Differences in Admission Rates Among Patients Treated with Apixaban Vs. Warfarin for Venous Thromboembolism in the Emergency Department

Background: With fixed-dose regimens, fast onset of action, and no requirement for routine monitoring, direct oral anticoagulants (DOACs) have facilitated the treatment of venous thromboembolism (VTE) on an outpatient basis. Little is known regarding the outcomes of patients who recieve DOACs in the emergency department (ED). Thus, in this study we evaluated hospitalization LOS and costs of patients who were diagnosed with VTE in the ED and treated with apixaban or warfarin

Methods: Adult patients (≥18 years of age) admitted into the ED with a primary discharge diagnosis code indicating VTE were identified from the Premier Hospital database (8/1/2014-5/31/2018). Patients who received apixaban or warfarin during the ED visit were identified and grouped into two study cohorts according to the oral anticoagulant received. Patients treated with warfarin were additionally required to have received ≥1 injectable anticoagulant during ED admissions. The first of such VTE ED admissions was defined as the index event, with the corresponding ED or hospital discharge date as the index date. Patient demographics and clinical and hospital characteristics were evaluated during the index event or a 12-month baseline period. The outcomes of ED discharge status, hospital LOS (ED visit alone has LOS = 0 day), cost of the index event, and rate of 1-month all-cause hospital readmission were compared for the two study cohorts. Multivariable logistic regression analyses were conducted to evaluate the impact of apixaban vs. warfarin treatment on the likelihood of being moved from the ED to the inpatient setting vs. discharged from the ED, as well as the likelihood of 1-month all-cause readmission. Generalized linear models were used to evaluate the impact of apixaban vs. warfarin treatment on index event hospital LOS and cost. Covariates in all analyses included age, gender, race, payer type, Charlson Comorbidity Index score group, baseline bleed status, baseline VTE status, index VTE type, and hospital characteristics.

Results: Of the overall study population, 30.5% (n=12,174; mean age: 59.7 years) received apixaban and 69.5% (n=27,767; mean age: 59.3 years) received warfarin for VTE in the ED setting. Mean Charlson Comorbidity Index score was lower for apixaban vs. warfarin treated patients (1.0 vs. 1.3; p<0.001). A significantly lower proportion of patients treated with apixaban, compared with those treated with warfarin, were admitted to the inpatient setting (31.0% vs. 67.1%, p<0.001). Unadjusted mean hospital LOS was shorter (0.9 vs. 2.6 days per patient; p<0.001) and mean index event cost lower ($2,389 vs. $5,348 per patient; p<0.001) for apixaban vs. warfarin patients. The unadjusted rate of 1-month all-cause readmission was also lower for apixaban vs. warfarin patients (10.3% vs. 12.4%; p<0.001). After adjusting for patient and hospital characteristics, apixaban treatment was associated with a significantly lower likelihood of admission to the inpatient setting vs. warfarin (Odds Ratio [OR]: 0.124, 95% Confidence Interval [CI]: 0.115 to 0.133; p<0.001). Correspondingly, mean index hospital LOS was 1.42 days shorter (95% CI: -1.47 to -1.36; p<0.001) and mean index event cost per patient was significantly lower ($3,732 [95% CI: $3,565 to $3,907] vs. $8,008 [95% CI: $7,676 to $8,355]; difference: -$4,276; p<0.001). After taking into account patient and hospital characteristics, the likelihood of all-cause 1-month readmission was significantly lower for patients treated with apixaban vs. warfarin (OR: 0.853, 95% CI: 0.793 to 0.917; p<0.001).

Conclusions: In the real-world setting, VTE patients admitted into the ED who are treated with apixaban had a lower likelihood than warfarin treated patients of being subsequently admitted into the inpatient setting, which was reflected in shorter average LOS and lower average index event cost. The risk of 1-month all-cause readmission was also lower for patients treated with apixaban vs. warfarin.