Predictors of interhospital transfer delays in acute surgical patient deaths in Australia: a retrospective study.

Background: Interhospital transfers in Australia facilitate access to acute surgical services, however transfer delays can occur. The aims of this study were to examine Australian mortality audit data on acute surgical patients who were transferred after presenting with a surgical emergency, and to identify modifiable predictors of transfer delay. Methods: Surgical admissions between 1 January 2001 and 18 August 2020 were retrospectively extracted from the Australian and New Zealand Audit of Surgical Mortality database. Relevant factors and themes of interest were collated. Results were presented as odds ratios (OR) and 95% confidence intervals (CI), with statistical significance defined as P <0.05. Results: After exclusion, a final 8270 cases were analysed. Non‐modifiable predictors identified were female gender (OR 1.34, 95% CI 1.05–1.70, P = 0.0184), comorbidities (OR 1.50, 95% CI 1.40–161, P <0.0001) and major non‐trauma non‐vascular specialty (OR 1.54 to 7.77, depending on specialty, P < 0.05). Modifiable predictors were inadequate clinical assessment (OR 49.48, 95% CI 32.91–74.38, P <0.0001), poor communication (OR 6.62, 95% CI 3.70–11.85, P <0.0001) and multiple transfers (OR 6.30, OR 95% 4.31–9.21, P <0.0001). Age, lack of bed and after‐hours transfer did not predict transfer delays. Metropolitan transfers was protective against transfer delays (OR 0.64, 95% CI 0.47–0.86, P = 0.0035). Conclusion: In the view of the receiving surgeon or assessor, all transfer delays potentially contributed to patient deaths, and may have been preventable. Strategies directed at modifiable factors could minimize delays. Increased surgical services in non‐metropolitan regions could reduce need for transfer. Prospective data is required to examine if the same predictors are observed in surgical patients who survive. [ABSTRACT FROM AUTHOR]