Evaluation of an Experimental Climate Information of Health Product for Vector-borne Diseases: An Initiative towards Climate Services for the Health Sector in India

Abstract

Aims: To evaluate the usefulness of the experimental Climate Information for Health (CIH) product issued by the India Meteorological Department (IMD) for malaria and dengue against observed disease-occurrence data for 2017. Study Design: A retrospective, observational verification study using categorical (contingency-table) analysis of forecast climatic suitability against reported vector-borne disease (VBD) occurrence. Place and Duration of Study: Office of Climate Research and Services, India Meteorological Department, Pune, India, covering all Indian states/meteorological subdivisions from May to December 2017. Methodology: As a requirement of the WMO Global Framework for Climate Services (GFCS), IMD initiated weekly Climate Information for Health (CIH) bulletins for malaria and dengue on an experimental basis. The bulletins used two-week-ahead maximum/minimum temperature forecasts from its Extended Range Prediction System (ERPS) and were issued every Friday from the second week of May 2017. States were classified into four categories according to whether climatic suitability for VBD occurrence was anticipated and whether the disease was subsequently reported, using monthly, state-wise disease data from the National Vector Borne Disease Control Programme, FluTrackers, state health departments and municipal corporations. Skill scores (probability of detection, false alarm ratio, miss rate, bias, critical success index and Heidke skill score) were computed from the resulting contingency tables. Results: CIH was found to be useful for malaria in most states and months from June to October 2017 (probability of detection [POD] ≈ 0.97; false alarm ratio [FAR] ≈ 0.03). For dengue, its usefulness was limited to the winter months of November and December (POD ≈ 0.38; FAR ≈ 0.11 for the full study period). This was attributable mainly to the absence of a validated maximum-temperature threshold for dengue transmission in the Indian context. Conclusion: The experimental CIH product provides a useful, low-cost early signal for malaria risk in India and a starting point for dengue, but region-specific temperature (and possibly humidity and rainfall) thresholds for dengue transmission need to be developed in collaboration with the health sector to improve its reliability outside the winter season.