Abstract: A disease network construction method includes the steps of: (1) organizing cohort data in time series; (2) stratifying or grouping the data organized in step (1) by confounding variables; (3) deriving a correlation of diseases within the stratification in step (2); and (4) constructing a disease network on the basis of the correlation derived in step (3). According to the present invention, a disease may be influenced by a variety of clinical and medical confounding variables such as age, gender, race, socioeconomic variables, and regional and national health care systems, and thus a method for more reliably deriving a correlation between diseases may be provided.