COVID-19 burden and hospital readiness by A M Zakir hussain

PREDICTING the number of cases and deaths from any disease requires some basic information, eg, (i) exposure potential — the probability of people to be exposed to the pathogen; (ii) the infectiousness of the pathogen (transmission capacity of the pathogen from host to host); (iii) the infectivity or capacity of the pathogen to cause the disease; (iv) the susceptibility of the host to infectivity or capability to resist infectivity. In the absence of these information, no model will be able to predict the incidence or outcome of an infection. Possible death – scenario 1: Let us assume that all of the 16 or so crore people in Bangladesh have been exposed and everyone is susceptible to COVID-19 although none of which is practically possible. If this were true, the total number of COVID-19-related mortality in Bangladesh would be as presented in Table 1, based on a direct standardisation method, applying Chinese age-specific mortality data. In Table 1, the total of 11,55,847 in the second but last column includes 4,25,238 deaths among 0–9 years old children, which were not accounted for among the total of 4,77,581 additional deaths from COVID-19, in the last column as there would be no death according to the Chinese estimate in the 0–9 years age group from COVID-19. This is corroborated by the data from European and American countries as well. Up to April 14, in the United States, of the 6,839 deaths, 5,151 had underlying conditions and only 137 were found without any underlying conditions. For another 1,551, this status was unknown. While of the 4,77,581 who might die of COVID-19 in Bangladesh, about 75 per cent, ie, 359,108 people, would be above the age of 60 years, and 1,18,473 people might die between the ages of 10 years to 59 years in Bangladesh (many of who would also have pre-existing medical conditions, especially at younger ages — the 40,239 deaths shown in the last column). These are the numbers above the deaths that would occur even in the absence of COVID-19. But again, the 4,77581 deaths due to COVID-19 will occur only if everybody is susceptible to SARS-CoV-2 and everybody is exposed to it, which, however, is not possible. Again, 70–80 per cent of these 4,77,581 hypothetical deaths would have critical co-morbid conditions. In actuality, therefore, 119,395 people might die in perfect health because of COVID-19 in Bangladesh. These figures are theoretical propositions. Possible cases and deaths – scenario 2: Let us now estimate COVID-19-related figures for Bangladesh more practically, from another perspective, ie using the figures of Wuhan first. The city has a population of 11 million and the number of reported COVID-19 cases was 50,333, of them 3,869 died. This means an infection rate of 4.6 per 1,000 population in the city — a case fatality rate of 7.7 per cent, and a mortality rate of 0.35 per 1,000 population in Wuhan. Applying these rates to the total population of Bangladesh, the total cases of infection would be 777,874; the total number of fatal cases (or mortality) would be 59,794. Let us consider now the scenario of Italy. The total population of Italy is about 60.36 million. It reported by April 16, a total of 178,972 cases of COVID-19, of whom 23,660 died of the diseaset. The infection rate of COVID-19 in Italy is 2.96 per 1,000 population and the case fatality rate is 13.2 per cent. The mortality rate is 0.4 per 1,000 population. If we apply these rates to Bangladesh, the total cases of infection would be 503,200 and the number of dead would be 66,422. The population size in the United States is 328.2 million. The United States by April 16 reported 770,564 COVID-19 cases with 41,114 deaths. The incidence rate of COVID-19 in the United States, therefore, is 2.35 per 1,000 population and the case fatality rate is 5.34. The application of these estimates will give the total number of COVID-19 cases in Bangladesh to be 399,500 and the total fatality from COVID-19 in light of the United States would be 21,316. The population size of the United Kingdom is 66.65 million; the COVID-19 cases reported were 120,067 and death 16,060 by April 16. The incidence rate is 1.8 per 1,000 population and the case fatality rate is 13.38. The application of these rates to Bangladesh would give the total incident cases of 306,000 and the total death would be 40,943. As figures in Table 1 are only a theoretical proposition, under extreme and unsubstantiated conditions and those of city/country-wise estimates given later are experiential, we will consider these later figures for further elaboration. According to the Wuhan estimates, Bangladesh might have a maximum of 777,874 infections of COVID-19 and 59,794 deaths. According to the Italian estimates, Bangladesh might experience incident cases of 503,200, of whom 66,422 might die. The application of US and UK estimates would give lower figures of infection and mortality. The maximum infection we get is 777,874 and the maximum deaths 66,422. As these estimates are based on rates, an increase in absolute numbers will not sway these rates much. The total numbers estimated for incidence and mortality for Bangladesh are, therefore, apt to remain the same. Optimistic estimates – scenario 3: The additional points we need to bear in mind are the effects of the environmental and weather conditions on the viral transmission (sunlight, humidity, atmospheric