<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Purnoor Sodhi on Medium]]></title>
        <description><![CDATA[Stories by Purnoor Sodhi on Medium]]></description>
        <link>https://medium.com/@purnoorsodhi?source=rss-52fa8be999eb------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*KIOvtLvQ0LuUDSI8_buawA.jpeg</url>
            <title>Stories by Purnoor Sodhi on Medium</title>
            <link>https://medium.com/@purnoorsodhi?source=rss-52fa8be999eb------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Wed, 18 Nov 2026 06:19:24 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@purnoorsodhi/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[We need to re-open the Indian economy soon: A decentralized approach]]></title>
            <link>https://medium.com/@purnoorsodhi/we-need-to-re-open-the-indian-economy-soon-a-decentralized-approach-f8271b908e67?source=rss-52fa8be999eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/f8271b908e67</guid>
            <category><![CDATA[economy]]></category>
            <category><![CDATA[coronavirus]]></category>
            <category><![CDATA[india]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[covid19]]></category>
            <dc:creator><![CDATA[Purnoor Sodhi]]></dc:creator>
            <pubDate>Sun, 19 Apr 2020 21:01:12 GMT</pubDate>
            <atom:updated>2020-04-20T04:14:46.962Z</atom:updated>
            <content:encoded><![CDATA[<p>This article will talk about the rationale behind why I think the Indian economy should re-open soon. The financial loss to the economy has been significant, estimated at <a href="https://www.thehindubusinessline.com/economy/covid-19-lockdown-estimated-to-cost-india-45-billion-a-day-acuit-ratings/article31235264.ece"><strong>~INR 30,000 cr per day</strong></a> of the lockdown. However, while the national lockdown has worked for India, a decentralized district-wise approach would need to be considered soon.</p><p><a href="https://medium.com/@purnoorsodhi/the-india-story-covid19-742c37a2b72b">When I last wrote on April 6</a>, we had 4298 confirmed infections with a 118 deaths. Today, we’re starting at a staggering number of 17139 cases and 559 deaths — this corresponds to a <strong>3.98x factor in number of infections and 4.73x in number of deaths</strong> over the last 2 weeks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/890/1*e4Hq5kBzOeuFuBZO4z4-_g.png" /><figcaption>Most data for this article has been sourced from the fantastic resources available at the <a href="https://www.covid19india.org/">Covid19India </a>website that also tracks state-wise metrics on cases, deaths and testing.</figcaption></figure><p>This looks like a scary graph. However, the rate of growth has seen some improvement — the 3 day CAGR has seen some deceleration across the country.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/616/1*JN-gmV6IcNH2_oQtH_s8Ag.png" /><figcaption>A 3 day CAGR calculated from the 15th of March onwards, estimated as the compounded growth rate over 3 days — this reduces local maxima due to spurts in testing/results and cluster identification</figcaption></figure><p>This also corresponds with a commensurate increase in the number of days the infection takes to double in the country — reaching close to an <strong>average of 8.5 days</strong> over the last week.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/944/1*WbQW3dsJjZ9JBMXdSILJxA.png" /><figcaption>Days to double are estimated as log(2)/log(1+3 Day CAGR). This now stands close to 8.5 days — far better compared to countries like United States (3.5 days) and United Kingdom (4 days)</figcaption></figure><p>This should give the Indian government and its citizens reason to cheer — the infection seems like it is in <strong>early stages of control</strong>, especially in many smaller states of the country.</p><p>This is also supported by the steadily improving ratio of positive infections to tests conducted, which have in turn also grown significantly.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/854/1*3eeUYmdgp9CmWpO9YGyPdA.png" /><figcaption>The number of tests conducted per 1000 people have grown — a total of ~410,000 tests have now been completed, translating to ~0.31 tests/1000 people. While this is still quite low compared to international averages, the number of positive infections/test stand at just about 4.18%</figcaption></figure><p>The news is even better in a number of states across the country — <strong>delivering a positive test ratio of less than 1.5% </strong>— far better than international benchmarks.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/738/1*96nNTnzj_rEyONsTsgDeAA.png" /><figcaption>The graph above only shows states that have more than 20 cumulative cases reported. Telangana is an exception to the table above — the state does not have reliable estimates on the number of tests it has done</figcaption></figure><p>This compares quite favorably with many countries across the world — the top 30 countries exhibit a ratio of positive cases/test of ~13.6%</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/946/1*QEBtA29ob1dds3hLudgI4g.png" /><figcaption>The top 5 countries exhibit an average of 17.2% positive cases/test, meaning that one in six people tested show up as a confirmed Covid19 infection — cities like NY show an unbelievable 52% in positive cases/test. A notable exception in this list is China — that does not report the number of tests it has completed.