Lessons learned from 108 doctor deaths in India:

Dr Rajeev Jayadevan, 16 July

(Link to original paper and other references are embedded/hyperlinked in the text below)

1. No age group appears to be immune to the pandemic. Unlike western countries where most of the deaths are among older age groups, published mortality data from Maharashtra and UP show that over 60% of deaths are below the age of 60.
The findings from this study of doctors correlates closely with that age trend, with 55.5% of the deaths among doctors being under the age of 60. More deaths occurred in the states with greater numbers of Covid-19 patients. The study was titled: A Hundred lives lost: Doctor Deaths in India during the times of COVID-19.
The study can be accessed at the following link:
In this analysis of 108 deaths among doctors from India (all systems of medicine including AYUSH, updated to 13 July), age was reported in 85 cases. From these, the average age at death was 56.3, excluding the four violent deaths. (If three accidents (age 53, 25 and 23) and a suicide (age 22) were included in the calculation, it is revised to 55)
Excluding the four violent deaths among doctors,
Percentage of deaths below 60 was 45/81 = 55.5%
Percentage of deaths below 50 was 24/81 = 29.6% Percentage of deaths below 40 was 17/81 = 21%
The average age among ten nurses was 49.6 years. [If the three violent deaths among nurses (two accidents and one suicide, all of whom were 22 years of age), were also included, the overall average age of death dropped to 42.7 years]

2. Although the individual mortality risk is lower, being young and apparently healthy is not a guarantee against COVID-19. Younger people are advised not to disregard standard precautions. Sudden deaths in the setting of COVID-19 requires further study. The exact reason why certain young people get severe disease is not known.

3. What causes doctor deaths?
Doctor deaths often get superficially blamed solely on PPE use by those who have not critically analysed the problem. To illustrate this concept with an example, it is similar to blaming ‘bad roads’ for all the deaths by road accidents in India.


Jayadevan, R. A Hundred Lives Lost: Doctor Deaths in India During the Times of COVID-19.


Preprints 2020, 2020070346 (doi: 10.20944/preprints202007.0346.v1).


Using the example of road deaths, if a root cause analysis is done, there are numerous other factors that cause accidents including bad behaviour such as rash driving, breaking of traffic rules, learning bad driving habits from peers, alcoholism, over-speeding, bad time management leading to chronic hurry, use of mobile phones while crossing the road or while driving, distracted driving, careless pedestrians, driving while sleepy or tired, needless use of the road, careless driving at night or during rains, refusing to wear helmets, overloaded two wheelers, unsafe vehicles without brakes or lights, animals on the road, inadequate enforcement, insufficient licensing criteria and unlicensed teen drivers, to name a few.

Likewise, if a root cause analysis is done, it can be seen that several of the doctor deaths in any country are the final outcome of defective COVID-specific policies and decisions at three levels:

a. Individual (doctor) level: e.g. inadequate awareness about the facts, denial, trivialising the pandemic, ignoring safety and social distancing guidelines, failure to use PPE according the individual work situation, taking unproven medications and feeling falsely secure, ignoring severe comorbidity, failing to be cautious outside workplace (doctors can get infected from outside the workplace too).

b. Hospital or clinic level: COVID-specific administrative and engineering policies on e.g. infection control committee, triage, crowd control, telemedicine, audits, PPE availability and quality, availability and access to testing, reengineering, barrier- building, improving ventilation, attention to airflow to limit aerosol spread, OT policy, HR and staff protection policy including stress management.

c. Government level: COVID policies on e.g. public awareness, transparency, availability and access to testing, surveillance, personnel (HR) policy including quarantine, law enforcement, prevention of crowds and group gatherings, infrastructure, ambulance, science-based preventive and treatment protocols.

The pyramid shows the relative effectiveness of infection control strategies. PPE is important, but represents only a small fraction of the measures needed to prevent infections and deaths. 2

Focusing on PPE alone might not reduce deaths significantly. All the other aspects need attention.


4. Operating on a patient who has COVID -19 involves risk of infection for the surgeon, anaesthetists and assistants. This is due to prolonged exposure time to aerosols in a closed space, and has been implicated in some of the deaths among surgeons in India. Comprehensive guidelines have been published on this topic in the current issue of The Indian Journal of Surgery by the Association of Surgeons of India (ASI) (click this link for the article).
The virus-laden aerosols need not necessarily come from the patient, but could be produced by an infected colleague who has no symptoms. This can happen during talking or breathing. Hence, COVID-specific protocols must be strictly followed in the operating theatre, ER and ICU’s. Limiting staff number in high risk exposure areas is important.

