Benchmarking performance of an automatic polysomnography scoring system in a population with suspected sleep disorders.

Aim: The current gold standard for measuring sleep disorders is polysomnography (PSG), which is manually scored by a sleep technologist. Scoring a PSG is time-consuming and tedious, with substantial inter-rater variability. A deep-learning-based sleep analysis software module can perform autoscoring of PSG. The primary objective of the study is to validate the accuracy and reliability of the autoscoring software. The secondary objective is to measure workflow improvements in terms of time and cost via a time motion study. Methodology: The performance of an automatic PSG scoring software was benchmarked against the performance of two independent sleep technologists on PSG data collected from patients with suspected sleep disorders. The technologists at the hospital clinic and a third-party scoring company scored the PSG records independently. The scores were then compared between the technologists and the automatic scoring system. An observational study was also performed where the time taken for sleep technologists at the hospital clinic to manually score PSGs was tracked, along with the time taken by the automatic scoring software to assess for potential time savings. Results: Pearson's correlation between the manually scored apnea-hypopnea index (AHI) and the automatically scored AHI was 0.962, demonstrating a near-perfect agreement. The autoscoring system demonstrated similar results in sleep staging. The agreement between automatic staging and manual scoring was higher in terms of accuracy and Cohen's kappa than the agreement between experts. The autoscoring system took an average of 42.7 s to score each record compared with 4,243 s for manual scoring. Following a manual review of the auto scores, an average time savings of 38.6 min per PSG was observed, amounting to 0.25 full-time equivalent (FTE) savings per year. Conclusion: The findings indicate a potential for a reduction in the burden of manual scoring of PSGs by sleep technologists and may be of operational significance for sleep laboratories in the healthcare setting.

Findings from a decade of experience following implementation of a Rapid Response System into an Asian hospital.

Aim: Rapid response systems (RRS) are present in many acute hospitals in western nations but are not widely adopted in Asia. The influence of healthcare culture and the effect of implementing an RRS over time are infrequently reported. We describe the introduction a RRS into a Singaporean hospital and the barriers encountered. The efferent limb activation rates, cardiac arrest rates and unplanned intensive care unit (ICU) admissions are trended over eleven years. Methods: We conducted a retrospective observational study using prospectively collected data derived from administrative and Medical Emergency Team (MET) databases. Results: The RRS used a MET with a single parameter track and trigger and physician led efferent limb. Barriers encountered included clinical leadership buy-in, assembling and equipping the efferent team, maintaining a non-punitive mindset, improving accessibility to MET and communicating the impact of the MET. Over an 11-year period with 488,252 hospital admissions, MET activation rates increased from 1.6/1000 admissions (2009) to 14.1/1000 admissions (2019). Code blue activations and unplanned ICU admission rates decreased from 2.9 to 1.7 and from 8.8 to 2.0/1000 admissions, respectively over the 11 years. There were associations between increasing MET activation rate and reduction in code blue activations (p = 0.013) and unplanned medical ICU admission rates (p = 0.001). Conclusion: Implementing, sustaining and continued improvement of an RRS in Singapore is possible despite challenges encountered. With increasing activation rates over a decade, there were reductions in cardiac arrest rates and unplanned medical ICU admissions.

Non-alcoholic fatty liver disease screening in type 2 diabetes mellitus: A cost-effectiveness and price threshold analysis.

INTRODUCTION: The cost-effectiveness of screening asymptomatic non-alcoholic fatty liver disease (NAFLD) patients remains debatable, with current studies assuming lifelong benefits of NAFLD screening while neglecting cardiovascular outcomes. This study aims to assess the cost-effectiveness of NAFLD screening among type 2 diabetes mellitus (T2DM) patients, and to establish a price threshold for NAFLD treatment, when it becomes available. METHOD: A Markov model was constructed comparing 4 screening strategies (versus no screening) to identify NAFLD with advanced fibrosis among T2DM patients: fibrosis-4 (FIB-4), vibration-controlled transient elastography (VCTE), FIB-4 and VCTE (simultaneous), and FIB-4 and VCTE (sequential). Sensitivity analyses and price threshold analyses were performed to assess parameter uncertainties in the results. RESULTS: VCTE was the most cost-effective NAFLD screening strategy (USD24,727/quality-adjusted life year [QALY]), followed by FIB-4 (USD36,800/QALY), when compared to no screening. Probabilistic sensitivity analysis revealed a higher degree of certainty for VCTE as a cost-effective strategy compared to FIB-4 (90.7% versus 73.2%). The duration of expected screening benefit is the most influential variable based on incremental cost-effectiveness ratio tornado analysis. The minimum duration of screening benefit for NAFLD screening to be cost-effective was at least 2.6 years. The annual cost of NAFLD treatment should be less than USD751 for NAFLD screening to be cost-effective. CONCLUSION: Both VCTE and FIB-4 are cost-effective NAFLD screening strategies among T2DM patients in Singapore. However, given the lack of access to VCTE at primacy care and potential budget constraints, FIB-4 can also be considered for NAFLD screening among T2DM patients in Singapore.