State of the art in clinical decision support applications in pediatric perioperative medicine

Abstract

Purpose of review The goal of this review is to describe the recent improvements in clinical decision tools applied to the increasingly large and complex datasets in the pediatric ambulatory and inpatient setting. Recent findings Clinical decision support has evolved beyond simple static alerts to complex dynamic alerts for: diagnosis, medical decision-making, monitoring of physiological, laboratory, and pharmacologic inputs, and adherence to institutional and national guidelines for both the patient and the healthcare team. Artificial intelligence and machine learning have enabled advances in predicting outcomes, such as sepsis and early deterioration, and assisting in procedural technique. Summary With more than a decade of electronic medical data generation, clinical decision support tools have begun to evolve into more sophisticated and complex algorithms capable of transforming large datasets into succinct, timely, and pertinent summaries for treatment and management of pediatric patients. Future developments will need to leverage patient-generated health data, integrated device data, and provider-entered data to complete the continuum of patient care and will likely demonstrate improvements in patient outcomes.

Quality improvement in New Zealand pediatric anesthesia: National quality direction, patient experience, equity, and collaboration

“…An electronic medical record positions an institution well to be able to present real-time, personalized metrics to individual anesthetists. 24,25 Key performance indicator (KPI) audit of five of the anesthesia indicators from the Australian council of healthcare standards. 27 The audit assessed feasibility of collecting this data nationally to ultimately allow benchmarking.…”

The state of adoption of anesthesia information management systems in Canadian academic anesthesia departments: a survey

“…9 Recent reviews have found evidence to support the use of near real-time and real-time CDS in AIMS, especially in the areas of surgical antibiotic prophylaxis, documentation compliance, and postoperative nausea and vomiting. [10][11][12] Despite documented benefits, several barriers have prevented the universal adoption of AIMS including cost, lack of technical support, inadequate integration with an existing EHR, medicolegal concerns regarding missing or outlier data, the inconveniences of rapid documentation and electronic data entry during short or emergency procedures, lack of clinician involvement during design and implementation, and resistance to changes in clinical workflow patterns. [13][14][15][16][17] There is evidence that most of these barriers are either exaggerated or can be overcome by targeting areas such as perceived usefulness and ease of use.…”

Anesthesia in the modern world of apps and technology: Implications and impact on wellness

“…Clinical decision support tools have progressed beyond automated alerts to more advanced algorithms that incorporate large, complex data sets to generate dynamic real-time decision support. 20 In pediatric anesthesia and critical care, machine learning methods and augmented intelligence to automate diagnostic and treatment algorithms promise to elevate EHR/AIMS beyond a vehicle for data entry. 21 These tools may substantially reduce cognitive load, particularly at times of high workload and stress, thus improving workflow rather than hindering it.…”

“…Continued advances may improve usability, shifting EHR from a hindrance to an asset. Clinical decision support tools have progressed beyond automated alerts to more advanced algorithms that incorporate large, complex data sets to generate dynamic real‐time decision support 20 . In pediatric anesthesia and critical care, machine learning methods and augmented intelligence to automate diagnostic and treatment algorithms promise to elevate EHR/AIMS beyond a vehicle for data entry 21 .…”