Measuring malaria
Accurately estimating the incidence and mortality from diseases in India, including malaria, is challenging. We lay out the available data sources and explain our reasons behind using selected sources
There is a significant lack of reliable and consistent data on malaria cases and deaths in India. While capturing direct estimates is challenging, modelled estimates can also be imprecise.
Malaria incidence is reported by the National Centre for Vector Borne Diseases Control (NCVBDC) and estimated by the Institute for Health Metrics and Evaluation's (IHME) Global Burden of Disease (GBD) study. Mortality data comes from the Medical Certification of Cause of Death (MCCD) system and the Sample Registration System (SRS) Cause of Death (CoD) survey. Each source has unique methodologies, strengths, and limitations.
Incidence
In the context of diseases, 'incidence' means the number of new cases registered, reported or estimated in a year.
The National Centre for Vector Borne Diseases Control (NCVBDC) under the Ministry of Health and Family Welfare (MoHFW), manages malaria surveillance and elimination programs in India. Its reports are based on both passive surveillance (cases reported at health facilities) and active surveillance (targeted blood sampling in high-risk areas).[1] Diagnostic methods such as microscopy and rapid diagnostic tests (RDTs) are conducted by health workers, including Accredited Social Health Activists (ASHAs) and Auxiliary Nurse Midwives (ANMs).[2]
The only other source of malaria incidence data is the IHME's Global Burden of Disease (GBD) study. For India, the study combines data from diverse sources, including government registries, surveys like DHS, entomological studies (primarily data and studies published on Malaria Atlas Project) and monthly reports from the Health Management Information System (HMIS). Using statistical models and covariates like climate and population density, it provides incidence estimates. The limitation of this source lies in its dependence on secondary data and uncertainties arising from assumptions used in statistical modeling.
We use the IHME's estimates of disease incidence as the NCVBDC estimates are by the government's own admission under-reported. For example, in 2019, NCVBDC reported 340,000 malaria cases, whereas modelled estimates from the Institute for Health Metrics and Evaluation (IHME) suggest close to 5.5 million cases - about 16x higher. This discrepancy arises from incomplete reporting from the private healthcare sector and poor surveillance in remote, forested areas where logistical barriers and reliance on passive reporting miss numerous cases.
Mortality
Accurate numbers of deaths from malaria are even more challenging to produce.
The most direct way to calculate the number of deaths from malaria would have been if the death of every Indian was medically certified. In India, the Medical Certification of Cause of Death (MCCD) system documents causes of death using standardised medical certifications provided by medical practitioners after a death. The consolidated report provides mortality data, helping identify disease patterns, allocate resources, and monitor health interventions effectively.
However, the dataset faces significant challenges, especially in rural areas where many deaths go unrecorded due to limited healthcare access and inadequate diagnostic tools. Only 21% of registered deaths in India are medically certified, with significant variation among states - from over 40% in Maharashtra and Tamil Nadu to just 3%-6% in Bihar and Jharkhand[3]. This underreporting affects the accurate assessments of mortality trends, particularly in high-burden malaria states where medical certification rates are among the lowest. In 2019, the MCCD reported 3,364 deaths from malaria.
NCVBDC mortality data is limited by the same underreporting challenges affecting its case data. For instance, in 2021, it recorded only 77 malaria deaths, significantly lower than all other estimates.
To address this gap in mortality data, India has been conducting the Sample Registration System (SRS) Cause of Death Survey since the 1960s. It is a national survey designed to estimate vital statistics, including causes of death. The 2017-19 (most recent) survey sampled 8.2 million people from about 4,000 urban and 5,000 rural units, representing India's diverse population[4]. For deaths recorded in sample areas, Verbal Autopsies (VA) are conducted - field workers interview family members about symptoms, medical history, and death circumstances. Physicians then assign causes of death using the International Classification of Diseases (ICD-10), including malaria codes (B50-B54)[5].
IHME, through its GBD study,uses statistical models that account for underreporting and misclassification, offering broader estimates of malaria mortality. The primary data sources include government registries and providers like Institute of Health Systems (India), Indian Council of Medical Research (ICMR) and State Bureau of Health Intelligence & Vital Statistics. However, it relies heavily on modelling. The dataset also provides upper and lower values of mortality and incidence estimates or the confidence interval. This indicates the range within which the true value is likely to fall. These intervals are crucial for understanding the reliability of the estimates and comparing trends across regions or time.
As a result, we use the SRS, a direct survey of a large sample of households, for estimates of mortality from disease for this piece. SRS provides the distribution of all-cause deaths across communicable, non-communicable, and other categories, including the percentage of deaths caused by malaria during a given period. To estimate the absolute number of malaria deaths, we combine this percentage with age and gender specific annual death estimates from the UN World Population Prospects[6].
Since SRS does not offer a state-wise breakdown of deaths, we turn to the IHME GBD study for state-level mortality data, despite its limitations.
[1] Strategic Action Plan for Malaria Control in India 2012-17
[2] National Centre for Vector Borne Diseases Control (NCVBDC) reporting format
[3] Report on Medical Certification of Cause of Death, 2020
[4] Sample Registration System - Cause of Death in India, 2017-19
