Juvenile Asthma in the U.S. Relate Asthma Incidence to Body Mass

Index

Linshuo Yang

, Zhijie Duan

and Jingchen Chai

McMaster University, Canada

University of North Carolina at Chapel Hill, U.S.A

University of Alberta, Canada

Keywords: BMI, Asthma, Adolescent, Sex, Ethnicity, NHANES.

Abstract: Asthma is one of the most common chronic diseases among humans, juveniles. Obesity, as many studies

suggest, is related to the development of asthma. This research study intends to identify the strength of

correlation between BMI and asthma in different gender and ethnic groups and explore whether the incidence

of asthma increases with higher BMI. Here, researchers first divided the data collected from the National

Health and Nutrition Examination Survey into two groups by the variable _Have_Asthma_. T-tests for the

difference in means were then employed to show that the average BMIs for the two groups differ in a

statistically significant way. Quantile analyses were also used to compare the proportion of people having

asthma in each BMI quantile. To quantify the strength of correlation, a bootstrap confidence interval was built

and logistic regression models for the data from 2015 to 2016, and 2017 to 2018. To avoid the collinearity

between gender and BMI, the researchers also built separate logistic models for each gender. The authors

concluded that juveniles with higher BMI have a higher risk of getting asthma. Male juveniles have an overall

higher risk of getting asthma than female juveniles. Non-Hispanic black juveniles are more likely to have

asthma than Mexican American juveniles and non-Hispanic white juveniles.

1 INTRODUCTION

Asthma, a prevalent and variable disease, can range

in severity from mild, with few obvious symptoms, to

severe, which is characterized by acute or subacute

progressive exacerbations of asthma symptoms

(Lloyd, Price and Brown 2007, Mukherjee et al

2016). There is a large amount of epidemiological

evidence showing that obesity is one of the most

influential factors of asthma (Forno 2020). However,

few studies are targeting specific age groups,

especially juveniles (under 18 years old).

Bidirectional Mendelian randomized studies in

children and adults have shown an association

between genetic risk score for obesity and subsequent

development of asthma, but not between asthma

genes and subsequent obesity (Chen, Fan, Huang,

Liou and Lee 2019, Xu, Gilliland and Conti 2019). In

addition, studies have shown that weight loss can

significantly improve asthma symptoms and lung

function (Okoniewski, Lu and Forno 2019).

Similarly, some studies have found that obese asthma

patients are difficult to manage, with poor lung

function and weak response to asthma medication

(Boulet and Franssen, Dixon, Shade, Cohen et al

2006). As juvenile asthma and obesity are becoming

pressing public health issues around the globe,

investigating the strength of correlation between BMI

and asthma in the under-age population may benefit

the prevention and prognosis of juvenile asthma.

Based on the literature review, researchers in this

study hypothesized that the incidence of asthma

increases with higher BMI among juveniles. T-tests,

quantile analyses, bootstrap confidence intervals, and

logistic regression models were employed to identify

and quantify the correlation between BMI and asthma

incidence among juveniles grouped by different

ethnicities and genders.

2 METHODS

2.1 Data Source and Screening

Our data set was downloaded from the National

Health and Nutrition Examination Survey. The

194

Yang, L., Duan, Z. and Chai, J.

Juvenile Asthma in the U.S. Relate Asthma Incidence to Body Mass Index.

DOI: 10.5220/0011241600003438

In Proceedings of the 1st International Conference on Health Big Data and Intelligent Healthcare (ICHIH 2022), pages 194-199

ISBN: 978-989-758-596-8

researchers merged the 2015-2016 data frame and

2017-2018 data frame, getting a sample size of

19225. Then select all rows that have a value of age

less than 18, yielding a subset that has 7377

individuals. Next, screening the data, excluding

individuals that have a missing value in the columns

of Age, Sex, Ethnicity, Have Asthma, and BMI. The

final complete data frame includes 5666 individuals.

Welch’s T-tests

To avoid ambiguity and ensure enough data, Non-

Hispanic White, Non-Hispanic Black, and Mexican

American ethnicities were selected for further study.

To control variables, the whole data was subsetted

based on _Ethnicity_ (Mexican American, non-

Hispanic whites, and non-Hispanic black) and

_Gender_ (Male and Female). Six subgroups (e.g.

Male+Mexican American) were then obtained. Each

subgroup was further split by the categorical variable

_Have_Asthma_ into two groups, which are the

individuals that have asthma and the individuals that

do not have asthma. Two sample t-tests were then

conducted to compare the average BMI of asthma

patients and people without asthma in each subgroup.

