Use of palivizumab is underestimated in the Swedish Prescribed Drug Register – implications for register-based drug studies

Abstract

Marie Linder,1 Camilla Byström,1 Helle Kieler,1 Gunnar Bergman,2 Ann Haerskjold3,4 1Centre for Pharmacoepidemiology, Department of Medicine, Karolinska Institutet, Solna, Stockholm, Sweden; 2Stockholm-Uppsala Department of Pediatric Cardiology, Astrid Lindgren Children's Hospital, Karolinska University Hospital and Department of Women's and Children's Health, Karolinska Institutet, Stockholm, Sweden; 3Child and Adolescent Clinic, Pediatric Department, 4Research Unit Women's and Children's Health, the Juliane Marie Centre for Women, Children and Reproduction, Copenhagen University Hospital, Rigshospitalet Hospital, Copenhagen, Denmark Background: Register studies are a valuable tool, when monitoring the safety of drugs. The Swedish Prescribed Drug Register (PDR) was established in 2005 and keeps records of all prescribed drugs dispensed in community pharmacies. Drugs prescribed in-hospital are not registered on an individual level, which may hamper the validity of register-based studies on drugs potentially administered in-hospital. Objective: The objective was to assess the ability of the PDR to identify children treated with the monoclonal antibody palivizumab, which is used for prophylaxis against respiratory syncytial virus (RSV) infection in children. Methods: Palivizumab exposure as filled prescriptions recorded in the PDR was assessed by indication of treatment (preterm-born children, bronchopulmonary dysplasia, or hemodynamically significant heart disease) and presented as numbers and proportions. For a random sample of children with an indication for treatment and without record of palivizumab exposure in the drug register, numbers and proportions by indication of treatment as noted in medical records were presented. The extent of underreporting in the drug register was estimated by indication for treatment. Results: Through the national health registers, 2,317 children were identified as being at risk for severe infection with RSV infection and 75% had no records indicating palivizumab exposure in the PDR. In a random sample of 176 children at high risk for RSV infection and with no records of palivizumab prescription fills in the PDR, 47% had been treated with palivizumab according to medical records. The PDR underestimated palivizumab treatment with 49% in children born preterm, 42% in children with bronchopulmonary dysplasia, and 23% in those with a hemodynamically significant heart disease. Conclusion: Our findings underline the need of improving the information in the Swedish national registers concerning drugs administered in-hospital. Keywords: validation, drug, respiratory syncytical virus, health register, medical records