HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:
Guest Access | Sign In via User Name/Password

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Weiss, K. B. Articles by Li, T. Search for Related Content PubMed PubMed Citation Articles by Weiss, K. B. Articles by Li, T.

The Effects of Asthma Experience and Social Demographic Characteristics on Responses to the Chicago Community Asthma Survey-32^*

Kevin B. Weiss, MD; Evalyn N. Grant, MD; Tao Li, PhD and for the Chicago Asthma Surveillance Initiative Project Team

^* From the Center for Health Services Research, Rush Primary Care Institute (Drs. Weiss and Li), and the Department of Immunology/Microbiology (Dr. Grant), Rush-Presbyterian-St. Luke's Medical Center, Chicago, IL. See Appendix for other members of the CASI Project Team.

Correspondence to: Kevin B. Weiss, MD, Center for Health Services Research, Rush Primary Care Institute, Rush-Presbyterian-St. Luke's Medical Center, 1653 W Congress Pkwy, Chicago, IL 60612

	Abstract

TOP Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
Introduction: Th e Chicago Community Asthma Survey (CCAS-32) is an instrument for characterizing the general public's knowledge, attitudes, and beliefs related to asthma. The purpose of this study was to examine the effects of asthma experience and social demographic characteristics on asthma awareness among the general public.

Methods: The CCAS-32 consists of 21 dichotomous items, designed primarily to test asthma knowledge, and 11 Likert-scale items, focusing on asthma attitudes and beliefs. From December 1997 through February 1998, a random-digit dialing method was used to administer the CCAS-32 via a telephone survey of Chicago-area (seven-county) residents 18 years. Each respondent's asthma experience was classified as "person with asthma," "family/household experience," or "no/low asthma experience." Demographic variables included sex, age, education, race/ethnicity, urban vs suburban residence, and income.

Results: Five hundred sixty-eight Chicago-area residents completed the survey (response rate of 40.6%). Of these, 43.3% were aged 35 to 64 years, 71.3% were women, 66.7% were white, and 71.3% had completed at least some college. Sixty-two percent had no or low asthma experience, 28.5% had family or household experience, and 9.5% were persons with asthma. The mean percentage (± SE) of correct, or desirable, responses to asthma knowledge questions was 71.9 ± 0.5%, with a range from 31.9 to 95.1%. The mean percentage of desirable responses differed significantly between persons with no or low asthma experience, family or household asthma experience, and persons with asthma (70.0 ± 0.6%, 74.0 ± 0.9%, and 77.7 ± 1.2%, respectively, p < 0.01 for trend). Social demographic factors also appeared to result in statistically significant differences in the responses to many items. Of the demographic variables studied, age and education appeared to have the strongest effect on responses to knowledge items, with statistically significant differences in responses seen for 10 (47.6%) and 8 (38.1%) of the 21 dichotomous items. Race or ethnicity and education were each associated with differences in responses for 7 of the 11 Likert-scale items (63%).

Conclusions: The results of this study suggest that the CCAS-32 can detect meaningful differences between groups with different degrees of asthma experience (ie, discriminative validity). Using the CCAS-32, it may be possible to identify subpopulations with differences in asthma awareness, thus providing guidance for the design of messages to target community and public awareness of asthma. (CHEST 1999; 116:183S–189S)

	Introduction

TOP Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
Since its inception in 1989, raising the asthma awareness of patients, health professionals, and the public has been an important goal of the National Asthma Education and Prevention Program.¹ Although early asthma education programs focused on persons with asthma, newer initiatives are targeting the general public.^{2 3} However, little is known about the general public's current knowledge, attitudes, and beliefs about asthma.

The Chicago Community Asthma Survey (CCAS)-32 is a 32-item instrument developed to evaluate asthma knowledge, attitudes, and beliefs of the general public. The initial development of this instrument (previously described) included processes to identify items with strong face validity, as well as suitable performance characteristics (specifically floor/ceiling distributions and variability of responses).⁴

The purpose of this study was to examine the discriminative validity of the CCAS-32 on the basis of the respondents' prior asthma experience. It was hypothesized that respondents with the most experience, through having asthma themselves, would have the greatest proportion of correct answers to an asthma knowledge survey compared with individuals with no personal experience with asthma. It was also hypothesized that the asthma knowledge level of individuals who have a family or household member with asthma would be intermediate between individuals with asthma and those with no asthma experience.