Welch’s t-test was chosen because the two samples

compared have different sample sizes.

2.2 Quantile Analysis

Different from the previous subdivision, this time the

data was subgrouped by each gender and each

ethnicity separately, yielding five subgroups in total,

which were used to conduct quantile analysis. The

sample sizes of the five ethnic groups are 1678(Non-

Hispanic White), 1086(Mexican American),

1312(Non-Hispanic Black), 1000(Other Race), and

589(Other Hispanic). The sample sizes of the two

gender groups are 2842 (male) and 2823(female).

Q1 (First 25%), Q2(First 25%-50%), Q3(50%-75%),

Q4(Last 25%).

2.3 Model Building

A statistical comparison is made between two

classified sets. A preliminary hypothesis was made by

directly comparing the number of people who have

asthma in Q1-Q4. Case resampling will then be

utilized to obtain a 95% bootstrap confidence interval

for the correlation between BMI and asthma. Logistic

regression models were then built to explore the

correlation between asthma and other selected

variables.

3 RESULTS

Welch’s t-test for the Mexican American male group

shows that the average BMIs of people with and

without asthma are 20.81 and 22.84, respectively. The

other t-test for the Mexican American female group

showed that the average BMIs of people with and

without asthma were 20.63 and 23.15, respectively.

The p-values were both less than 0.05, indicating the

differences in means were statistically significant

(Figure 1).

Figure 1: T-test Result for Mexican American Samples.

Welch’s t-test for the non-Hispanic white male

group showed that the average BMIs of people with

and without asthma were 19.06 and 20.61,

respectively. The other t-test for the non-Hispanic

white female group showed that the average BMIs of

people with and without asthma were 19.11 and

21.73, respectively. The p-values were both less than

0.05, indicating the differences in means were

statistically significant.

Welch’s t-test for the non-Hispanic black male

group showed that the average BMIs of people with

and without asthma were 19.38 and 20.62,

respectively. The p-value was less than 0.05,

indicating the difference in means was statistically

significant. The other t-test for the non-Hispanic

black female group showed that the average BMIs of

people with and without asthma were 20.76 and

21.70, respectively. The p-values were greater than

0.05, indicating that the difference in means was not

statistically significant.

The BMI-based distribution quantile analyzed for

each gender group were in Figures 4 and 5. In general,

our male sample had a higher overall asthma

incidence (18.0%) than the female sample (13.5%).

Though the male sample had a higher proportion of

asthma population in all quantiles, there was not

much difference between the two samples in the

fourth quantile (23.7% vs. 21.2%), revealing the fact

that the incidence of asthma was nearly the same for

males and females with high BMI. In short, the

Juvenile Asthma in the U.S. Relate Asthma Incidence to Body Mass Index

195

incidence of asthma increased with higher BMI for

both genders, and male juveniles were more likely to

get asthma than female juveniles did (figure 2).

Figure 2: Total Male Samples .

Figure 3: Total Female Samples.

In all four BMI quantiles, the Non-Hispanic Black

sample exhibits a higher asthma incidence rate than

the other two ethnic groups. In the Non-Hispanic

Black sample, all the quantiles except Q1 had an

asthma incidence rate higher than 20% (Figure 3).

Figure 4: Non-Hispanic Black Samples.

The 95% bootstrap confidence interval is

(0.09507, 0.10764), indicating that 95% of the time

the population correlation between BMI and asthma

fell within the range of 0.095 and 0.107. In the

univariable logistic regression model (Figure 4), the

p-value was smaller than 0.05, suggesting that it is

significantly associated with asthma status. In the

multivariable logistic regression model (Figure 5),

variables that were significantly associated with

asthma include sex, age, BMI, and Mexican

American. The rest of the variables had an

insignificant association with asthma. Researchers in

this study found that sex had the lowest p-value,

indicating a strong association between gender and

the probability of having asthma. The positive

coefficient of this predictor suggested that with all

other variables being equal, males were more likely

to have asthma. Being male increased the log odds by

0.3474. The accuracy of this model showed 0.82. The

dataset was then split into two subsets by gender. P-

values in logistic regression models for males and

females, respectively, were both less than 0.05.

Figure 5: Univariable and Multivariable logistic regression

analysis, NHANES, 2015-2016,2017- 2018(5,666)

Figure 6 was created to help explain the results

from the logistic regression. From the graph, it was

obvious to see that people who did not have asthma

were significantly higher than people who have

asthma in the low to medium range of BMI from 15-

25. Although there were fewer data within the range

from 25-30 compared with the range from 15-25, it

was still clear to see that the people with asthma

occupy a larger proportion of the whole.

