| 221 |
Clinical Research
|
Using real clinic data to develop benchmarks for performance improvement Sarah Brown, BSc(Hons), MSc, CSTAT, Peter Cooper BSc(Hons), CSTAT, Smith & Nephew Wound Management, UK; Roberto Penne, MD, Medical Multiplex, Inc., Louisville, KY Problem.Without an understanding of outcomes, wound clinics have no sound basis for reviewing and improving performance. Using real clinical data requires a large data set to make reasonable evaluations of outcomes. Since wound care is multi-factorial, one must use statistical analysis to make appropriate conclusions. Benchmarking has been shown as an effective tool for adopting best practices. It is a cyclical series of events which involves documenting information, self-assessment of performance and implementing improvements. Solution.Using a tracking application, a dataset of 4628 wounds, comprised of Venous Leg Ulcers (VLU) Diabetic Foot Ulcers (DFU) and Pressure Ulcers (PU) has been aggregated from a group of US clinics. Using this data we generated benchmarks for healing taking into account wound area at presentation.. .Results.The prognostic importance of wound area has been illustrated by generating benchmarks of healing accounting for wound area at presentation, using summary statistics, Kaplan-Meier plots and output from logistic regression models. These show that for VLUs; the percentage of wounds healed at week 12 decreases with an increase in baseline area; there is a clear separation in baseline wound area category for the percentage of wounds healed through time; and the probability of wound healing at week 12 decreases with an increase in baseline area. The clinic's performance was determined on percentage of VLUs healed at week 12 relative to average of all other clinics, after adjusting for baseline wound area and showed clear difference in the odds of healing among clinics. Conclusions.Benchmarks enable interpretation of results from clinical research (non-comparative clinical studies) and clinical practice (assessing a wound clinic's performance). Prognostic factors including ulcer area need to be taken into account whenever results are reported or interpreted A comprehensive set of patient, wound and treatment prognostic factors for the healing of Venous Leg Ulcers are recorded within tracking software including patient age and gender, wound duration, location of wound, exudate level, wound bed condition, debridement and treatments applied. Further work is ongoing to generate benchmarks that also take these factors into account. Montori VM, Guyatt GH. What is evidence-based medicine and why should it be practiced? Respir Care J. 2001;46(11):1201Ð1211 Brown S, Cooper P, Miller L. Using real clinic data to develop benchmark healing rates for venous leg ulcers. EWMA Symposium 2005, Session V41. |
|