The COVID-19 (the disease that the coronavirus causes) outbreak in India is a disaster in the making akin to several disasters that have occurred in the past, e.g., 1984 Bhopal gas tragedy, 1999 Ersama cyclone. They start as a health crisis and quickly morph into a humanitarian and an economic crisis.
To understand the COVID-19 pandemic, the following questions need answers:
- What will COVID-19 spread in India?
- How is India coping with COVID-19? Is India prepared for a COVID-19 outbreak?
- Is the COVID-19 pandemic a black swan or a gray rhino event? How can their impacts be mitigated?
- Why do such events recur? What important policy shifts are required?
PART 1: COVID-19’s POSSIBLE SPREAD IN INDIA
The coronavirus is like any other hazard. It takes a certain pathway and causes receptor injury. The receptor for the virus is invariably humans. For some hazards, e.g., a tornado, the receptor could also be the environment to which it causes damage. Risk management is the understanding of the interplay between a hazard, its pathway and the receptors it may strike and the impact it causes to them, in order to reduce risk the hazard poses to the receptors (see Box 1 for an explanation of the dynamics between hazards, pathways and receptors).
The coronavirus moves inside an infected person and moves with him. It is transmitted largely through his sneeze or cough droplets. Most hazardous situations usually have a single point for the hazard location, e.g., a volcano, with the receptors located in a single area, e.g., a 10 km radius around the volcano. Since an infected person is mobile and transmits the infection to people wherever they go, viral spread is a multi-point hazard and receptor situation with an almost infinite number of pathways between the hazard and receptors—a nightmare for risk controllers.
As people from Wuhan travelled, they carried the virus to other Chinese provinces and countries. Places better connected to Wuhan were affected earlier, e.g., Europe, then spread the disease to other onward connected places. There is a visible correlation between COVID-19 incidence and population mobility (and carbon emissions). In India, states with greater population mobility–Maharashtra, Delhi and Tamil Nadu, have a significantly large caseload compared to the Northeastern states. In the US, coastal states with more mobile populations and are better connected to the world, have higher caseloads.
Once a country is seeded with ‘imported cases’ in Stage 1 of the disease, it progresses in a short time in three more stages. In Stage 2, ‘local transmission,’ an imported case passes the virus to a local contact. In Stage 3, ‘community transmission,’ the infection spreads in a community and can no longer be traced to an imported case or their contacts. Stage 4 is an epidemic.
On 14 April 2020, the number of detected cases worldwide was 2 million with a doubling time of 13 days, and deaths attributed to coronavirus were 1.2 lakhs. Detected cases and deaths are in exponential growth with no sign of slowing down.
|Date||Total confirmed cases||Daily new confirmed cases||Total confirmed deaths||Daily new confirmed deaths|
|1 January 2020||27||0||0||0|
|1 February 2020||11,946||2,120||259||46|
|1 March 2020||67,024||1,821||2,979||58|
|1 April 2020||851,308||73,512||41,885||4,614|
|14 April 2020||2,082,418||73,966||126,601||6,983|
Bean counting: How many Indians will catch the coronavirus in the near future?
Just before India locked down on 25 March, two modelling studies were published that forecast the incidence of COVID-19 cases in India in the near future with and without non-pharmaceutical interventions (NPI), e.g., social distancing, for optimistic and pessimistic scenarios (see Box 2 for details). All studies state that their results have considerable uncertainty, which is inherent to risk studies where little is known about the virus.
The results vary widely—with no NPI, COVID-19 incidence range from a low of 2 million to a high of 800 million. And with NPI the variation in case incidence is 650-242,000. Such a wide difference in their predictions is due to the variation in the equations used, input data used, and the assumptions made. Some of these studies have already gone wrong in their predictions of incidence with NPI as the number of cases in India is now over 10,000 reported cases and rising.