</figcaption></figure><p><strong>Summary</strong>: It’s evident that the lockdown has worked in India. We’ve done well in limiting the growth in the number of cases compared to most countries across the world — some states are better off than others. The rate of increase of the infection has slowed, the testing has ramped up and the number of positive infections/test are at a manageable low.</p><p>The current lockdown ends on the 3rd of May. Does this mean that we should open up the national economy immediately?</p><p><strong>Not quite.</strong></p><p>If you look at the data for the infections at a district level for all the 700+ districts in the country, it’s evident that most of the infections are <strong>currently concentrated in the urban zones</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/960/1*2muYBrYnPZCBfXjZiqdpgw.png" /><figcaption>This is a cluster graph depicting the district wise number of Covid19 cases. The X-axis indicates the urbanization index (number of urban households/total households) and the Y-axis is the number of confirmed Covid19 cases on a <strong><em>logarithmic scale. </em></strong><em>The dotted line indicates a positive correlation between the two variables.</em></figcaption></figure><p>In fact, the same trend is visible if we plot a similar graph of the number of confirmed infections to the “Economic index” at a district level.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*j42yoeGsjmgJeEAnxATIVQ.png" /><figcaption>The dots again represent the coordinates between the Economic Index of a district and the number of Covid19 confirmed infections. While the district wise domestic product was difficult to retrieve, I have used surrogates like Household income, Two and Four Wheeler ownership, Urbanization, TV/other appliance ownership to estimate the Economic Index. Data for this was available from the <a href="https://howindialives.com/gram/index.php">How India Lives</a> website. The X-axis indicates the Economic Index and the Y-axis indicates the number of Covid19 cases on a <strong>logarithmic scale</strong></figcaption></figure><p>Clearly, the infection seems to be prevalent in urban centers and more importantly, centers of economic activity.</p><p>In fact, <strong>states contributing to a larger share of the GDP are worse hit than others</strong>. If we classify all districts with a simple colour code based on the following logic:</p><p><strong>Red</strong>: With more than 100 cases so far; <strong>Orange</strong>: Between 20 and 99 cases so far; <strong>Yellow</strong>: Between 1 and 19 cases so far and finally, <strong>Green</strong>: No confirmed cases so far</p><p>The following table indicates states that are the worst hit — incidentally, these 11 states also contribute to ~71% of India’s GDP.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*gjDx6Dx4RNJwUrm0hlnIrw.png" /><figcaption>The above table indicates states that have a share of Red districts &gt;0 (more than 100 confirmed cases). The situation is increasingly better with states at the bottom of this list — with a higher share of Yellow and Green districts.</figcaption></figure><p>The residual states and UTs that contribute to the remaining 29% of the economy are in far better shape.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/731/1*bnzwSQx5k8EeyZNFBoc1uQ.png" /></figure><p><strong>Summary</strong>: Although the infections are concentrated in states/districts of high economic activity, more than 300 districts have not had a single case of Covid19.</p><p><strong>This brings us to an important point:</strong> Although there are multiple urban centers that have a high number of infections, a national lockdown blocking ~85% of economic activity across the 720 districts of the country is <strong>inherently inefficient</strong>. The decision on continuation of a lockdown beyond the 3rd of May needs to be decentralized — the state governments should be given control on which districts can recommence economic operations and which should continue to remain shut.</p><p>However, there would need to be some <strong>ground rules and principles</strong> that state governments could use to restart the economic engine without compromising on the health and safety of its citizens.</p><p>a) <strong>Identification of the Go/No Go Districts</strong>: While the above analysis focuses on the number of cumulative cases, a better way of identifying districts would be on the number of active cases within each district. State governments are far better equipped to decide and manage this.</p><p>b) <strong>Deciding upon the conditions for Restart:</strong> Critical success factors to be identified to restarting economic and social activity. This could also mean that states should be free to lockdown districts again that do not exhibit these success factors.</p><p>c) <strong>Timely and accurate monitoring, case isolation and contact tracing to continue</strong>: This is important to control and limit future outbreaks and future spread.</p><p>d) <strong>Control on inter-district travel</strong>: Personal and non-essential travel would still need to be controlled to avoid spreading the infection beyond its borders. This is easier-said-than-done, but one of the few ways we could avoid spreading the infection further.