COVID-19 Statewise R0 Analysis for India

This project is being archived due to discontinuation of reliable data streams.

Link to Dashboard

Data Refresh

The chart is generated daily multiple times using GitHub actions. It is scheduled to run every 6th hour as per UTC timezone.

What is R_t?

In any epidemic, is the measure known as the effective reproduction number. It's the number of people who become infected per infectious person at time t. The most well-known version of this number is the basic reproduction number: when . However, is a single measure that does not adapt with changes in behavior and restrictions.

How does it help?

As a pandemic evolves, increasing restrictions (or potential releasing of restrictions) changes . Knowing the current is essential. When , the pandemic will spread through a large part of the population. If , the pandemic will slow quickly before it has a chance to infect many people. The lower the : the more manageable the situation. In general, any means things are under control.

The value of helps us in two ways.

1. It helps us understand how effective our measures have been controlling an outbreak
2. it gives us vital information about whether we should increase or reduce restrictions based on our competing goals of economic prosperity and human safety. Well-respected epidemiologists argue that tracking R_t is the only way to manage through this crisis.

Limitations

• Few States and United Territories might be missing in the chart. That is due to either insufficient data or very few confirmed cases in those areas.

News Coverage

This work was cited in the below news

• How dangerous is COVID-19 in the grand scheme of things, Deccan Herald
• Covid slowing in Maharashtra and MP, shows infection rate analysis of govt data, The Print

Social Media Coverage

This work has been appreciated in social media by epidemiologists and public health professionals-

• Dr. Giridhar R Babu, Professor, Head, Life course Epidemiology, Physician
• Lipika Nanda, Vice President, Multisectoral Planning in Health, PHFI. Public Health Professional

Credits

• This work is based on Estimating COVID-19's R_t in Real-Time By Kevin Systrom
• Special thanks to Thejesh GN for collecting and maintaining the State wise COVID-19 data from the Ministry of Health and Family Welfare, Government of India website.

Disclaimer

I'm not an Epidemiologist or scientist. I have only applied a modified version of a solution created by Bettencourt & Ribeiro 2008 to estimate real-time using a Bayesian approach. While this paper estimates a static R value, here we introduce a process model with Gaussian noise to estimate a time-varying .