SESSION TITLE: New treatments for COPD
SESSION TYPE: Original Investigation Slide
PRESENTED ON: Monday, October 28, 2013 at 01:45 PM - 03:15 PM
PURPOSE: To compare the 30-day hospital readmission rates between patients with chronic obstructive pulmonary disease (COPD) hospitalization who used roflumilast versus those who did not in a managed care population.
METHODS: Data were from MarketScan, a large US commercial health insurance claims database. Patients aged ≥40 years with ≥1 hospitalization for COPD (ICD-9: 491, 492, or 496) between 3/1/2011-6/30/2012 were included. Roflumilast group included patients who used roflumilast within 14-days after hospitalization (index), while the comparison group (non-roflumilast) included patients who did not use roflumilast during the study period. Continuous enrollment at least 6-months before and 30-days after the index date was required. Patients with cystic fibrosis or lung cancer were excluded. 30-day hospitalization rate was calculated both before and after the index hospitalization. Logistic regression with difference-in-difference method was employed to assess the difference of 30-day hospital readmission rates between roflumilast and non-roflumilast groups adjusting for the pre-index 30-day hospitalization rate, baseline characteristics (age, gender, regions, insurance, co-morbidities, exacerbation frequency, COPD medications, and COPD-related and all medical service utilizations), and COPD medication use within the 14 days after index.
RESULTS: A total of 181,311 COPD patients met sample selection criteria: 367 in roflumilast and 180,944 in non-roflumilast group. Mean (SD) age was 70.8 (12.5) years with 51.6% female. From pre- to post-index hospitalization, the all-cause 30-day hospitalization rate decreased from 10.4% to 7.9% in roflumilast group, but increased from 5.2% to 5.5% in non-roflumilast group. Similarly, the 30-day COPD hospitalization rate decreased from 10.1% to 7.6% in roflumilast group, but increased from 2.1% to 2.3% in non-roflumilast group. Logistic regression showed trends of decreased all-cause and COPD 30-day readmission rates (OR=0.61, 95% CI [0.33-1.12] and OR=0.54, 95% CI [0.27, 1.07], respectively), though not reaching statistical significance, for roflumilast patients relative to non-roflumilast patients.
CONCLUSIONS: This study showed, in a real-world setting, that the use of roflumilast may reduce the 30-day hospital readmission rate among patients who are hospitalized for COPD.
CLINICAL IMPLICATIONS: Pharmacotherapy such as roflumilast treatment may be an effective option to decrease the 30-day hospital readmission rate for patients with COPD exacerbations.
DISCLOSURE: Alex Fu: Other: Research support from Forest Research Institute Shawn Sun: Employee: Employee of Forest Research Institute Xingyue Huang: Employee: Employee of Forest Research Institute Alpesh Amin: Other: Research and speaker for Forest Research Institute
No Product/Research Disclosure Information