A secondary purpose of this study was to examine the role of social demographic factors, eg, age, sex, race or ethnicity, educational attainment, income, and urban residence, on responses to items in the CCAS-32.

	Materials and Methods

TOP Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
This study involved a cross-sectional random-digit dialing telephone survey of adults in the Chicago area.

Survey Instrument
The CCAS-32 consists of 21 dichotomous items (true/false or yes/no), and 11 Likert-scale items. The scale for the Likert items was as follows: 1, never true; 2, rarely true; 3, sometimes true; 4, often true; and 5, always true. In general, the dichotomous items tested factual asthma knowledge, and the Likert items tended to assess attitudes and perceptions. The survey covered nine domains: (1) symptoms, (2) stigma/acceptability, (3) seriousness/severity, (4) perceptions of susceptibility, (5) consequences, (6) barriers to care, (7) perceptions of quality of life, (8) treatment/utilization of health care, and (9) triggers/environmental risk.

The survey introduction included one question about the respondent's general health. In addition to the 32 core items, the survey contained 4 items used to understand the respondent's experience with asthma. These questions were as follows: (1) "Does anyone in your family, including yourself have asthma (yes/no)?" If yes, "Is that person (2) yourself, or (3) a family member (choose all that apply)?" (4) "Thinking only of those people currently living in your household, including yourself, have any of the adults or children in your household ever had asthma (yes/no)?" Demographic questions were asked after completion of the CCAS-32, and included age, gender, race or ethnicity, education, and area of residence (zip code).

Survey Methods
Participants were adults ( 18 years), residing in Chicago and the seven surrounding counties (suburban Cook, Du Page, Kane, Lake, McHenry, Will counties, IL, and Lake County, IN). The survey was conducted from December 1997 through February 1998, via random-digit dialing of residential telephone numbers purchased from a commercial vendor (SDR Sampling Services; Atlanta, GA).

The survey was administered (in English only) by experienced and supervised telephone interviewers; the respondents remained anonymous. Calls were made between the hours of 9:00 AM and 9:00 PM. Any adult answering the telephone was considered a potential respondent and asked to participate. For telephone numbers that were not answered or busy, the callers made up to seven attempts at contact.

The pool of telephone numbers obtained from the vendor included 2,958 telephone numbers. Of these, 952 (32.2%) were not active residential numbers and were therefore considered ineligible. An additional 253 (8.6%) were considered ineligible because of technical telephone problems or because of language barriers. An additional 670 were of unknown eligibility because of repeated busy signals, no answer, answering machines, or requests to call back. Using methods described by Groves and Lyberg,⁵ it was estimated that 353 of the telephone numbers with unknown eligibility were likely to be ineligible. This resulted in a total of 1,558 (52.7%) estimated ineligible numbers. Of the remaining 1,400 eligible numbers, 568 interviews were completed, for a response rate of 40.6%. An additional 30 interviews (1.9%) were partially completed (and are not included in this analysis).

Statistical Analysis
Data analyses were conducted using SAS software (version 6.12; SAS Institute; Cary, NC). Tests of significance were performed using ² or nonparametric analysis of variance (ANOVA). Means are reported with the SE. Tests of trend significance were performed using regression models. The effect of the social demographic characteristics was additionally evaluated by multiple ANOVA, using a simple summary score of desirable responses to the 21 dichotomous items as the dependent outcome. Unlike the dichotomous items, the Likert-scale items were not amenable to a simple summary score, and multiple ANOVA was not performed.

Asthma experience was defined by the following criteria: "person with asthma" if the respondent reported themselves to have asthma; "no/low asthma experience" if neither the respondent, a family, or household member was reported to have asthma; and "family or household experience" if a family or household member, but not the respondent, was reported to have asthma.

Respondents were not asked to report their income. Instead, census data were used to identify the median household income for each zip code. United States census data were obtained using the software program Census CD and Maps (Geolytics; East New Brunswick, NJ). Respondents whose zip codes had median incomes > 50th percentile for the seven-county area were classified as "high income"; those who resided in zip codes with median incomes 50th percentile were classified as "low income."

This study was approved by the Institutional Review Board of Rush-Presbyterian-St. Luke's Medical Center.

	Results

TOP Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
Figure 1 presents the rates of completed calls per 100,000 population by county. The sampling fraction ranged from 1.3 completed calls per 100,000 population in DeKalb, (a county not in the target area, but with some overlap in telephone exchanges) to 10.9 completed calls per 100,000 population in McHenry County.

