Comparative, cross-sectional study describing agreement and accuracy of emergency centre triage using either a mobile application or manual triage
Date
2016-03
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Stellenbosch : Stellenbosch University
Abstract
ENGLISH SUMMARY: Background and Purpose: Unacceptable patient waiting times and misidentification of critically ill patients are significant problems within Emergency Centres (ECs). Triage, when performed correctly, is acritical process to address this. The South African Triage Scale (SATS) is widely used. Manually performed triage may be prone to inaccuracies and prolonged triage time. A mobile tablet application to facilitate use of the SATS by automating triage calculation and guiding nurses has been developed.
Methods: This is a comparative cross-sectional study to observe the accuracy of triage using the mobile
tablet application compared with triage performed manually. Under classroom examination conditions,
nurses calculated triage scores on written case scenarios of typical EC presentations. A total of 59 nurses across five hospitals in the Western Cape Province, South Africa were randomized into an ‘app group’ and a ‘manual group’. Results: The app group scored a 23% higher level of agreement with the expert-validated results than the manual group. Kappa of 0.735 (0.719 - 0.770) and 0.597 (0.545 - 0.656) were found respectively. One in five patients are triaged more correctly using the app. Sensitivity for emergency cases was 65.5% and 53% respectively, with and without the app. Conclusions: Nurses triaging written scenarios with the aid of the app were observed to have a higher agreement with the expert-validated results than nurses performing traditional manual triage. The effect size is considerable and of practical relevance. The app could have significant benefit in busy Emergency Centres of public sector hospitals. A larger study involving real patients is recommended.
AFRIKAANSE OPSOMMING: Geen opsomming beskikbaar.
AFRIKAANSE OPSOMMING: Geen opsomming beskikbaar.
Description
Thesis (MSc)--Stellenbosch University, 2016.
Keywords
Decision support systems, Artificial intelligence, Triage (Medicine), UCTD