A large-scale analysis of electronic hospital records, that sheds light on how diseases commonly progress, is reported in Nature Communications this week. The study, which incorporated over 100 million datasets covering all hospital encounters in Denmark over a 15 year period, could help uncover previously unknown links between diseases or inform the design of future clinical trials.
Population-wide analyses of healthcare data are challenging and thus often focus on a specific group of diseases or comparatively short time scales.
Soren Brunak and his team performed the biggest analysis of this kind so far, working with 15 years of data from the Danish National Patient Registry, which covers all hospital encounters of the entire Danish population from 1996 to 2010. They then grouped over 100 million individual diagnoses into so-called ‘disease trajectories’, which describe the likelihood of one diagnosis, or disease, leading to another one.
The resulting 1,171 disease trajectories provide a comprehensive global picture of the most common diseases and their clinical development-for example from angina to chronic ischemic heart disease to cardiac arrest-and may be useful to clinicians, epidemiologists, as well as ‘big data’ researchers.
Technology: Soft robots catch more flies with magnetsCommunications Materials
Technology: Self-driving cars drive more safely with new algorithmNature Machine Intelligence
Ecology: Fast-growing trees die young and could affect carbon storageNature Communications
Epidemiology: US COVID-19 cases may be substantially underestimatedNature Communications