Main Article Content
Objectives: The primary aim of this quality improvement project was to develop, implement and evaluate an automated bed assignment algorithm (ABAA) which can offer objective and consistent bed assignment recommendations that comply with the unique operational constraints and prioritization rules of a tertiary hospital in Singapore.
Methods: Using the classical process improvement framework of Plan-Do-Study-Act (PDSA), the quality circle workgroup first developed and tested the ABAA prototype to confirm its feasibility and reliability to meet all hospital operational constraints and prioritization rules. PDSA framework was then also employed in the user interface design and integration of ABAA into existing system setup. The staff satisfaction level of the ABAA was subsequently assessed via an anonymized online survey.
Results: In the prototype development phase, the workgroup was able to conclude after nine rounds of review meetings that the ABAA prototype was able to perform bed assignments like hospital staff using data in 64 operational scenarios. Among the 10 eligible staff who completed the online survey, up to 90% of them reported that ABAA was able to generate bed assignment recommendations which met the hospital operational requirements. 90% of these staff also reported that ABAA was easy to use and navigate, while all respondents reported using ABAA before attempting to assign beds manually. 80% of staff felt ABAA was able to reduce human error, while 50% of staff felt ABAA had reduced their time taken for bed assignments by 30 minutes to 2 hours per shift.
Conclusions: Evidently, the user-centric design of ABAA has enabled its high adoption and acceptance rate among staff. Overall, it has allowed the staff to make faster, consistent and objective bed assignment decisions which complied with hospital operational constraints and prioritization rules so that newly admitted patients received the most appropriate care at their point of admission.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.