MULTIVARIATE COX ANALYSIS REGARDING TREATMENT SWITCH ON CHRONIC HEPATITIS B (CHB) DISEASE MANAGEMENT BASED ON A 2-YEAR PROSPECTIVE STUDY IN 5 EUROPEAN (EU) COUNTRIES
Author(s)
Arama V1, Zeuzem S2, Leblebicioglu H3, Simon K4, Klauck I5, Morais E5, Lescrauwaet B6, Kamar D7, Zarski J81Carol Davila University of Medicine and Pharmacy, Prof. Dr. M. Bals National Institute of Infection Diseases, Bucharest, Romania, 2J.W. Goethe University Hospital, Medicine Departements, Frankfurt, Germany, 3Medical School Samsun, Ondokuz Mayis University, Samsun, Turkey, 4Wroclaw University of Medicine, Wrocalw, Poland, 5Bristol-Myers Squibb, Rueil Malmaison, France, 6Xintera Consulting, Leuven, Belgium, 7DOCS, Sèvres, France, 8Clinique Universitaire d'Hépato-Gastroentérologie, CHU de Grenoble, Grenoble, France
OBJECTIVES: To explore the major determinants associated with treatment switch in CHB treated patients from a cohort across 5 EU countries. METHODS: Multivariate Cox proportional hazard regression model was used to analyze determinants of treatment switch by a backward selection. Data analyzed were derived from an observational cohort study, consisting of diagnosed adult CHB patients managed in outpatient clinics, prospectively enrolled from March-2008 to October-2009, in France, Germany, Poland, Romania and Turkey. Patients with HIV and/or HCV co-infection and hepatocellular-carcinoma or liver failure at baseline were excluded. Factors considered candidates to enter the multivariate model (with univariate test at p-value<0.20) were: country, ALT level before switch, VL level before switch, liver biopsy test performed, Hepatitis B e-antigen status, CHB status, age and agent received before switch. The statistical criteria for exclusion at each step of the model building process is p-value>0.05. Results are presented as hazard ratio(HR), [95% confidence interval] and p-value. RESULTS: Data were collected for 1267 patients followed for up to 2 years with a median (months) ranging from 12.5 (Poland) to 14.5 (Romania). A total of 567 patients received a treatment at baseline and were included in the model to analyze the determinants of treatment switch. Time on last treatment before switch was 1.9 years (median). There is a higher probability to treatment switch in France (5.19;[1.76-15.24];p=.0027), Germany (4.80;[1.60-14.39];p<.0050), and Poland (4.42;[1.41-13.83];p<.0106) compared to Turkey. No statistically difference in Romania. There is a higher probability to treatment switch when VL≥2000IU/ml (4.35;[2.52-7.52];p<.0001) compared to Undetectable and VL<2000IU/ml. Analysis of interaction showed that VL when associated with a performed liver biopsy test increase probability of switch. CONCLUSIONS: This observational cohort identified VL level as a major clinical outcome driving the treatment switch decision in CHB treated patients. Results were statistically different across countries, possibly due to different healthcare policies.
Conference/Value in Health Info
2012-11, ISPOR Europe 2012, Berlin, Germany
Value in Health, Vol. 15, No. 7 (November 2012)
Code
PIN80
Topic
Health Service Delivery & Process of Care
Topic Subcategory
Treatment Patterns and Guidelines
Disease
Infectious Disease (non-vaccine)
Explore Related HEOR by Topic