</p><p>e) <strong>Augmenting medical capacities at a district level:</strong> I wrote about the <a href="https://medium.com/@purnoorsodhi/the-india-story-covid19-742c37a2b72b">“Breaking point”</a> the last time, while we have started flattening-the-curve, medical infrastructure and capacities at a district level are still quite unevenly distributed. An exercise to augment and re-purpose existing infrastructure would be required immediately.</p><p>f) <strong>Continuation of “social-distancing” at places of work, business, school etc</strong>: Resumption of social and economic activity should not mean that we go back to the way things were before the epidemic. Remote working where possible should be encouraged.</p><p>g) <strong>Evaluating a stack-rank of activities that a high-risk v/s low-risk:</strong> Working in an office that calls its employees to work via an “odd-even” rule could be far less riskier than attending an IPL game. It therefore makes more sense to resume activities that deliver a higher benefit with respect to the potential cost.</p><p>The burn-rate for the Indian economy is much higher than we can afford. India has done well in significantly reducing poverty and enhancing personal disposable incomes over the last 10–15 years — with a grim outlook expected for the short term, the authorities would need to do all in their power to ensure that these gains are not erased in a matter of months. However, the last thing India needs is a second-wave of infections. Let’s do our best to make it right.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=f8271b908e67" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The India Story — Covid19]]></title>
            <link>https://medium.com/@purnoorsodhi/the-india-story-covid19-742c37a2b72b?source=rss-52fa8be999eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/742c37a2b72b</guid>
            <category><![CDATA[pandemic]]></category>
            <category><![CDATA[coronavirus]]></category>
            <category><![CDATA[covid19]]></category>
            <category><![CDATA[data]]></category>
            <category><![CDATA[india]]></category>
            <dc:creator><![CDATA[Purnoor Sodhi]]></dc:creator>
            <pubDate>Mon, 06 Apr 2020 17:04:38 GMT</pubDate>
            <atom:updated>2020-04-07T08:22:25.772Z</atom:updated>
            <content:encoded><![CDATA[<h3>Covid19 in India — The Breaking Point</h3><p>I wrote about the <a href="https://medium.com/@purnoorsodhi/the-r0-effect-6a556e554d94">R0 effect</a> nearly 10 days ago, on the 26th of March (a few days after the national lockdown was announced) — India then had 681 confirmed infections with 12 deaths.</p><p>Since then, the number of confirmed infections has exploded to 4298 with 118 deaths in just 10 additional days. This translates to a <strong>multiple of 6.3x for infections and 9.8x for deaths</strong>. The sheer number of confirmed deaths compared to the reported number of infections in such short duration has thrown up a number of interesting insights. For example:</p><p>a) Although the number of tests delivering a positive confirmation have been around 3–3.5% for India, have we indeed tested enough?</p><p>b) We expected that the number of new daily cases would have reduced over time, at least 10 days after the announcement of the lockdown. Why has this not happened?</p><p>c) With this rate of explosive growth, does India have the necessary medical infrastructure to take care of its citizens?</p><p>d) Which states would be under the most amount of pressure from this point onward?</p><p>This brings me to the first point of discussion.</p><h3>1. Estimating under-reporting via Delay-Adjusted CFR</h3><p>An easy way to estimate under-reporting that has been used in multiple reports has been to evaluate the total number of active cases via the total deaths that have occurred due to Covid19.</p><p>The model evaluates under-reporting based on the following</p><p>i) Dividing deaths-to-date by cases-to-date leads to a biased estimate of the <strong>case fatality ratio (CFR)</strong></p><p>ii) Deaths today would mean that the infected person(s) had caught the infection at least 18 days prior to death. In this duration, the person would have likely infected a few more people.</p><p>iii) Using the distribution of the delay from hospitalization-to-death for cases that are fatal, we can estimate how many cases so far are expected to have known outcomes (i.e. death or recovery), and hence adjust the naive estimates of CFR (total deaths/total cases) to account for these delays</p><p>The fundamental reasons why this methodology is effective is because:</p><p>a) <a href="https://www.forbes.com/sites/brucelee/2020/03/18/what-percentage-have-covid-19-coronavirus-but-do-not-know-it/#69babc8c7e90">Nearly 20%</a> of all Covid19 infections are asymptomatic. This means that many people continue to live their lives without displaying any symptoms like fever, cough etc and continue to infect an average of 2.4 more people.</p><p>b) It is easy to under-report the number of active cases, but <strong>extremely difficult to under-report deaths due to Covid19</strong>.