PMU146 VALIDATION AND REPRESENTATIVENESS OF THE SPANISH BIG-PAC DATABASE: INTEGRATED COMPUTERIZED MEDICAL RECORDS FOR RESEARCH INTO EPIDEMIOLOGY, MEDICINES AND HEALTH RESOURCE USE (REAL WORD EVIDENCE)

Abstract

Databases that integrate information from primary and specialist care medical records offer an information source that may be useful for studies of epidemiology, therapeutic adaptation and health/non-healthcare resource use and associated costs. The objective of the study was to determine the validity and representativeness of the BIG-PAC database electronic medical records in usual clinical practice in Spain. A retrospective descriptive study was carried out, based on review of health records obtained from health providers in various primary and hospital centres in Spain, unified in the BIG-PAC anonymized database since 2012 (1.9 million patients). Validation of the quality of the records was made through: a) indicative comparison of population prevalences (hypertension, diabetes, hypercholesterolemia, asthma) in patients aged 35-74 years, and b) comparison of biochemical/anthropometric parameters (systolic blood pressure [mmHg], body mass index [BMI, kg/m2], baseline-glycemia [mg/dL], cholesterol [mg/dL]) and other parameters, with the literature consulted. Representativeness was measured by comparison with the Spanish age pyramid in 2018. Descriptive statistical-analysis. We analysed the records of 589,385 patients aged 35-74 years attended during 2018. The mean age was 43.5 years and 52.4% were female. The population prevalences of the indicators studied were 6.2%, 9.1%, 31.0% and 32.8% respectively. Complete information was obtained for 72.5% of the parameters. The means (SD) of the indicators were: 129.1 (25.3) mmHg, 28.8 (14.2) kg/m2, 95.2 (25.7) mg/dL and 197.5 (33.5) mg/dL. Comparison of the age distribution and sex of the BIG-PAC population with the Spanish population showed they were similar. Various subanalyses were made. The Spanish BIG-PAC database is acceptable in terms of the quality of the records and the representativeness compared with the Spanish population, according to the indicators analysed. Currently, other types of qualitative/quantitative comparisons are being carried out with other national and international databases to ensure the quality of the records.