Figure 6: Distribution of asthma outcomes on the range of

BMI.

4 DISCUSSION

A study shows that treating obese asthma patients has

become a challenge since these patients have worse

lung function and are less responsive to asthma

medications. For these patients, the best and easiest

treatment is to lose weight. Significant weight loss

can improve asthma symptoms, lung function, or the

rate at which asthma worsens (Dhabuwala and

Cannan 2000). It was shown that bariatric surgery had

significant effects on asthma controls, which led to an

approximately 60% reduction in having asthma

ICHIH 2022 - International Conference on Health Big Data and Intelligent Healthcare

196

exacerbation (Peters, Dixon and Forno 2018). Their

research has proved that the incidence of asthma is

highly related to obesity, and losing weight is the best

way to suppress asthma attacks, which also supports

our conclusion that weight gain increases the risk of

getting asthma. Moreover, some dietary habits and

nutrition they absorb are also risk factors that might

have impacts on children (Peters, Dixon and Forno

2018). Lack of vitamin D is a risk factor for the

development of asthma (Peters, Dixon and Forno

2018). Infant feeding also plays an important role in

the development of asthma among children (Figure 7)

(Miliku and Azad 2018). Breastfeeding had also been

considered as a factor, children who had been

breastfeeding while they were infants were associated

with a lower risk of asthma (Yan, Liu, Zhu, Huang

and Wang 2014). This is due to the fact that there

exists components in breast milk that can protect

against allergies (Figure 8) (Oddy 2017). Therefore,

there are multiple facets in life associated with having

asthma. Children are also expected to develop good

habits and diet to prevent asthma. Asthma

exacerbations can take place at any time, however,

there is a seasonal pattern among children (Herman,

Hannah, Moshe, Erez and Ran 2014) , it was shown

that there is an increased risk during the autumn

(Sears and Johnston 2007). However, our research

also has some flaws. Our study used BMI as the only

indicator of obesity and did not delve into some other

features that are shared by those obese people but

cannot be quantified by BMI. Another study also

pointed out that although the BMI indicator is a

measure of physical factors, it does not identify body

composition, adipose tissue, or metabolic

disturbance. The criteria represented by BMI are too

broad to capture a specific understanding of the

nature of the disease (Forno 2020).

Figure 7: The impact of breastfeeding on lung health

(Miliku and Azad 2018).

Figure 8: Factors that indues and protects against food

allergies (Friedman and Zeiger 2005).

In this study, it was proven that the positive

correlation between BMI and asthma incidence is

significant, probably because obesity may affect free

air movement in the lungs and thus can have a severe

impact on the respiratory system (Azizipour and

Yosra et al 2018). However, we still cannot define a

causation relationship between BMI and asthma

incidence. One possibility people need to consider

was that having asthma might also influence BMI

through a confounding variable, which is exercise.

Though the value of exercise as a treatment aid was

emphasized by many studies, it is also pointed out

that vigorous physical activity can trigger bronchial

narrowing and may result in bronchospasm

(McFadden and Illeen 1994). Rapid breathing may

cause evaporation of mucosal surface water and an

increase in osmolarity, resulting in mast-cell

degranulation and contraction of airway smooth

muscle (Anderson 1984). In general, exercise-

induced asthma is seen more commonly among

children and young adults because of their high levels

of physical activity. Given the controversy of the

pathogenesis of exercise-induced asthma, some

physicians would advise asthma patients not to

overexert themselves. Thus, asthma patients may on

average have a lower level of physical activity than

people without asthma, putting them at a higher risk

of having higher BMI. In short, the authors can only

reveal a positive correlation at present. Whether there

is a causation relationship between BMI and asthma

incidence is still an unsolved question that needs

further research.

In an epidemiological related study, obese

individuals with a BMI greater than 30kg/m2 had a

92% increased risk of asthma (Beuther and

Sutherland 2007). This research supports our results

sufficiently. In a separate article of the effect of

obesity on the incidence of asthma, the authors noted

that analysis of both adults and children showed there

are no prominent differences between females and

males, which is consistent with our findings as well

(Julia st al 2015). Since BMI is a comprehensive

Juvenile Asthma in the U.S. Relate Asthma Incidence to Body Mass Index

197

measurement to evaluate an individual’s body

condition, to better study the relationship between

obese condition and asthma, a more precise definition

is needed. For example, to explore and reduce the risk

of asthma in obese individuals, nutritional indicators

can be used, which better describe the inner changes

of obese patients. Studies have shown that dietary

restriction in obese asthmatic patients can improve

bronchial hyperresponsiveness, airway inflammation

and other related diseases (Dixon et al 2011). In

contrast, a high-fat diet can increase airway

neutrophilia and impair bronchodilator recovery in

obese asthmatics (Wood, Garg and Gibson 2011).