</p><p>We can use the infections and death data for the states in India to evaluate the accurate CFR via the following:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/271/1*yNPxxgfBGDVpyov68iYGRQ.png" /><figcaption>ut represents the underestimation of the known outcomes and is used to scale the value of the cumulative number of cases in the denominator in the calculation of the cCFR, ct is the daily case incidence at time, t and ft is the proportion of cases with delay of t between confirmation and death.<a href="https://cmmid.github.io/topics/covid19/severity/global_cfr_estimates.html"> A fabulous study</a> has been conducted by the CMMID nCov working group to establish this.</figcaption></figure><p>The best estimates of CFR (post-adjustment) are in the 1%-1.5% range taken from a detailed study in China which is around <strong>1.38%</strong> (95% CI: 1.23–1.53%). This means that if the adjusted CFR is around 8.5%, which is the case for India, then the actual under-reporting is 1.38%/8.5% = 16.2%.</p><p>This is how the situation looks like for India at the moment, with the reported cases:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/654/1*Fl7nKS1dKlxv4bw1FwbHVg.png" /><figcaption>For the purpose of legibility and clarity, I have focused all the analysis henceforth on the top 22 most populous states in the country — these states currently contribute to 98.9% of our population and contribute to 98.7% of the confirmed infections with 100% of the reported deaths</figcaption></figure><p>However, if we now apply the adjusted CFR for the Indian states and determine under-reporting, this is what shows up:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/698/1*h8GURR0y1KwARwr_Ra9Lrw.png" /></figure><p>Note how states like Maharashtra, Madhya Pradesh, Gujarat and Telangana now have a <strong>massive number of under-reported cases</strong> — this is primarily since the number of deaths in these states are extraordinarily high compared to the number of reported infections.</p><p>Maharashtra seems to skew this graph a lot, so lets look at it without MH:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/697/1*laAvxihl0r786USWdudiFg.png" /></figure><p><strong>Summary</strong>: With a little bit of math, we are able to estimate the number of unreported cases of Covid19 in India. It seems like <strong>we have ~26606 actual infections in the country </strong>— instead of the currently reported number of 4298 (as on 06th Apr 2020).</p><p><strong>The Silver Lining</strong>: So far, we’ve all heard stories of the excellent contact-tracing conducted in India, opening up of Covid testing to private players and the aggressive approaches taken up by the government to tackle the situation both at the national level and at the state level. We can hope that the number stated above is an over-estimation.</p><p>The next part evaluates how well prepared India’s medical infrastructure is to tackle this crisis.</p><h3>2. Healthcare Capacity — The Breaking Point</h3><p>We’ve just seen that there are massive disparities between different states in terms of the number of Covid19 cases — both reported and potentially, the under-reported ones. There are some additional complications:</p><p>a) The medical infrastructure available in each of these states are also <strong>unevenly distributed</strong>.</p><p>b) Populations of the vulnerable, especially the 65+ population in each of these states, are also varied.</p><p>c) With an exponential rise in cases, we could expect many of these systems being stressed to potentially a <strong>breaking point</strong>.</p><p>To start, lets assess the number of hospital beds available in each of these states. A lot of this data is available in the public domain, multiple reports like <a href="http://www.cbhidghs.nic.in/WriteReadData/l892s/Chapter%206.pdf">this one by the National Health Profile 2018</a>, carry decent information on the state-wise infrastructure available. I have made a few estimates to supplement this data with additional information on private hospitals and augmented capacities in current ones.</p><p>Here’s the number of beds available in India state-wise:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/810/1*FTb2eT9PD-AXMo5pu3nHYQ.png" /></figure><p>India has 0.07% beds/person — this translates to ~930,000 beds for the population of 1.33b or <strong>0.7 beds per 1000 people</strong> (or 1 bed/1419 people). This compares poorly with international standards, especially in countries which have shown a high number of Covid19 cases.</p><p>For example, the United States has ~<strong>2.9 beds per 1000 people</strong> (this is roughly the same for the United Kingdom as well). In addition, Germany <a href="https://data.worldbank.org/indicator/SH.MED.BEDS.ZS?view=chart">boasts </a>of <strong>~8.3 beds per 1000 people</strong>.</p><p>Why is this important? Even with superior numbers, we’re hearing reports of hospitals running short of beds in many places in the US, especially in New York, calling in hospital ships to supplement resources.</p><p>Even in India, the number of beds/person is not evenly distributed — the southern states of TN, Karnataka and Kerala do have above-normal capacity.</p><p>Let’s look at another stat: <strong>Beds/Population of 65+</strong></p><p>This is important since the population that is <strong>at the most risk to this virus are the elderly</strong>.