How Kerala’s Death Audit Committee classifies deaths under COVID-19

The state has been criticised for under-reporting deaths due to COVID-19.
COVID-19 death representative image
COVID-19 death representative image
Written by:

For many months now, Kerala has been criticised for its count of COVID-19 deaths, several doctors and public health experts alleging under-reporting. When the questions rose, the State Level Death Audit Committee has since August 2020 begun publishing monthly reports of its audit of COVID-19 related deaths. It includes details of all the COVID-19 and communicable diseases audited during a month, definitions, methods, reasons for classifying a death as non-COVID-19 related.

The number of COVID-19 deaths reported by the Health Department on Saturday evening is 198, taking the total official toll to 8,455. Kerala which had reported a maximum of 35 COVID-19 deaths on a single day in all of last year, began recording more deaths as cases suddenly surged into tens of thousands every day. The peak reached in May, when the per-day cases crossed 40,000 and for the first time, the state recorded more than a hundred deaths on a single day. Doctors told TNM that the increase in deaths is an expected phenomenon since the cases had multiplied so much.  Dr Aravind Reghukumar, member of the Death Audit Committee (DAC), in this interview to The News Minute, talks about the increased mortality rate, the criticism on under-reporting and other concerns.

“Even one death is worrisome but when the surge happened, the number of patients in B and C categories (more serious conditions) increased and death rate, as expected increased. So when denominator (number of cases) increases, numerator (number of deaths) also increases,” Dr Aravind says.

Case fatality rate calculated as number of deaths by number of cases is now 0.33%.

“Over the last two months we were dealing with a more infective virus, the B.1.1.7, which was first detected in the United Kingdom. But now this strain has been relegated to the second place by the double mutant – B.1.617. The transmissibility of the UK variant itself was 40 to 90% more than the strain which was prevalent in our state or country. The newer variant is 50% more transmissible than the UK strain. Therefore you see more number of infections, family clustering and ultimately more deaths. In Kerala, stats indicate that three to five of 100 patients develop moderate to severe illness, of which 1.5 reach Intensive Care Units (ICU),” Dr Aravind adds.

Another reason for the increase in deaths is also an indirect result of the surge in cases. Hospital beds get full, patients stay at home and come to hospitals very late which can cause death. “It needn’t therefore be due to the virulence of the new variants. But this is what happened in the rest of the world, not just in Kerala," he adds.

Deaths of those with co-morbidities left out?

One of the main criticisms pointed at the state’s reporting of COVID-19 deaths was that infected persons with comorbidities who die were not counted. The reason was instead attributed to the particular comorbidity such as cancer. Another criticism was that a patient who has tested negative by the time of death is not counted among COVID-19 deaths.

However, Dr Aravind says that Kerala follows the guidelines of the World Health Organisation (WHO) and the Indian Council of Medical Research (ICMR). "The ICMR guidelines are based on the WHO’s. These guidelines say that deaths of people with COVID-19, irrespective of comorbidities, shall be declared as COVID-19 deaths, if the treating doctor thinks that COVID-19 has contributed to the death. Suppose a patient has advanced malignancy and he is developing COVID-19, which leads to pneumonia. His death may be due to pneumonia or respiratory failure or such. It will be counted as a COVID-19 death despite the advanced malignancy," he says.

“Before COVID-19, the patient might have lived for, say a month longer. But he died a month earlier because of COVID-19. So that should be counted. That’s the WHO’s definition. If you think there is a contribution to death by the virus, whatever be the comorbidities, it has to be declared a COVID-19 death. His test negativity has nothing to do with it,” Dr Aravind says.

“The graphs in our audits show that 15 to 20% of the COVID-19 deaths are of patients with chronic kidney diseases. Basic issue is however that the audit team has to do a catching up job. The reports up to March are now out. We cannot do daily auditing. We are a scientific community that collects data from districts and then do the auditing once a week. During the peak, we had two sessions a week. But we can reach a decision only once we have the complete data,” Dr Aravind adds.

Tricky situations

However, the doctor adds that certain situations can be very difficult or tricky. “Suppose a patient is on dialysis. He can become breathless due to many reasons such as fluid overload or COVID-19 pneumonia. It may not be possible to take a CT scan then and the treating doctor may not be able to determine the cause of death. Usually, such cases are declared as COVID-19 deaths. The practice may vary from district to district and situation to situation. That’s why you have a Death Audit Committee in place.”

All death reports come to the DAC. “We go through it with a very fine toothcomb to see which is a COVID-19 death and which is not. The WHO has said that in only a few situations can you not count a COVID-19 patient’s death as a non-COVID death. That’s when you have alternative diagnosis like a road traffic accident or suicide, or bleeding in the brain etc.”

Another situation is that of a patient who has improved, and has been totally free of symptoms for 24 to 48 hours. If that patient later dies, the WHO guidelines say that it needn’t be counted as a COVID-19 death. “Various international authorities differ on this aspect. It is true that such a situation could also be attributed to COVID-19. So you cannot say even WHO guidelines are perfect. That’s why developed nations have gone back to look at all-cause mortality.”

A report by the DAC says that there has been a 5.5% reduction in the absolute number of deaths in Kerala in 2020 when compared to pre-pandemic year 2019. This number could have come down as deaths from accidents and other causes decreased, and also because of the preventive measures followed by people such as wearing of masks and physical distancing. Many developed countries had reported a noticeable increase in the number of all-cause deaths during the pandemic year.

Excluding and including deaths during audit

Sometimes the deaths reported as COVID-19 by the treating doctor is declassified as non-COVID by the DAC. Some other times the reverse happens, when a death not included under COVID-19 is moved to the COVID-19 deaths after the audit.

“In every case, we specify clear-cut reasons for the decision. If a treating doctor didn’t count it as a COVID-19 death and during the audit we see that the patient had symptoms as fever, cough and evidence for pneumonia, it will be taken as a COVID-19 death. Reclassification of deaths happen at the level of the DAC. But this is not a real time process and takes time," he says.

District-wise disparity

Yet another criticism was on the disparity in the number of deaths in districts. Districts such as Ernakulam which has the most number of cases has reported much lesser number of deaths than Thiruvananthapuram.

“This is because details from certain districts may not be complete. Usually in districts where Medical Colleges are there, everything gets reported scientifically. We audit deaths from those districts where details are complete. That’s why in certain districts, the number of audited deaths is less,” Dr Aravind explains.

Death audit report for March 2021 can be found here.

Related Stories

No stories found.
The News Minute