Implementation and evaluation of a 24/7 virtual intensive care support model in regional Australia: A controlled quasi-experimental study.

BACKGROUND

The COVID-19 pandemic accelerated implementation of virtual care models to enhance rural intensive care capacity. However, rigorous evaluation of such models during routine critical care operations remains limited.

LOCAL PROBLEM: Two regional level 4 intensive care units (ICUs) in New South Wales, Australia required sustainable capacity enhancement to manage critically ill patients while addressing workforce limitations and geographic isolation.

METHODS

We implemented a multiphase virtual intensive care support model progressing from daily videoconference rounds to 24/7 remote monitoring by experienced ICU nurses with specialist physician backup. Using a controlled quasi-experimental design, we compared two intervention sites with two matched control sites across four time periods (January 2019-June 2022). Segmented interrupted time series and difference-in-differences analyses evaluated effectiveness (standardised mortality ratios, length of stay, and adverse events) and efficiency (admission rates and transfer rates). Economic evaluation followed Consolidated Health Economic Evaluation Reporting Standards (CHEERS) 2022 guidelines. Qualitative methods explored clinician and patient experiences following consolidated criteria for reporting qualitative research standards.

INTERVENTIONS: The virtual support model included continuous remote patient monitoring, daily multidisciplinary videoconference rounds, tiered escalation protocols, bidirectional staff education, and formal governance structures with defined accountability frameworks.

RESULTS

Standardised mortality ratios remained within 95% confidence limits throughout (range: 0.72-1.18). The interrupted time series analysis demonstrated significant level changes in transfers following 24/7 implementation (site 1: -2.8 transfers/month, p = 0.003; site 2: -2.1 transfers/month, p = 0.01). Difference-in-differences estimates confirmed intervention effects independent of concurrent care model changes. An economic analysis revealed cost-benefit ratios of 1:7.1 and 1:5.9 (per AUD invested). Twenty-one clinicians described enhanced collaboration, improved access to expertise, and increased confidence managing complex patients. Implementation fidelity exceeded 92% across monitored components.

CONCLUSIONS

Virtual intensive care unit support model represents a safe, cost-beneficial model for enhancing regional ICU capacity during routine operations. Benefits extend beyond pandemic response to sustainable improvements in rural critical care delivery. Implementation requires robust governance, reliable technology, dedicated staffing, and systematic fidelity monitoring.