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/826/1*64Izeo7lQS-iUcXTD23ACA.png" /></figure><p>Here we see a few states have slightly better capacity to tide over the crisis — like West Bengal, TN, Karnataka, Delhi etc.</p><p>Where are we going with this?</p><p>The virus has an exponential spread —<strong> the number of cases have been doubling every 4–5 days in India for the last few days</strong>. Assuming that hospitalizations also increase similarly (exponentially), we could be seeing a breaking point very soon in many states. Keep in mind that I’m only counting the “official” cases for now — the situation could be much worse if we counted all the cases.</p><p>The next graph estimates the breaking-point using the following:</p><p>a) <strong>Exponential </strong>spread of cases (similar to what we’re seeing right now)</p><p>b) Assuming that <strong>25% of the beds can be utilized</strong> for these patients — others would continue to be utilized for regular patients.</p><p>c) Mapping the current number of official cases with the beds available at a state level (not even at a city level — which could be much more accurate, and unfortunately, much more worse)</p><p>d) Beds would be unoccupied once a patient recovers after ~20 days</p><p>So, in how many days would our medical capacities hit the “breaking point”?</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/738/1*irYpfnQ1RRHCemc0rq_eDQ.png" /><figcaption>This graph signifies the number of days we would run out of beds in each of these states. The Y axis is the number of days.</figcaption></figure><p><strong>Summary</strong>: We are soon reaching a breaking-point in terms of the number of active Covid-19 cases that we can effectively treat primarily due to the shortage in medical infrastructure. <strong>States like Maharashtra, AP, TG and Delhi would be in trouble within the next 3 weeks, </strong>unless things improve dramatically.</p><p>This is why <strong>“flattening-the-curve”</strong> is important. Even a large scale expansion in capacity, say 30%, would only give us <strong>only 3–4 additional days</strong>, primarily due to the exponential growth.</p><p>So that’s that — stay at home, eliminate contact, practice strong social distancing and protect the elderly. Staying at home won’t just save your life — it’ll save a lot more. It’s the least we can do.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=742c37a2b72b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[The R0 Effect]]></title>
            <link>https://medium.com/@purnoorsodhi/the-r0-effect-6a556e554d94?source=rss-52fa8be999eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/6a556e554d94</guid>
            <category><![CDATA[coronavirus]]></category>
            <category><![CDATA[epidemic]]></category>
            <category><![CDATA[math]]></category>
            <category><![CDATA[pandemic]]></category>
            <category><![CDATA[covid19]]></category>
            <dc:creator><![CDATA[Purnoor Sodhi]]></dc:creator>
            <pubDate>Thu, 26 Mar 2020 18:02:58 GMT</pubDate>
            <atom:updated>2020-03-27T04:52:41.076Z</atom:updated>
            <content:encoded><![CDATA[<p>The Covid19 pandemic is a full fledged global threat today. At the time of writing, we now have more than 472,907 cases across the planet with 21,315 deaths. For now, India has 681 confirmed infections with 12 deaths. Although it seems that while India is on the cusp of an exponential breakout, the Prime Minister has announced a “21 day Lockdown” to take the virus head-on.</p><p>This is a great move, albeit a little too late. That’s because</p><p>a) India hasn’t tested enough — granted that only ~2.5% of all tests have come back positive, compared to a much higher number for other countries. But I would much rather feel safer with the knowledge that there are ~3100 Indians that have tested positive and have been isolated (with additional tests completed for say 100,000 people) than with the knowledge that there are only 681 infections and an unidentified number of carriers more.</p><p>b) 17.9% of people testing positive for the virus are asymptomatic (95% credible interval for this number of 15.5% to 20.2%). This could possibly mean that there are far more carriers of the infection that haven’t been close to a testing centre yet. With an R0 (this is <strong>important </strong>— this is the Reproductive factor of a typical infection) of 2.4, each asymptomatic person is now infecting an average of 2.4 more people.</p><p><em>(Source: https://www.forbes.com/sites/brucelee/2020/03/18/what-percentage-have-covid-19-coronavirus-but-do-not-know-it/#5a9acb5a7e90)</em></p><p>c) A lockdown for 3 weeks is frankly not enough. This will be apparent shortly.</p><h3><strong>The Epidemic Model</strong></h3><p>This brings us to the epidemic model. The model that’ll be the difference between life and death.