Study of long-acting reversible contraceptive use in a UK primary care database: validation of methodology.

OBJECTIVES To develop and validate algorithms to identify new users of long-acting reversible contraceptives (LARCs) in a primary care database, The Health Improvement Network (THIN). METHODS Women in THIN aged 12 to 49 years in 2005 were studied. THIN was searched using Read and MULTILEX codes to identify new users of copper intrauterine devices (Cu-IUDs), the levonorgestrel-releasing intrauterine system (LNG-IUS) and progestogen-only implants. Validation was undertaken for a randomly selected sample of 398 LARC users, in which their primary care physicians were asked to complete a questionnaire detailing LARC use. RESULTS Questionnaires were received for 379 patients (95%), confirming 316 (83%) as new LARC users. Confirmation rates for Cu-IUDs, the LNG-IUS and progestogen-only implants were 64%, 94% and 89%, respectively. The use of Read codes alone had the lowest confirmation rate, particularly for Cu-IUD users. Confirmation rates increased by using MULTILEX codes when available, or by examination of computerised medical records. CONCLUSIONS Computer algorithms were used to identify new LARC users. While THIN is a useful resource for studying LARC uptake, steps to gather additional information are necessary to ensure the validity of LARC classification.