Methodology for Big Data Analysis Using Data from National Health Insurance Service: Preliminary Methodologic Study and Review about the Relationship between Sinus Surgery and Asthma

Abstract

Background and Objectives：Sinus surgery has been reported to improve pulmonary function and decrease the use of asthma medications in patients with chronic rhinosinusitis and asthma. Most studies, however, have used a small number of patients and were conducted over a short period. To demonstrate a causal relationship between sinus surgery and asthma, a suffi cient amount of patient data observed over a long period is required. To address the limitations of the existing approaches, we conducted a preliminary methodological study for large-scale data analysis using data from the National Health Insurance Service (NHIS) to suggest a basis for the effect of sinus surgery on asthma. Materials and Methods：The data from NHIS consisted of unidentified medical histories of a sample cohort representing the whole nation over a period of nine years. We select ed the following types of study samples: 1) patients with surgical codes for sinus surgery; 2) patients with disease codes for sinusitis; 3) patients with disease codes for asthma; and 4) patients with medication codes for asthma treatment. Results：In this study, we applied a methodology for selection of subjects from big data to investigate the effect of sinus surgery on improving asthma in the future. We could include 152 subjects after the four-stage selection method from 1,025,340 patients. Conclusion： We could establish a method to select patients with history of sinus surgery and asthma treatment from a big data. This methodology using big data may contribute to identify relationship between sinus surgery and asthma treatment in the future.