Wound Risk Home
What is the Wound Risk Model
The Signature HealthCARE wound risk assessment model uses relevant skilled nursing facility resident data to predict the likelihood of each resident developing a wound. This model calculates the probability that one resident will acquire a wound and categorizes the probability as low, medium, or high risk. To use Signature HealthCARE’s predictive model, simply click the “Conduct the Assessment” button at the bottom of this page. There you will complete the wound risk assessment, which takes less than five minutes on average, and receive the model results for one resident.
DETAILS BEHIND THE MODEL
This study was a retrospective analysis of five years of Signature HealthCARE data in 115 skilled nursing facilities representing more than 97,000 admissions. After a literature search and interviews with subject matter experts, a logistic regression analysis was performed on 100 variables from the resident population to predict the presence or absence of a wound. The final model consists of 37 different variables and was tested on historical data to judge the goodness-of-fit to actual outcomes on residents of different age groups, diagnoses, facilities, and geographies. The likelihood ratio chi-square with a p-value of 0.0001 tells us that our model is a very good fit to the data and each of the variables used in our model are statistically significant.
Practical Model Use
Because effective interventions to reduce wounds for skilled nursing facility residents are often expensive and time-consuming to implement, a score to predict wounds will help stratify the resident population and target those most likely to benefit. Resident demographics, functional status, active diagnoses, and medications were retained in the final wound predictive model.