Non-obvious correlations to disease management unraveled by Bayesian artificial intelligence analyses of CMS data

Artificial Intelligence in Medicine. 2016 Nov;74:1-8. doi: 10.1016/j.artmed.2016.11.001

Vijetha Vemulapalli, Jiaqi Qu, Jeonifer M. Garren, Leonardo O. Rodrigues, Michael A. Kiebish, Rangaprasad Sarangarajan, Niven R. Narain, Viatcheslav R. Akmaev

Objective: Given the availability of extensive digitized healthcare data from medical records, claims andprescription information, it is now possible to use hypothesis-free, data-driven approaches to mine medi-cal databases for novel insight. The goal of this analysis was to demonstrate the use of artificial intelligencebased methods such as Bayesian networks to open up opportunities for creation of new knowledge inmanagement of chronic conditions.

Materials and methods: Hospital level Medicare claims data containing discharge numbers for mostcommon diagnoses were analyzed in a hypothesis-free manner using Bayesian networks learningmethodology.

Results: While many interactions identified between discharge rates of diagnoses using this data setare supported by current medical knowledge, a novel interaction linking asthma and renal failure wasdiscovered. This interaction is non-obvious and had not been looked at by the research and clinicalcommunities in epidemiological or clinical data. A plausible pharmacological explanation of this link isproposed together with a verification of the risk significance by conventional statistical analysis.

Conclusion: Potential clinical and molecular pathways defining the relationship between commonly used asthma medications and renal disease are discussed. The study underscores the need for further epi-demiological research to validate this novel hypothesis. Validation will lead to advancement in clinical treatment of asthma & bronchitis, thereby, improving patient outcomes and leading to long term costsavings. In summary, this study demonstrates that application of advanced artificial intelligence meth-ods in healthcare has the potential to enhance the quality of care by discovering non-obvious, clinicallyrelevant relationships and enabling timely care intervention.

Learn more: [weblink]