Aims: Delivering high-quality care in long-term care facilities (LTCF) is a persistent international challenge, with ongoing variation in outcomes. Benchmarking identifies under-performance but offer limited guidance on what level of improvement is realistically achievable and through which levers. This study estimated potential improvements in LTCF resident outcomes if modifiable risk factors changed under realistic, expert-informed scenarios.
Methods: A population-based simulation study using the Registry of Senior Australians National Historical Cohort was conducted, including 53124 LTCF entrants (≥65 years) of 2367 LTCFs in four states in 2019. Risk-adjusted quality indicators of fractures and emergency department (ED) presentations within 365 days were examined. Established Fine-Gray competing-risk models were refitted and simulated expert-informed “what-if” scenarios, including changes in medication-related factors (less polypharmacy, sedative load, proton pump inhibitors, selective serotonin reuptake inhibitors), and a provider-level proxy for best-practice care (more people accessing government-run LTCF). Expert scenarios were constrained to factors realistically modifiable under existing intervention guidelines and best-practice models. Absolute changes in predicted annual events were estimated.
Results: Predicted baseline events were 2590 fractures and 22472 ED presentations annually. Under the expert scenarios, fractures changed by -2.1% (-56 events/year, 95%CI=-96;-22), and ED presentations by -1.4% (-304 events/year; 95%CI=-555;-28). Medication-related factors accounted for reductions in ED presentations, whereas increasing the share of care provided within government-run LTCF contributed to improvements across both outcomes.
Conclusions: Modest but achievable improvements in LTCF outcomes may be realised by optimising medication use and strengthening best practice care models. Simulation-based estimates complement benchmarking by providing evidence-informed targets for quality improvement.