Why Can’t India Get its Heatwave Mortality Data Right?
India’s rising heatwave deaths expose a tracking system that gives conflicting figures and undercounts fatalities.
advertisement
(As extreme heat claims more lives across India, independent climate reporting has never been more essential. Become a member and power our journalism.)
Last week, Anujkumar Shah, a 30-year-old migrant worker in Surat, collapsed on his way home from the textile unit where he worked. He died shortly afterwards at a nearby hospital. His body was severely dehydrated, likely due to prolonged exposure to direct sunlight and extreme heat, a resident medical officer at the hospital told the media.
Surat, like much of Gujarat, is currently reeling under severe heatwave conditions that are expected to continue through the week. Similar heat-related deaths have been reported from other parts of India as well.
In April, two schoolteachers in Odisha reportedly died of heatstroke while carrying out Census-related work. Four deaths were also reported during polling in West Bengal. In Karnataka, a 20-year-old man in Bidar district and a 35-year-old government officer in Belagavi district are both suspected to have died from heatstroke in March. And these are only the cases that made it to the news.
In India, according to National Crime Records Bureau (NCRB) data, heat and heatwaves are, in most years, the second leading cause of weather-related deaths after lightning. In the rest of the world, it is the leading cause.
So, why is India not tracking how many are dying—and why is it so difficult to get an accurate count?
The Fault in Our Numbers
India doesn't maintain a nationally representative source for data on heatstroke and heat-related deaths, forcing researchers to stitch together figures from multiple fragmented datasets—figures that are often widely inconsistent with one another.
The most commonly cited government sources are NCRB, the National Centre for Disease Control (NCDC), and the India Meteorological Department (IMD).
But each source tells a strikingly different story.
Source: NCDC
Data obtained by PTI under the RTI Act show 3,812 heatstroke deaths between 2015 and 2020.
Source: NCRB
During the same period, NCRB figures, cited in Parliament questions many times by Union Earth Sciences Minister Jitendra Singh, put the number at 8,171 deaths from "heat/sunstroke".
Source: IMD
The IMD's annual reports record 3,436 deaths due to "heatwave" from 2015 to 2020.
Several independent studies in recent years have tried to bridge this data gap. A report by non-profit HeatWatch released underscores the discrepancies between official government death counts and those collected through a systematic review of national and regional media reports.
The report found that at least 84 suspected deaths were reported by the media between February and July. NCDC, by contrast, reported a total of 7,192 suspected heatstroke cases in 2025, but only 14 confirmed deaths due to extreme heat between 1 March and 24 June that year.
Maharashtra and Uttarakhand recorded the highest confirmed toll at three deaths each. Several other states reported hundreds of suspected cases, but zero confirmed deaths. NCRB and IMD figures for 2025 are not yet available.
State-wise data is available for some states through sources such as heat action plans and state disaster management authorities (SDMAs). However, these figures, too, often differ from the state-level data reported by NCRB.
Also, as Abhiyant Tiwari, a Heat Health and Climate Resilience Expert based in Delhi, points out, the data we get from all these sources "is only for heatstroke deaths and not all heat-related deaths."
For instance, the NCRB data for 2024, released by the Ministry of Home Affairs, stated that Telangana reported 116 heatstroke and sunstroke deaths. However, the state’s Heatwave Action Plan 2026, prepared by the Revenue (Disaster Management) Department, recorded only 10 heatwave deaths in the same year.
Another state that has a relatively active SDMA that tracks heatwave deaths in near real-time during summer is Odisha. By the end of May, the SDMA in Odisha confirmed 34 heatstroke deaths and 149 suspected deaths, while NCRB recorded 139 for the full year. It's unclear how this final count was arrived at.
On the other hand, Bihar recorded 422 heat-related deaths per NCRB—the highest of any state—but does not operate an extensive automated weather network or publish a dedicated heatwave action plan, making independent verification of that figure difficult.
Why Can’t We Get the Numbers Right?
It is a combination of several factors, but experts say the root of the issue begins with how deaths are recorded in India.
"India doesn't have a robust system of registration for deaths, not only for heat, but in general. The cause of death analysis is poor as well. Partly because a proper system has not been developed," explains Dr Mavlankar, who's the Director of the Indian Institute of Public Health, Gandhinagar.
Moreover, different systems use different definitions, methods of assessment, and reporting mechanisms. “IMD, NCDC, and NCRB all rely on different data sources and methodologies," says Tiwari, which helps explain the differences in the data.
NCRB, which falls under the Ministry of Home Affairs, records deaths based on police reports. "If a death is registered at a police station and heatstroke or sunstroke is cited as the cause, it is included in the national database,” says Dr Mavlankar, who helped develop India’s first heat action plan in Ahmedabad in 2013.
Dr Mavlankar points out that the IMD is fundamentally a weather forecasting agency, and its heatstroke death figures are drawn, not from any surveillance system, but from media reports and SDMA data.
Lastly, NCDC collects data through the Integrated Disease Surveillance Programme, tracking both suspected and confirmed heat deaths reported by hospitals and state public health authorities.
