Costs and Benefits of the Melbourne Mobile Stroke Unit Compared With Standard Ambulance: Causal Analysis Using Observational Linked Data

• DOI: 10.1161/strokeaha.124.048403
• Publication Date: 2025-03-24T18:08:48Z
• Authors: Dominique A. Cadi...; Mulugeta Molla Bi...; Leonid Churilov; Henry Zhao; Skye Coote; Bruce C.V. Campbell; Francesca Langenberg; Stephen M. Davis; Geoffrey A. Donnan; Karen Smith; Joosup Kim

BACKGROUND: Evidence of the cost implications and health outcomes associated with the use of mobile stroke units (MSU) is required to support their utilization. We aimed to evaluate the causal effect of the use of an MSU compared with a standard ambulance on hospitalization costs and 90- to 180-day health outcomes.

METHODS: Causal effect estimation was performed using patient-level data from a cohort of patients with stroke in 2018 identified from the Australian Stroke Clinical Registry (Victoria) and Melbourne MSU. These data were linked to Ambulance Victoria and government-held administrative data sets. In total, linked data from 8657 patients were available. Propensity score matching was used to define comparator groups within a target trial framework. Costs included emergency department and hospital admission costs in the first 180 days after stroke. Multivariable regression analyses of the matched data were used to compare costs and outcomes (mortality and modified Rankin Scale) between MSU and standard ambulance groups.

RESULTS:

The target trial sample included 96 patients transported by the MSU (intervention) and 198 patients transported by standard ambulance services (control). Of these, the mean age was 76 years and 157 (53%) were men. A greater proportion of patients received mechanical thrombectomy in the intervention group than the control group (40% versus 23%; P <0.001). The adjusted hospital costs were $17 949 greater in the intervention group than the control group (95% CI, $4682–$31 214; P =0.01). Patients in intervention group doubled the odds of achieving nondisability (modified Rankin Scale scores of 0–1, adjusted odds ratio of 2.11 [95% CI, 1.07–4.18]) and halved the mortality rate (adjusted hazard ratio, 0.53 [95% CI, 0.32–0.86]) within 90 to 180 days poststroke compared with the control group.

CONCLUSIONS: There are important cost implications and improved outcomes from using the MSU that are likely related to increased provision of reperfusion therapy.