View larger version (56K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Rates of CASI CCAS-32 survey calls completed within the Chicago area by county (n = 568).

Effect of Asthma Experience on Responses to the CCAS-32
Of the respondents, 352 (62.0%) had no or low asthma experience, 162 (28.5%) had family or household asthma experience, and 54 (9.5%) were self-reported to have asthma. As seen in Table 2 , asthma experience was associated with a trend effect for differences in the proportion of the sample with desirable responses to many of the dichotomous items in the CCAS-32. The mean percentage of desirable responses differed significantly between persons with no or low asthma experience (70.0 ± 0.6%), family or household asthma experience (74.0 ± 0.9%), and persons with asthma (77.7 ± 1.2%; p < 0.01 for trend). For 10 of the 21 dichotomous items, the trend demonstrated that persons with asthma had the highest proportion of desirable responses, and persons with no or low experience, the lowest proportion. The proportion of correct responses from persons with family or household members with asthma generally fell somewhere between the two other groups. For one item, "when a person with asthma is doing well, they do not need to go to the doctor," this trend was reversed, with the desirable response (false) chosen by 84.9% of persons with no or low experience as compared with a slightly lower percent (81.5%) of persons with asthma (p < 0.05 for trend).

Asthma exposure was also associated with differences in mean responses to 5 of the 11 Likert-scale items (45.5%). For example, the mean response to "asthma is a serious disease" was 4.2 among persons with no or low asthma experience, 4.4 among those with family or household exposure, and 4.6 among persons with self-reported asthma. (p < 0.05 for trend).

Effect of Social Demographic Characteristics on Responses to the CCAS-32
Statistically significant differences between men and women were seen for 5 of the 21 dichotomous items (23.8%). For all five of these items, female respondents had a higher proportion of desirable responses than male respondents. Only 1 of the 11 Likert-scale items ("the emergency room is the best place to get treated for an asthma attack") exhibited significant but modest sex-specific differences, with a mean response of 3.1 for men, compared with 3.4 for women (p < 0.05).

The age of the respondent was also associated with differences in desirable responses to the dichotomous section of the survey instrument (Table 3 ). Ten individual items (47.6%) showed significant trend effects for age. For six of these items, older individuals had fewer correct responses, and for four items, the trend was significant for more correct responses with increasing age.

Race or ethnicity also had an impact on responses to selected dichotomous items, with responses to two items displaying significant differences between white and Hispanic or Latino respondents, and one item with differing responses between white and African-American respondents. For example, "asthma cannot be cured (true/false)" was answered correctly (true) by 52.2% of white respondents, compared with only 28.2% of Hispanic or Latino respondents. White respondents differed from African- American respondents in their answers to the question "a vaporizer is a good treatment for asthma (true/false)" (37.0% vs 17.9% correctly answering false, p < 0.01).

Seven of the 11 Likert-scale items displayed some variability in responses according to the respondent's racial or ethnic group, and for 3 of these 7, there were multiple differences. For example, in response to the statement "the emergency room is the best place to get treated for an asthma attack," the mean response for the white respondents was 3.2, compared with 3.8 for African- American respondents, and 3.7 for Hispanic or Latino respondents (p < 0.001).

Median zip code income was associated with different proportions of correct responses to 5 of the 21 dichotomous items (23.8%) (Table 3) . For four of the items, respondents living in high-income communities had higher proportions of correct responses than respondents in low-income communities. For example, "if you had asthma, you would know where to go for treatment" was answered in the affirmative by 82.4% of the respondents in high-income communities compared with 73.6% of those in low-income communities (p < 0.05). Similarly, 95.0% of high-income respondents answered that "asthma is mainly an emotional illness" is false, compared with 84.5% of low-income respondents (p < 0.001).

Four of the 11 Likert-scale items (36.4%) showed a response difference related to income. For example, among persons living in high-income areas, the mean response to the item "people with asthma who get relief from over-the-counter drugs like Primatene Mist still need to see their doctor" was 4.4, compared with 4.2 for those in low-income communities (p < 0.05).

Few differences emerged related to urban vs rural residence of the respondent. For three dichotomous items, residents of suburban areas were more likely than urban residents to choose the desirable response. Urban vs suburban differences were also seen for three of the Likert-scale items.

