SCREENING CHARACTERISTICS AND DIABETES BIOMARKERS IN FRENCH AND UK PATIENT-LEVEL DATABASES
Author(s)
Cabrera C*1;Le Jeunne P2;Lee A3;Thomas F4;Mian N3, McDonell A5 1AstraZeneca, Mölndal, Sweden, 2IMS Health, Paris, France, 3AstraZeneca GMD UK, Cheshire, England, 4AstraZeneca France, GMD, Paris, France, 5IMS, London, England
OBJECTIVES: Describe the patient characteristics and diabetes markers of type 2 diabetes (T2D) patients in the newly launched IMS LifeLink Diabetes Cohort in France. Monitor the data quality through comparison of measures with a French national survey of diabetes patients (ENTRED), the UK National Diabetes Audit (NDA) where possible, and the “gold standard” UK Clinical Practice Research Datalink (CPRD). METHODS: The IMS Diabetes cohort in France supplements patient-level data from general practitioners with additional patient data via pop-up technology within electronic medical records to facilitate robust epidemiological studies. This study compares the T2D patients in the Diabetes Cohort with similar populations in terms of gender, age, disease duration, and proportion of patients meeting key disease-specific targets (data presented as summary statistics). RESULTS: Patient characteristics and diabetes markers were analyzed for T2D patients in the Diabetes Cohort (n=5,142), ENTRED (n=3,894), UK NDA (n=1,909,494), and UK CPRD (n=268,618). The mean (SD) age in years was 66 (12.58) in the Diabetes Cohort, 66 in ENTRED, and 61 (15.35) in CPRD. The mean (SD) time since T2D diagnosis was 8 (7.7), 11, and 7.7 (6.2) years in the Diabetes Cohort, ENTRED, and CPRD, respectively. Mean (SD) BMI was 30.5 (5.76) in the Diabetes Cohort, 29.5 in ENTRED, and 30.8 (6.51) in CPRD. HbA1c target level of ≤6.5% was met by 32.5%, 34%, 23.5%, and 26% of the patients in the Diabetes Cohort, ENTRED, CPRD, and NDA, respectively. CONCLUSIONS: Health care systems and data collection methods vary across EU countries in general. Data monitoring helps assess data quality and robustness. Based on comparison of patient characteristics and diabetes markers, the IMS Diabetes Cohort population does not appear to differ from the ENTRED population. Although some differences between the Diabetes Cohort and CPRD data were noted, this unique French Diabetes Cohort appears appropriate for epidemiological research.
Conference/Value in Health Info
2013-11, ISPOR Europe 2013, The Convention Centre Dublin
Value in Health, Vol. 16, No. 7 (November 2013)
Code
PRM42
Topic
Real World Data & Information Systems
Topic Subcategory
Reproducibility & Replicability
Disease
Diabetes/Endocrine/Metabolic Disorders