Integrating Contextual Factors into Clinical Decision Support to Reduce Contextual Error and Improve Outcomes in Ambulatory Care

点击翻译按钮获取中文摘要

Project Summary/Abstract A patient contextual factor refers to a patient circumstance or behavior that is essential to address when planning effective care. For instance, a patient’s inability to pay for costly medication is a contextual factor when presenting with poor medication adherence, particularly when a less costly alternative is available. Inattention to contextual factors results in contextual errors when care plans are inappropriate because of inattention to patient context. Contextual errors are associated with diminished health care outcomes and with overuse and misuse of medical services. Contextual errors and their adverse consequences may be averted if physicians are provided and prompted to use contextual information at the point of care. The aim of the proposed research, submitted in response to Special Emphasis Notice (SEN) NOT- HS-16-015, is to assess the potential of clinical decision support (CDS) enhanced with patient contextual information to reduce contextual error rates, improve health care outcomes and reduce unnecessary care. The first phase of the study consists of designing and introducing a passive and active CDS intervention informed by contextual information provided by patients through a web-portal as well from the electronic medical record, in two widely employed EMRs, Cerner and Epic. In the second phase, patients will be randomized to receive care either with or without enhanced contextualized CDS. To assess the efficacy of the intervention on physician behavior during the encounter, participating patients will audio record their visits, and the data will be audio and EMR coded using the validated contextual error tracking system, Content Coding for Contextual Error, or “4C.” 4C tags each contextual factor and classifies the encounter as contextualized when contextual factors are addressed, and containing a contextual error when not addressed. Four to six months following the index visit a structured chart review will assess outcomes as defined by response to the clinical presentation at the index visit. For instance, resolution of an elevated HgB A1c in a patient whose care was informed by contextualized CDS indicating that the patient required pre-filled syringes because of difficulty dosing their insulin, would constitute a favorable response. In addition, unannounced standardized patients (USPs), who are actors trained to portray standardized scripts customized to assess response to the intervention, will present with cases containing complicating contextual factors that if overlooked result in overuse and misuse of medical services, to assess the impact of the intervention on reducing unnecessary care. This study, which will recruit 480 patients and engage 20 USPs in 80 visits, has sufficient power to detect clinically meaningful reductions in contextual error rates, improvement in health care outcomes and cost reductions following the introduction of contextualized CDS.