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
Studies of asthma have used medical records and maternal reports, but analyses of each source tend to be based on different samples of children. Although these different studies permit comparison of overall prevalence estimates and some insight into distribution (the effects of demographic or other predictors) on asthma, they cannot be used to directly compare estimates from the different sources because the designs of the studies vary. For example, a study of medical records will typically include different children than a health survey against which it has been compared.

What cannot be determined by a comparison of prevalence estimates based on two or more separate study samples is the extent to which the sets of children identified as cases by the two sources overlap. Furthermore, with different samples, it is not possible to say with confidence how the two sets of cases differ in terms of other characteristics. With data from both reporting sources for a consistent sample, we are able to address both questions.