Asthma prevalence estimates for the years spanning the NHIS redesign are shown in this table. Data from the last 2 years of the previous version of the NHIS (1995 -1996) were combined to obtain adequate sample sizes for all race/ethnicity and age subgroups. The asthma attack prevalence estimate from the 1997 NHIS, 5.44 percent, is 21 percent lower than the asthma prevalence estimate from the 1995 -1996 NHIS, 6.86 percent. For all race/ethnicity and age groups, the difference was approximately 20 percent, except for Hispanic children for whom the difference was 33 percent. In addition, in contrast to 1995 -1996 when Hispanic children had higher asthma period prevalence than did non-Hispanic White children, 1997 asthma attack prevalence was similar for Hispanic and non-Hispanic White children.

(For a summary of the impact of the 1997 redesign on age and race/ethnicity subgroups of children, see references 7 and 8).

Because there may have been different impacts among subgroups, most notably among Hispanic children who had the greatest change in asthma prevalence estimates between the old and redesigned NHIS, the analysis to assess the impact of the 1997 redesign was stratified by age and race/ethnicity.

Using the 1988 NHIS, the estimates from the standard questionnaire and the CHS were compared for each of nine race/ethnicity/age groups: non-Hispanic White, non-Hispanic Black, and Hispanic, each stratified by age (0-4 years, 5-10 years, and 11-17 years). Other race/ethnicity groups were not included because of insufficient sample size and therefore also were excluded from the overall analyses so that the sample size remained uniform for all analyses.