Does improved patient care lead to higher treatment costs? A multicentre cost evaluation of a blunt chest injury care bundle.

Background: Blunt chest injury is associated with significant adverse health outcomes. A chest injury care bundle (ChIP) was developed for patients with blunt chest injury presenting to the emergency department. ChIP implementation resulted in increased health service use, decreased unplanned Intensive Care Unit admissions and non-invasive ventilation use. In this paper, we report on the financial implications of implementing ChIP and quantify costs/savings.

Methods: This was a controlled pre-and post-test study with two intervention and two non-intervention sites. The primary outcome measure was the treatment cost of hospital admission. Costs are reported in Australian dollars (AUD). A generalised linear model (GLM) estimated patient episode treatment costs at ChIP intervention and non-intervention sites. Because healthcare cost data were positive-skewed, a gamma distribution and log-link function were applied.

Results: A total of 1705 patients were included in the cost analysis. The interaction (Phase x Treatment) was positive but insignificant (p = 0.45). The incremental cost per patient episode at ChIP intervention sites was estimated at $964 (95 % CI, -966 - 2895). The very wide confidence intervals reflect substantial differences in cost changes between individual sites Conclusions: The point estimate of the cost of the ChIP care bundle indicated an appreciable increase compared to standard care, but there is considerable variability between sites, rendering the finding statistically non-significant. The impact on short- and longer-term costs requires further quantification.