Exploring the impact of route of administration on medication acceptance in hospitalized patients: Implications for venous thromboembolism prevention

Background Non-administration of venous thromboembolism (VTE) prophylaxis contributes to preventable patient harm. We hypothesized that non-administration would be more common for parenteral VTE prophylaxis than oral infectious disease or cardiac prophylaxis or for treatment medications. The primary study goal was to determine if non-administration of parenteral VTE prophylaxis is more frequent than other prophylactic or treatment medications. Methods In this retrospective cohort study of consecutive admissions we used descriptive statistics and risk ratios (RR) to compare the number of non-administered doses of VTE prophylaxis, oral infectious disease and cardiovascular prophylaxis and treatment medications. To quantify the influence of demographic and clinical characteristics on non-administration, we estimated incidence rate ratios from Poisson regression models. Results 645 patients were admitted from July 1, 2014 through March 31, 2015. Median age was 52 years (Interquartile range 43–57) and 365 (56.6%) were male. Subcutaneous VTE prophylaxis doses were not administered nearly 4-fold more frequently than oral infectious disease and cardiovascular prophylaxis (RR = 3.93; 95% CI 3.36–4.59) and 3-fold more frequently than treatment medications (RR = 3.06; 95% CI 2.91–3.22). Ninety percent of non-administered doses of VTE prophylaxis were refused. Risk factors for non-administration included younger age (age 18–35 years), male sex, uninsured status, HIV-positivity and high VTE risk status. Conclusions Subcutaneous VTE prophylaxis is not administered more frequently than oral infectious diseases or cardiac prophylaxis and treatment medications. These data suggest that availability of an oral medication could improve the effectiveness of VTE prophylaxis in real world settings.