Digitized clinical data analysis can help to improve patient care and increase medical quality in regional clinical systems. Specifically, it enables population health queries and clinical research and improves clinical processes through the secondary usage of the data available. In many regions, clinical pathways help to shorten hospital stays and reduce redundant treatments. Here, the authors use the available data to improve clinical pathways.
The paper presents an optimization technique based on k-means clustering to evaluate optimal clinical pathways on digitally available data. According to the authors, clinical pathways are usually developed at “multi-department expert meetings,” based on the statistics available. The proposed optimization technique would replace these meetings, while also providing a timely answer for continuous improvements.
The results are encouraging. For example, in the presented experiment, average medical expenses decreased. However, since the datasets used are not available, it is difficult to assess whether the results are generic or tailored to the authors’ region.
Indeed, the idea of using statistics and neural networks to improve clinical pathways is a worthwhile read for policymakers. Once data is standardized in a region--for example, via the Nationwide Health Information Network (NHIN) or Carequality in the US--the result is more precise statistics. The paper’s audience includes policymakers and welfare services providers interested in improving regional pathways.