Items With Multiple Demographic Influences
Individual items varied in the extent to which the responses were influenced by multiple social demographic variables. Responses to 7 of the 32 items (21.9%) were unaffected by sex, age, education, race, urban vs suburban residence, and income. In contrast, responses to 18 items (56.3%) were influenced by two or more social demographic variables. Responses to only one item, "the emergency room is the best place to get treated for an asthma attack," were influenced by all the above social demographic variables. All six social demographic covariates proved to be significant effect modifiers on the summary score of the 21 dichotomous items.

	Discussion

TOP Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
The CCAS-32 was developed in part to identify specific areas of low knowledge and uncertainty related to asthma among the general public. This telephone survey of Chicago-area adults provides new insight into how an instrument such as the CCAS-32 may be used. Responses to most of the dichotomous items indicated only a fair degree of knowledge about the asthma disease process, signs and symptoms, triggers, etc. (mean correct responses, 71.9%). Responses to the Likert-scale items also indicated some uncertainty (nearly half the items had means near the middle of the scale). This was, in part, an expected finding, as items were chosen for inclusion in this survey on the basis of these performance characteristics for smaller Chicago-area samples.⁴ The demonstration of similar item-performance characteristics in this large sample serves to confirm the earlier work related to the appropriateness of the items in the CCAS-32.

Although the overall responses suggest a fair degree of uncertainty, it was notable that for 10 of the 21 true/false items, correct responses were chosen by > 80% of the respondents. This occurred despite a development process designed to reject items with this degree of a ceiling effect. One plausible explanation is that the respondents in this sample appeared to have higher levels of education than the samples used in earlier testing.⁴ Future development of this survey will require examining the performance of these items in diverse populations, perhaps resulting in a reevaluation as to the usefulness of these items.

These findings also highlight several important misconceptions held by the Chicago-area adults surveyed. Among the most striking of these misconceptions was that fewer than half of respondents knew that asthma cannot be cured, and fewer than one third correctly characterized the statement "a vaporizer is a good treatment for asthma" as false. Similarly, the survey identified other possible areas of limited knowledge, such as incorrect responses to the following: nocturnal cough as a sign of asthma, cockroaches as an asthma trigger, and asthma as a common cause of school absences. There also appeared to be a fair amount of uncertainty among the respondents about whether asthma hospitalizations are preventable, and whether asthma medications are potentially addictive.

In a validation study of an asthma knowledge questionnaire designed for parents and patients with asthma, Fitzclarence and Henry⁶ noted differences in scores between parents of children with asthma and other adults having little contact with asthma. Similarly, in this study, the demonstrated effect of asthma experience on item performance suggests the discriminative validity of the instrument. Because there is no "gold standard" for asthma knowledge, the assumption was made that individuals with greater personal experience with asthma should have higher levels of asthma knowledge as compared with individuals with less asthma experience. The trend effect seen in this study suggests that many of the CCAS-32 items have sufficient ability to detect meaningful degrees of difference between groups.

Little is known about the effect of social demographics on asthma knowledge at the community level. However, it would be surprising if, in a survey of knowledge, attitudes, and beliefs, no differences were detected by social and cultural demographic variables. The findings of this study suggest that age and education were the most frequently identified social demographic variables affecting responses to dichotomous items. For Likert items, age and education remained important in terms of the frequency of items with significant differences, but race or ethnicity also appeared to play a role. Of note, the items in the CCAS-32 appeared to differ in the degree of social demographic influence.

Other studies have shown that risk for asthma morbidity and mortality is related to various social demographic factors (ie, age, race, ethnicity, income, and geography).^{7 8 9 10 11} It is possible that the differences in asthma awareness between different subpopulations, as seen in the responses to the CCAS-32, may underlie some of the differences in community risk for asthma morbidity. Therefore, it is hoped that this instrument will be used to gain additional knowledge about the relationship between social demographic variables and asthma knowledge. Individuals or groups planning to use the CCAS-32 should be aware of the findings related to age, education, and race or ethnicity when surveying diverse populations. Comparing responses between two demographically dissimilar populations may require adjustment for the effect of social demographic variables.

