid-19 in Pakistan, and in June it looked very much like the worst was about to come true. But then everything changed: infections went down, the pressure on hospitals eased up and the number of fatalities remained low. This was excellent news. But do we understand what happened?



In the middle of June, doom and anxiety hung clearly in the air. Those were the days of the infamous “twin peaks”, the two days on the 13th and the 19th, when the country saw the highest number of new infections from Covid-19 since the outbreak began — 6,825 and 6,604 respectively. Hospitals were filled up, reports circulated of shortages of oxygen and critical medicines and, on WhatsApp groups, people anxiously shared stories of friends and relatives who had tested positive — in some cases entire families — and of people shuttling from one hospital to another searching for space.

It was in those days that a study was released by the Imperial College of London, which projected that Pakistan could see up to 2.2 million fatalities by end June if the trend continued without any mitigation measures. Planning Minister Asad Umar, who was tasked by the prime minister to chair meetings of the National Command and Operations Centre (NCOC), from where the government was coordinating its response to the pandemic, cautioned that the number of people infected could double by the end of the month, and reach 1.2 million by end of July. Only a few weeks earlier, Prime Minister Imran Khan had addressed the nation, defending his decision of a month ago to lift the lockdowns and move towards reopening the country on economic grounds.

“This virus will spread more” he had said in that address. “I have to say it with regret that there will be more deaths.” But Pakistan could not afford to remain shut any longer, he had argued, and people “will have to learn to live with the virus ... we cannot feed the poor indefinitely.”

The words had brought relief to many who were anxious for business to restart but has also fuelled anxiety about the future the country was moving into. The lockdowns had been lifted in the first week of May and the country saw rampant disregard for social distancing guidelines and advisories in the days leading up to Eidul Fitr on May 24. Images and videos circulated of mosques packed shoulder to shoulder with worshippers and markets crammed with shoppers. In one such video, a hapless woman police officer was shown after being roughed up by a crowd of worshippers in a working class locality of Karachi, when she had gone to try and enforce social distancing regulations. Debate raged in the country around this model of “self policing” and “smart lockdowns”, fuelled by the spike in infections that began after Eid.

The post-Eid spike began after May 26, soon after the number of daily new infections reported in the country dropped to a low of 1,356, after having touched a peak of 2,603 a few days before Eid. People noted that testing had dropped in those days as well, something the government explained by saying technicians have to go on Eid holidays too. From 16,387 tests done on May 21, the country saw only 7,252 tests performed on May 25, the day after Eid.

And then came the deluge.

Daily reported infections climbed relentlessly, as did daily tests, and both figures hit all-time highs by the time of the ‘twin peaks’ of June 13 and 19. The spikes, the projections and the government’s advice to “learn to live with the virus” all stirred massive anxiety around the country, as an anxious citizenry wondered whether they had been abandoned by their government.

During lockdown, shopkeepers at Lahore's Anarkali Bazaar sell products on the streets while keeping their stores shuttered to avoid confrontations with the police | Arif Ali/White Star
During lockdown, shopkeepers at Lahore's Anarkali Bazaar sell products on the streets while keeping their stores shuttered to avoid confrontations with the police | Arif Ali/White Star

Alarmed by the emerging scenario, the World Health Organisation (WHO) wrote a letter to all the four provincial governments, advising caution against lifting of the lockdowns.

“As of today, Pakistan does not meet any of the prerequisite conditions for opening the lockdown,” the letter said, pointing out the virus had spread far and wide within the country. “The pandemic has reached almost all the districts of Pakistan while major cities contributed the highest number of cases nationally.”

It advised raising the testing rate to 50,000 per day and an alternating two-weeks-on, two-weeks-off type of lockdown.

But the peaks proved to be short-lived. From June 15, both figures — daily tests and new infections — fell sharply even though none of WHO’s advice was followed in earnest. Within a month, new cases, as well as tests performed dropped to one-third of the peaks they had hit in mid-June, declining continuously till today, when fewer than 500 new cases are being reported on a daily basis.


