Predicting acute myocardial infarction (AMI) 30-days mortality: Using standardised mortality ratio (SMR) as the hospital performance measure

• Published in: International journal of medical informatics (Shannon, Ireland) Ireland), 2022-12, Vol.168, p.104865-104865, Article 104865
• Main Authors: Kamarulariffin Kamarudin, Mohd, Tan Jen Ai, Calyn, Lisa Zaharan, Nur, Yahya, Abqariyah
• Format: Article
• Language: English
• Subjects: Acute myocardial infarction; Funnel plot; In-hospital mortality; Standardised mortality; Variable life-adjusted display
• ISSN: 1386-5056; 1872-8243
• DOI: 10.1016/j.ijmedinf.2022.104865

•A funnel plot with an overdispersion limit could identify outlier SMRs with higher confidence levels.•A VLAD with Vmask signals aids in the identification of performance deviation from averages.•AMI outcomes in hospitals without PCI facilities can be comparable to the referral centres. A standardised mortality ratio (SMR) is the hospital observed mortality divided by its predicted mortality and has been used as an indicator to monitor hospital performance. This study developed a model that predicted 30-day mortality for acute myocardial infarction (AMI) and compared the SMR among 41 Malaysian public hospitals using statistical process control charts. Data from referral centres and specialist hospitals with cardiology services were analysed. Both referral centres and specialist hospitals had comparable mortality, except for Hospitals A and B, which the study considered outliers. Two-thirds of the remaining hospitals had an SMR of above one (SMR 1.05–1.51), but the indices were still within the expected variations. The SMR coupled with a funnel plot and variable life adjusted display (VLAD) can identify hospitals with potentially higher than expected mortality rates.