One of these studies that did city-specific predictions for Delhi, Mumbai, Kolkata and Bengaluru may yet come closer to what may happen in the coming weeks. The prediction for Delhi, the city with the maximum likely cases amongst these 4 cities, is 14.5-110 lakhs without NPI, and 2-97 lakhs with NPI. At the other end, in Bengaluru which is likely to have the least number of cases amongst these cities, the predicted number of cases are 3.5-23 lakhs without NPI and 0.7-22 lakhs with NPI. All the studies agree that with no NPI, the case numbers will be very high.
Even though the Imperial College study, published on 16 March, made predictions for Great Britain (GB) and USA, it may have influenced the Indian government’s thinking. The study predicted that with no NPI, 81% of the GB and US populations would be infected, and GB and USA would have 0.51 and 2.2 million deaths, respectively. Their public health care systems would be overwhelmed and not be able to cope with this case load, hence a strategy shift was required. GB shifted gears drastically and jumped from no restrictions to national lockdown a week after the Imperial College study was published. Two days later India followed GB.
To understand why GB and India did such a U-turn at short notice, it is important to understand how risk mitigation is done.
There are six ways of removing or reducing the risk from natural or manmade hazard: 1) remove the hazard, 2) remove humans and other receptors, 3) enclose the hazard, 4) enclose humans or other receptors from its pathway, 5) increase the pathway length between the hazard and the receptors so that the intensity of the hazard decreases when it hits the receptor, 6) break the pathway (see Box 3: Risk mitigation).
Not all six measures are available for every hazard, e.g., a cyclone cannot be removed or enclosed nor can humans be removed from its pathway. But humans can be enclosed in a shelter or evacuated before the cyclone makes landfall. The only method not available for a coronavirus hazard is removing humans and surfaces, e.g., chairs, as both are found everywhere (method #2 in Box 3). All four other methods work only for a limited time period and in a limited space.
Flattening the curve
In a virus outbreak, the disease tends to spread quickly in a population after it is seeded by an infected person. A rapid rise of cases in the ascendant phase of the disease can often overwhelm even a good health care system, leaving many patients without adequate attention. ICU beds and ventilators ran out in Italian hospitals in March 2020, and doctors had to make choices about whom to deny a ventilator. To avoid overstretching the health care system, disease control strategies attempt to flatten the curve by beating down the peak number of cases and spreading them over a longer time period.
Mitigation and suppression strategies
The curve can be flattened using either mitigation or a suppression strategy, or sometimes a combination of both where a mitigation strategy is used countrywide and a suppression strategy is used in local hotspots. A mitigation strategy uses NPI not to interrupt transmission completely, but to slow it down, reduce health impact and provide health care for vulnerable populations. It does this by home-isolating mild cases, quarantining suspect cases, their families and contacts, hospitalizing severe cases, using social distancing, and possibly shutting educational institutions, and closing borders with some or all countries. The specificity of a circumstance determines the combination of mitigation measures deployed.
A suppression strategy breaks the virus’ pathway and reversing its spread by locking down communities, towns or even a country, and ordering everyone, barring essential service workers, to stay home. This measure is in addition to all the measures used in a mitigative strategy.
Mitigation and suppression strategies and their limitations
|Hazard control method||Actions||Strategy||Limitations|
|Remove hazard||Washing hands, sanitizing surfaces||Mitigation, Suppression||Limited in time & space. Person & surfaces can be re-contaminated, Other persons & surfaces not sanitized may carry infection|
|Remove humans||Not possible||—||—|
|Enclose hazard||Isolate cases, quarantine suspected cases & persons they contacted||Mitigation, Suppression||Limited in time & space. Other persons outside isolation/ quarantine may carry the infection|
|Enclose humans||Personal protective equipment||Mitigation, Suppression||Can be used only by health workers|
|Increase pathway length||Keep social distance of 1-2 m between people, no physical contact||Mitigation, Suppression||Everyone may not follow social distancing or may not be able to do it they live & travel in crowded houses and transport|
|Break pathway||Lockdown community/ cities/ districts/ country– Everyone, except essential services workers are isolated at home||Suppression||Social life and the economy will be impacted as supply chains of products & services are broken.|
A mitigation strategy allows for the disease to spread in a controlled way in the population thereby allowing the population to gain herd immunity, thus reducing the probability of a repeat outbreak. If the disease transmission is moderately high, and the number of cases exceeds the maximum capacity of health services, it may be badly stretched or even breakdown.
