- As temperatures rise, India must place public health at the centre of how heat impact is measured, understood, and acted upon.
- This will help deliver effective, equitable, and evidence-based responses at scale.
- Heat action plans respond to heat that is already here but do little to address the drivers of warming or the heat yet to come.
- The views in this commentary are those of the author.
Every summer, as temperatures rise, Heat Action Plans are invoked, labour woes are highlighted, and mortality counts are contested. Yet, nothing changes in the long run. The central question: Are we adequately protecting people’s health? remains conspicuously unaddressed. It is not for lack of evidence. The failure, repeated every year, lies in failure to translate health science into coherent strategy and actions.
Across surveillance, research, intervention design, implementation, and governance, health science and public-health principles are often overlooked, distorted, or lost in translation.
The quantification trap
Expecting public health surveillance to fully account for every heat-related illness (HRI) before mounting a response is a fundamental misunderstanding.
First, surveillance systems, by continuously gathering data of selected serious, preventable, and reliably reportable health conditions, are meant to detect signals in time to trigger action. The goal is situational awareness, not a census. As such, the National Heat-Related Illness and Death Surveillance (NHRIDS) focuses on capturing heatstroke, the most severe heat-related illness, in near-real time.
Secondly, India’s heat surveillance carries structural limitations. Reporting still depends on manual data entry, which is inherently prone to underreporting. Additionally, unlike malaria or tuberculosis, there are no confirmatory tests for heat-related illnesses to add diagnostic certainty.
High-income countries bypass this technologically by extracting syndromic data from electronic health records, and still suffer from undercounting. India’s digital health mission is years behind reaching that level of modernisation. As such, extending surveillance reporting to all heat-related illnesses in our current system adds noise and burden rather than actionable insights.
Epidemiologically, measuring excess hospital admissions and mortality provides better quantification of heat’s impact. NHRIDS captures such indicators but lacks the digital infrastructure for real-time analysis.
Ultimately, in the case of heat-related illnesses, simplistic case counting is not an evidence-based practice or effective response strategy; it must give way to established epidemiological methods and must never negate the imperative to act.
Excluding health from health research
Health-related illnesses are clinical diagnoses which require a doctor to assess heat exposure, symptom combinations, physiological status, and exclude alternative causes. They cannot be reliably diagnosed solely on the basis of symptom reporting.
Yet a growing strand of short-term heat-health field research appears to bypass both clinical diagnosis and rigorous epidemiological methods that help establish heat-related illnesses in quantitative studies. Such studies report participant-identified heat-related illnesses or classify self-reported individual symptoms such as headache, fatigue, nausea, or muscle pain during summer as ‘heat-related’ simply because they occurred during hot weather. Participants’ perceptions, narratives, and recall are treated as equivalent to clinical judgement.
While such data provide valuable insights into lived experiences, they do not establish whether symptoms are attributable to their heat exposure, underlying health conditions, medications or drug use, or other common summer infections. The result is the production of numbers without a diagnostic or attributional foundation.
Studies can adopt clinical verification or validated epidemiological tools, such as HOTHAPS, which capture health impacts relatively objectively through extensive data collection on exposure, physical activity and physiological measurements, and provide a more robust basis for characterising heat exposure and identifying heat-related illnesses. Even simple comparative approaches, such as establishing baseline health status or comparing symptom patterns across seasons, can also reduce certain biases. The absence of such practices, in favour of quick, small sample surveys, can introduce substantial biases, inflate disease burden, generate spurious associations, normalise exclusion of health expertise, and ultimately misinform policy and undermine legitimate evidence.
The heat mortality conundrum
The inconsistent heat-mortality figures reported across agencies stem from the absence of a uniform definition and categorisation of heat-related deaths. Different departments have been collecting heat mortality data under different mandates and therefore rely on different reporting systems, data sources, and classification criteria.
The National Crime Records Bureau (NCRB) has been recording accidental deaths in its annual reports through its police network, largely outside the health system, with little public information on how a heatstroke death is identified. The India Meteorological Department (IMD) has historically compiled heatwave mortality figures from media reports. Hence, their validity is as good as those media reports. Neither system is equipped to apply clinical criteria for diagnosing heatstroke or attributing deaths to heat. By contrast, the health sector can investigate and classify heatstroke and heat-related deaths using established clinical definitions and surveillance protocols, but only for deaths that occur within or are reported to the health system.
Ultimately, without a common case definition, differing mortality counts are inevitable. More importantly, these figures are not directly comparable and cannot be meaningfully combined.
Again, for understanding broader population-level mortality impacts, all-cause mortality data offer a more consistent basis for analysis than heat-death counts. The Central Registration System (CRS), with its long-term mortality records, is therefore a better resource to assess excess all-cause mortality and validate NHRIDS. However, it is neither designed for real-time heat-mortality surveillance nor supported by sufficiently robust cause-of-death coding to reliably identify heat attributed deaths.
The hidden burden of heat-induced illnesses
Beyond heat-related illnesses lies a much larger and largely invisible burden of heat-related diseases. Heat exacerbates a wide range of cardiovascular, kidney, respiratory, neurological, maternal and mental health outcomes. However, existing large health verticals, including the Non-Communicable Disease Programme, the Integrated Disease Surveillance Programme, and others, have yet to integrate temperature and other environmental exposures into their surveillance and response frameworks. Analysing long-term trends in disease and temperature would reveal the broader burden of heat and support more policy-relevant responses through those health programmes. Although heat’s additional health burden remains invisible in statistics and in health adaptation policies currently, health services manage it continuously.