Because the three sources are measuring different subsets, there is likely some overlap between the numbers, but it hasn't been formally quantified, and "the absence of a unified death registry means there's no way to know how much double-counting or undercounting exists across all three," explains Dr Mavlankar.
A study co-authored by Tiwari in 2024 analysed all-cause mortality across 10 Indian cities between 2008 and 2019, comparing death rates on heatwave days against non-heatwave days using satellite temperature data. He says:
What Counts as a Heat-Related Death?
Another major obstacle when it comes to counting heat-related deaths is classifying them.
Heatstroke deaths typically occur when the body overheats and is unable to regulate its temperature in extreme heat and humidity, leading to severe dehydration, low pulse, and dangerously high body temperature.
"Only a few cases can be directly attributed to heat exhaustion. For instance, when a construction worker with no prior health conditions that spends long hours in the sun presents with clear signs of heatstroke and dies, the cause can be clearly linked to extreme heat," Dr Suresh Reddy Challamalla, senior consultant (neurology) at Gleneagles AWARE Hospital, Hyderabad, tells The Quint.
However, he adds, "Heatstroke on its own causing death, without complications, is relatively uncommon."
Heatstroke deaths represent a small fraction, only about 10 percent of all heat-related deaths, say experts.
What is much more common is that heat-related deaths occur among people who are already vulnerable, such as the elderly, diabetics, smokers, or those with underlying cardiovascular conditions.
As per guidelines from the National Programme on Climate Change and Human Health (NPCCHH) and NCDC, a death can be clinically classified as heat-related when there is high body temperature, extreme environmental heat, and a clear history of heat exposure.
Exceptions exist that often slip through the cracks.
Heat-related fatalities can occur on “non-alert” days as well. Even when body temperature is not high, like in cases where cooling was attempted before arrival at the hospital, doctors shouldn't rule out heatstroke if there is strong evidence of previous heat exposure and no other clear cause of death.
"In such cases, patients often deteriorate over several days from organ or respiratory failure, and their deaths frequently are not recorded as heat-related," says Dr Mavlankar.
Clinical diagnosis of heat-related death is done principally through epidemiological investigation, death audit or autopsy findings, but that rarely happens in India.
He says, "Many simply note the immediate cause, such as cardiorespiratory failure, instead of identifying the underlying trigger."
Tiwari adds that heat-related deaths often occur outside hospitals and without medical examination, making it even harder to accurately attribute the cause of death.
Excess Mortality and Why the Data Matters
Because it's difficult to diagnose and classify each case of heat-related death accurately, experts argue that “excess mortality” is a more effective way to estimate deaths caused by extreme heat.
"Excess mortality is the difference between how many people actually died during a given period as opposed to how many would normally be expected to die (of all causes) based on historical averages," explains Tiwari.
"Daily all-cause mortality data should be recorded at the city and village levels by local registrars, and should be made public," says Dr Mavlankar.
He adds:
All-cause mortality data is something that other countries have long had in place. EuroMOMO (European Mortality Monitoring network), for instance, tracks excess deaths across 27 European countries in real time, publishing weekly bulletins that show spikes in all-cause mortality during heatwaves, cold waves, and disease outbreaks.
India has the raw infrastructure to do the same.
Under the National Programme on Climate Change and Human Health, NCDC tracks heat-related illness and death through its National Heat-Related Illness and Death Surveillance system since 2023 (an implicit acknowledgement that this was a gap that needed filling).
It collects daily data on heatstroke cases and deaths, emergency attendance, and all-cause deaths from health facilities at the primary health centre level and above. But none of this data is made public. What it lacks is not the data but the political will to publish it.
"No one is publicising this. I don't know why India doesn't do it—if not the whole country, at least each city should be recording and publishing this data. It is not rocket science," says Mavlankar.
Why Does This Data Matter?
As per NPCCHH, if global mean temperature continues to increase toward 2°C of pre-industrial levels, annual heat-related deaths are projected to increase by 370 percent by mid-century without adaptation measures.
"If we under estimate the deaths, we are also going to under estimate the urgency of the response," says Tiwari, adding, "And the data we are currently getting from official sources really doesn't adequately quantify the burden."
You can't design effective interventions without knowing who is getting affected, where, and under what circumstances, he adds.
Without reliable granular data, heat action plans end up being generic rather than targeted. For instance, most Indian states have elaborate plans, but little evidence of their effectiveness in preventing deaths and illness.
Also, as heatwaves have not been classified as a national disaster, there is no Central government ex gratia compensation for victims. However, some states have declared heatwaves as "state-specific disasters," making victims’ families eligible for ex gratia payouts—but only if the death is officially confirmed as heat-related.
Several experts and organisations, including HeatWatch, have called for legislation mandating heat protection for workers. Legislative pushes like that are much harder to sustain without data establishing that heat is killing people in workplaces at scale.
(The Quint has reached out to NCDC, and IMD with detailed questionnaires. Their responses will be added once they are received.)