Venous thromboembolism (VTE) is a common complication in hospitalized patients. To reduce the incidence of this potentially preventable adverse event, we incorporated computerized clinical decision support-enabled “smart order sets” in our electronic order entry system which have significantly increased the prescription of risk-appropriate VTE prophylaxis. Despite our performance improvement, hospital-acquired VTE remains a common complication at our institution. The purpose of the current study was to examine the clinical characteristics of patients developing VTE during their hospital stay and identify potentially modifiable causes for these events.
We conducted a retrospective review of an administrative database of objectively-confirmed VTE among patients hospitalized at the Johns Hopkins Hospital during fiscal year 2011. Demographic and clinical data including VTE risk category, ordered VTE prophylaxis and the number of missed doses of VTE prophylaxis were retrieved from our computerized provider order entry system. VTE were attributed to central venous catheter (CVC) if they occurred in the same vascular distribution. VTE were deemed preventable if risk-appropriate prophylaxis was not ordered or if any doses of pharmacologic prophylaxis were missed. CVC-VTE were not classified as preventable.
Between July 1, 2010 and June 30, 2011, 139 hospitalized adult patients were classified as developing potentially preventable VTE. The mean age is 61.3 (SD: 17.7) years, 54.0% were male, the mean BMI was 28.2 (SD: 7.6) kg/m2 and 70.5% had a surgical procedure during hospitalization. Median length of hospitalization was 15 days and median time to VTE was 6 days. Eighty nine patients (64.0%) were classified as being at high or very high risk for VTE; 52 patients admitted to a surgical service and 37 admitted to the medical service. Fifty patients (36.0%) were judged to be at moderate risk, 23 patients admitted to a surgical service and 27 admitted to the medical service.
Eighty five patients (61.2%) developed DVT, 45 patients (32.4%) developed PE, and 9 patients (6.5%) had superficial thrombophlebitis (ST). Forty nine events were CVC-VTE; 38 CVC- DVT, 3 CVC- PE and 7 CVC-ST.
Among 139 patients who developed in-hospital VTE, 117 patients (84.2%) experienced events that were not preventable by best practice prophylaxis (49 CVC-VTE, 2 ST, 58 received all doses of risk-appropriate prophylaxis). Among 88 patients with potentially preventable VTE events, best-practice prophylaxis was prescribed for 76 (86.5%). The most common reason for potential preventable VTE was missed doses of VTE prophylaxis. For those ordered best-practice prophylaxis, 18 patients (23.7%) missed one or more doses of pharmacologic prophylaxis, 12 patients on the medical and 6 patients on a surgical service. For 7 patients (38.9%), the provider determined that their clinical condition was inappropriate for one or more doses of prophylaxis administration; 4 patients on the medical service and 3 patients on a surgical service. Eight patients (25%) refused one or more doses of prophylaxis; 7 patients on the medical service and 1 patient on a surgical service.
This retrospective consecutive case series of hospital-acquired VTE reveals that a substantial percentage of VTE among hospitalized patients are not preventable even with best practice prophylaxis. Therefore, process measures (prescription of risk-appropriate VTE prophylaxis) rather than outcomes (VTE) is the appropriate quality improvement metric for assessing provider and institutional performance. In addition, we noted that missed doses of VTE prophylaxis were the most common finding among patients suffering potentially preventable hospital-acquired VTE. Patient refusal and provider omission were the two leading reasons for missed doses of VTE prophylaxis. These findings underscore the need to track rates of ordered and administered doses of best-practice VTE prophylaxis and identify reasons and solutions for these quality deficits.
Streiff:sanofi-aventis: Consultancy, Honoraria; BristolMyersSquibb: Research Funding; Eisai: Consultancy; Janssen Healthcare: Consultancy; Daiichi-Sankyo: Consultancy.
Asterisk with author names denotes non-ASH members.