Õhutemperatuuri ja haiglates teostatud protseduuride seoste uurimine standardiseeritud terviseandmetel

Abstract

Extreme temperatures are an increasing health risk, but their specific impact on hospital workloads has not been studied in detail. This bachelor’s thesis analyzes how daily air temperature in Estonia, and its delayed effects up to ten days later, are related to the frequency of medical procedures during 2012–2019. The analysis uses health data standardized to the OMOP CDM format and daily temperature data from nearby weather stations. A Distributed Lag Non-Linear Model (DLNM) was applied to estimate how temperature affects the number of procedures over time. Based on this, an interactive Shiny application was developed. Users can select a hospital and procedure, set lag days, and see how the relative risk changes with temperature. They can also view how a specific temperature affects procedure frequency in the days that follow. The application also includes a heatmap that highlights statistically significant links between temperature and specific hospital–procedure combinations. A demonstration analysis is included to show how the tool works and how it helps identify temperature-related patterns.