By isolating everyone, a suppression strategy reduces disease transmission, and cases to low levels, allowing the health services to cope. Without a vaccine or a first-time infection no herd immunity is acquired by the population. There is then the possibility of repeated disease outbreaks which will require suppression to be done consecutively till a vaccine becomes available or the virus weakens in time, The Imperial College study illustrates this process for GB and USA (see figures and table below).
|Action||ICU beds required per 1 lakh population|
|Mitigation strategy||Suppression strategy|
|Close schools & universities||245|
|Case isolation + household quarantine||130|
|Case isolation + household quarantine + social distancing of >70 yr olds||90|
|Case isolation + household quarantine + general social distancing||16|
|School & university closure + case isolation + general social distancing||6|
GB has a surge capacity of 0.08 ICU beds for 1,000 population. But the requirement for ICU beds for a combination of mitigation actions for the COVID-19 exceed GB’s ICU bed surge capacity by a factor of 10-35 times. However, GB’s ICU beds surge capacity met the projected ICU bed demand if a suppression strategy is used, the downside being that these measures must be in place for 5 months.
Box 1: Risk management
Risk management is the understanding of the interplay between a hazard, its pathway and the receptors it may strike and the impact it causes to them, in order to reduce risk the hazard poses to the receptors.
A hazard is a substance—toxic chemical, virus; or energy—ionizing radiation, earth’s, e.g., as is released in an earthquake, that may cause injury to humans, or their life support systems prevalent in the environment. Hazards may be natural, e.g., tornadoes or man-made, e.g., explosions. Hazardous may strike suddenly and without much warning, e.g., earthquakes, giving no time for doing hazard control, or may act over time to cause a slow deterioration of the life support systems in the environment, e.g., global warming,
A hazard travels through a pathway before it strikes a receptor. Pathways may be of through a single medium, e.g., as happens in an explosion when an energy release is transmitted through air, or through several media, e.g., as happens in the release of mercury from thermal power plants, where the mercury travels first through the air from the power plant stack to grass, which is eaten by cattle, where it goes into cattle milk, and from there into humans.
A receptor is an object that the hazard hits after travelling through the pathway. A receptor may be a human or any part of the environment. In the case of the Coronavirus, the receptor is a human.
Most hazards are point sources, i.e., they emanate at one point, e.g., a volcano or a flammable gas leak. And most receptors, whether human or environmental, also point or area receptors, i.e., they are in one area. Coronavirus is a multi-point hazard as it can be present at innumerable places and its receptors are also multi-point as people and surfaces that may be contaminated are spread all over the world.
