Background
Asthma is a leading chronic pediatric health condition affecting approximately 6.2 million children under age 18 years in the United States (US) [
1]. Current asthma is reported in 10% of adolescents (2.5 million) age 12–17 years [
1], and the alarming rates of asthma and its increasing morbidity in urban young people are particularly concerning [
2‐
5]. The disproportionate burden of asthma morbidity in urban children has been attributed to their exposure and heightened sensitivity to certain indoor allergens such as cockroaches or mice [
2,
6]. Exposure to some indoor and outdoor allergens has been identified as a major culprit in the development and exacerbation of asthma in pediatric patients [
7]. Evidence has consistently shown that children with severe asthma tend to report greater allergic burden [
8‐
11]. Similarly, a large epidemiologic study reported a major impact of allergen exposures and sensitivity on symptom severity and acute healthcare utilization in children with asthma [
6].
Despite the wealth of literature elucidating the intricate links between inhalant or food allergen sensitization and asthma morbidity in pediatric populations, the majority of evidence is based on either young children or mixed age groups of young children and adolescents. Little is known about common allergies and their relationships with asthma morbidity specifically among adolescents. Furthermore, a number of studies target only specific allergens (e.g., cockroach or mice), yet there is limited literature comparing various types of allergies and their associations with asthma morbidity. Thus, the purpose of this study was to examine the prevalence of a broad spectrum of indoor/outdoor allergies among urban adolescents and to assess their relationships to asthma control and healthcare utilization.
Discussion
This study demonstrates the ubiquitous prevalence of a wide range of self-reported allergies among urban adolescents with asthma, and their associations with asthma morbidity. Dust mites were identified as the most common allergen, reported by more than a half of this population, and the prevalence was strikingly higher, nearly 70%, among adolescents who had been tested for sensitivity. The rate of dust mite allergy in our study is substantially greater than the 35% reported in a previous large study of inner-city children [
6]. The difference might be due in part to differences in data sources between our finding and the earlier report, self-report vs. skin testing, respectively. In our study, the self-reported allergy information was based on prior sensitivity testing as well as their experience of allergic reaction to dust mites, thus resulting in a higher rate. Nonetheless, given that the high proportion of adolescents (> 82%) reporting dust mite allergy had been tested for sensitivity, our estimate of mite allergy may closely align with that of sensitivity testing.
We found that allergies to cockroaches and mice were linked to poor asthma control and more frequent ED visits, hospitalization, or specialist visits. Similarly, Busse et al. [
6] identified cockroach allergen as a dominant and sole factor of other allergens linked to greater asthma morbidity and urgent healthcare use in young inner-city children. In their study, sensitivity to cockroaches was found in about 37% of the inner-city children; that is slightly higher than our 31% based on self-report. In many urban dwellings, cockroach and mouse infestation is widespread [
2,
6,
13], which is often linked to suboptimal living conditions with compromised structural integrity (e.g., cracks and holes in the wall, water leaks) of the house [
14,
15]. Cockroach allergens were detected in over 85% of inner-city houses [
2], and even when there was no apparent exposure to cockroaches at home, children can become sensitized to the pest from exposures at other places. African American children, 6–16 years of age, are 2.5 times more likely to have cockroach sensitivity than their white counterparts [
16]. This alarming rate underscores the importance of a pest-reduction intervention targeting the broader urban community instead of individual houses. Regional differences in the rates of mouse allergy in our study could be due in part to differences in the types of housing. Compared to other sites, Baltimore’s predominant use of multifamily dwellings with shared walls and/or floors/ceilings may provide an ideal environment for the widespread infestation of cockroaches and mice.
The prevalence of self-reported food allergies in our adolescent sample was 30%, a rate slightly higher than that of a previous report in younger children with asthma, 24% [
10]. However, our food allergy rate is considerably higher than the prevalence of self-reported food sensitization, 17%, in the general population based on a large U.S. national survey [
11]. Similar to the previous reports [
11], we found peanut to be the most common food allergen in adolescents with asthma. This is the first study documenting the extent of peanut allergy in adolescents with asthma. In contrast to the previous study [
11], in which food allergy was associated with the increased risk of ED visits and asthma exacerbations, we found no significant associations between food allergy and urgent healthcare utilization or exacerbation. This discrepancy may have been in part due to the differences in the sample in the two studies. While the earlier report was based on the general population of all ages, our sample is limited only to urban adolescents. Another reason might have to do with asthma severity. Liu et al. study [
11] included those of a wide range of severity, and revealed that food allergy became more prevalent in those with higher levels of asthma severity, thus resulting in ED visits more often. In contrast, all of our participants had persistent asthma and over 85% had not-well controlled or very poorly controlled asthma based the NAEPP classification. The lack of variability due to the relative homogeneity regarding disease severity might have prevented us from detecting any differences in healthcare utilization associated with food allergy. Also, without assessment of specific serum IgE levels, our study is unable to quantify the varying degrees of food sensitivity. If only severe food allergy is linked to asthma morbidity, possible mild cases of food allergy in many of our sample might have contributed to non-significant relationships between the allergy and asthma morbidity in this study.