THE DECLINING NUMBERS

A new debate briefly flared alongside these declines. At first, people argued that the low number of tests was responsible for the declining infections, and that large numbers were opting to suffer the virus at home. Different theories were posited for this. People were afraid to be tested for fear of the aggressive contact tracing that was triggered by a positive result, some said. People feared they would be deprived of a proper funeral in case they died of Covid-19 in a hospital, argued others. The government was deliberately concealing the numbers to justify its policy of opening up the country, went yet another theory. But the declines were relentless and across the board, and no discernible spike could be seen in the graveyard data to justify the argument that an ongoing catastrophe was being deliberately understated.

Most importantly, the declines in infections and tests were mirrored in a rise in the number of people recovering from infection. The recoveries began in large numbers on June 12 but, at that time, were overshadowed by the rising tide of daily new infections being reported at the hospitals, and did not receive much attention.

Government data showed 9,809 people recovered on that day, compared to 6,472 new cases of infection reported. From then on, the recovery data oscillated wildly, touching one peak after another with short, sharp troughs in between.

But each peak was larger than the preceding one. On July 3, more than 11,400 people recovered, while just over 4,000 new infections were reported. On July 17, more than 14,700 people recovered, while less than 2,000 new infections were reported. A week later, July 24 saw more than 16,800 recoveries against 1,487 new cases. The tide was changing rapidly. People were being discharged from hospitals in droves, while new arrivals slowed to a trickle.

The most important indicator — active cases — peaked at a level slightly above 100,000 between June 19 and July 4, after which it fell sharply. This was the high watermark for the infection, and this was the time period when the hospitals saw peak caseloads and it seemed as if the prophecies of doom were materialising before our eyes. But, in fact, this was it, and active cases fell rapidly from their peak of 108,642 on July 1 till they reached around 9,000 today. Consider that in a country like the United States, where Covid-19 became a health catastrophe, active cases climbed in the same time period from 1.2 million to beyond 1.6 million by July 4. By August 24, the US had more than 2.5 million active cases.

Men gather outside Data Darbar's new sanitisation gate | Arif Ali/White Star
Men gather outside Data Darbar's new sanitisation gate | Arif Ali/White Star

The NCOC began putting out figures showing how broad-based the reversal of the tide really was. Demand for pressurised oxygen had dwindled, ventilators were surplus, ICUs and Covid wards in the hospitals emptied out, and demand for tests was also diminishing. Moreover, the test positivity ratio — which measures the number of people that test positive out of all the tests that are performed — was also dropping rapidly. As this trend continued, the ranks of those who doubted the numbers initially also dwindled, until today hardly anybody is left who is willing to publicly argue that the government is deliberately understating the numbers.

Beyond the infection numbers, fatality rates showed an even more surprising trend. When the pandemic began, provincial health authorities, who had taken the lead in crafting the response in their respective jurisdictions, were preparing to treat people in the tens of thousands within a month, and anticipating fatalities on a large scale. But where the models said fatalities could rise to the millions by end June, less than 4,400 Covid-19 deaths were recorded in the country by June 30. There is no doubt that there were many unregistered Covid-19 deaths in the same time period, but it is not possible that these could have been in the tens of thousands.


GLOBAL TRENDS AND PAKISTAN

Epidemiologists look at the spread of an infection such as Covid-19 along two dimensions: incidence and severity. Incidence is a measure of how far and rapidly the infection has or will spread, among what clusters of the population and along what geographical vectors. Severity is what clinical outcomes this spread will produce, ranging from those who will be asymptomatic carriers to those who will die, to everything in between. In Pakistan, both the incidence and the severity have been far below what every model projected. Complex measurements are involved in measuring incidence and severity, but ultimately two indicators are of crucial importance: number of cases and fatalities, seen as a proportion of the total population.

