COVID-19 and all-cause-mortality; differential trends between countries and selected states of India

Jaideep C Menon (  menon7jc@gmail.com ) Amrita Institute of Medical Sciences, Amrita Vishwa Vidyapeetham Kochi P Suseela Rakesh Amrita Institute of Medical Sciences, Amrita Vishwa Vidyapeetham Kochi Omesh K Bharti State Institute of Health and Family welfare Kaushik Mishra Sri Jagannath Medical college and hospital Basanta K Swain State Surveillance unit, IDSP Sunil K Raina Dr. Rajendra Prasad Government Medical College Denny John Amrita Institute of Medical Sciences, Amrita Vishwa Vidyapeetham Kochi Chandrasekhar Janakiraman Amrita Institute of Medical Sciences, Amrita Vishwa Vidyapeetham Kochi Amitava Banerjee University College Hospital


Introduction
All-cause mortality-lessons from current and past pandemics Health systems worldwide, including those of developed economies are struggling to prevent spread and mortality due to COVID-19, without neglecting other health and societal priorities. It is clear that  has direct effects through infection, and indirect effects through changes and strains on health systems and behaviours of individuals, whether patients or health professionals. Despite increasing recognition of impact on acute and long-term morbidity, mortality is the most frequently used measure of impact and of comparison between countries and over time. All-cause mortality is an important measure of overall health, relevant across a wide spectrum of diseases 2,3 . Through a "syndemic" lens, COVID-19, NCDs (noncommunicable diseases) and social determinants of health are interplaying with health system preparedness and political decision-making.All-cause, COVID-19 and non-COVID-19 mortality need to be understood to estimate and mitigate direct and indirect impact, and to inform actions required at system level 4 . In India, with a higher NCD burden that any other country, this approach is especially important, and comparison with other countries severely affected by the pandemic, may also be informative.
Previous pandemics have been associated with increased all-cause mortality, as well as deaths not directly attributable to the pandemic, especially NCDs and other infections. For example, the Spanish u pandemic of 1917-1918, in uenza, Severe Acute Respiratory Syndrome (SARS) and Middle-Eastern Respiratory Syndrome (MERS) have followed these patterns. During the in uenza pandemic globally between 1957 and 1959, excess respiratory mortality rate was 1.9/10 000 population (95% con dence interval 1.2-2.6 cases/10 000 population). 5 Excess mortality rates varied 70-fold across countries; Europe and Latin America experienced the lowest and highest rates, respectively.The H1N1 pandemic

Methods
The Indian scenario-India, a country of 1.37 billion, has the largest total NCD burden in the world with signi cant health and social inequities.Given the scale of infection in India, and the increased risk of mortality associated with underlying NCDs, COVID-19 and non-COVID-19mortality is expected to be high. In India, health and healthcare are organised at the state level within the federal system of the country, with signi cant differences between states in terms of health-related indices, but mortality has not been investigated in India compared with other countries, or between states in India in the COVID-19 context, to our knowledge. (Table 1) Study Design: The study was an observational study.
Methods: We compared population size, COVID-19 and non-COVID deaths, crude fatality rate (CFR), prevalence of NCDs (from Global Burden of Disease 2016 data) 10 and proportion of the population > 65 years of age in India, USA(National Vital Statistics System, NVSS) 11 , UK(O ce for National Statistics), 12 Spain (Instituto Nacional de Estadistice, INE) 13 and Italy(Worldometer). 14 For the Indian states of Himachal Pradesh, Kerala, and Odisha, birth and death data were collated from Local Self Government websites and State Epidemiology o ces of respective states 17 , comparing 2020 with the preceding ve years. (Table 1)The three Indian states were selected on the basis of data availability.
India has faced lower burden of COVID-19 cases and deaths per million (758 and 11) than other countries (e.g. USA: 6649 and 112, and UK: 4526 and 120 respectively), resulting in a relatively lower CFR (1.4%), compared with USA (1.7%) and UK (2.6%). 15 The proportion of the population aged >65 years is lower, and the burden of NCDs per million population per year is also lower, compared with other countries 16 ( Table 2).

Results
( Figure 1)In the northern state of Himachal Pradesh,there was a 1.9% increase in deaths, with 2.8% decrease in births in 2020, compared to 2019. In the southern state of Kerala, all-cause mortality and total birth rate reduced by 13.3% and 9.2% respectively in 2020, compared to 2019.All-cause mortality and births decreased by 16.7% and 21.4% in the eastern state of Odisha. Himachal Pradesh has the higher CFR (1.7%) than Kerala (0.4%) and Odisha (0.6%).

Conclusion Lessons from India
Many potential causes have been proposed to explain the low CFR in India, including extensive BCG vaccination coverage, enteric biomes, relatively low population density in rural India and younger population. O cial statistics suggest the role of the relatively younger age of the Indian population and the lower prevalence of NCDs,compared with many developed countries worse affected by the pandemic.
The decrease in all-cause mortality in Kerala and Odisha has been underlined by signi cant reductions in emergencies in specialties from paediatrics andcardiology to neurology and orthopaedics,as reported by respective clinical associations and media, although under-reporting is possible.Decreases in clinical interventions and surgical procedures, as well as sales of antibiotics have also been reported as low, which tally with reduced burden of disease.
Kerala, like the majority of states in India, has had low case fatality rates, compared with the USA, UK, Spain and Italy, despite signi cant COVID 19 case-load. The response of Kerala to COVID-19 with good political commitment, active community engagement, proactive care of elderly and vulnerable and through a social equity lens has been highlighted previously.
The present pandemic has ushered certain drastic changes in our social and behavioural habits, which is now considered the new normal. The lockdown, physical distancing, face masks, frequent hand washing and other sanitary measures are likely to continue as a behavioural change even after the pandemic recedes.
Reduced all-cause mortality could berelated to reductionsin deaths due to myocardial infarctions, strokes, road tra c accidents, non-COVID lower respiratory tract infections, and acute exacerbations of chronic respiratory diseases.Other contributing factors may include behavioural changes, such as use of masks, hand washing and sanitisation, decreased alcohol and tobacco consumption due to lack of availability and healthier dietary habits. Moreover, lockdown has led to lower environmental pollution, potentially better care of vulnerable groupsand greater attention to health at individual, community and media level.On the other hand, indirect effects of the pandemic on treatment of acute and chronic non-COVID disease, including cancer, cardiovascular disease andrehabilitation caremay lead to morbidity and mortality in coming months and years.
All-cause mortality gures need to be supplemented with breakdown by cause of death and ageadjustment.We only present data from three states in India,which have a good record in implementingpublic health programmes.Data from other states is required to before generalising for the rest of the country.
There are signi cant differences in incidence, deaths and CFR due to COVID-19 between populations and countries. In terms of spread of a pandemic, COVID-19 is the most extensive with all nations affected. Unlike previous pandemics,all-cause mortality has actually been lower or on par with previous years in three states of India,compared with signi cantly higher mortality rates in many high income countries.High quality data from all states is required to facilitate further research regarding the low COVID 19 case fatality rate in Kerala and India, as well as breakdown of cause of death and age-speci c mortality.  Trend of births and deaths across Himachal Pradesh, Kerala and Odisha