Although the results cannot give strong evidence to

demonstrate that BMI plays a decisive role in the

prevalence of asthma, other studies reveal that an

unusual high BMI (>25kg/m2) often indicates an

individual is obese in most cases, and strict diet

control is proved efficient to reduce the incidence of

asthma. Therefore, our study can better suggest obese

individuals reduce the risk of having asthma

according to their BMI.

There are also some studies stating that girls with

high BMI would have a higher risk of having asthma

(Ulrik, Lophaven , Anderson, Sørensen and Baker

2018). However, in our study, the authors find that

there is no significant difference in asthma incidence

between boys and girls with high BMI. Apart from

sex and BMI, there are also other risk factors,

ethnicity is also considered as a risk factor of having

asthma. This is consistent with the results of our study

that the non-Hispanic Black set has a larger

proportion of people having asthma than others.

While many scientific research studies show that

more physical exercise is needed, clinical trials

should also be taken into action, as obese asthmatics

have multiple consequences related to mechanical or

physiologic effects, and immune or metabolic effects

(Baffi and Cynthia et al 2015).

5 CONCLUSIONS

In conclusion, this study reveals a significant positive

correlation between BMI and asthma incidence

among juveniles no matter if they are male or female.

Although the incidence of asthma is nearly the same

for males and females with high BMI, male juveniles

have an overall higher risk of getting asthma than

female juveniles, probably because boys are more

likely to have an inconsistent growth of their airway

diameter and their lung volume in their early life

(Fuseini, Hubaida and Dawn 2017). Additionally,

other factors such as ethnicity also exert some

influence on asthma outcomes. Non-Hispanic black

juveniles are more likely to have asthma than

Mexican American juveniles and non-Hispanic white

juveniles as their weight increases. However, as

mentioned previously, there are some limitations in

this study. For example, BMI is the only indicator of

obesity used in this study. Possible improvements can

be made by including more obesity indicators and

even other measurements of body condition. This

study can provide a reference for future prevention

and treatment of juvenile asthma. One of the

takeaways is that obese juveniles should be aware of

the importance of losing weight given our result that

the risk of having asthma increases with higher BMI.

Additionally, having balanced nutrition and a good

eating habit is also a key to preventing asthma. Future

research studies should also pay more attention to

both asthma prediction and prevention. Developed to

tackle asthma problems, personalized prediction

models not only can prevent attacks but can also

reduce attacks (Fleming 2018).

REFERENCES

Azizpour, Yosra et al. “Effect of childhood BMI on asthma:

a systematic review and meta-analysis of case-control

studies.” BMC pediatrics vol. 18,1 143. 26 Apr. 2018,

doi:10.1186/s12887-018-1093-z

Anderson, S. “Is There a Unifying Hypothesis for Exercise-

Induced Asthma?” Journal of Allergy and Clinical

Immunology, vol. 73, no. 5, 1984, pp. 660–65.

Crossref, doi:10.1016/0091-6749(84)90301-4.

Beuther, D. A. & Sutherland, E. R. Overweight, obesity,

and incident asthma: a meta-analysis of prospective

epidemiologic studies. Am. J. Respir. Crit. Care Med.

175, 661–666 (2007).

Baffi, Cynthia Wilson et al. “Asthma and obesity:

mechanisms and clinical implications.” Asthma

research and practice vol. 1 1. 4 Jun. 2015,

doi:10.1186/s40733-015-0001-7

Boulet LP, Franssen E. Influence of obesity on response to

fluticasone with or without salmeterol in moderate

asthma. Respir Med. 2007;101(11):2240 -2247.

doi:10.1016/j.rmed.2007.06.031

Chen YC, Fan HY, Huang YT, Huang SY, Liou TH, Lee

YL. Causal relationships between adiposity and

childhood asthma: bi-directional Mendelian

Randomization analysis. Int J Obes (Lond).

2019;43(1):73-81.

Dixon AE, Shade DM, Cohen RI, et al. Effect of obesity on

clinical presentation and response to treatment in

asthma. J Asthma. 2006;43(7):553-558.

doi:10.1080/02770900600859123

Dhabuwala A, Cannan RJ, Stubbs RS. Improvement in co-

morbidities following weight loss from gastric bypass

surgery.Obes Surg2000;10:428–435.

