BACKGROUND: Heterogeneity is observed in the care of patients with chest pain. We investigate the association of physician specialty and diagnostic testing among patients admitted for suspected acute coronary syndrome (ACS). METHODS: This is a secondary analysis of the CMR-IMPACT multicenter randomized controlled trial in which patients with suspected ACS were admitted and randomized to undergo invasive angiography or non-invasive CMR stress imaging. Admitting physician was dichotomized to interventional cardiologist (IC) or not (e.g. hospitalist). We describe adherence to protocol and angiography during the index visit by treatment arm and admitting physician specialty. A generalized estimating equation accounting for clustering within physician was used to evaluate significance and adjusted for randomization arm. RESULTS: The 258 enrolled patients from 2013 to 2018 had a mean age of 60.7 (SD ± 10.9) years, 40.3 % (104/258 were female), and 64.7 % (167/258) were white race. ICs were the admitting physicians for 50.4 % (130/258) of the patients. Index angiography was performed more often among patients admitted by IC versus other physicians, 65.4 % (85/130) versus 53.1 % (68/128), respectively
aOR 1.75 (95 % CI 1.14-2.68). Among patients randomized to an invasive strategy, higher protocol adherence was observed in those admitted by IC [85.3 % (58/68)] versus other physicians [64.5 % (40/62)]
OR 2.82 (95 % CI 1.08-7.38). For patients randomized to the CMR-based strategy, adherence to protocol was similar for IC [67.7 % (42/62)] and other physicians [66.7 % (44/66)]
OR 0.82 (95 % CI 0.35-1.94). CONCLUSION: Invasive angiography was more frequent among patients admitted by interventional cardiologists compared to other physicians.