temperature, etc); genetic pre-disposition of the people of Bangladesh, which will determine people’s susceptibility to the virus to cause disease (manifestation of infectivity); exposure potential which might be controlled through adoption of the preventive measure, eg isolation, lockdown, quarantine; and the infectiousness of the viral strain running through the population; development of immunity, in addition to the comorbidities. Exposure to homologous (similar genes) viruses/pathogens may also give people some partial protection. There is also a possibility that exposure to sunlight would help people develop vitamin D, which also is contemplated to give some resistance in people against SARS-CoV-2. The actual numbers, however, are difficult to estimate as the parameters are unknown. In the United States, at least 50 times more people have been found to develop antibody to the virus already, without any trace, while 20 per cent to 80 per cent people did not show any post-infection clinical features in China. There is no reason why the same phenomenon will not be observed in Bangladesh. This means of the 777,874 cases of infection, actually 518,583 cases may be reported if the asymptomatic cases are estimated to be a third of the total infections. But if the US data are considered, only 15,557 cases will be reported in Bangladesh for 777,874 infections. As the number of deaths in the South Asian countries is markedly low in comparison with those in Europe, we believe that the 66,422 deaths that have been estimated for Bangladesh, may not actually be experienced, because of some or all of the factors mentioned above. Hospital readiness for COVID-19 A DIVISION-WISE hospital based management of COVID-19 cases is suggested below. The number of 777,874 infections and 66,422 deaths, as estimated above theoretically, may be divided by division and cities hypothetically based on the relevant population size. The distribution of the cases of infection and mortality would be as in Table 2. It needs to be noted that, the actual number of incidence and death reported by the government by location is different from what has been estimated in Table 2. This difference may be due to the actual exposure potential and other factors, mentioned above. We only give a theoretical possibility here. Experientially, about 15 per cent of the infected (critical patients) would require hospitalisation; 5 per cent (a third of the critical patients) of whom would require ICUs, ie 38,894, in total. This may be broken down into the eight divisions of the country commensurate with the population size. This would be approximately 11,886; 7,125; 4,632; 3,928; 3,901; 2,851; 2,485; 2,089 ICU beds in Dhaka, Chattogram, Rajshahi, Rangpur, Khulna, Mymensingh, Sylhet and Barishal divisions. The total number of hospitals in Bangladesh is more than 6,500, ie six ICU beds on average per hospital — varying according to the size of hospitals and number of hospital beds, which usually follow the population size by location and are apt to be proportionate to the critical patients by location. This may range between 2–50 ICU beds as per the hospital size. But as may be surmised, we do not need these ICU beds all at a time. The average hospital bed occupancy, as the experience tells us, should not be more than 14 days on average. This means, over the next two months, we would in fact require 2,972; 1,781; 1,158; 982; 975; 713;621; and 522 (9,724 in total) ICU beds in Dhaka, Chattogram, Rajshahi, Rangpur, Khulna, Mymensingh, Sylhet and Barishal divisions. As many COVID-19 patients practically would die within a few days after hospital admission and some would not be able to reach hospitals in time, so about a half of the 9,724 beds might be actually optimal for COVID-19 patients in Bangladesh. This would mean 1,486; 891; 579; 491; 488; 356; 311; 261 ICU beds (4,862 in total) for Dhaka, Chattagram, Rajshahi, Rangpur, Khulna, Mymensingh, Sylhet and Barishal divisions. As the use of mechanical ventilators has found to be rather injurious to COVID-19 critical patients. This has not been considered. It may be of interest to note that the European Union countries have 11.5 ICU beds per 100,000 people, on average, in a normal condition. According to EU standards, we need 19,550 ICUs for Bangladesh under similar conditions. Life expectancy in the European Union is 81 years and in Bangladesh it is about 73 years while 60 per cent of the deaths in Bangladesh occur from non-communicable diseases (eg, heart diseases, diabetes, chronic obstructive pulmonary diseases, cancer and injury). Almost all of these might require ICUs. So, even keeping the near future needs into consideration, we should establish at least 9,724 ICU beds, mentioned above, in the country now. It might be kept in mind also that adequacy of ICUs in Bangladesh could increase the life expectancy of Bangladeshis further. The clinical aspects have not been discussed here as the treatment processes are changing rapidly globally. It is expected that treatment should be commensurate with the pathogenesis of the disease. Ventilators are not considered a panacea any more.

 AM Zakir Hussain is Working Group member of Bangladesh Health Watch. Dr. Hussain is also former director, Primary Health Care and Disease Control; former director, IEDCR, DGHS; former regional adviser of SEARO, WHO and former staff consultant, Asian Development Bank, Bangladesh.