Rabies and the missing deaths: Why do India’s official numbers differ so sharply from estimates? | Explained

The story so far: India is working towards eliminating dog-mediated human rabies by 2030. Yet the country does not have a single, comprehensive national count of human rabies deaths. While the Health Ministry’s disease surveillance system recorded 54 deaths in 2024, an ICMR-led study estimates that 5,726 people die of rabies in India every year.Is rabies a notifiable disease in India?Rabies is not a notifiable disease nationwide. Unlike diseases for which there is a clearly established national notification mechanism covering every State and Union Territory, rabies notification has historically depended on State-level notification requirements and the disease surveillance system.The National Centre for Disease Control (NCDC) has been working to make human rabies a notifiable disease across the country. Communications have been sent to States to notify the disease under the relevant State legislation.However, a disease being under surveillance does not necessarily mean that every case or death occurring in the country is automatically captured in one central database.Rabies is a particularly difficult disease to count because a large proportion of deaths can occur outside the formal health system. Once clinical symptoms appear, rabies is almost invariably fatal, but patients may die at home or in facilities that do not routinely report the death through disease-surveillance channels.How is a disease chosen for notification in India?Disease notification is largely a public-health surveillance and State-level regulatory function, while the Union government maintains national surveillance systems such as the Integrated Disease Surveillance Programme (IDSP).A disease is generally prioritised for notification or surveillance when it has significant public-health importance — for example, high mortality or morbidity, epidemic potential, rapid spread, risk of international transmission, emergence of a new pathogen, or where early detection can trigger effective control measures. India’s disease-prioritisation exercises have considered epidemiological, clinical, economic and public-health factors, with expert consultation.Once a disease/condition is designated for surveillance, health facilities, laboratories and health workers report cases through the surveillance system. Under IDSP, reporting includes suspected (S), presumptive (P) and laboratory-confirmed (L) cases. Data move from reporting units to District and State Surveillance Units and ultimately to the Central Surveillance Unit at the National Centre for Disease Control (NCDC).A rising or unusual cluster triggers investigation by Rapid Response Teams, which can lead to laboratory confirmation, containment measures and further reporting. The exact legal notification requirement can depend on the State/UT and the particular disease. Leprosy, for instance, was declared a notifiable disease nationally in 2025.So, how many people die of rabies in India?There are different answers, depending on what is being measured. The Integrated Disease Surveillance Programme/Integrated Health Information Platform (IDSP–IHIP) recorded 21 human rabies deaths in 2022, 50 in 2023 and 54 in 2024.The Registrar General of India’s Medical Certification of Cause of Death (MCCD) system, meanwhile, recorded 66 medically certified rabies deaths in 2022 and 271 in 2023.But neither set of numbers represents all rabies deaths in India. The MCCD figures are based on medically certified causes of death. The system does not cover every registered death in the country. At the national level, medically certified deaths accounted for only about 22% of registered deaths in both 2022 and 2023.The IDSP–IHIP figures serve a different purpose. They are generated through disease surveillance and are intended to identify and track cases and deaths reported through the surveillance network.This is why the figures cannot simply be added together or compared as if they were measuring the same thing. Then where does the figure of 5,726 deaths come from?An ICMR-led study published in The Lancet Infectious Diseases estimated that India has about 5,726 human rabies deaths every year.The researchers did not arrive at this number by counting death certificates or hospital records.They conducted a large community-based survey involving 337,808 people in 78,807 households across 60 districts in 15 States. The study estimated that India sees about 9.1 million animal bites every year, including around 5.6 million dog bites.The researchers then combined information from the community survey and laboratory data with a probability decision-tree model to estimate how many people bitten by animals would develop rabies and subsequently die.The 5,726 figure is therefore a modelled estimate of annual mortality, rather than a tally of deaths reported to the government.Why is the difference so large?This is because each system is designed to capture its own unique data.IDSP–IHIP is primarily a disease-surveillance system. It depends on cases and deaths being detected and reported through the health system and surveillance network.MCCD records medically certified causes of death. It provides valuable information on the causes of deaths that are medically certified, but it does not cover every death.The ICMR-led study, on the other hand, attempts to estimate the actual burden in the population, including deaths that may never enter either surveillance or medical-certification systems.This means that a low number in a surveillance database does not necessarily mean that the disease burden is low.In rabies, the problem is compounded by the nature of the disease itself. Patients may not reach a hospital after symptoms begin; some deaths occur at home; and the animal bite that initiated infection may have happened weeks or months earlier.There is also a critical point between the animal bite and a possible rabies death: post-exposure prophylaxis (PEP). If a person receives appropriate wound care, rabies vaccine and, where indicated, rabies immunoglobulin promptly after exposure, death can be prevented.What happens to a case of a notifiable disease?India’s disease surveillance system works through multiple levels.At the frontline are healthcare facilities, laboratories and other reporting units. When a suspected, probable or confirmed case of a disease covered by surveillance is detected, information is reported through the appropriate surveillance mechanism.The information moves through district and State surveillance structures and is consolidated nationally through the IDSP/IHIP platform. The purpose is not simply to count cases. Surveillance is intended to detect unusual increases, identify outbreaks, understand geographical patterns and enable public-health authorities to respond.But notification depends on detection, diagnosis and reporting. A person who never reaches a reporting facility, a death that occurs outside a medical setting, or a case that is not recognised as rabies may not enter the surveillance database.This is why notification should not be confused with a census of every person who has the disease.Does that mean the 5,726 figure is the “real” number?The 54 deaths recorded by IDSP–IHIP in 2024 tell us how many deaths were reported through that surveillance system.The 5,726 estimate attempts to answer a different question: how many people may actually be dying of rabies each year, including deaths that are missed by routine systems.The study itself is therefore important not because it replaces surveillance data, but because it draws attention to what routine surveillance may be missing.A senior Health Ministry official has also pointed out that the two figures are generated through different methods and should not be treated as directly equivalent. Together, however, they point to a substantial gap between reported surveillance deaths and estimated mortality.Why does this matter when India wants to eliminate rabies by 2030?Reliable surveillance is essential to rabies elimination. India cannot determine whether it is moving towards elimination if it cannot accurately establish where human rabies deaths are occurring, where dog-mediated transmission continues and which populations are not receiving timely post-exposure care.The ICMR-led study also found important gaps after animal bites. About one in five dog-bite victims had not received anti-rabies vaccine, while only 66.2% had received at least three doses. Among those who received only one dose, nearly half did not complete the vaccination course. Only about one in 10 eligible victims received passive immunisation.The gaps are therefore not confined to counting deaths. They extend across the entire rabies-control chain — from preventing dog bites and vaccinating dogs to recognising exposures, completing human post-exposure prophylaxis and recording outcomes.As medical scientist and former ICMR Director-General Nirmal Kumar Ganguly puts it, the absence of consolidated data remains a challenge for rabies elimination. Cases and deaths can occur in places where they are difficult to measure, making stronger surveillance, dog vaccination and timely human post-exposure treatment critical.Rabies is almost entirely preventable, but once symptoms appear, it is almost invariably fatal. That makes every missed exposure, incomplete vaccination course and unrecorded death significant.For India, therefore, the question is not simply why one database reports 54 deaths while a study estimates 5,726.The larger question is how many deaths the health system is failing to see — and whether India can close that surveillance gap before it attempts to declare dog-mediated human rabies eliminated by 2030.