Global crisis, local response: Emphasis must be on Covid-19 data disaggregated by geographies, vulnerabilities

The Indian Express, 

October 08, 2020

At 1.55 per cent, India’s COVID-19 mortality (case fatality rate) is an important talking point. It’s been hailed as among the lowest in the world, “less than many nations in Europe”, and the government is targeting to bring it down to less than 1 per cent. However, The Lancet cautioned about “the dangers of false optimism” in its September 26 editorial on the Indian situation.

Overall, Asian countries have experienced much lower COVID-19 mortality as compared to those in Europe. The initial high mortality in Spain and France has been attributed to several reasons: More than 75 per cent of the deaths were among those aged more than 75 years and living in care homes; high prevalence of co-morbidities, obesity and smoking among those infected; disproportionate number of health care providers infected early on (15 per cent of total cases) rendering the health services weak; and, shortage of diagnostic tests and personal protective equipment.

Three leading factors are reported to strongly influence the risk of dying from COVID-19: Male sex, advanced age (more than 60 years), and the presence of comorbidities such as diabetes, hypertension, chronic respiratory diseases, cancer and cardiovascular disorders. One analysis observed that the presence of these factors conferred a two-fold higher risk of death for Italian COVID-19 patients as compared to Chinese patients. The Indian age distribution is skewed in favour of the young, and contributes to the overall lower mortality.

Central to all this is “what counts as a COVID death”. In an epidemic situation, the public health approach is likely to attribute a death with complex causes as being caused by the disease in question, termed ascertainment bias. As early as April 16, the WHO clarified in unequivocal terms: “A death due to COVID-19 is defined for surveillance purposes as a death resulting from a clinically compatible illness, in a probable or confirmed COVID-19 case, unless there is a clear alternative cause of death that cannot be related to COVID disease (e.g. trauma). A death due to COVID-19 may not be attributed to another disease (e.g. cancer) and should be counted independently of pre-existing conditions that are suspected of triggering a severe course of COVID-19”. Many states have set up expert committees to re-examine and verify COVID-19 deaths; the official mortality figures reflect the “final” figures after vetting by the experts. While a state has full authority to review deaths, and indeed should do rigorous mortality analysis, it should be guided by the WHO position on this matter.

A recent National Bureau of Economic Research (a private non-profit research organisation) Working Paper used age-specific COVID-19 mortality rates from 17 comparison countries, coupled with India’s distribution of cases (mirroring the age composition), to “predict” what India’s CFR would be with those age-specific rates. In most cases, the predictions are lower than India’s actual performance, leading them to suggest that India’s CFR is, if anything, too high rather than too low and caution against misplaced complacence. The critical point being that an overwhelming concern with the overall CFR focuses on crude aggregates (and not age specific rates) and may cloud the fact that the news may not be as comfortable it appears to be.

The WHO estimates that 14 per cent of infected cases are severe and require hospitalisation, 5 per cent of infected cases are very severe and require intensive care admission and that 4 per cent of the infected will die. The Ministry of Health and Family Welfare reports the current case fatality rate at 1.55 per cent. It was nearly 4 per cent in Punjab and Maharashtra a week ago. District-level figures are instructive. Mumbai and Ahmedabad, the high-incidence districts (predominantly metropolitan and suburban/peri-urban), had mortality of 4.89 per cent and 5.20 per cent respectively while an urban, industrialised district such as Ludhiana, Punjab reported 4.31 per cent. Nanded and Sangli in Maharashtra with about three-fourth rural population had mortality of 5.25 per cent and 4.99 per cent respectively in the last fortnight, and now declined to 3.94 per cent and 3.86 per cent.

India’s mortality figures should be bench-marked against the WHO’s 4 per cent (and not the very high European figures), and closely watched as cases continue to spread in rural areas. Are the health services able to cope with the current levels of cases? An emerging indicator signalling a crisis seems to be the story of oxygen supplies. The industry has indeed quadrupled production to 2,700 tonnes per day from 750 tonnes a day in the last six months. Yet, ground reports from Madhya Pradesh, Gujarat and Mumbai have indicated, as one of them put it, a “ . . . panic mode, even forcing a few (hospitals) to marginally reduce the oxygen flow to their patients to optimise use”.

The COVID-19 new cases are currently on a declining trend, notwithstanding the caveat of a relatively high proportion of rapid antigen testing with higher false negative rates; surges shall remain a constant challenge as all activities other than education, hospitality and entertainment sectors have resumed, with very little real restrictions. The WHO cautioned on August 4 that “we have seen broad variations in naïve estimations of CFR that may be misleading”. Adjusting for the stage of the pandemic in each setting is essential to correctly interpreting COVID-19 statistics.

The imperative now is to have open and transparent data sharing with scientists, public health professionals and indeed the public at large. The emphasis should be a lot more on data disaggregated by geographies and vulnerabilities, looking beyond broad-brush summary aggregates. This will enable framing responses “as local as possible”.

https://indianexpress.com/article/opinion/columns/covid-19-global-crisis-local-response-6707887/