Predicting Patient Outcomes Via the Wound Healing Index
Virtually every other specialty has developed a risk stratification for its most common, serious conditions. Risk stratification is a corrective tool that enables the reporting of patient outcome in a fair way so that physicians caring for the sickest patients are not penalized by appearing to have worse outcomes than their peers who care for less sick patients. Anesthesiology has a risk classification that is important for evaluating whether a patient should even undergo anesthesia and to understand anesthetic complications in relation to their predicted likelihood. Trauma surgeons, cardiologists and oncologists use risk classification to help patients and families understand the likelihood of patient survival in high mortality situations. Stratification also serves as way to demonstrate clinical excellence and the impact of certain therapies. Unless you know what was predicted to happen, it’s not possible to demonstrate the impact of superior care or a novel therapeutic agent. Unfortunately, since the field of wound care has lacked a risk stratification system, clinicians, hospitals, and wound center management companies have used a different method to depict wound healing rates – fabrication.
That’s why we created the Wound Healing Index (WHI): to act as a predictive model for patient outcomes based on real-world clinical data.
The USWR collaborated with Dr. Susan Horn of the Institute for Clinical Outcomes Research (ICOR) to create such a model. The project was funded through ICOR by KCI which contributed $500,000. We had hoped to create one model that worked for all wounds and but in the end, we developed 7 models, one for each major ulcer category, which we called the Wound Healing Index (WHI). It was painstaking work, analyzing the structured data from almost 70,000 wounds, identifying individual factors that were associated with failure to heal, and then creating models from those factors.
We now have a predictive model for many wound variations and their related conditions based on real-world data, rather than clinical trials of sample sizes a fraction of the size and using subjects with no related conditions.