Triaging patients into and away from preoperative assessment clinics remains a challenge. Anaesthesia Preoperative Assessment Tool (APAT) is a web application that delivers an online 22 question survey to patients at home, and uses an artificially intelligent algorithm to stratify patient risk and identify the need for non routine preoperative investigation and intervention. We assess APATs accuracy and patient acceptability in this prospective observational study. Patients were recruited at preoperative assessment clinic, where they were assessed by a consultant anaesthetist. Anaesthetist (ASA) grade, need for nonstandard investigation and intervention were recorded (gold standard). Patients were invited to complete an APAT assessment on their PC or smartphone at home, and the results of both assessments compared. 22 patients completed conventional clinical assessment by consultant anaesthetist and online assessment by APAT. APAT score correlates with clinicians ASA grade (rτ=0.6075, p=0.0008). APAT predicts patient risk group (misclassification rate of 0%, Area Under the Curve (AUC)=0.9825). APAT predicts the need for additional investigation (AUC=0.8077) and preoperative intervention (AUC=0.7193). Online assessment was acceptable to 92% of patients. Our findings support the hypothesis that APAT accurately predicts patients perioperative risk and predicts the need for investigation and intervention. Further studies are needed to confirm that APAT may be used to identify ASA 1 and 2 patients who could safely bypass preoperative assessment clinic.
- assistive technology
- medical apps
- preoperative assessment
Statistics from Altmetric.com
Contributors JO: first Author, built website, designed study protocol, obtained ethical approval, obtained funding from INE Academic Track Program, primary data collector, statistician, wrote paper. MaC: provided consulant anaesthetist assessments of paitents in POAC. SM: provided consulant anaesthetist assessments of paitents in POAC. MeC: consented patients, study protocol design and write up. GC: departmental author, study protocol design, write up.
Funding This project was funded and supported by the Health Service Executive INE Academic Track Programme (restricted grant). The funding was used to pay for application server hosting only, no other expenses were incurred.
Competing interests None declared.
Patient consent for publication Not required.
Ethics approval Ethical approval was granted by the HSE Mid-Western Regional Hospital Research Ethics Committee.
Provenance and peer review Not commissioned; externally peer reviewed.
Data availability statement Data are available on reasonable request. Our dataset is available on reasonable request from the lead author. Requests for specific data shall be discussed with the University of Limerick Research Ethics Committee on a case by case basis prior to release of data.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.