Abstract
We examined variation in hospital treatment and its relationship to clinical outcome in a large population-based cohort of ACS patients within a single payer-government funded health care system.
Patients hospitalized in 106 hospitals in Alberta, Canada with a primary diagnosis of ACS were included (July 1, 2010-March 31, 2013) with comparisons made across the three cardiac catheterization-capable hospitals (Sites A-C). Cox proportional-hazard regression models were used to examine the multivariable-adjusted association between site and 1-year death or repeat cardiovascular (CV) hospitalization (primary endpoint).
Of 14,155 patients, 1938 (13.7%) were admitted to a community hospital without transfer to an invasive hospital (10.7% in-hospital death). The remaining were admitted (n=4514, 36.9%) or transferred (n=7703, 63.1%) to an invasive hospital (A:5480; B:3621; C:3116) where 11,247 (92.1%) underwent catheterization. Comorbidities and angiographic disease burden differed across sites. Variation in 30-day revascularization (PCI: 71.3%, 72.0%, 68.7%, p<0.001; CABG: 6.2%, 6.4%, 9.3%, p<0.001) and drug-eluting stent use for PCI (24.3%, 54.6%, 50.5%, p<0.001) were observed. After adjustment for patient demographics and comorbidities, variation in rates of 1-year death or CV hospitalization was observed among those with 30-day revascularization (p(interaction)<0.001; B versus A: HR 0.78, 95%CI 0.66-0.91; C versus A: HR 0.77, 95%CI 0.65-0.91; B versus C: HR 1.01, 95%CI 0.84-1.21).
Despite a government funded health system, we have shown variation in hospital treatment exists. Following adjustment hospital site was associated with differences in clinical outcome within 1year. Hence, further efforts may be warranted to help address potential disparities in ACS care.