Vol. 13 •Issue 10 • Page 14
Allergy & Asthma
Interstate Asthma Prevalence Variability Exists, but Is It for Real, and Does It Matter?
Asthma prevalence varies dramatically depending on the population and age group studied, research shows. The same also can be said about geographic location.
Various national surveys have placed the prevalence of “current asthma” in U.S. adults between 7 percent and 8 percent.1,2 These estimates come from the Centers for Disease Control and Prevention’s Behavioral Risk Factor Surveillance System (BRFSS) — the only national surveillance survey that provides state-by-state data on various health risk factors, including asthma.
In 2002, asthma prevalence ranged from 5.8 percent in South Carolina to 10.0 percent in Maine.3 (See Figure, page 16.) This Figure demonstrates that the prevalence of adult “current asthma” varies from state to state and that this has changed slightly over the three years displayed.
IS IT REAL?
An important question is if these observed differences are real. The first step in addressing this is to better understand the nature of the BRFSS data and the means by which they’re acquired.
The BRFSS data are accumulated through state run, random-digit-dialed telephone surveys. The purpose of this program is to determine the prevalence of many health risk behaviors among the noninstitutionalized, civilian U.S. population ages > 18 years and their association with premature morbidity and mortality.
Emphasis is on collection of data related to actual behaviors, rather than on attitudes or knowledge of disease states.
The asthma module consists of the following two questions: “Have you ever been told by a doctor, nurse or other health professional that you have asthma?” and “Do you still have asthma?”
Data for all 50 states, the District of Columbia and three territories have been available since 2000, although some states had asked about asthma prevalence in earlier years.
While the general survey provides an estimate of adult asthma prevalence and current incidence, it doesn’t assess severity, current level of therapy or exacerbations. These data are acquired in an eight question Adult Asthma History Module that was used in 19 areas in 2003.
In the absence of clinical information, it’s not possible to determine whether “lifetime” or “current” asthma, as measured by the BRFSS, truly captures physician-diagnosed asthma. What can be determined, however, is whether this measure is similar from year to year.
Over the three years depicted in the Figure (2000 to 2002), there has been a slight increase in “current” asthma prevalence. The data also appear to be consistent, in that states tend to stay in the same or adjacent prevalence categories from one year to the next.
Another means of explaining whether these differences are real is to look at other factors that both are related to asthma and vary by state.
Thus, variability in prevalence may be related to interstate variability in population age, racial mix, cigarette smoking, body mass index, early childhood exposures, airborne pollution exposure or socioeconomic status. Some of these factors, such as age, class, gender, race/ethnicity, smoking status, body mass index, education class and income level, are part of BRFSS, whereas air pollution exposure, atopy and early life exposures aren’t.
Another likely source of this variation and, perhaps, the most difficult to measure, relates to physician practices. (How likely is it that a physician, given a certain clinical presentation, will diagnose a person as having asthma instead of a number of other diagnoses?)
Similarly, the correlation between the physician’s diagnosis and the patient’s understanding of that diagnosis is another source of variation that’s difficult to assess.
The geographical patterns shown in the Figure can help generate potential hypotheses for why these rates may vary between states. For example, some air pollution studies suggest geographic clustering of asthma, and air pollutant levels tend to be higher in the Northeastern United States, where asthma rates are also higher.4 Conversely, air pollutant levels are high in California, which has lower asthma rates.
Another intriguing explanation of state-by-state variability of asthma prevalence is the character of early childhood exposures to allergens and tobacco smoke and active smoking as adults.5-7
Again, though, the findings are inconsistent, with some high smoking states (Kentucky, Nevada) having high asthma rates and others (North Carolina, Louisiana) having lower rates.
States also vary in their demographic and socioeconomic status (SES) characteristics. As observed above, though, there’s no consistent pattern between expected SES in the states and asthma prevalence (i.e., with higher SES states having consistently more or less asthma than lower SES states).
Turning from known risks for asthma, one could attribute state-based differences in asthma prevalence to diagnosis differences. In this manner, one would have to postulate that there are regional differences in making asthma diagnoses or access to care. While this explanation is appealing, there’s no available data to support such widespread variation in health care delivery.
A final consideration is that self-reported “current asthma” isn’t an accurate marker of true asthma in the population. The purpose of the BRFSS is to collect data related to actual behaviors, rather than on attitudes or knowledge of disease states. It’s entirely possible that asthma prevalence may be underreported in some areas because individuals are poorly aware of their disease.
DOES IT MATTER?
The second key question is whether interstate variations in asthma matter. While these differences make for interesting hypotheses, the reality is that whether 5.8 percent or 10.0 percent of a state is affected, asthma still affects a significant proportion of the population.
Furthermore, for any individual, his or her risk of asthma is either 0 percent or 100 percent. We have no “primary prevention program” for asthma. People with asthma, though, should receive appropriate medical care that results in their maintaining a healthy life, and this needs to be done no matter what the prevalence of asthma is in any state, county or city.8
In conclusion, self-reported asthma prevalence has some variability between states, and this phenomenon isn’t entirely explainable by currently available data. This variability should remain an active area of investigation because it may foster a better understanding of the factors leading to the development of asthma.
More importantly, however, is that patients receive maximal treatment to control symptoms and maintain their quality of life. Asthma has a significant effect on society and is likely modifiable based on adherence to National Asthma Education and Prevention Program guidelines.
Dr. Hiestand is a senior fellow in pulmonary and critical care medicine at the University of Kentucky Medical Center, Lexington. Dr. Mannino is an associate professor of medicine in pulmonary and critical care medicine at the same facility.
For a list of references, please call Mike Bederka at (610) 278-1400, ext. 1128, or visit www.advanceweb.com/respmanager.