In terms of incidence, Pakistan has seen 1,326 cases per million of population, where the world average is 3,072. The figure is much higher in those countries that have seen Covid-19 turn into a catastrophe. For example, in the US or Brazil, the same figure is above 17,000. Fatalities today are 28 per million of population in Pakistan, where the world average is 104 [see graph for comparison].

Worldometer data, accessed on August 24, 2020
Worldometer data, accessed on August 24, 2020

Not only that, the Situation Report for Pakistan — issued on August 23 by the MRC Centre for Global Infectious Disease Analysis at Imperial College — projects continuous declines in the rate of infection, fatalities, demand for hospital beds or patients requiring high-pressure oxygen, critical care or mechanical ventilation — all indicators used to measure and project disease severity — over the next 28 days, if present-day mitigation measures continue. The projections show a slight increase if these measures are relaxed and a sharp decline if the measures are tightened. Clearly, the incidence and severity of the virus has not turned into a catastrophe as in other countries.

When the pandemic began, the big question on everybody’s minds was how bad is this going to get. Today, the big question is this: what happened? How is it that the catastrophe that the models were projecting never materialised? Even when there was widespread disregard for the government’s advice on social distancing in the days leading up to Eidul Fitr on May 24, the spike in infections that followed flattened out long before reaching what the models had projected.

Most customers and shopkeepers have stopped wearing face masks in crowded markets in Karachi | Fahim Siddiqui/White Star
Most customers and shopkeepers have stopped wearing face masks in crowded markets in Karachi | Fahim Siddiqui/White Star

When asking this question — “what happened?” — the most important thing to bear in mind is that Pakistan is not alone in seeing its Covid-19 curves flatten out so rapidly. In terms of the fatality rate, there are 88 countries that have fared better than Pakistan. In terms of total cases (per million of population), there are 103 countries that are faring better than Pakistan out of a total of 215 countries (these figures are at the time of writing and vary slightly from day to day). So, in both dimensions — incidence and severity — Pakistan has fared well, but is one country out of more than a hundred that have fared equally well, if not better.

Compare those pairs of countries that are nearby to each other, share similar demographics and climate, as well as similar capacity to craft a response, yet have seen sharply divergent outcomes. In the Caribbean region, for example, the Dominican Republic has been hit hard, with 146 deaths per million (far above the world average of 104) and total cases per million of population at 8,488, where the world average is 3,090. Next door Haiti, on the other hand, has registered minimal impact, with deaths per million at 17 and cases per million at 710, making it one of the most lightly touched countries in the world.

When the pandemic began, the big question on everybody’s minds was how bad is this going to get. Today, the big question is this: what happened? How is it that the catastrophe that the models were projecting never materialised?

Closer to home, take the example of three countries whose death rates are below one (per million of the population): Vietnam, Thailand and Sri Lanka. In Vietnam, the first lockdown barely lasted 22 days, starting on April 1 and eased on April 23, when some businesses were allowed to reopen. Building on their experience with the SARS outbreak in 2003 (Vietnam was the first country to be declared free of SARS back then), the Vietnamese authorities put in place one of the world’s most aggressive and far-reaching contact-tracing efforts and contained local transmission at a time when the rest of the world was still grappling with managing their lockdowns. Since then, there have been localised lockdowns in the country, but no large scale re-emergence of local transmission.

By the middle of May, domestic tourism and flights had resumed in Vietnam and all schools reopened. The country saw 288 infections and zero deaths till then. By end June, international bodies were holding the country up as an example for other developing countries to follow. “Early on, the prime minister prioritised health above economic concerns,” the IMF said in a note about Vietnam published on June 29. Today, Vietnam has seen 10 cases per million of population, compared to Pakistan’s 1,324, and 421 active cases (meaning those who are still showing symptoms) where Pakistan has more than 9,000.

