Validating childhood asthma in an epidemiological study using linked electronic patient records
Open Access
Abstract
Objective To investigate the performance of parent-reported data in identifying physician-confirmed asthma. Design and setting Validation study using linkage between the Avon Longitudinal Study of Parents and Children (ALSPAC) and electronic patient records held within the General Practice Research Database (GPRD). Participants Participants were those eligible to participate in ALSPAC who also had a record in the GPRD; this included 765 individuals, just under 4% of ALSPAC-eligible participants. The analysis was based on 141 participants with complete parent-reported asthma data. Primary and secondary outcome measures The main GPRD outcome measure was whether a child had a diagnosis of asthma before they were nine. Parent-reported measures were doctor diagnosis of asthma (before mean age 7.5 years), various outcomes based on wheezing and breathlessness recorded longitudinally between 6 months and 8.5 years. Secondary outcomes were bronchial hyper-responsiveness (BHR), forced expiratory volume in 1 s/forced vital capacity ratio and skin prick test responses. Results Among the 141 participants with complete parent-reported data, 26 (18%) had an asthma diagnosis before age nine. Using general practitioner (GP)-recorded asthma as the gold standard, the question ‘Has a doctor ever diagnosed your child with asthma?’ was both sensitive (88.5%) and specific (95.7%). ‘Ever wheezed’ had the highest sensitivity (100%) but low specificity (60%). More specific definitions were obtained by restricting to those who had wheezed on more than one occasion, experienced frequent wheeze and/or wheezed after the age of 3, but these measures had low sensitivities. BHR only identified 50% of those with a GP-recorded diagnosis. Conclusions Parental reports of a doctor's diagnosis agree well with a GP-recorded diagnosis. High specificity for asthma can be achieved by using detailed wheezing questions, although these definitions are likely to exclude mild cases of asthma. Our study shows that linkage between observational studies and electronic patient records has the potential to enhance epidemiological research.Keywords
This publication has 24 references indexed in Scilit:
- Epidemiological measures of childhood asthma: Cross-sectional and longitudinal consistencyRespiratory Medicine, 2012
- Cohort Profile: The ‘Children of the 90s’—the index offspring of the Avon Longitudinal Study of Parents and ChildrenInternational Journal of Epidemiology, 2012
- Verifying a questionnaire diagnosis of asthma in children using health claims dataBMC Pulmonary Medicine, 2011
- Validity of diagnostic coding within the General Practice Research Database: a systematic reviewBritish Journal of General Practice, 2010
- Performance of the ISAAC Questionnaire to Establish the Prevalence of Asthma in Adolescents: A Population-Based StudyJournal of Asthma, 2010
- British Guideline on the Management of AsthmaThorax, 2008
- International Study of Asthma and Allergies in Childhood (ISAAC): rationale and methodsEuropean Respiratory Journal, 1995
- Asthma and Wheezing in the First Six Years of LifeNew England Journal of Medicine, 1995
- Epidemiological study of wheeze, doctor diagnosed asthma, and cough in preschool children in Leicestershire.BMJ, 1993
- Toward a Definition of Asthma for EpidemiologyAmerican Review of Respiratory Disease, 1992