prev next front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |19 |20 |21 |22 |review
Estimates of asthma prevalence differ substantially depending on which data source is used: Weighted to national levels, the prevalence estimates for asthma in the sample were 7.7% based on medical records or 10.0% based on maternal reports. Restricting the definition of a case to those for whom both mother and medical provider mention asthma yields a prevalence of 4.1%; broadening the definition to include any child for whom either the mother or medical provider reported the condition yields a prevalence of 13.6%.

These findings are important because they help disentangle issues of study design and population from those of reporting patterns by source of information. Most published studies can report prevalence based on either mother’s reports (regardless of medical records, which are unavailable in most health surveys, which corresponds to our “mother’s reports” estimate of 10.0%), or on medical records (without knowing mother’s reports, which corresponds to our “medical records” estimate of 7.7%).