![\"Figure](\"https:\/\/blogs.bmj.com\/bmjgh\/files\/2020\/05\/Capture.jpg\")
Note: Mumbai and Mumbai suburban are merged in COVID-19 reporting, we followed same for other characteristics. We considered Delhi as a place only for analysis because districts were not mentioned for Delhi COVID-19 data. Kancheepuram district was bifurcated in Chengalpattu and Kanchipuram in 2019, we used undivided Kancheepuram district information.<\/span><\/em><\/p>\n Mumbai has the highest burden of COVID-19 cases and more than 31% households of Mumbai reported more than four individuals sleeping in a single room. Apart from COVID-19 prevalence, death per million population is a good indicator to measure health system response, containment measures and overall management of COVID-19 [6]. In India, COVID-19 national death per million (DPM) is 3.4, it places India in a good position in comparison of Belgium (805 DPM), Italy and United Kingdom (545 DPM), France (437 DPM) and many other countries. Total 130 countries have death per million more than Indian average [7]. However, death per million, in ten out of top 20 districts is higher than national average and maximum in mega cities like Ahmedabad (87 DPM) and Mumbai (85 DPM), which is more than 25 times of national average. It may reflect that few cities have COVID-19 burden completely different from rest of the country.<\/p>\n Without any vaccine\/treatment for COVID-19, non-pharmaceutical interventions are only measures to control the COVID-19 epidemic. However, in crowded large cities, it is challenging to maintain one-meter physical distance due to lack of proper space within and outside the home. We find that very few districts with large cities are having an overwhelmingly large proportion of COVID-19 cases in the country. In our analysis, high urban proportion and population density were significantly correlated with the COVID-19 burden in districts having the highest burden of COVID-19. It seems COVID-19 is spreading only in a limited number of large cities\/districts with high population density. A tiny proportion of the cases are in rural areas or small cities and towns. Similar findings were reported in another study [8]. Indian districts are large, with about 65% of Indian districts have more than one million population. At present, only district-level data is available. It gives only a very gross picture of the districts as well as country. Declaring an entire district as a Red zone (high case burden) or Orange ( moderate case burden), and locking down the whole district may risk other health services as well as closing all social and economic activities will have major financial burden [9]. We argue that the availability of more granular data at the ward level in cities and sub-district level in the rural area,together with ;ocal policy flexibility at a lower administrative level, might help achieve better control. It is the right time to focus on a robust public health system to manage COVID-19 and to prepare for any future pandemic as well as to achieve sustainable development goals [10].<\/p>\n Editorial Note : BMJ Global Health Blogs is not peer-reviewed.\u00a0<\/em><\/span><\/p>\n About the authors:<\/strong><\/p>\n Ashish Awasthi is Assistant Professor, Biostatistics at Centre for Chronic Conditions and Injuries, Public Health Foundation of India. He is an INSPIRE Faculty supported by Department of Science and Technology, Government of India, New Delhi. He tweets @AwasthiAshish15<\/em><\/p>\n Dileep Mavlankar has been heading the Indian Institute of Public Health Gandhinagar as its Director since 2012. He is the Chair of Technical Working Group on Maternal Health by Government of India and a member of the Steering Group on Health for the 12th Five-year Plan of the Planning Commission of India. He tweets at @DMavalankar<\/em><\/p>\n Competing Interests<\/strong><\/p>\n We have read and understood the BMJ Group policy on declaration of interests and declare the none interests.<\/p>\n Acknowledgment: <\/strong><\/p>\n Ashish Awasthi is supported by Department of Science and Technology, Government of India, New Delhi through INSPIRE Faculty Program, for some other work.<\/p>\n References:<\/strong><\/p>\n 1 \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 CSSE. COVID-19 Dashboard by the Center for Systems Science and Engineering (CSSE) at Johns Hopkins University. https:\/\/coronavirus.jhu.edu\/map.html<\/p>\n 2 \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Lewnard JA, Lo NC. Scientific and ethical basis for social-distancing interventions against COVID-19. Lancet Infect Dis<\/em> 2020.<\/p>\n 3 \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 World Health Organization. Modes of transmission of virus causing COVID-19: implications for IPC precaution recommendations: scientific brief, 27 March 2020. World Health Organization 2020.<\/p>\n 4 \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Dubey M, Rastogi S, Awasthi A. Hypertension prevalence as a function of different guidelines, India. Bull World Health Organ<\/em> 2019;97<\/strong>:799\u2013809. doi:10.2471\/BLT.19.234500<\/p>\n 5 \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Prenissl J, Jaacks LM, Mohan V, et al.<\/em> Variation in health system performance for managing diabetes among states in India: a cross-sectional study of individuals aged 15 to 49\u2009years. BMC Med<\/em> 2019;17<\/strong>:92. doi:10.1186\/s12916-019-1325-6<\/p>\n 6 \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Deaths Per Million Key Indicator Of COVID Containment & Case Management. https:\/\/www.indiaspend.com\/deaths-per-million-key-indicator-of-covid-containment-case-management\/<\/p>\n 7 \u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0\u00a0 Total confirmed COVID-19 deaths per million people, May 26, 2020. 2020.https:\/\/ourworldindata.org\/grapher\/total-covid-deaths-per-million<\/p>\n