Personalized Pressure Injury Prevention Planning: Clinical Practice Guidelines, The Electronic Health Record And Big Data
Kath M. Bogie1, M Kristi Henzel2, GQ Zhang3, Steven Roggenkamp3, Jiayang Sun1, Arielle Bloostein1, Jacinta M. Seton2, Youjun Li1, Mary Ann Richmond2, Monique Washington2.
1Case Western Reserve University, Cleveland, OH, USA, 2Louis Stokes Cleveland VA Medical Center, Cleveland, OH, USA, 3University of Kentucky, Louisville, KY, USA.
Background: Clinical practice guidelines (CPG) aid clinicians in pressure injury (PrI) prevention. However, there is limited guidance on prioritization. This can be overwhelming and impractical to address, negatively impacting care and intervention planning. Effective clinical tools to prioritize the multiple recommendations in the CPG has been identified as a need by experts in the field. Methods: Bioinformatics enables data extraction, storage, and analysis for clinical decision support and user-interface development for complex clinical challenges such as PrI. Electronic health records (EHR) provide a rich information resource. Over 200 ICD-9 codes related to PrI risk factors were identified. EHR data over a 5 year period was collated for 36,626 Veterans with spinal cord injury (SCI). Natural language processing (NLP) was used to parse and analyze clinical narratives. Results: Hierarchical clustering and network analysis indicate that neurogenic bowel and bladder are the comorbidities most strongly associated with PrI development. Smoking and diabetes are lower tier risk comorbidities. Over 6 million text notes were analyzed using NLP. Many risk factors were recorded in the clinical notes but not coded. Conclusion: Understanding the relative impact of comorbidities is important for primary and secondary PrI planning for persons with SCI. The clinical notes provide a rich alternative source of clinical information related to PrI risk complementing ICD-9 coding. Systemic analysis of PrI risk factors recording in the EHR by ICD coding and clinical notes enable development of a personalized care planning tool. Each individual’s risk factor profile can provide the basis for adaptive personalized PrI prevention care planning based on CPG prioritization.
Back to 2019 Abstracts