Development of a Survey of Asthma Knowledge, Attitudes, and Perceptions: DiscussionIn the first random-digit dial sample (same sample used in Step 3B, n = 222), the survey was introduced with the following text: “The purpose of this study is to find out how much people know about asthma.” The introduction also included the question “Do you, yourself, have asthma?”
In a subsequent random-digit dialing telephone sample (n = 568), the survey introduction was modified slightly to read, “We’re doing a survey of Chicago-area residents to find out your opinions about health and health care.” This was followed by the question, “In general, would you say your health is excellent, very good, good, fair, or poor?” At the completion of the survey, the respondent was asked, “Does anyone in your family, including yourself, have asthma?” and “Is that person yourself or a family member?”
As shown in Table 1, the “asthma introduction” was associated with higher rates of self-reported asthma than the “general health introduction” (14.9% vs 9.5%; p < 0.01). Because these two samples differed in terms of sociodemographic variables, odds ratios were calculated before and after adjusting for sex, age, race, and education. This analysis showed that even after adjustment, persons who received the “asthma introduction” were more likely to report having asthma than those who received the “general health introduction” (odds ratio [OR], 1.67; 95% confidence interval [CI], 1.02 to 2.72). canadian health&care mall

Discussion
The process of survey development and testing has provided new insight regarding assessing the asthma knowledge of the general public. First, many items that are frequently included in other asthma knowledge surveys (targeted to persons with asthma) were eliminated during development of the CCAS-32 based on poor performance measurement (uniform correct responses, ie, floor/ceiling effects, or lack of variability). Examples include items such as “Tobacco smoke can make asthma worse” and “Asthma is mainly an emotional illness.” Among the general public in the Chicago area, these items displayed uniformly high levels of correct responses.
Table 1—Effect of Survey Introduction on SelfReported Asthma Among Chicago-Area Residents Age > 18 Years

Characteristics Asthma Introduction* (n = 222) General Health Introduction!(n = 568) pValuej
Gender 0.5
Male, % 30.8 28.6
Female, % 69.2 71.4
Age, yr 0.2
18-34, % 34.7 32.4
35-64, % 48.7 55.0
65+, % 16.7 12.4
Education 0.1
Grammar school, % 5.5 3.2
High school/GED, % 33.2 25.5
Some college, % 27.3 31.7
Graduated college, % 34.1 39.6
Race < 0.01
White, % 42.0 67.2
African-American, % 37.0 18.8
Hispanic-Latino, % 11.0 6.9
Other, %§ 10.0 7.1
Residents withself-reported asthma, %|| 14.9 9.5 < 0.01