This study has several important limitations. As it was designed to test instrument performance, and not to determine population knowledge, the protocol did not maximize attempts to reach households by calling to completion. Thus, the results from this sample should not be interpreted as representative of all Chicago-area residents. The extent to which the response rate of 40.6% limits the generalizability of these findings is not known. Random-digit dialing methods have been shown to be both valid and reliable; however, exclusion of households without telephones is always a source of bias.¹² Also, there are demographic differences in willingness and ability to be interviewed. In this study, the respondents appeared to reflect a higher proportion of women and individuals with higher levels of education compared with the general population of the Chicago area. Also, the method used to classify asthma experience was a relatively simple measure that may have resulted in misclassification; persons labeled as "no experience" may have had experiences that were not captured using questions about personal, family, and household experience.

It is hoped that the CCAS-32 will provide guidance for the design of asthma-related community education and public awareness interventions. We believe that the CCAS-32 is a useful tool for assessing community knowledge, attitudes, and beliefs about asthma, and for examining how these attributes may differ in certain segments of the population.

	Appendix 1

TOP Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 
Other members of the CASI team include (in alphabetical order): Claudia Baier, MPH, Steven Daugherty, PhD, Edward Eckenfels, Christopher Lyttle, MS, Anita Malone, MPH, and Karen Turner-Roan, MPH, of Rush-Presbyterian-St. Luke's Medical Center, Chicago, IL; Michael McDermott, MD, and James Moy, MD, of Cook County Hospital, Chicago, IL.

	Acknowledgements

 
The authors would like to thank the following individuals for their work in data collection: Bob Sprengel, Mary Marre, Cynthia Ortega, Marilyn Bradshaw, Arline Wilson, and Nada Smith. We would also like to thank Ms. Robin Wagner for her assistance in manuscript preparation

	Footnotes

 
CASI is funded by a grant from the Otho S.A. Sprague Memorial Institute.

Abbreviations: ANOVA = analysis of variance; CASI = ChicagoAsthma Surveillance Initiative; CCAS = Chicago Community Asthma Survey

	References

TOP Abstract Introduction Materials and Methods Results Discussion Appendix 1 References

 

1. Lenfant, C, Hurd, SS (1990) Special report: National Asthma Education Program. Chest 98,226-227[Free Full Text]
2. Action against asthma: a strategic plan for the Department of Health and Human Services. Washington, DC: Department of Health and Human Services, 1999
3. Campbell, M, Cormier, J, Daglish, S, et al (1994) Consideration of public programs and techniques for public/community health education. Chest 106(suppl),274S-278S
4. Grant EN, Turner-Roan K, Daugherty SR, et al. Development of a survey of asthma knowledge, attitudes and perceptions: The Chicago Community Asthma Survey-32 (CCAS-32). Submitted for publication
5. Groves, RM, Lyberg, LE (1988) An overview of nonresponse issues in telephone surveys. Groves, RM Biemer, PP Lyberg, LEet al eds. Telephone survey methodology ,191-211 John Wiley & Sons New York, NY.
6. Fitzclarence, CAB, Henry, RL (1990) Validation of an asthma knowledge questionnaire. J Paediatr Child Health 26,200-204[ISI][Medline]
7. Mannino, DM, Homa, DM, Pertowski, CA, et al (1998) Surveillance for asthma: United States, 1960–1995. MMWR CDC Surveill Summ 47(SS-1),1-28[Medline]
8. Lang, DM, Polansky, M (1994) Patterns of asthma mortality in Philadelphia from 1969 to 1991. N Engl J Med 331,1542-1546[Abstract/Free Full Text]
9. Weiss, KB, Wagener, DK (1990) Changing patterns of asthma mortality: identifying target populations at high risk. JAMA 264,1683-1687[Abstract]
10. Gergen, PJ, Mullaly, DI, Evans, R, III (1988) National survey of prevalence of asthma among children in the United States, 1976–1980. Pediatrics 81,1-7[Abstract/Free Full Text]
11. Sly, RM (1989) Mortality from asthma. J Allergy Clin Immunol 84,421-434[CrossRef][ISI][Medline]
12. Massey, JT (1988) An overview of telephone coverage. Groves, RM Biemer, PP Lyberg, LEet al eds. Telephone survey methodology ,191-211 John Wiley & Sons New York, NY.

This article has been cited by other articles:

				K. B. Weiss and E. N. Grant The Chicago Asthma Surveillance Initiative: A Community-Based Approach to Understanding Asthma Care Chest, October 1, 1999; 116(suppl_2): 141S - 145S. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Free Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Article Archive Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Weiss, K. B. Articles by Li, T. Search for Related Content PubMed PubMed Citation Articles by Weiss, K. B. Articles by Li, T.