</p><p>A typical infectious disease typically follows the following model:</p><p><strong>S</strong>usceptible → <strong>E</strong>xposed → <strong>I</strong>nfected → <strong>R</strong>emoved</p><p>The dynamics of this model are characterized by the following set of four ordinary differential equations that correspond to the stages of the disease’s progression</p><p><em>dS/dt</em>​=−(R<em>t</em>​​/<em>T</em>inf​)⋅<em>IS</em></p><p><em>dE/dt</em>​=​(R<em>t</em>​​/<em>T</em>inf)⋅<em>IS</em>−​(<em>E/T</em>inc)</p><p><em>dI/dt</em>​=(​<em>E/Tinc)</em>−(<em>I/Tinf)</em></p><p><em>dR/dt</em>​=(<em>I/Tinf)</em></p><p>In these equations:</p><p>Rt= R0 factor (the multiplicative factor for the infection) at a certain period of time</p><p>Tinc= Length of incubation period</p><p>Tinf =Duration the patient is infectious</p><p><em>(Source: </em><a href="https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)30260-9/fulltext"><em>https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(20)30260-9/fulltext</em></a><em>)</em></p><p>This now brings us to a couple of simulations of how this would progress in India.</p><h3><strong>Simulation 1: If we never had a lockdown</strong></h3><p>The Indian Government announced a massive lockdown starting 25th of March 2020 for a period of 21 days. Had this not been announced and people continued their lives, we would be facing a pandemic of epic proportions.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*36eSiCU5Lv-DwOcfsJWfxQ.png" /><figcaption>Scenario 1: No Lockdown at all (<a href="http://gabgoh.github.io/COVID/index.html">http://gabgoh.github.io/COVID/index.html</a>)</figcaption></figure><p><em>(Hat-tip to Gabriel Goh for the Epidemic Calculator — the graph above has been based on this calculator)</em></p><p><strong><em>Assumptions</em></strong><em>: a) A person who recovers once will not contract the virus again b) Other assumptions are the standard medical/clinical values plugged in for realism — eg morbidity ~2%, average recovery time ~11 days, Tinc ~5.2 days, Tinf ~2.9 etc</em></p><p><strong><em>Summary</em></strong>: This scenario predicts that within the Indian population of ~1.33B, the virus runs amok unchecked with an R0 of 2.4, putting an enormous stress on the hospital system, leading to more than <strong><em>22 million fatalities</em></strong> with a <strong><em>peak hospitalization of ~62 million</em></strong> on day 166. Another estimate — of the 62M hospitalizations, at its peak, an approximate <strong><em>15 million people would need ventilator support</em></strong> (20–25%). At present, India has only 30,000–40,000 ventilators available.</p><p>This is obviously a) Just unimaginable and untenable and b)Hypothetical, since we have already announced a lockdown.</p><p>That brings us to the next scenario.</p><h3><strong>Simulation 2: Lockdown for 21 days</strong></h3><p>This is where things get interesting.</p><p><strong>The single biggest factor in helping reduce the impact of this virus is the R0. </strong>The larger the R0, the faster the infection spreads. If R0&gt;1, the virus will continue to spread until everyone is infected at least once. If R0&lt;1, people would continue to infect a much smaller group, thereby leading to the end of the spread of the infection.</p><p><strong>The biggest purpose of this lockdown is reducing the R0 so that it reaches &lt;1</strong>. If the R0 does not fall &lt;1, the lockdown would have failed.</p><p>Here’s an example of some of the types of lockdowns:</p><p>Status 0 Green: Business as Usual</p><p>Status 1 Blue : Case Isolation, Medium Social Distancing, Closure of group activities &gt;50, Limited mass-transit stoppages, Personal sanitation</p><p>Status 2 Yellow: <strong><em>Blue </em></strong>+ Household quarantine, Closures of Schools/Universities, Work-From-Home, Restrictions on travel</p><p>Status 3 Orange: <strong><em>Yellow</em></strong>+Stoppage of deliveries, maids/cooks in societies, Strong Social Distancing, Groups activities &gt;5, Complete Travel lockdown</p><p>Status 4 Red: <strong><em>Orange</em></strong>+Complete isolation, Complete distancing (especially for people aged&gt;70), No Personal travel, Large-scale Govt intervention</p><p>It is quite evident that the R0 factor would be inversely proportional to the Emergency status — the higher the alert, the lower the reproductive factor R0. The R0 for an Orange lockdown would &lt;&lt; R0 for a Yellow lockdown. Additionally, there would exist multiple versions of an R0 even within a single Stage.</p><p>For example, if India with its current format of a lockdown were at a Status 4 Red, but given its dense population and general inability to implement a complete lockdown, we could assume the new R0 (lets call it Rt) to be close to 0.79 post implementation.</p><p>With this, here’s how the new infection curve would look like (assuming a complete <strong><em>indefinite </em></strong>lockdown)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*QkOy_2zMAuMfVgnCK6RTVA.png" /><figcaption>Scenario 2: An indefinite lockdown</figcaption></figure><p>In an attempt to simulate the conditions for India, this scenario assumes that we have implemented a lockdown after 7 deaths (much like what happened here)</p><p><strong><em>Summary</em></strong>: The results are quite surprising — By Day 218, the <strong><em>total number of fatalities would be close to 266</em></strong> and a total number of infected individuals would not cross ~13500. This also assumes that the <strong><em>peak number of hospitalizations would be close to 473, with a peak of ~118 requiring ventilator support</em></strong>. Quite manageable?</p><p>Again, this assumes that we have an <strong><em>indefinite </em></strong>lockdown — unlike what has been announced by PM Modi. Lets simulate what would happen if the lockdown was indeed lifted in 21 days.