Purpose
Stock et al. (Eur Respir J 25:47–53, 2005) recently estimated asthma prevalence in Germany using claims data on prescriptions and hospital diagnoses and found high prevalence peaks in infants. Our objective was to critically assess and discuss various aspects of identifying children with asthma using prescription data.
Methods
We replicated the selection procedure of Stock et al. using data on 290,919 children aged 0–17 years insured in the Gmünder ErsatzKasse (GEK) in 2005. Asthma prevalence was also estimated in a sample of 17,641 children aged 0–17 years participating in the German Health Interview and Examination Survey for Children and Adolescents (KiGGS) from 2003 to 2006.
Results
In children aged 0–4 years insured in the GEK, prevalences were found to range from 11.7 to 17.7% for boys and from 7.2 to 11.1% for girls when the criteria of Stock et al. were applied. A steady decline in prevalences was observed in older age groups. Asthma prevalence estimated in the KiGGS data showed a quite different distribution. In the age group 0–4 years, prevalences were found to range from 0 to 2.6% in boys and from 0 to 1.0% in girls; in children >4 years, prevalences were found to increase with increasing age.
Conclusions
When additional validation studies were taken into account, asthma medications were found to be prescribed not only for asthma but also for other respiratory diseases. In addition, not all children with current asthma had prescriptions. We therefore conclude that asthma medications are therefore not a good proxy for the disease.