Background: Estimating burden of disease, injuries and risk factors is crucial for health policy decision making. The Burden of Diseases (BoD) studies provide data about the magnitude and distribution of health problems among the population at national and sub-national levels. The BoD studies are designed to use secondary data for estimating prevalence and incidence of diseases, injuries and risk factors. However, due to the scarcity of data sometimes it becomes unavoidable to collect data from medical records. Among all needed source of data, including surveys, registries, censuses, inpatient and outpatient data, hospital data are an essential source for BoD studies. Hospital Data Survey (HDS) aims to estimate the prevalence and incidence of diseases and injuries that led to admission to hospitals. This paper aims to describe the required steps for data gathering, sampling, analytical methods, and other needed procedures for HDS.
Study design: The designed questionnaire includes demographic data, current health status, diseases, injuries and co-morbidities with their ICD10 codes, curative procedures, and treatment. A pilot study was conducted on 302 medical records from 6 hospitals to evaluate the validity and reliability of the questionnaire. Sampling frame was designed and probability proportional was used after being tested in the pilot study. In the next step, we will collect 367500 medical files from 863 hospitals (0.5% of all inpatient records in hospitals from1996 - 2013). The HDS is the first national study in Iran that is gathering data through an online-offline web-based system based on electronic version of the questionnaire which makes the process of data cleaning and analyses more comfortable.