Sri Lanka has also seen less than one death per million since the first Covid case was detected on January 27. Lockdowns were lifted on June 28, when the government declared local transmission had ended, and all schools, hotels, cinemas and businesses were reopened. Today, Sri Lanka has 136 active cases, and its total cases per million of population is 138 — roughly one-tenth of Pakistan’s.

Likewise with Thailand, which also has less than one death per million of population, and currently has 115 active cases and total cases per million of population at 49. Lockdowns in Thailand — a country which relies heavily on tourism to sustain its economy — were lifted on July 1 and schools reopened fully, without any social distancing rules, by early August, by when the authorities said the country had gone two months without a case of local transmission being reported anywhere. However, in early August, Thailand once again closed schools in some parts of the country, after two new cases from international arrivals were detected.


ACROSS THE BORDER

Perhaps the most poignant example is that of India, which imposed the most brutal lockdown seen anywhere in the developing world, leaving tens of millions of migrant workers to walk hundreds, and in some cases thousands, of miles back to their hometowns and villages. In India, infections have continued to rise, but at 42 (per million of population), fatalities there remain far below the world average.

India is a big country though, and a closer look is necessary to make sense of their numbers, because many different things are happening within it. For example, the rate of infections is continuing to fall in Delhi and Mumbai, the two largest and most densely populated cities in the country, both with large crowded slums where social distancing is next to impossible.

In fact, similar conversations are taking place in India and Pakistan about why the catastrophe that has befallen many countries has not landed here, at least not yet. India has a far higher infection rate compared to Pakistan, and their situation report projects this to rise sharply in the next 28 days. But mortality rates there have still remained far below the world average [see graph]. “This is the puzzle” tweeted Vincent Rajkumar, editor in chief of the Blood Cancer Journal.

“Something is happening that we must try and understand more. What’s happening in India suggests some protection from severe disease being afforded by cross-immunity from prior corona viral infections.”

In his Twitter thread, reflecting on the puzzle — why we are not seeing similar mortality outcomes in our part of the world compared to those that have catastrophic levels of death from Covid-19 — he points towards two things. One is “cross-reactive immunity” from “prior coronavirus infections”, which he argues cannot grant immunity but can help curb severity and reduce ICU rates and mortality. The other is “genetic predisposition” which he himself says is “much less likely.”


WHAT THE STUDIES SAY

A paper published by 33 co-authors in Nature magazine on July 29 comes to a similar conclusion. “Clinical manifestations of Covid-19 vary, ranging from asymptomatic infection to respiratory failure,” the authors say. “The mechanisms determining such variable outcomes remain unresolved,” and they go on to explain the method by which they investigated why there is such large variation in clinical outcomes among those infected by Covid-19.

Their research led them towards what they call S-cross-reactive T-cells, a type of cell that is part of the body’s immune system, that was present in the blood of 83 percent of those patients in their sample who had recovered from Covid-19. They found that, in their sample, 35 percent of healthy donors (those not infected by the virus) also had a similar cell, “probably generated during past encounters with endemic coronaviruses.”

Pakistan has averted the sort of catastrophic outcomes that many countries have seen in the fight against Covid-19, but it would be a mistake to declare victory and consider the matter settled.

“[T]he presence of S-cross-reactive T-cells in a sizable fraction of the general population may affect the dynamics of the current pandemic,” they said, though pointed their findings more towards the search for a vaccine rather than an answer to why certain populations were more susceptible to the virus than others.

In another study, published on July 22 in the Journal of Human Genetics, five Japanese co-authors looked at the same problem. “[N]umbers of infected cases, deaths, and mortality rates related to Covid-19 vary from country to country,” they said. They studied 12,343 genome sequences of the virus gathered from infected people, from across six geographic areas, containing 28 countries that they clustered into three groupings, based on their fatality rates.

“Our findings suggest that SARS-CoV-2 mutations as well as BCG-vaccination status and a host genetic factor, HLA genotypes might affect the susceptibility to SARS-CoV-2 infection or severity of Covid-19,” they said. What “host genetic factor” might be at play is difficult to say.