ICHIH 2022 - International Conference on Health Big Data and Intelligent Healthcare

198

Dixon, A. E. et al. Effects of obesity and bariatric surgery

on airway hyperresponsiveness, asthma control, and

inflammation. J. Allergy Clin. Immunol. 128, 508–

515.e2 (2011).

Fuseini, Hubaida, and Dawn C Newcomb. “Mechanisms

Driving Gender Differences in Asthma.” Current

allergy and asthma reports vol. 17,3 (2017): 19.

doi:10.1007/s11882-017-0686-1

Fleming L. Asthma exacerbation prediction: recent

insights. Curr Opin Allergy Clin Immunol.

2018;18(2):117-123.

doi:10.1097/ACI.0000000000000428

Forno E. Moving Beyond the Confines of Body Mass Index

in the Quest to Understand Obese Asthma. Am J Respir

Crit Care Med. 2020;201(3):271-272.

doi:10.1164/rccm.201910-2031ED

Forno E. Moving Beyond the Confines of Body Mass Index

in the Quest to Understand Obese Asthma. Am J Respir

Crit Care Med. 2020;201(3):271-272.

doi:10.1164/rccm.201910-2031ED

Herman Avner Cohen, Hannah Blau, Moshe Hoshen, Erez

Batat, Ran D. Balicer; Seasonality of Asthma: A

Retrospective Population Study. Pediatrics April 2014;

133 (4): e923–e932. 10.1542/peds.2013-2022

Julia, Valerie et al. “The impact of diet on asthma and

allergic diseases.” Nature reviews. Immunology vol.

15,5 (2015): 308-22. doi:10.1038/nri3830

Lloyd A, Price D, Brown R. The impact of asthma

exacerbations on health-related quality of life in

moderate to severe asthma patients in the UK. Prim

Care Respir J. 2007;16(1):22-27.

doi:10.3132/pcrj.2007.00002

Mukherjee M, Stoddart A, Gupta RP, et al. The

epidemiology, healthcare, and societal burden and costs

of asthma in the UK and its member nations: analyses

of standalone and linked national databases. BMC Med.

2016;14(1):113. Published 2016 Aug 29.

doi:10.1186/s12916-016-0657-8

McFadden, E. R., and Ileen A. Gilbert. “Exercise-Induced

Asthma.” New England Journal of Medicine, vol. 330,

no. 19, 1994, pp. 1362–67. Crossref,

doi:10.1056/nejm199405123301907

Miliku K, Azad MB. Breastfeeding and the Developmental

Origins of Asthma: Current Evidence, Possible

Mechanisms, and Future Research Priorities. Nutrients.

2018;10(8):995. Published 2018 Jul 30.

doi:10.3390/nu10080995

Okoniewski W, Lu KD, Forno E. Weight Loss for Children

and Adults with Obesity and Asthma. A Systematic

Review of Randomized Controlled Trials. Ann Am

Thorac Soc. 2019;16(5):613-

625.doi:10.1513/AnnalsATS.201810-651SR from

gastric bypass surgery.Obes Surg2000;10:428–435.

Oddy WH. Breastfeeding, Childhood Asthma, and Allergic

Disease. Ann Nutr Metab. 2017;70 Suppl 2:26-36.

doi:10.1159/000457920

Peters U, Dixon AE, Forno E. Obesity and asthma. J

Allergy Clin Immunol. 2018;141(4):1169-1179.

doi:10.1016/j.jaci.2018.02.004

Sears MR, Johnston NW. Understanding the September

asthma epidemic. J Allergy Clin Immunol.

2007;120(3):526-529. doi:10.1016/j.jaci.2007.05.047

Ulrik CS, Lophaven SN, Andersen ZJ, Sørensen TI, Baker

JL. BMI at school age and incident asthma admissions

in early adulthood: a prospective study of 310,211

children. Clin Epidemiol. 2018;10:605-612. Published

2018 May 25. doi:10.2147/CLEP.S156310

Wood, L. G., Garg, M. L. & Gibson, P . G. A high-fat

challenge increases airway inflammation and impairs

bronchodilator recovery in asthma. J. Allergy Clin.

Immunol. 127, 1 133–1 140 (2011).

Xu S, Gilliland FD, Conti DV. Elucidation of causal

direction between asthma and obesity: a bi-directional

Mendelian randomization study Int J

Epidemiol2019;48:899–907.

Yan J, Liu L, Zhu Y, Huang G, Wang PP. The association

between breastfeeding and childhood obesity: a meta-

analysis. BMC Public Health. 2014;14:1267. Published

2014 Dec 13. doi:10.1186/1471-2458-14-1267

Juvenile Asthma in the U.S. Relate Asthma Incidence to Body Mass Index

199