Evaluation of data sources and approaches for estimation of influenza‐associated mortality in India

Abstract

BackgroundNo estimates of influenza‐associated mortality exist for India.ObjectiveTo evaluate national mortality and viral surveillance data from India for assessing their appropriateness in estimating influenza‐associated mortality using varied analytic approaches.MethodsWe reviewed influenza virus surveillance data from a national influenza surveillance network. We also reviewed national mortality data from Civil Registration System (CRS), Medical Certification of Cause of Death (MCCD) and the Sample Registration System (SRS). We compared and scored the different sources of mortality data using specific criteria, including the process of cause of death assignment, sample size, proportion of ill‐defined deaths, representativeness and availability of time series data. Each of these 5 parameters was scored on a scale from 1 to 5. To evaluate how to generate an influenza‐associated mortality estimate for India, we also reviewed 4 methodologic approaches to assess the appropriateness of their assumptions and requirements for these data sets.ResultsThe influenza virus surveillance data included year‐round sample testing for influenza virus and was found to be suitable for influenza mortality estimation modelling. Based on scoring for the 5 mortality data criteria, the SRS data had the highest score with 20 of 25 possible score, whereas MCCD and CRS scored 16 and 12, respectively. The SRS which used verbal autopsy survey methods was determined to be nationally representative and thus adequate for estimating influenza‐associated mortality. Evaluation of the modelling methods demonstrated that Poisson regression, risk difference and mortality multiplier methods could be applied to the Indian setting.ConclusionDespite significant challenges, it is possible to estimate influenza‐associated mortality in India.

Burden of influenza-associated respiratory and circulatory mortality in India, 2010-2013

“…Estimating influenza mortality is complicated by the fact that influenza is often not listed as underlying cause on death certificates, thus influenza-associated mortality estimation is often conducted using indirect estimation methods with statistical modeling [5,6]. These indirect methods rely on national vital records systems as well as consistent surveillance data, which has also hampered many estimates in LMICs such as India due to issues with the quality or availability of necessary data sources [6][7][8]. However, estimates from the lower resource and tropical climate settings can be important to understand the true burden of influenza on high risk age groups and populations.…”

“…The SRS is a routine annual demographic survey, serving as the primary system for collecting data on births and deaths in India since 1971. Details of the SRS, including design, sampling scheme, physician assignment of the underlying cause of death and the methodology have been described elsewhere [8,16,17]. In brief, the SRS divided each state or union territory of India into one million sample units based on the population using the 2001 census.…”

“…SRS teams visit communities that are part of this system throughout the year to identify deaths and collect information on possible causes of death. VA reports are sent through a web-based platform to two trained physicians who assign the most probable underlying cause of death based on guidelines from International Classification of Diseases (ICD) coding system [8,16,17]. For our analysis, deaths were categorized into groups based on codes from the tenth revision of ICD (ICD-10): circulatory diseases (ICD-10 codes I00-I99) and respiratory diseases (ICD-10 codes J00-J99).…”

Advances in measuring influenza burden of disease

“…Two studies in this Special Edition, from India and South Africa, evaluated various data sources and methods and determined the best approaches for developing disease burden estimates in their country. The study from India found that while the Sample Registration System provided the most appropriate national mortality data set, other mortality data sources could be used for subregional estimates . The South African study showed that weekly proportion and influenza subtype‐specific proxies provided the best model fit with non‐significant differences in the estimates …”

“…The outcomes also varied across studies. The most common outcomes were deaths attributable to influenza, used in nine studies, and severe acute respiratory infections (SARI) or its equivalent, used in eight studies . Another seven studies measured excess or absolute hospitalizations due to influenza, and five studies measured community or outpatient visits .…”