No tropical or demographic dividends from Covid-19

From just 3 cases and hundreds of tests in early March, Bangladesh now has 1,56,391 Covid-19 cases detected from 8,20,347 tests as on July 3, 2020. The stock of cases has grown at a daily rate of 10 percent, doubling every seven days. The daily infection rate per hundred tests has grown from less than 5 percent in March to between 22-24 percent of late. At least one is found infected in every five tests. The virus has spread to all 64 districts.

Heat and humidity are no defence

The number of cases has grown proportionately more than the number of tests while heat and humidity increased from March to June. Rapidly rising infection cases in Bangladesh, India Indonesia and Thailand has demonstrated the resilience of the virus to the most hot and humid months (March to May). The ongoing resurgence of cases in Florida, the Sunshine state of the United States, puts the nail in the coffin of the theory that heat, and humidity constitute robust protection from the virus. 

One can argue endlessly that the situation could be worse without heat and humidity, citing various quantitative analyses using Chinese provincial level data in particular. However, the point is those other factors, such as poorly managed "holidays", assumed equal in the regressions and simulations seem to matter more. Bangladesh has been experiencing a spike in cases since June after the government lifted mobility restrictions on May 31. It had been imposed on March 26. 

Virus spread ahead of tests. This is illustrated by the responsiveness of the total number of infected cases to total number of tests. It stood at 1.23 on the 105th day of the pandemic in Bangladesh, rising from 1.1 on the 57th day. A 10 percent increase in test was associated with increase in infected cases by 11 percent on the 57th day while the same percentage increase in tests found infected cases rising by 12.3 percent on the 112th day – a change in the wrong direction (see chart). It is evidence of the virus spread curve not even reaching the point of inflection towards what appears to be a shifting apex, notwithstanding peak summertime, and the growing inadequacies of testing relative to the community transmission of the virus. 

Despite increases, testing coverage is still lower than India, Nepal, Sri Lanka, Thailand and Malaysia, among others. Official claims of a decline in the virus reproductivity rate from 2.5 in April to 1.5 currently lacks corroborative facts.

Younger is not safer

Fatalities have grown to nearly 2,000. New deaths per week has increased from less than 100 until April to 308 in June. Bangladesh's apparent overall Case Fertility Rate (CFR), based on deaths reported and attributed to Covid-19, is estimated at 1.26, rather low compared to many countries, developed as well as developing. Leaving aside issues related to under-reporting, undercounting of deaths and evidence on "excess deaths", what do we make of this low CFR?

With younger population structures as in Bangladesh, the per capita incidence of clinical cases is expected to be lower than in countries with older population structures. Consequently, demographic differences are assumed to play an important role in explaining differences in fatality rates across countries. The virus is known to have the worst consequences for older age groups who suffer from underlying medical conditions. As plausible as these propositions appear to be, a deeper look at the details reveals that the virus more insidious than we think.

The distribution of both cases and fatalities across ages in Bangladesh shows expected concentration of deaths in older ages (see table). The CFR in 60 plus age group is about 8.2 percent, nearly 2.5 times that in the 51-60 and 7.4 times in the 41-50 age group. CFR ranges between 6.3 percent in the 60-70 age group to 13 percent in the 80 plus age groups in South Korea, 1.9 to 15.6 in Spain, 3.6 and 14.8 in China and 3.5 to 20.2 in Italy, well above the CFRs below 60 years of age in all these countries.

Bangladesh's CFR in the 60 plus age group can by no means be considered low. However, less than 44 percent of the fatalities are concentrated in the 60 plus group, which is significantly lower than the 94 percent of fatalities found uniformly concentrated in the population over 60 years of age in other countries. The proportion of death (56 percent of total) among the young (less than 50 years) is high compared to other countries even though they recover better relative to the old.

A double-edged sword

The young get infected more. The 60 plus group account for only 6.7 percent of the cases while the less than 50-year-old account for 82 percent of the cases. Note that the share of 60+ in total population is 7.8 percent. Transmission among the young have whole-population effects if the young are major contributors to community transmission rates. In theory, given the fraction of the population that are young, the effect depends on the contacts they have with other age groups, their susceptibility to infection and their infectiousness if infected.

The susceptibility is high as evident from the dominance of the young in total cases. The contact can be assumed to be high too in the presence of extended family living in dense physical spaces. As many as 97 percent of Bangladeshi households have working age adults while 28 percent have an elderly. Overcrowding with high housing occupancy rates and room densities in infection hotspots such as Dhaka, Narayanganj and Chattogram implies high contact potential. It is biologically plausible that milder cases are less transmissible, for example, because of an absence of cough, but direct evidence is limited. If those with no or mild symptoms, as the young tend to be, are as or more efficient transmitters of infection compared with those with fully symptomatic infections, the overall burden increases with increased infection of the young.

Belief in the myth that young people are not as much at risk as older people and people with comorbidities means that the young are often unaware or oblivious to the fact that they may have the disease or the risk that they pose to others. This naturally causes the tendency to be lax about masks, distancing and engaging in super-spreading behaviour. They may do fine with the disease while still being the dominant vector in spreading it. Sadly, the young do not appear to do fine with the disease either.