5. Working in crowded conditions appears to be a common risk factor. Crowds are almost inevitable in a populous country, but administrative and engineering controls can make a difference to healthcare workers. Token or appointment system, telemedicine for non-essential visits, staggered timings for patients by extending hours and adding shifts of doctors, enforcement of physical distancing and hand hygiene by non-medical personnel, limiting the number of bystanders, adding exhaust fans and windows to consulting rooms that are sometimes cramped and poorly ventilated, construction of makeshift waiting areas in open-air settings, and triage system where symptomatic patients are seen in a separate section are some workable solutions.
Nurses, pharmacists, technicians, physiotherapists, ambulance drivers, housekeeping and security personnel, billing and reception staff are exposed to high concentration of aerosols from people whose virus status is unknown. Identifying areas that are prone to crowding and installing administrative and engineering solutions to reduce crowds are important. Plexiglass barriers are helpful at the reception and billing section.

6. Doctors must assume leadership roles to protect the personnel who are working in their establishments. Periodic reminders or virtual meetings with all the staff will be helpful, preferably in regional language.

7. Building awareness alone is not enough. It takes substantial effort to convert awareness into meaningful action, and to avoid some of the common mistakes:

a. To avoid touching the face and mask with fingertips. Achieving this is harder than most people think. Urge to touch one’s face is a subconscious behaviour (not voluntary) and occurs at least 23 times an hour.

b. To keep finger-tips clean at all times. They frequently come into contact with surfaces and therefore are the most likely contaminated exposed body part.

c. To avoid prolonged meetings, especially in closed rooms.

d. Being aware that the virus can spread through microscopic droplets generated
through conversation. The longer the conversation, the greater the droplet load. 3


e. Being aware of the direction of airflow in a room while interacting with others, and keeping away from it. Suspended microdroplets travel in the direction of airflow and get inhaled, leading to spread of infection.

f. Being aware that the fit of the mask is more important than the type; air leak along the sides of the mask will defeat the purpose of wearing an N95 mask.

g. Taking extra care not to get infected while removing PPE after a long day’s work.

h. Being aware that a well-looking colleague or friend might be carrying the virus.

i. To not speak or allow others to speak with the mask down.

j. Keeping car windows down for better ventilation, while traveling with others.

8. Invariably there will be a gap between what is the protocol and what is actually practised on the ground. The greater that gap, the riskier it gets for everyone. Audit will keep that gap to the minimum. An audit is essentially a friendly survey done by any one of the staff members using a checklist, to observe the working pattern of the other staff members. Conducting an audit will not only enhance compliance due to embarrassment of being singled out, but also prompt people to become more aware and responsible.

9. Violent deaths might have stress or mental illness as the root cause, which are preventable and treatable. There were 8 COVID-related violent deaths among 138 healthcare worker deaths in India so far, with an average age of 27.8 years. These included two suicides and six road accidents. Suicides in the setting of COVID-19 are increasingly reported, especially among young healthcare professionals. If anyone is showing signs of stress, it is important to intervene early through counselling or other established means.
Mental fatigue and physical exhaustion contribute to road accidents. Three doctors, two nurses and one ambulance driver in this series died in accidents.

10. Lack of hospital beds and ICU beds affected the care of several doctors who were critically ill. In several instances they had to go to multiple hospitals to receive a bed.

11. Healthcare workers acquire the infection not only from patients and colleagues, but from outside the workplace too. Hence universal social distancing must be followed, even after work.

12. Doctors have succumbed to the disease after caring for family members with the infection. Unfortunately, there are at least two reported instances on this list where both the doctor and his wife died from COVID-19. In both these instances, the couple were under the age of 60.

13. Universal precautions will help reduce the risk, even if it is done at home. Instances of multiple family members becoming infected are documented in this list. The chance of a person with the virus infecting another person is called secondary attack rate. Published secondary attack rate of the SARS-Cov-2 virus among close household contacts ranges from 10 to 80%. These numbers indicate that it is


possible to care from someone with the disease without getting infected in the process. However, the risk is worse when precautions are not followed.

14. Studies have shown that the infection rate and mortality decreased as soon as infection control protocols were in place, and awareness of the disease improved. Many of the early deaths among doctors and nurses across the world were from lack of awareness of the virus being present in their community. In the absence of widespread testing, it is impossible to precisely determine the individual risk in each area, hence universal precautions are a must. It is safer to assume that everyone else has the virus, and take appropriate measures during work.