Box 2: Expected number of COVID-19 cases in India—results of 2 studies published before the 25 March lockdown and 3 studies published after the lockdown
Studies published before the lockdown
|Study title||Predictions and role of interventions for COVID-19 outbreak in India||COVID-19 for India Updates|
|Authors||D Ray, R Bhattacharyya, L Wang, M Salvatore, S Mohammed, A Halder, Y Zhou, P Song, S Purkayastha, D Bose, M Banerjee, V Baladandayuthapani, P Ghosh, B Mukherjee||E Klein, G Lin, K Tseng, E Schueller, G Kapoor, R Laxminarayan|
|Published on||22 March 2020||24 Mar 2020|
|Predicted cases||Interventions||No of expected cases & dates||Scenarios||No of expected cases & dates|
|No interventions||2.2 million cases by mid-May||High trajectory-no effect of current lockdowns and a rapid spread||Total peak number of cases-25 crores by end-Apr|
|Travel ban only||6.6 lakh cases by mid-May||Medium trajectory-No effect of lockdown or temperature/ humidity sensitivity, consistent with data from Italy (more likely scenario)||Total peak number of cases-18 crores by mid-May|
|Travel ban + Social quarantine||55,200 cases by mid-May||Low trajectory-Decreased transmission, potentially due to temperature/ humidity sensitivity||Total peak number of cases-12.5 crores by mid-June|
|Travel ban + Social quarantine + Lockdown||13,800 cases by mid-May|
Studies published after the lockdown
|Study title||Age-structured impact of social distancing on the COVID-19 epidemic in India||Healthcare impact of COVID-19 epidemic in India: A stochastic mathematical model|
|Authors||R Singh, R Adhikari||K ChatterjeeK Chatterjee, A Kumar, S Shankar|
|Published on||26 March 2020||2 Apr 2020|
|Predicted cases||Interventions||No of expected cases & dates||Interventions||No of expected cases & dates|
|Without mitigation||Total of 0.9 bill cases, with a peak of 167 mill cases by end-June. Total mortality of ~3.6 mill persons.||Uninterrupted epidemic in India||Resulted in over 364 million cases & 1.56 million deaths, peak by mid-July|
|3 consecutive lockdowns of 21, 28 & 18 days starting 25 Mar, with two 5-day unlocked intervals between lockdowns; or a single 49-day lockdown starting 25 March||657 cases||Immediate institution of NPIs||Epidemic might still be checked by mid-April 2020. It would then result in 241,974 total infections, 10,214 hospitalizations, 2,121 ICU admissions and 1,081 deaths.|
Expected number of COVID-19 cases in 4 major Indian metro hubs with and without intervention
|Study title||Prudent public health intervention strategies to control the coronavirus disease 2019 transmission in India: A mathematical model-based approach|
|Authors||S Mandal, T Bhatnagar, N Arinaminpathy, A Agarwal, A Chowdhury, M Murhekar, R R Gangakhedkar, S Sarkar|
|Published on||28 March 2020|
|Predicted peak number of cases||Intervention||Scenarios||Predicted peak number of cases in lakhs (approx)|
|Without intervention||Pessimistic||110 lakhs in 45 days||47.5 lakhs in 50 days||38 lakhs in 55 days||23 lakhs in 50 days|
|Optimistic||14.5 lakhs in 210 days||7 lakhs in 300 days||5 lakhs in 300 days||3.5 lakhs in 300 days|
|With intervention||Pessimistic||97 lakhs in 45 days||45 lakhs in 55 days||33 lakhs in 55 days||22 lakhs in 55 days|
|Optimistic||2 lakhs in 620 days||1 lakh in 725 days||0.7 lakhs in 725 days||0.7 lakhs in 725 days|
Not all 6 methods are available for all hazards, e.g., a cyclone cannot be enclosed nor can humans be removed from the pathway of viruses. Five actions are available to reduce the risk of the Coronavirus:
- Remove hazard: The virus can be removed by washing hands or sanitizing surfaces. These measures work only briefly as re-contamination is possible, besides it is impossible to sanitize all surfaces.
- Remove humans: This action is not possible as humans are very widely spread.
- Enclose hazard: Infected persons are put into isolation wards and those suspected to be infected are quarantined so that the infection is enclosed in the containment facility. But persons outside the isolation ward or quarantine may be infected.
- Enclose humans: Personal protective equipment (PPE) may be used to enclose health workers, but it is impossible to have everyone use PPEs.
- Increase pathway distance: Social distancing by 1-2 m increases the pathway between humans and minimizes the chances of the virus in an infected person’s cough or sneeze droplets reaching the other person.
- Break pathway: Closing educational institutions is a way to break the pathway. But the break is partial as there are other ways that people meet, e.g., at work, in the market.
- Lockdowns are a more complete way of breaking the pathway as they minimize contact between infected persons and surfaces and uninfected persons by isolating everyone except those involved with essential services. Lockdowns include shutting educational institutions, factories, commercial establishments, markets, public transport services, etc.
The author is an environmental engineer with specialization in risk analysis