</p><p>In this simulation, after 21 days of a lockdown (on Day 75), the R0 changes back from 0.79 to say, 1.8 (possibly because people are still quite careful, washing their hands etc but just not good enough?). On this day, India still has 505 infectious citizens actively circulating the disease with a total of 59 fatalities.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*HNWsA1uVzXWkKVI3Tl23rQ.png" /><figcaption>Chart 3: Eventual rise in the number of infections after the 21 day lockdown ends, with an R0 of 1.8</figcaption></figure><p><strong><em>Summary</em></strong>: Here we are back again — the virus infecting ~75% of the entire population, leading to <strong>18m fatalities</strong>, <strong>peak hospitalization of 39.7m</strong> on Day 220 (54+166) and a peak of <strong>~9.9m ventilators</strong> required on the same day.</p><p>If you thought that just a 21 day lockdown would end Covid19’s run in India, I hope this chart helps you understand the situation a little better.</p><p>So what do we do?</p><h3><strong>Simulation 3: Indefinite and Complete Lockdown</strong></h3><p>There is no other way. The virus will have a resurgence, primarily since its hosts (at least most of them) show symptoms much later — thereby infecting many others in this duration. What we need to do is make that R0 as small as possible. Here’s what India could do to make this a short affair and reduce R0 immediately:</p><p>a) <strong>Test, test, test</strong>. No better way to identify infections, isolate them and treat them. This is one of the BEST ways to reduce R0.</p><p>b) <strong>A complete and a total lockdown</strong>. Extend the current one until the number of active infectious citizens drops to 0. This might take at least ~160 days (fewer days if the R0 is even lower than 0.79). So its in our best interests to make this lockdown a successful one (at least until a treatment is found)</p><p>c)<strong> Build/Acquire ventilators.</strong> This is again something that goes without saying. People with pre-existing unrelated conditions need ventilator support as well. Its impossible to turn away these patients.</p><p>d) <strong>Find Treatments</strong>. While a number of drugs have shown promise, even with expedited approvals, human trials to evaluate long-term effects would take at least 12–18 months. We need to buy time.</p><p>e) <strong>Stimulate the economy.</strong> Its great that business leaders are talking about supporting their employees during these tough times. But if this goes on for more than 2 months, we could expect many, many more people to lose their jobs. This is where the Government would need to step in and support payments, rent/other liabilities, direct cash transfers to stimulate demand.</p><p>f) <strong>Educate</strong>. Educate your neighbors, family, friends and other citizens on proper social-distancing and quarantining themselves — to help lower R0.</p><p>Lives are at stake. Time to act.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=6a556e554d94" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Covid-19: The second-order effects]]></title>
            <link>https://medium.com/@purnoorsodhi/covid-19-the-second-order-effects-5e1aa33f8b1c?source=rss-52fa8be999eb------2</link>
            <guid isPermaLink="false">https://medium.com/p/5e1aa33f8b1c</guid>
            <category><![CDATA[business]]></category>
            <category><![CDATA[coronavirus]]></category>
            <category><![CDATA[covid19]]></category>
            <category><![CDATA[economics]]></category>
            <dc:creator><![CDATA[Purnoor Sodhi]]></dc:creator>
            <pubDate>Thu, 12 Mar 2020 13:10:23 GMT</pubDate>
            <atom:updated>2020-03-12T13:10:23.426Z</atom:updated>
            <content:encoded><![CDATA[<p>The Coronavirus is here to stay. Hopefully, for a short time.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*_sjMsVwN1soZBn3tnCBsUQ.jpeg" /></figure><p>While some of the consequences of the virus are frankly quite obvious, its some of the unintended consequences (usually called second-order effects) that make for much interesting reading. <em>For example, the price of electricity shot up significantly post-Fukushima, which in turn changed consumer behaviour in Japan. The resultant deaths from freezing was in fact, much higher than that of the nuclear disaster.</em></p><p>This post is not meant to cause panic — its just however, in the realm of conjectures and extrapolations. Therefore, prepare to be disappointed if you don’t find facts, figures and data to support or refute your own hypothesis. And yes, this is by no means an exhaustive list.</p><ol><li><strong>General Employment Trends</strong></li></ol><p>While work-from-home has been implemented in a number of economies and encouraged by many organizations, we could see this form of employment a new way of life in previously un-imagined sectors. This could also mean that livelihoods of many <em>daily-wage/hourly-wage workers, gig employees</em> that work in sectors that involve physical participation could be affected. Bankruptcy is a real threat. Households with high-debt shares could contemplate serious steps.