Resolving these data-related challenges requires systemic changes and political will. Nevertheless, the imperative for non-health sectors to effectively and urgently address the upstream drivers of heat exposure through shade, water, rest, and cooling, and the need for health systems to manage the downstream health consequences, remain inseparable priorities.
Forgotten physiology
Heat affects health through predictable physiological responses that begin before discomfort is even perceived. Understanding these responses is essential for designing effective interventions. Early warnings, heat risk indices, occupational advisories, personal cooling guidance, and cooling technologies should therefore be grounded in thermal physiology and health outcomes, but most current approaches in India are not.
A physiological perspective highlights a critical reality: in India, the margin between tolerable and potentially lethal heat exposure is often narrow due to heat’s widespread and intense presence. Many people experience sustained heat strain, yet heat policy is increasingly being designed around vulnerable groups. Heat risk is determined not only by who people are, but also by the level of heat exposure and exertion they experience at a given time. Exposure is dynamic; vulnerability categories are relatively static. A teacher conducting fieldwork, a pilgrim walking long distances, an auto driver and a homeless in outdoor conditions, or a traffic police officer managing congestion during extreme heat may face risks comparable to those of recognised vulnerable groups such as gig workers.
A cooling shelter reserved for a particular occupational sub-group reflects a narrow vulnerability-oriented, demand-based approach rather than a broader exposure-oriented, proactive approach. Such interventions can exclude others facing similar levels of heat stress, particularly when formalised through HAPs, and may divert prioritisation from population-wide measures that provide more equitable protection at scale.
Protecting vulnerable populations remains essential, but most such vulnerabilities should first be addressed through basic safeguards, regulatory protections, and targeted support. In a country where extreme heat is already widespread and basic provisions and safeguards remain critically inadequate, universal protection must form the foundation of heat response, with vulnerability-based measures as a complement, not its organising principle. Otherwise, heat response risks becoming a siloed exercise in identifying, advocating and protecting one vulnerable subgroup after another. With looming El Niño, that fragmentation will be stress-tested at scale, and it will not hold.
An index that fails to land
Heat vulnerability assessments (HVAs) are designed to rank geographic areas along a continuum of relative risk. While conceptually useful, their operational value remains limited unless they are standardised, validated, tailored to sector-specific decisions, and linked to explicit response frameworks.
There are no standardised criteria for indicator selection, weighting, and methodology, making these assessments sensitive to subjective choices and distorted risk representations, and most are not validated through ground-truthing. As a result, different assessments produce conflicting risk maps for the same geography, with no objective basis for determining their accuracy or utility for decision-making. In practice, heat vulnerability assessments in heat action plans, such as those of Delhi or Jodhpur, remain weakly connected to targeted, long-term action and are, at best, used for seasonal action. Without an evidence-based response framework, most heat vulnerability assessments remain exercises in risk mapping rather than tools for action.
Their utility is further constrained by two conceptual limitations. A generic heat vulnerability assessment cannot meet the distinct needs of different sectors and most provide static snapshots that fail to capture changing weather, urban development, and population mobility.
Adaptation without mitigation
There are a few translation failures that operate at a scale far beyond the rest. Heat action plans respond to heat that is already here but do little to address the drivers of warming or the heat yet to come. Heat action plans are inherently adaptation tools, so they remain largely silent on ambient heat’s linkage with greenhouse gas mitigation and lack mechanisms to assess whether current heat responses are themselves contributing to future warming.
This disconnect creates a risk of maladaptation. Cooling demand powered by coal-fired electricity, air-conditioning uptake without energy-efficiency or fan-first strategies, and interventions such as misting asphalt roads or Miyawaki forests are not climate-neutral responses. Some may reduce heat exposure locally while increasing emissions, energy demand, water use, or other environmental pressures elsewhere.
While climate mitigation is driven by Nationally Determined Contributions (NDCs), net-zero targets, and climate action plans, heat response is centred on HAPs. These two architectures often operate on different time scales, with different administrative mechanisms and funding streams. The link between reducing emissions and responding to heat remains broken.
Perhaps the most striking feature of this nexus is that the health sector remains peripheral to both climate mitigation and heat adaptation. Neither climate mitigation nor heat adaptation support systemic health sector strengthening, despite the urgent need to strengthen India’s public health services and infrastructure, digital and physical. Frameworks like the National Solar Mission or EV policy implementation also do not treat health as a ‘critical’ sector. The absence of specific health-sector targets within India’s NDCs institutionalises this gap, leaving the sector responsible for managing climate-related harms largely as an afterthought of climate action.
India’s heat response remains hampered by gaps in definitions, planning, accountability, and action. Across sectors, the challenge is not simply understanding heat risk but translating that understanding into interventions that measurably reduce exposure, illness, and death. Heat exposure may be shaped by climate, labour, urban planning, and energy systems, but its consequences ultimately manifest in human health. Unless public-health outcomes become the organising principle of heat governance, rising temperatures will continue to expose the limits of our response.
Banner image: An elderly woman suffering from a heat-related illness is brought to an overcrowded government district hospital in Ballia, Uttar Pradesh in 2023. (AP Photo/Rajesh Kumar Singh, File)
Dr. Purvi Patel is a public health professional with a focus on climate change and health, and a former senior consultant at the National Programme on Climate Change and Human Health, National Centre for Disease Control, Ministry of Health and Family Welfare.
Credits
