2015 Child Asthma Data: Technical Information

Viewing and Printing Tables and Maps

Set the view option for your browser to the smallest available text size. This will ensure that columns of numbers are properly formatted and that part of the table is not lost when printing.

Tables are provided in both HTML and PDF format. The PDF files will load more quickly, but cannot be used with ADA assisted technology. Tables printed in portrait orientation using Adobe Acrobat may produce more pleasing results than printing the HTML-formatted tables.


Behavioral Risk Factor Surveillance System (BRFSS)

In 2015, the BRFSS survey was conducted in all 50 states, the District of Columbia, Guam, and Puerto Rico.

In each area where interviews were conducted, respondents were randomly selected from non-institutionalized civilian adults (18 years of age or older) living in households with a telephone (The BRFSS Data User Guide: http://www.cdc.gov/brfss/data_documentation/PDF/UserguideJune2013.pdf Cdc-pdf[PDF – 605 KB]).

Child asthma prevalence data are produced from the responses to questions on BRFSS Random Child Selection and Childhood Asthma optional modules. A randomly selected adult provided responses to the child asthma questions. If the adult reported that one or more children aged 17 years or younger lived in the household, then one of the children was randomly selected as the “Xth” child. Questions on the BRFSS Random Child Selection and Childhood Asthma optional modules were asked of the “Xth” child.

The child prevalence tables include data for the thirty-one areas, and one territory that used both the BRFSS Random Child Selection and Childhood Asthma optional modules on both the cell phone and the landline samples in 2015. The included areas were:

California, Connecticut, District Of Columbia, Georgia, Hawaii, Illinois, Indiana, Kansas, Louisiana, Maine, Massachusetts, Michigan, Minnesota, Mississippi, Missouri, Montana, Nebraska, Nevada, New Hampshire, New Jersey, New Mexico, New York, Ohio, Oklahoma, Oregon, Pennsylvania, Rhode Island, Texas, Utah, Vermont, Wisconsin, Puerto Rico

The following 4 states were not included in the child asthma prevalence tables because, while they did the Random Child Selection module, they did not do the Childhood Asthma module in 2015:

Maryland, North Carolina, South Carolina, South Dakota

The following two asthma questions pertinent to prevalence estimation were asked on the BRFSS Childhood Asthma prevalence optional module.

Lifetime asthma: Question number 23.1
“Has a doctor, nurse or other health professional ever said that the child has asthma?” (Variable name: CASTHDX2 [Yes = 1 / No = 2])

Current asthma: Question number 23.2
“Does the child still have asthma?” (Variable name: CASTHNO2 [Yes = 1 / No = 2])

Subgroup tables are based on data from the following questions of Random Child Selection optional module.

Survey Questions, Variable Names and Codes Used in Subgroup Tables

Survey Questions, Variable Names and Codes Used in Subgroup Tables
Survey Question Variable name Codes used
22.1 What is the birth month and year of the “Xth” child? RCSBIRTH* Grouped into age categories
(in years):0 – 4, 5 – 9, 10 – 14, 15 – 17
22.2 Is the child a boy or a girl? RCSGENDER 1 = Boy
2 = Girl
1.01 Child Hispanic, Latino/a, or Spanish origin calculated variable _CHISPNC 1 = Yes
2 = No
1.04 Child Non-Hispanic Race including Multiracial _CRACE1 Grouped into race categories:White, Black, Other/Multiracial races

* The child’s age was calculated by (IDATE-RCSBIRTH)/365.25, where IDATE is the interview date and RCSBIRTH is the child’s birth month and year with 15th as the day. The values for RCSBIRTH are not reported in the public data, but they are in each state’s internal data.  Top of Page

BRFSS Sampling and Weighting Changes

The 2011 BRFSS data collection changed from a landline sample survey to a dual-mode survey, landline and cellphone. The 2015 BRFSS data continues to reflect the changes initially
made in 2011 in weighting methodology (raking) and the addition of cell phone only respondents. The aggregate BRFSS combined landline and cell phone dataset is built from the landline and cell phone data submitted for 2015 and includes data for 50 states, the District of Columbia, Guam, and Puerto Rico. More information about the changes to the 2011 BRFSS can be found at the Web location: https://www.cdc.gov/surveillancepractice/reports/brfss/brfss.html.

For data analysis, the 2011 BRFSS data should be considered a baseline year and are not directly comparable to previous years of data because of the changes in weighting methodology and the addition of the cell phone only respondents. Please see the BRFSS Frequently Asked Questions document for additional information: https://www.cdc.gov/surveillancepractice/reports/brfss/brfss_faqs.html

 Top of Page

Survey Design and Sample Weights

The survey design and the child-weighting methods are described in the technical document, “OVERVIEW: BRFSS 2015”, which can be found at the Web location: http://www.cdc.gov/brfss/annual_data/2015/pdf/Overview_2015.pdf Cdc-pdf[PDF – 120 KB].


Data Analysis

Data Used:

The BRFSS 2015 survey dataset (http://www.cdc.gov/brfss/annual_data/annual_2015.html) was used to calculate prevalence estimates for all states that had the Random Child Selection & Childhood Asthma optional modules included in the common questionnaire.

Prevalence and standard error estimates were calculated using SAS callable SUDAAN Release 11.1 (Research Triangle Institute, Research Triangle Park, NC).

Data Management:

Responses of “don’t know/not sure,” and “refused” were re-coded as missing values. Children older than 18 years were excluded from further analysis.

Confidence Intervals and Prevalence Numbers:

The 95% confidence intervals were calculated using the following formulas:

Lower 95% confidence interval = % prevalence – tv (sep)
Upper 95% confidence interval = % prevalence + tv (sep)

Where sep is the standard error of the prevalence percent and tv represents the student’s t distribution with v degrees of freedom.

Small Sample Size and Failure of Distributional Assumptions

When sample sizes for a particular state or territory was smaller than 50, the calculated standard error of the prevalence estimate might have been large relative to the point estimate of the prevalence value, which caused a wide 95% confidence interval. This sometimes caused the lower limit of the 95% confidence interval to be negative. In which case, the lower limit of the interval was set to zero.

In situations where the normal distribution approximation to the binomial distribution did not hold, values for the standard error and the 95% confidence interval of estimates were not provided, because these estimates are not reliable.

Table Conventions:

In the tables, states and territories are listed in Federal Information Processing Standard (FIPS) code order.

 Top of Page

Page last reviewed: November 27, 2017