An Expert Knowledge Algorithm and Model Predicting Wound Healing Trends for a Decision Support System for Pressure Injury Management in Home Care Nursing: Development and Validation Study.

The algorithm was tested on 12 records-based clinical vignettes, with 4 vignettes per round, rather than in prospective patient care. Only one expert nurse rated the algorithm and expected healing, so other experts may not agree. The study reported algorithm agreement and prediction coverage, not patient healing outcomes. The prediction model used 76 records from 18 pressure injuries, and records from the same pressure injury were not split into training and test data, which may make performance look better than it is. It was developed for nonterminally ill older patients at home and excluded lower limb injuries. The 90% prediction intervals included actual changes in 23 of 29 assessments, so the model underestimated uncertainty.

This work was supported by the Japan Society for the Promotion of Science Grants-in-Aid for Scientific Research. The authors declared no conflicts of interest.

Do not read this as proof that the algorithm improves pressure injury healing in home care. It was tested on 12 vignettes and 18 pressure injuries, and no patient outcomes were measured.