Can increased incidence of deep vein thrombosis (DVT) be used as a marker of quality of care in the absence of standardized screening? The potential effect of surveillance bias on reported DVT rates after trauma

Background: Deep vein thrombosis (DVT) is a significant cause of morbidity and mortality in trauma patients, even with appropriate prophylaxis. Many national agencies (Agency for Healthcare Research and Quality, Joint Commission, National Quality Forum) have suggested DVT incidence as a measurement of health care quality, but none has recommended a standardized screening approach. Duplex ultrasound serves an important role as a noninvasive diagnostic tool for detection of DVT. However, screening of asymptomatic patients for DVT is somewhat controversial and these practices vary widely among trauma centers. We hypothesized that as the number of screening duplex examinations in trauma patients increases, the rate of DVT identification will also increase. Methods: Retrospective cohort study of 21,961 patients from an urban, university-based Level I trauma center for more than 11 years (1995-2005). We grouped patients according to admission at the trauma service either before or after implementation of a written practice management guideline for DVT prophylaxis and duplex ultrasound surveillance in 1998. We compared duplex, DVT, and pulmonary embolism rates per 1,000 trauma admissions using Fisher's exact test. Results: The proportion of trauma patients having a duplex ultrasound increased significantly (20.9-81.5 per 1,000 trauma admissions, p < 0.0001). The rate of DVT reported increased 10-fold (0.7-7.0 per 1,000 admissions, p = 0.0024), significantly, between the two periods. The pulmonary embolism rate increased almost fivefold (0.7-3.2 per 1,000 admissions, p = 0.15), although this difference was not statistically significant. Conclusions: Increasing the number of duplex screening exams resulted in an increased rate of DVT identification. In the absence of standardized surveillance, DVT rates may be more influenced by how often caregivers look for these events rather than the quality of care provided.