Globally, experts are looking at “excess mortality” or mortality gaps as another measure of the coronavirus death toll. This approach compares the number of deaths during a certain period to the historical average.
By Atisha Kumar
To quantify the effect of a shock in real-time is fraught with measurement and data issues. As India grapples with this pandemic, estimating the toll is even more challenging. Is the disease progressing slower in India than in many developed nations, or is the data reflecting testing gaps? Indian states’ mortality figures highlight that fewer deaths occurred in March compared to the same month in previous years, and historical monthly averages. The absence of missing deaths suggests that, for March, India’s official coronavirus mortality tally may be close to the actual toll, and the disease may not have progressed widely, at least until March 31. But, data gaps also emphasise that timely and accurate data will be important to combat the crisis.
Initially, limited access to testing restricted Covid testing to those exhibiting severe symptoms or travel exposure. Today, even as more people are tested, the sample is not representative of the population. Further, the false-negative rate on tests is up to 30%. The number of cases may likely be higher than the number of positive tests. Despite being a lagging indicator, the death toll due to coronavirus offers a more reliable estimate than the confirmed case count. But the official death toll attributed to the virus may also be lower than the number of actual deaths due to misdiagnoses and lagged reporting.
Globally, experts are looking at “excess mortality” or mortality gaps as another measure of the coronavirus death toll. This approach compares the number of deaths during a certain period to the historical average. For instance, a New York Times article compared the number of confirmed Covid-19 deaths with the number of burials taking place in Jakarta, Indonesia in March. Jakarta’s official death toll for March was 84. Yet, the city witnessed 1,600 more burials in March relative to the past. These deaths could likely be Covid-19 related.
Using India’s mortality data from the National Health Management Information Portal, I focus on 10 states with the highest number of coronavirus cases as of April 23: Maharashtra, Delhi, Gujarat, Rajasthan, Tamil Nadu, Madhya Pradesh, Uttar Pradesh, Telangana, Andhra Pradesh and Kerala. For each state, I compare total monthly deaths for individuals aged six and above in March 2020 and previous years. Deaths considered as infant and maternal mortality cases are excluded.
Unlike other middle-income countries such as Indonesia and Turkey, India witnessed fewer reported deaths in March this year. Seven of these ten states saw 46% fewer reported as compared to last year. In contrast, the decrease in deaths between March 2019 and March 2018 was 7%. For example, Maharashtra witnessed 14,804 deaths in March 2020, lower than the 18,584 deaths in March 2019.
The decrease in deaths due to other causes maybe falling owing to reasons unrelated to Covid-19. Thus, I exclude deaths from probable causes such as malaria, previously known illnesses, HIV/AIDs, encephalitis and accidents/burns. As some Covid-19 deaths may have been misattributed to other respiratory illnesses, I retain reported deaths due to tuberculosis, respiratory diseases including infections (other than TB), and other fever-related causes, heart disease/hypertension-related or unknown causes. Again, though, the same set of states witness fewer monthly deaths in March 2020 compared to historical averages. Maharashtra experienced 2,163 fewer deaths in March 2020, a 19% decline.
What does the absence of “excess mortality” suggest? Besides, showing that India’s official tally may be close to the actual reported deaths, it reinforces that India was affected at a later stage relative to other countries. But there could also be other factors at play. First, delayed reporting may be playing a role.
Second, physical distancing increased vigilance and targeted medical care due to the lockdown could have driven fewer deaths in March. Third, many deaths are not recorded or noted late in developing countries. Despite these issues, the state-level data remains important for studying the pandemic.
State health departments, which are already working round the clock to fight Covid-19, should prioritise reporting health indicators in an accurate, granular and timely manner. This will enable policymakers to follow the disease’s progression and assess the effectiveness of measures.
The author is Research scholar, Columbia University. Views are personal