Venous thromboembolisms (VTE) are considered preventable events with appropriate mechanical or chemical prophylaxis. However, chemical prophylaxis is frequently delayed or interrupted during hospitalization. We investigated the impact of delayed initiation and interruption of chemical prophylaxis on VTE rates. The incidence of VTE at an urban academic medical center was measured in patients hospitalized for >2 days between November 2013 and May 2014. Patients receiving prophylaxis were grouped as complete (started within 24 hours of admission and no interruptions), delayed (started >24 hours and no interruptions), and interrupted (interruption for >24 hours with or without delay). There were 9961 hospital admissions and 33 VTE (3.3 per 1000 admissions). 25.2 per cent had complete, 16.4 per cent had delayed, and 11.8 per cent had interrupted prophylaxis. 36.8 per cent received no prophylaxis. Interrupted prophylaxis was associated with more VTE than complete (10.2 vs 2.0 per 1000, P < 0.01) and 5.2 greater odds. Admission to a surgical service and prolonged hospital stay were independently associated with increased likelihood of VTE. There was a lower likelihood of getting complete prophylaxis among patients admitted to orthopedic, transplant, cardiac, plastic, and vascular surgery. Surgical patients are at higher risk for VTE and interruptions in VTE prophylaxis significantly increase the risk of VTE.

No abstract

Background-To enhance quality improvement, we created a unique statewide collaboration among 3 organizations: the Virginia Health Quality Center (Virginia's Medicare Quality Improvement Organization), the American College of Cardiology, and the American Heart Association. The goal was to improve discharge measures for acute myocardial infarction and heart failure. Methods and Results-In 2004, 29 hospitals participated in the collaborative initiative. Using Medicare data submitted from 2004 through the second quarter of 2006, we analyzed adherence to individual discharge measures and all-or-none appropriate care measures for acute myocardial infarction, heart failure, and both. To control for differences in hospital characteristics, we were able to match 21 of the participating hospitals with 21 similar nonparticipating hospitals. In this paired analysis, the total appropriate care measure increased from 61% to 77% in participating hospitals compared with an increase from 51% to 60% in nonparticipating hospitals (PϽ0.0001). A generalized linear mixed model examining the full data set at the patient level failed to show a clear advantage among participating hospitals. Participating hospitals had higher baseline rates for most quality measures, suggesting a possible effect of a prior collaborative. Further analysis of only hospitals that participated in a prior collaborative showed that participants in the current collaborative initiative had higher rates of improvement for 7 of 10 quality measures and appropriate care measures for heart failure, acute myocardial infarction, or both (all PϽ0.05). Conclusions-We report a unique collaboration of a Medicare Quality Improvement Organization and 2 national organizations to address quality of care for acute myocardial infarction and heart failure. A composite measure of quality (the total appropriate care measure) improved more in the participating hospitals during the timeframe of the intervention, although the greater improvement in this and other measures in the participating hospitals appeared to be dependent on participation in a prior collaborative initiative.