The strokes India does not count

As the world prepares to mark World Stroke Day on October 29, governments and health systems will once again pledge to cut stroke deaths. This year's campaign, "Every Minute Counts", asks people to #ActFAST: To recognise a drooping face, a weak arm and slurred speech, and to call for help in time. India should add a harder question to its pledge: How would we know if stroke deaths fell?

How many Indians died of stroke last year? India does not know. In 2024, the Registrar General's report on Medical Certification of Cause of Death (MCCD) recorded 86,300 deaths from cerebrovascular disease, just 4.2% of all certified deaths. That is a fraction of the real toll: The Million Death Study estimated that stroke killed about 0.4 million Indians aged 30-69 alone in 2015. That gap is the difference between a stroke epidemic the state can see and one it cannot. A pledge without a baseline is only a hope.

Of 89.4 lakh deaths registered in 2024, only 20.7 lakh, or 23.1%, carried a medically certified cause; nearly 69 lakh went into the record without one. In 2015 the figure was 22.0%: A decade of almost no progress. The national average hides a chasm. Goa certified 100% of registered deaths; Bihar certified 3.6%, Haryana 6.4% and Uttar Pradesh 6.7%. Where a doctor attended the dying person, 63% of deaths were certified, and nine states and Union Territories, including Maharashtra, Tripura and West Bengal, certified all of them. The deepest silence surrounds deaths that receive no medical attention.

Where certificates exist, their quality is uneven. A review of 20 Indian studies covering 17,106 certificates found that doctors recorded the mode of death, typically "cardiopulmonary arrest", instead of the underlying cause in 8.9% to 86% of certificates. Every heart stops at death; recording that it stopped tells the system nothing. The 2024 report bears the marks: 10.7% of certified deaths were assigned only to symptoms and abnormal findings, and another 18.7% to "other forms of heart disease", an ICD-10 block that also holds vague entries such as cardiac arrest and heart failure.

Even in this imperfect record, circulatory diseases, the group that includes stroke and hypertensive disease, rose from 29.8% of certified deaths in 2010 to 38.8% in 2024. Stroke is the outcome most tightly coupled to blood pressure. Globally, high systolic blood pressure accounts for 57.9% of the stroke burden. In India, 74.5% of stroke patients in ICMR's hospital registry had hypertension, yet only 12.6% of hypertensive adults have their blood pressure controlled. In 2024, certificates attributed 86,300 deaths to stroke and another 81,928 to hypertensive diseases; 15% of certified stroke deaths were in people under 45, and two-thirds (66.9%) in people under 70. Stroke deaths are the most direct measure of whether hypertension programmes are working.

Without reliable cause-of-death data, the National Programme for Prevention and Control of Non-Communicable Diseases (NP-NCD) can count how many people it screened, but not how many strokes it prevented. Districts cannot see their own stroke mortality, and budgets cannot be tied to it. In the North-East, where a third of India's premature stroke deaths occur among a sixth of its population, Arunachal Pradesh certified only 8.7% of registered deaths in 2024, Nagaland 13.8% and Assam 14.4%.

The World Stroke Organization urges governments to build national systems for monitoring stroke through registries, electronic health records and vital statistics. India already has the legal foundation. The Registration of Births and Deaths (Amendment) Act, 2023 makes it mandatory for medical institutions to send the registrar a cause-of-death certificate for every death on their premises, with a copy to the nearest relative, and creates a national database of registered deaths.

Six steps to count every stroke:-

Enforce Section 10 of the Act, as amended in 2023. Every hospital, public and private, must certify and report every death. Yet under Section 23(3), failing to issue a certificate is punishable by a fine of at most ₹50. That penalty needs updating, and states should publish facility-wise reporting rates and audit non-reporting hospitals.

Make certification digital. At one tertiary centre in Puducherry, moving from handwritten to software-generated certificates cut forms with a major error from 88% to 42%. Electronic MCCD with built-in checks should reject "cardiac arrest" and other modes of death as the underlying cause.

Train every certifier. Make MCCD a mandatory, assessed competency in MBBS, internship and in-service training; every intervention study in the Indian review found that training reduced errors.

Count deaths outside hospitals and without a doctor. Integrate structured verbal autopsy by trained health workers into civil registration for home deaths, starting with low-certification states such as Bihar, Uttar Pradesh and the North-East.

Publish stroke locally. The 2024 report already counts cerebrovascular and hypertensive deaths nationally. Publish them every year by district, age and sex, so that NP-NCD can track stroke mortality alongside blood-pressure control.

Link life to death, and fund it. With consent and data-protection safeguards, connect the National NCD Portal, through ABHA, to the national database of registered deaths that the Registrar General maintains under the 2023 amendment, so that a person screened at 18 under a University Health Programme, or at 30 under NP-NCD, can be followed to outcome. Give MCCD dedicated budget lines in state health plans for training, ICD coders, software and audit.

A death certificate is the last clinical document a person receives, and the first data point for preventing the next death. India is screening millions for high blood pressure; it must invest as seriously in knowing how many of them still die of stroke. The pledges made on World Stroke Day will be judged by stroke deaths averted, and India can claim that success only if it counts every stroke death. What gets counted gets prevented.

(The views expressed are personal)

This article is authored by Dr Om Prakash Bera, senior policy researcher and public health specialist working at the intersections of health, public policy and data analytics.