</p><p><strong>2. Domestic Consumption</strong></p><p>There’s no question that the economy will bounce back eventually. At the time of writing, investors have already seen an erosion of nearly 20% of their investments in the Indian capital markets. However, as stated earlier, workers in the non/less-formal sector that involve physical presence could be affected seriously — This could in turn affect <em>domestic consumption of products and services meant for the bottom-of-the-pyramid</em>. 2-wheeler sales, while previously showing some green-shoots of recovery, could be impacted for a longer period.</p><p>However, since the <em>virus affects seniors and predominantly men</em>, many companies would be under significant threat of their customer base being targeted. What brands do you think could be affected? (That’s what the comments section is for!)</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*pbHc0AkiKpILEZZlnLYvYQ.jpeg" /></figure><p>Industries like <em>retail, transportation, restaurants, bars/clubs </em>would be impacted by their inability to congregate — other industries like <em>research, design, outsourcing, software development</em> will obviously be less affected.</p><p><strong>3. Entertainment Sector</strong></p><p>Some win, some lose here.</p><p>While multiplexes, music event organizers and comedians will struggle to fill up seats for their events, <em>OTT brands like Netflix, Prime, Hotstar</em> <em>and DTH services will thrive</em> — especially with more people now ‘working-from-home’. Youtube viewership metrics could rise. All of this in turn, could drive up consumption of <em>home data and internet services</em>, possibly to a point of stretching their bandwidth capability to the limit. Media buying prices could rise as well, especially in the non-peak slots. (Of course, it could even out since brands would limit investments in such a period). In addition, v<em>ideo game sales (digital copies)</em> could also rise significantly, especially with kids now spending a lot more time at home and having skipped their annual exams. <em>Event organizing firms</em> will increasingly have to dip in to their war-chests and wait for the water to subside.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/275/1*CXB5sfWi_LoXXFRrqITWTw.jpeg" /></figure><p><strong>4. Education</strong></p><p>We’ve all heard of how Bjyu’s has transformed the education sector. With Byju and many similar <em>tech-based learning organizations announcing free courses </em>on their app, fence-sitters would also start exploring these services. This, in fact, could transform the entire education industry — especially with mass-quarantines becoming the rule around the planet, parents preferring their children staying home for a longer period and with patrons experiencing and enjoying their experience on learning apps. Expect more brands to enter this space as well.</p><p><strong>5. Human Psychology</strong></p><p>While we have already seen examples of<em> racist behaviors</em> in multiple countries, limited education levels and surfacing of underlying biases are expected to exacerbate in the near future as well. In addition, a positive outcome could be that the Indian government spends a lot more on programmes for <em>personal cleanliness, sanitation and food safety</em>. For example, we could expect better standards of cleanliness in Indian road-side markets, travellers continuing to wear masks at airports, railway stations and bus-stops etc.</p><p><strong>6. Pollution</strong></p><p>Again, we’ve seen satellite images of reduced pollution metrics in the manufacturing hubs of Central-Eastern China. This could happen in India as well, with <em>high-population density industrial zones reducing output</em> in favour of reduced work-days, fewer shifts and declining demand. This could in turn lead to <em>better air quality</em> in major cities, <em>increased fish and wild animal </em>population and unfortunately, <em>uncontrollably high livestock population</em>. Owners could be forced to cull them in order to avoid high costs of maintenance.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/632/1*TZmmhlDOGkqKn-PS8SmohQ.png" /></figure><p><strong>7. Insurance</strong></p><p>While there will continue to be demand for better health insurance, insurance companies will be forced by governments to <em>include provisions for Covid-19</em> in proving cover. On the flip side, since most people would avoid travelling and generally stay at home, there would be <em>lesser traffic density and fewer accidents</em> — thereby reducing potential liabilities.</p><p><strong>8. Diversification of Supply Chains</strong></p><p>With China virtually at a stand-still, companies and now increasingly governments are realizing the importance of <em>diversifying their sourcing strategies, especially for critical resources and products</em>. Local production could be given higher incentives, especially in a post-Corona world (with governments expected to dole out multiple stimuli packages), thereby lending credence to the ‘<em>nationalistic’ movement,</em> espoused by world leaders like Trump, Modi etc.</p><p>That’s that, then. There would be scores of other consequences, some completely un-imagined previously. Go berserk in the comments section.</p><p>Till then, here’s wishing all of you lots of good health!</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=5e1aa33f8b1c" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>