15. A number of people carry the virus without symptoms, and infect others. Sentinel testing of staff is an essential tool in assessing local risk. Sentinel testing is systematic testing of a random number of staff who are working in high traffic areas, in the absence of symptoms. Once found positive, they can be safely isolated. This ensures safety of colleagues as well as patients. Without sentinel testing, it is impossible to estimate the amount of virus penetration in any community.

16. There are reports of false negative initial testing all over the world, even when COVID-19 is clinically suspected. This can lead to a missed early diagnosis, and a false sense of security. Where clinically indicated, repeated testing must be done. Discussing the types of testing is beyond the scope of this article.

17. There are unconfirmed verbal reports of surgeons dying after being infected by their colleagues who were present during surgery, even when the patient was COVID- negative. If all operation theatres and personnel followed a standard COVID protocol, this is avoidable.

18. Even if the initial symptoms are mild, there needs to be careful monitoring for any worsening. Hypoxia need not be always experienced by the patient as dyspnoea. It might only be detected using a pulse oximeter. As in the case of Dr Bahulekar from Mumbai, he had no symptoms for four days and relaxed at home. On day 5, he quickly became short of breath and deteriorated, by which time a CT scan showed severe bilateral pneumonia.

19. While treating patients or colleagues with severe COVID-19, it is important not to suffer from fatigue, visions of futility or pessimism. With diligent care, numerous individuals have escaped otherwise certain death.

20. As it is a new disease with emerging scientific evidence, the treatment varies considerably from other established illnesses. For instance, prone ventilation and happy hypoxia are concepts that are new. The criteria for ventilatory support are different. It is important to stay updated with the latest guidelines and follow them, rather than insist on treating based on past experience alone. Empirical preventive or treatment measures based on random individual opinions could be dangerous.


Transparency in reporting and sharing of clinical experience by publishing regularly are important, so that colleagues and patients receive the benefit of locally relevant data and experience.

21. Doctors who have worked extensively in COVID care programs in India feel that healthcare workers should focus more on their physical and mental well-being during the pandemic, and worry less on matters such as weekly OPD patient numbers or monthly oncome. Psychiatrists say that while the season is optimal for planning for the future, it is not the right time to make important decisions.

22. The large number of young deaths among doctors (55.5 % doctor deaths are below 60 years of age) and the general population (60% deaths are below 60 years age) is a reflection of the relatively young workforce in India. Although greater numbers of younger people are succumbing to the illness, the absolute mortality risk per head is higher in older age groups and among those with comorbidities like heart disease, uncontrolled diabetes, severe obesity and COPD. (See graphs on page 7,8)
Those with high risk medical conditions or advanced age must discuss it with local authorities and modify their style of work accordingly, until the community infection risk level is lowered.

23. Even though the number of reported deaths (numerator) among doctors is substantial, it must be remembered that the denominator in a country like India is also large. In other words, there is a large number of healthcare workers who were able to discharge their duties without suffering the disease or dying from it. Therefore, the overall message here is not to be pessimistic, but to be armed with useful life-saving information in the days ahead.

24. Denial or trivialising of the pandemic is a surprisingly common problem even among healthcare professionals. Individuals with such an attitude are not only a danger to themselves, but to colleagues, family and society at large because they are at risk of flouting guidelines. A doctor has recently been suspended from practising in the UK because of his attempts to mislead society into believing that the pandemic was a hoax, and that social distancing was unnecessary. Deaths on this list have occurred after the person reportedly refused to believe that the infection was a problem.
The fact that over a hundred colleagues have succumbed in India within five months would serve as a grim reminder that the pandemic is real.

25. Healthcare workers are prone to attacks by the general public. Violence and abusive behaviour are reported all over the world. This requires customised solutions.




Mortality among older age groups across the world is at least 4 times the overall population mortality or Case Fatality Rate (CFR). This graph shows age-specific mortality rates in India.
Source; Times of India, report by Pushpa Narayan and Rema Nagarajan

Note: CFR involves a numerator (number of deaths) and denominator (number of people falling ill as a result of virus infection), both of which can vary depending on multiple factors including locally implemented testing strategy, accuracy in reporting of deaths, lag-time error (the deaths that happen today are mostly the outcome of patients who became ill about two weeks ago), quality of available healthcare, mean age and health status of the population under study, and quality of data collection and entry. For instance, including asymptomatic contacts who test positive in the denominator will artificially lower the CFR. Therefore, CFR should not be used to compare between states or countries, and is best utilised to monitor the progress in the same region over a period of time.



Age-specific mortality data from India, published by Partha Mukhopadyay, The Hindustan Times, June 11

Dr Rajeev Jayadevan, Cochin




Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: