Evaluating the utility of the CPRD GOLD-HTI linkage: Anticoagulant prescribing at the GP practice compared to hospital dispensed medication at discharge date

Background: There is a lack of information on hospital medications in UK electronic healthcare data. CPRD and IMS Health have been collaborating to make linked data combining hospital pharmacy prescribing records with longitudinal primary care data available for public health research. Objectives: This study aims to demonstrate the utility of the linkage between CPRD primary care data (GOLD) and the HTI (Hospital Treatment Insights) which links hospital dispensing information with events in hospital. Anticoagulants (novel oral anticoagulants (NOACs) heparin and warfarin) have been chosen as an example medication that may be initiated in hospital and continued by the GP. Methods: Patients were limited to those eligible for the linkage between HTI and GOLD, with at least one of the study anticoagulants dispensed. ICD-10 codes associated with the anticoagulant dispensing were examined for specific treatment indications (Atrial Fibrillation, Venous Thromboembolism (VTE) and Stroke). GOLD therapy records were examined for further anticoagulant treatment following discharge. Results: From a population of 328,930 patients in the HTI database that have been individually linked to GOLD, 19,421 had at least one record for rivaroxaban, apixaban, heparin or warfarin dispensed in hospital. Atrial Fibrillation was the most commonly recorded indication (18%); followed by VTE (5%) and Stroke (2%). 59.4% of patients had further anticoagulation prescribing at the GP; 2.4% within 30 days of discharge. Conclusions: In this study we were able to show that further anticoagulation was continued in primary care in over half of hospitalisations discharged with these medications. Individual linkage between hospital dispensing information and existing primary care data offers the opportunity to fill the gaps in longitudinal data and follow the care pathway for patients starting treatment in secondary care in the UK.