During their investigations, the researchers led by Peter Klimek and Stefan Thurner documented a total of 1,642 conditions, spanning the entire spectrum from A to Z. "Our results enable us to produce fairly accurate "disease demographics" for Austria," explains Thurner. "It's possible to see very clearly, for example, what other diseases, with what sort of probability and even at what stage of their lives 25-year-old diabetics, for example, will develop in ten years' time."
The results have enabled researchers to create a mathematical model to predict how great the future risk of illness is for each individual disease in different sections of the population, depending on the patient's age and gender.
In a further study, Klimek and Thurner joined forces with Alexandra Kautzky-Willer from the Gender Medicine Unit at the Medical University of Vienna to investigate the extent to which personalised disease risks for diabetic patients differ from those for the rest of the population. The researchers were able to identify more than a hundred "disease pairings", known ones and also less well-known ones, allowing them to confirm, for example, a previously disputed connection between diabetes and Parkinson's disease. The risk of this is more than twice as high. Moreover, the risk of coronary heart disease is seven times higher, while the risk of the lung condition COPD is three times higher and the risk of depression five times higher.
This could lead to doctors treating patients who now have this documented risk score offering them personalised, preventative treatment in future or preventively "investigating" possible follow-on conditions that may not actually be considered in the first instance. Says Thurner: "If diabetes is associated in rare cases with sleep disturbance, for example, the doctor could nevertheless preventively ask the patient whether or not he is sleeping badly in order to pre-emptively head off any severe sleep disturbance." The risk of developing sleep disturbance later on is twice as high for young diabetic patients, for example.
Costs predictable for the healthcare system
At the same time, this data allows Austria's healthcare politicians to predict very accurately for the first time just what costs the healthcare system, statistically, is likely to incur in the future and where exactly it would be useful to begin prevention programmes. Says Klimek: "But this is of course assuming that nothing changes in relation to Austria's fundamental quality of medical care."
Service: New Journal of Physics
„Spreading of diseases through comorbidity networks across life and gender.“ Anna Chmiel, Peter Klimek, Stefan Thurner. New Journal of Physics 16 (2014) 115013
PloS Computational Biology
„Quantification of diabetes comorbidity risks across life using nation-wide big claims data.” Peter Klimek, Alexandra Kautzky-Willer, Anna Chmiel, Irmgard Schiller-Frühwirth, Stefan Thurner. PloS Computational Biology (2015).