Plant-based allergies (grasses, trees, and ragweed) were highly prevalent in our urban adolescent sample. Consistent with other studies [
17,
18], we found no associations between plant allergies and asthma morbidity, except that those with a plant-based allergy were more likely to receive specialist care for asthma. Pet allergies were also common: however, unlike a previous study demonstrating positive associations between pet allergies and asthma severity [
19], we found no relationships between pet allergies and asthma morbidity.
It is noteworthy that the magnitude and directions of associations between certain allergies and healthcare utilization differed by sensitivity testing status. We demonstrated that dust mite allergy was associated with substantially increased urgent office visits due to asthma among the tested group compared to their untested counterparts. Likewise, the positive associations between pest allergies and specialist visits appeared more pronounced among the tested group than their untested counterparts. These findings may simply reflect common clinical practice that sensitivity testing is more often indicated when a patient is presented with severe or uncontrolled asthma necessitating frequent acute office visits or specialist care. When we considered only the subgroup with prior sensitivity testing, however, the relationships between those pest allergies and acute healthcare utilization (ED and hospital admission) became weak and nonsignificant. This may have been due in part to diminished power (sample size) or the possibility that the pest allergies reported by those never tested for sensitivity could be more strongly linked to the acute healthcare utilization than the tested adolescents. The latter speculation is somewhat supported by our finding that the association between mouse allergy and ED visits was negative among those who were tested, indicating fewer ED visits associated with mouse allergy by the tested subjects than those untested. This seemingly counterintuitive finding might have been reflecting the influence of intervening effects, such as treatments (e.g., immunotherapy) or modifications of the home environment to eliminate the pest that were offered to those who tested positive to mouse allergen, resulting in fewer acute asthma episodes requiring emergency care. On the other hand, the untested patients with mouse allergy were left untreated, continuing to suffer greater morbidity represented by more frequent ED visits. Such implication underscores the importance of conducting sensitivity testing for asthma patients reporting pest allergies or living in pest-infested areas to adequately address the issue and ultimately improve asthma outcomes. Future research is warranted to investigate the implications of sensitivity testing for treatment choices and its impact on asthma outcomes in urban adolescents.
This study has several limitations that warrant caution. First, this study relies on self-report data rather than sensitivity testing. Therefore, we cannot rule out the possibility of reporting bias in our prevalence estimation, particularly for about 40% of participants who had never been tested for sensitivity. Those untested adolescents might not be aware of their sensitivity status unless they had prior allergic reactions to certain inhalants or foods, resulting in underreporting allergies. Second, because of the absence of information about specific IgE levels, we were unable to assess how varying degrees of allergic sensitivity play a role in the relationships between certain allergies and asthma morbidity. Third, although this study is based on a relatively large number of urban adolescents representing three discrete regions in the U.S., the sample is by no means representative of urban adolescents in other parts of the country. Moreover, selection bias toward those with higher asthma severity may have occurred as a large proportion (34%) of the sample were referred by clinicians. Future research using a representative sample of urban adolescents is needed to minimize the selection bias. Finally, this study was based on cross-sectional survey data, which limits our ability to making inference about causal links between allergies of various types and asthma morbidity.
Despite the identified limitations, this study offers important insight into the relative prevalence of self-reported allergies of various types in urban adolescents with asthma and the relationships between each type of allergy and asthma morbidity. In general, allergies appear to increase the likelihood of asthma-related healthcare utilization in urban adolescents. Particularly, pest allergies are linked to uncontrolled asthma and exacerbation as well as frequent use of acute healthcare services. Our findings underscore the importance of identifying and eliminating cockroaches and mice that have particular implications for asthma morbidity. Interventions aimed at reducing the level of indoor allergens have been found effective in improving asthma outcomes [
20]. Environmental interventions focusing on reduction of cockroaches have resulted in a decrease in the level of the allergens and improved asthma symptoms in urban residents [
21,
22]. However, extermination alone may have only minimal or fleeting effects on asthma morbidity unless it is accompanied by patient education and behavior modification that can augment and sustain the effects of environmental interventions. Behavior changes in parents/adolescents to control the pests and to increase treatment adherence in combination with environmental interventions would provide the best chance to achieve enduring optimum asthma control and contain healthcare costs [
2].
Authors’ contributions
HR as the principal investigator of the study made substantial contributions to conception and design of the study as well as acquisition, analysis and interpretation of data. HR played a leading role in developing and writing the manuscript. TL made substantial contributions to data analysis and interpretation and drafting the analysis and results sections. DH was involved in data analysis and drafting the result section. AG contributed to data management and drafting and critically reviewing the manuscript for important intellectual content. All authors read and approved the final manuscript.