Temporal associations of early patient transfers and mortality with the implementation of a regional myocardial infarction care model

Background

In order to reduce the delays encountered through patient transfer, regional care models have been developed that directly transport subsets of acute myocardial infarction (AMI) patients to hospitals with percutaneous coronary intervention (PCI) facilities. Calgary is a Canadian city that implemented this type of model in 2004.

Methods

The study population included 9768 AMI patients admitted to Calgary hospitals between 1997 and 2007. Administrative data were used to define patients who were directly admitted to the PCI hospital and those transferred there after initial admission to a hospital without specialized cardiac care. The differences in clinical characteristics and mortality trends of patients grouped by hospital delivery site and transfer practice are described.

Results

The proportion of patients directly admitted to a PCI hospital has increased with the implementation of a regional care model. Among patients admitted to non-PCI facilities, the patients who are transferred are younger, more likely to be male, have a shorter length of stay, and have lower proportions of several comorbid conditions. The risk-adjusted in-hospital mortality odds ratio for patients who received care at the PCI hospital postmodel relative to those treated at non-PCI hospitals premodel was 0.38 (95% confidence interval, 0.31-0.47). The corresponding adjusted odds ratio was 0.60 (0.47-0.76).

Conclusions

Our results suggest changing care over time and trends toward improved outcomes. Patients' clinical characteristics appear to play a major role in the decision to transfer. Avoidance of the risk treatment paradox through refinement of regional transfer protocols ought to be a priority.