DISCUSSION
Because all six hospitals included in the study were closely located geographically and were managed within the same healthcare system, it was surprising to find significant differences among them in the frequency of variables that could increase D2B time. Identified in the study were differences in age, ambulance versus walk-in, scenario of admission, and patient-related variables (cardiac arrest, need for intubation, need for radiologic evaluation, and inconclusive ECGs). However, none of these variables were consistently different between the higher and lower performing hospitals. It is possible that differing socioeconomic variables (such as poverty and level of education) affected outcomes among the six hospitals.
The American Heart Association has recommended that at least 75% of STEMI patients receive the first ECG < 10 minutes of admission.3 Collectively, the percent of adherence to this recommendation among the six hospitals was 83.1% (median 4 minutes, IQR 1-8). The six hospitals included in this study collectively performed above the minimal adherence score for a D2B time ≤90 minutes (87.8% adherence as opposed to 75%, respectively).3Nonetheless, there was considerable variation among the six hospitals; the two highest performing hospital complied in 94.9% and 97.1% of the cases, while the lowest performing hospital complied in 79.1% of the cases. It is difficult to explain why this wide range of compliance occurred; as there was no consistent pattern in the distribution of suspect variables among the six sites.
An expected finding was that patients with D2B times >90 minutes had a significantly higher mortality than those with D2B times< 90 minutes. However, the data on mortality is not conclusive because the study lacked information describing whether the mortality was cardiac or non-cardiac related as well as there was no documentation on comorbidity or underlying cardiac disease such as left-ventricular function, valvular heart disease, prior congestive heart failure, prior stent or coronary artery bypass graft.
Perhaps the most important finding of our study was the effect of the scenario of admission to the CCL. Scenarios B and C were associated with significantly better D2B times than was Scenario A. This is reasonable, given that these scenarios involve EMS admission with ECGs performed in the field. Scenario B is best in that it bypassed an ED admission and patients are transported directly to the CCL. An expected finding was that the presence of confounders significantly affected D2B time. While there were insufficient data to assess the significance of non-patient related confounders on D2B time, the presence of at least one patient-related confounder greatly increased D2B time.