A METHOD TO INVESTIGATE SEASONAL VARIATION IN HOSPITALIZATION FOR COPD IN A MIDWESTERN US STATE
Author(s)
Su W1, Zhao R1, Heins-Nesvold J2, Carlson A1
1University of Minnesota, Minneapolis, MN, USA, 2American Lung Association of the Upper Midwest, St. Paul, MN, USA
OBJECTIVES: To investigate the impact of seasonal variation on daily number of COPD hospital admissions using 6 years of hospital discharge data from the Iowa Hospital Association. METHODS: Inpatient admissions from 2006 to 2011 for persons 30 years of age or older were included if there was a principal ICD-9-CM diagnosis of COPD (491.xx; 492.xx; 496.xx).Generalized linear models extending Poisson regression and Negative Binomial model (to account for overdispersion) were fitted. Admission date was converted into an angle called Ө and treated as a continuous variable with a circular distribution. SinӨ and cosӨ together were proposed to be used as the covariates representing seasonal effect. Subgroup analyses of age and sex were applied. RESULTS:
Conference/Value in Health Info
2014-09, ISPOR Asia Pacific 2014, Beijing, China
Value in Health, Vol. 17, No. 7 (November 2014)
Code
PRS5
Topic
Epidemiology & Public Health
Topic Subcategory
Safety & Pharmacoepidemiology
Disease
Respiratory-Related Disorders