Another study, also published in Nature magazine in June, focussed on gender differences and their relation to fatality rates. Just like some populations in the world have shown a higher susceptibility to the virus than others — as evidenced by their higher fatality rates — the authors of this study argued that men, especially those aged over 30, are more likely to die from the virus than women. What explains this divergence between men and women in the fatality data?

A market is full of shoppers ahead of Eid, most customers can be seen without masks and social distance is not being maintained | Arif Ali/White Star
A market is full of shoppers ahead of Eid, most customers can be seen without masks and social distance is not being maintained | Arif Ali/White Star

Their research points in some directions — the importance of genetics or how hormones interact with the immune system — without any conclusive answers. But the directions are critical to understanding why fatality rates differ so widely, between geographic areas, classes or gender. “[T]aking a sex-informed approach to Covid-19 research and medicine will uncover novel features of the host immune response to SARS-CoV-2 and, ultimately, result in more equitable health outcomes,” the authors say.

Others looking at the same problem have pointed out that those countries with younger populations could have reduced susceptibility to the virus, though why exactly this should be the case often leads us back to the studies that are looking carefully at the genetic or physiological reasons for how the human immune system is able to fend off an attack from the virus, reducing susceptibility and mitigating mortality. The best evidence is provided from Africa, described as “the youngest continent on the planet” — because 60 percent of the population there is below the age of 25. Most African countries have seen very low rates of incidence and severity. On August 16, for example, the WHO Africa office reported the largest drop in new cases being reported daily, as well as deaths globally, with South Africa, Kenya, Algeria, Nigeria and Ghana leading the way.

At its peak, countries under WHO Africa saw daily new cases just above 16,000 compared to almost 80,000 in Asia or Latin America. These are very low numbers and, even though there is some divergence among the countries — with some reporting continuous but gradual increases and others reporting sharp declines — the continent as a whole has escaped the catastrophe that Covid-19 has produced in North America or Europe. Like Pakistan and India, the low levels of mortality could owe themselves to a youthful demographic, but the specific reasons for why youth would be less susceptible are still under study.

In terms of the fatality rate, there are 88 countries that have fared better than Pakistan. In terms of total cases (per million of population), there are 103 countries that are faring better than Pakistan out of a total of 215 countries (these figures are at the time of writing and vary slightly from day to day). So, in both dimensions — incidence and severity — Pakistan has fared well, but is one country out of more than a hundred that have fared equally well, if not better.

Likewise, others looked towards climatic reasons, but countries such as Brazil and Peru, that have similar tropical climates to South Asia, have seen very large incidence of infections as well as mortality rates, so that explanation requires a great deal of more scrutiny before it can be considered.

Pakistan has averted the sort of catastrophic outcomes that many countries have seen in the fight against Covid-19, but it would be a mistake to declare victory and consider the matter settled. The most critical metric by which to measure success is the number of new cases being reported on a daily basis. In Pakistan, this number has come off its mid-June peaks when it was above 6,800 new cases per day. Today, it has fallen to less than 500, but compared to the rest of the world, while the curves may have flattened, Pakistan now stands near the top in the list of countries still reporting new infections on a daily basis, standing at number 15 in a list of 215 countries. Most other countries have brought local transmission under control far better than Pakistan.

It is premature to declare victory because we do not yet understand what exactly has happened here, considering so many other countries have seen similar, if not better results. Secondly, so long as even a small number of cases of local transmission remain, the danger lurks and can burst into the open again.

An authority no less than Dr Faisal Sultan, the government’s key point person on Covid-19, has cautioned against seeing the declining numbers as a victory. Until we have a clearer idea of why the virus did not produce the kinds of fatalities here that it has in other countries, and until we have firmly eliminated all cases of local transmission, it would be a mistake to consider this fight finished.


Header image by Mohammad Asim/White Star

The writer is a member of staff