We conducted a cross-sectional analysis of 378 hospitals (reflecting a 70% response rate among 543 eligible hospitals surveyed between January and November 2013), and a longitudinal analysis of a subsample of 72 hospitals (reflecting a 67% response rate among the 107 hospitals that had also responded to a survey between April and December 2010). We used the data to examine the current prevalence of commonly recommended hospital strategies used to lower mortality rates among patients with AMI as well as features of organizational culture.

For the 2013 survey, we contacted a random sample of 600 hospitals that were part of the American College of Cardiology’s Acute Coronary Treatment and Intervention Outcomes Network (ACTION) registry and treated at least 12 patients with ST-segment elevation acute myocardial infarction per year [9]. This sample size had been previously determined based on the conservative assumption of a 58% response rate in order to have 80% power to detect a difference in RSMR of 1.0 percentage points between 2 groups of 10% of responding hospitals. We sent a letter of invitation to participate in the study to the chief executive officer at all hospitals. We asked the chief executive officer to identify the person most involved in AMI quality improvement efforts, who we contacted to participate in a Web-based survey. Respondents were instructed to coordinate with other relevant staff to complete a single survey reflecting a hospital-level response. Of the 600 hospitals, 54 had ended their participation in the ACTION registry and 3 additional hospitals had closed, leaving 543 hospitals eligible for the follow-up survey. Of these, 378 hospitals (70%) completed a survey between January and November 2013. Additionally, 72 hospitals among the 378 respondents had been previously surveyed as part of our earlier study from April to December 2010 [5]. Among hospitals previously surveyed (N = 107), 35 were non-respondent at follow up (response rate 67%). The 72 hospitals previously surveyed did not differ statistically from the other respondent hospitals in our full sample with respect to teaching status, number of staffed beds, census region, urban/rural location, ownership type, or multihospital affiliation (all P-values > 0.05). We used the subsample of hospitals to explore changes in hospital strategies and organizational culture characteristics between 2010 and 2013.

Hospitals responded to a modified version of the survey previously described [5] (See Additional file 1 for questionnaire). Measures included a set of strategies (e.g., employing quality improvement teams focused on post-hospital mortality, physician and nurse champions, pharmacist rounding) as well as features of organizational culture (e.g., communication and coordination across departments, creative problem solving). We also obtained data on hospital characteristics from the 2010 American Hospital Association (AHA) annual survey. Variables included hospital size (total number of hospital beds), teaching status (Council of Teaching Hospitals (COTH)/has accredited residency program/non-teaching), ownership (for-profit/nonprofit/government) and multihospital affiliation (yes/no). We determined census regions from the U.S. Census Bureau and ascertained area type (urban/suburban/rural) using the 2003 Urban Influence Codes.

We generated frequencies to describe hospital characteristics among our overall and subsample of hospitals. We compared characteristics of these two samples using chi-square tests. We then generated frequencies to describe the current use of hospital strategies and features of organizational culture among the overall sample of hospitals. Last, we generated frequencies of hospital strategies used among the subsample of hospitals for 2010 and 2013 and compared these differences using McNemar’s chi-square tests. A complete case analysis was conducted because only a low proportion of data was missing. We used a significance threshold of P-value < 0.01 given the multiple comparisons. All analyses were completed with SAS 9.3 (Cary, NC). We obtained Internal Review Board exemption (protocol number 1207010622) for our study; participant consent was waived because no identifying participant information was obtained.

Almost 40% of respondents for which data were available on hospital characteristics (N = 358) were teaching hospitals, 73% had fewer than 399 staffed beds, and 90% were in urban locations (Table 1). The hospitals were distributed among all Census regions. The subsample of 72 hospitals that was surveyed in both 2010 and 2013 was not significantly different from the overall sample in terms of teaching status, number of beds, region, geographic location, ownership type, or multihospital affiliation.

In terms of strategies pertaining to quality improvement and monitoring, the vast majority of hospitals had quality improvement teams devoted to improving inpatient mortality in patients with AMI, had a designated person or group to review deaths of patients with AMI that occurred during hospitalization, and belonged to a regional effort or consortium to improve AMI care (Table 2). In contrast, less than half of hospitals had a quality improvement team focusing on post-discharge deaths or a review process for deaths that occurred within 30 days of admission. In terms of strategies for pre-hospital and inpatient care, several of the strategies were used by less than half of the hospitals, including meeting at least monthly with Emergency Medical Services (EMS) providers to review AMI care, not cross-training nurses to cover in the catheterization laboratory, and having pharmacists rounding on patient with AMI.

Several features of organizational culture were prominent in the full sample of hospitals surveyed in 2013. For instance, 87% of hospitals reported that clinicians were encouraged to creatively solve problems related to AMI care processes, and easy communication among departments caring for patients with AMI was reported by 91% of hospitals (Table 2). In terms of collaboration and coordination with partners outside of the hospital, 39% of hospitals held monthly meetings between clinicians and EMS providers to review the care of AMI patients. Somewhat more hospitals (55%) provided training to Emergency Medical Service (EMS) providers monthly or quarterly.

Between 2010 and 2013, the use of many strategies increased but only three increased significantly; the percentage of hospitals that had a quality improvement team to improve post-discharge mortality in patients with AMI increased from 23.6% to 43.5% (P-value = 0.024), and the percent of hospitals providing training to EMS providers about AMI care at least monthly or quarterly increased from 36.1% to 60.9% (P-value < 0.001). The percentage of hospitals using computerized assisted physician order entry increased from 33.3% to 82.6% (P-value < .001) (Table 3). A couple of features of organizational culture worsened, notably coordination among different departments (4.2% to 16.2% of hospitals reported they never, rarely, or sometimes versus usually or always had good coordination, P-value = 0.012), and communication across departments caring for patients with AMI (5.6% to 17.9% of hospitals reported they never, rarely, or sometimes versus usually or always had good coordination, P-value = 0.035) (Table 3).