Wolters Kluwer Health, Waltham, Mass, will integrate its POC Advisor real-time clinical surveillance and analytics platform with the Natural Language Processing (NLP) solution from Hiteks Solutions Inc, New York, for inclusion of structured and unstructured documentation to make them immediately actionable.
The collaboration will make possible real-time analysis of physician notes about infection sources, chronic comorbidities, and other clues to the presence of sepsis. Doing so will improve the efficacy of sepsis alerts and patient-specific treatment advice that POC Advisor delivers to clinicians at the point of care.
“The advanced algorithms and NLP built into Hiteks’ Insight platform parses clinical notes in less than one second to identify infection sources from real-time [electronic health record] data feeds, allowing us to build upon the considerable success POC Advisor has already achieved in our quest to improve sepsis outcomes,” says Sean Benson, vice president and general manager of specialized surveillance at Wolters Kluwer Clinical Software Solutions. “Inclusion of clinical documentation, such as progress, consult, and admit notes, which are often unstructured, as well as nursing assessments, pathology, and radiology reports, enhances our prescriptive analytics and helps bridge clinical data silos that hinder clinical decisions and patient care. This enhances the speed and accuracy of the alerts and actionable information generated by POC Advisor.”
POC Advisor aggregates, normalizes, and analyzes patient data from disparate clinical systems to deliver timely and accurate sepsis alerts and advice. Hundreds of rules built into the platform account for possible comorbidities and medication abnormalities, and trigger alerts that have demonstrated high levels of sensitivity (95%) and specificity (82%).1 POC Advisor was also found to reduce sepsis mortality and sepsis-related 30-day readmissions by 53% and 30%, respectively.
“Our goal with Insight: VigilantQA solution is to enable efficient information extraction matched to clinical workflow, so that clinicians do not need to change how they document or work,” says Gerry Petratos, MD, CEO of Hiteks Solutions. “Rather, we help bring attention to the specific patient needs and timing of intervention for quality conditions like sepsis, so that guidelines can be followed and cases can be reported on with ease. Our NLP solution has a high predictive value for sepsis, severe sepsis, and septic shock, which aligns with the objectives Wolters Kluwer has established for POC Advisor.”
The power of the integrated platform derives from Hiteks’ quality and reimbursement-focused NLP, initially leveraged for sepsis, along with POC Advisor’s prescriptive analytics, to improve and accelerate diagnosis. Health systems and providers also benefit from improved clinical documentation and core measures abstracting functionality, which reduces the resources required to exceed the 75% core measures threshold and improves specificity of ICD-10 claims codes, optimizing revenue cycle.
- Manaktala S, Claypool SR. Evaluating the impact of a computerized surveillance algorithm and decision support system on sepsis mortality. J Am Med Inform A. 2017;24(1):88–95; doi: 10.1093/jamia/ocw056.