Predictors of Hospital Length of Stay among Patients with Low-risk Pulmonary Embolism

Background: Increased hospital length of stay is an important cost driver in hospitalized low-risk pulmonary embolism (LRPE) patients, who benefit from abbreviated hospital stays. We sought to measure length-of-stay associated predictors among Veterans Health Administration LRPE patients. Methods: Adult patients (aged ≥18 years) with ≥1 inpatient pulmonary embolism (PE) diagnosis (index date = discharge date) between 10/2011-06/2015 and continuous enrollment for ≥12 months pre- and 3 months post-index were included. PE patients with simplified Pulmonary Embolism Stratification Index score 0 were considered low risk; all others were considered high risk. LRPE patients were further stratified into short (≤2 days) and long length of stay cohorts. Logistic regression was used to identify predictors of length of stay among low-risk patients. Results: Among 6746 patients, 1918 were low-risk (28.4%), of which 688 (35.9%) had short and 1230 (64.1%) had long length of stay. LRPE patients with computed tomography angiography (Odds ratio [OR]: 4.8, 95% Confidence interval [CI]: 3.82-5.97), lung ventilation/perfusion scan (OR: 3.8, 95% CI: 1.86-7.76), or venous Doppler ultrasound (OR: 1.4, 95% CI: 1.08-1.86) at baseline had an increased probability of short length of stay. Those with troponin I (OR: 0.7, 95% CI: 0.54-0.86) or natriuretic peptide testing (OR: 0.7, 95% CI: 0.57-0.90), or more comorbidities at baseline, were less likely to have short length of stay. Conclusion: Understanding the predictors of length of stay can help providers deliver efficient treatment and improve patient outcomes which potentially reduces the length of stay, thereby reducing the overall burden in LRPE patients.