Objective. To test whether obesity is associated with decreased peak expiratory flow rates (PEFR), increased asthma symptoms, and increased health service use.
Design/Methods. Secondary analysis of data from a cross-sectional convenience sample.
Setting. Emergency departments (EDs) and primary care clinics in 8 inner-city areas in 7 cities.
Participants. One thousand three hundred twenty-two children aged 4 to 9 years with asthma.
Measures. Obesity was defined as a body mass index (BMI, weight/height2) >95th percentile. Nonobese children were those with a BMI between the 5th and 95th percentile. Underweight children with a BMI <5th percentile were eliminated from the study. Demographic and anthropometric data were obtained during a baseline interview with the primary caretaker and the child. Symptoms, health service use data and measurements of PEFR were obtained by parental report during the baseline interview and at 3-month intervals by telephone interview over the following 9-month period.
Results. Obese (n = 249) and nonobese (n = 1073) children did not differ in terms of age, gender, family income, passive smoke exposure, caretaker's mental health, and skin test reactivity to indoor allergens. Obese children were more often Latino (28% vs 17%) and, in the 3 months before the baseline interview, were more likely to have used oral steroids (30% vs 24%). There were no differences between groups in terms of baseline PEFR scores. During the 9 months after baseline assessment, the obese group had a higher mean number of days of wheeze per 2-week period (4.0 vs 3.4), and a greater proportion of obese individuals had unscheduled ED visits (39% vs 31%). There were no differences between the groups in terms of frequency of hospitalization, or in nocturnal awakening.
Conclusions. In our sample of inner-city children with asthma, obese children used more medicine, wheezed more, and a greater proportion had unscheduled ED visits than the nonobese children.
Over the last 2 decades, the prevalence of both childhood asthma and childhood obesity have increased markedly in the United States.1–3 However, whether childhood obesity is associated with a greater degree of asthma morbidity has not been extensively studied. Luder et al studied a group of African-American and Latino children referred to a pulmonary clinic for asthma and found that obesity was associated with a greater medication use, a higher relative risk of low peak expiratory flow rates (PEFR) scores, and more missed school days.4 In addition, adverse effects of obesity on the pulmonary function of healthy children without asthma have been noted.5–7 A study of 13 morbidly obese children found that both the forced vital capacity (FVC) and the forced expiratory volume in 1 second (FEV1) were substantially below predicted values.8 A significant proportion of the morbidly obese children examined in that study experienced improvement in pulmonary function test results after receiving bronchodilators. Exercise-induced bronchospasm has also been found in obese but otherwise healthy children.6 These studies suggest that obesity may be associated with increased asthma morbidity.
Using the cohort of 1528 children recruited in the National Cooperative Inner-City Asthma Study (NCICAS),8 we examined whether obesity was associated with lower PEFR scores, greater health service use for asthma, and more asthma symptoms.
The study group consisted of a convenience sample of 1528 children aged 4 to 9 years with asthma recruited from emergency departments (EDs) and primary care clinics in 8 inner-city areas in the United States (Bronx, NY; East Harlem, NY; St Louis, MO; Washington, DC; Baltimore, MD; Chicago, IL; Cleveland, OH; and Detroit, MI). Participants were recruited during ED visits for asthma or other acute illnesses or injuries, or from primary care clinics during visits for routine care or asthma follow-up care. Recruitment for the study began in November 1992 and was completed by October 1993. Participants had to live in census tracts served by the study hospital in which 20% to 40% or more of the households were below the 1990 federal guidelines for poverty.
Definition of Asthma
At recruitment, each study child was required to meet at least 1 of the following definitions of asthma: 1) having been told by a physician that the child has asthma, in combination with cough, wheezing, shortness of breath, or whistling or tightness in the chest lasting for >3 days within the past 12 months or 2) cough, wheezing, or shortness of breath that lasted >6 weeks during the previous 12 months and 3 out of the 5 following conditions: a) cough, wheezing, or shortness of breath present more than half the days and nights during the 6-week period, b) cough, wheezing, or shortness of breath aggravated by exercise or cold air, c) a parent or sibling with asthma, d) no history of antibiotic therapy for sinusitis, accompanying the cough, or e) cough, wheezing, or shortness of breath that resulted in disturbance of the child's sleep.
After the child removed shoes and heavy clothing, the height and weight were measured at the study center during the baseline interview. Weight and height measurements were made using a balance beam scale and sliding L-shaped arm in accordance with standard clinical techniques.
Definition of Obesity
Based on reference data collected as part of the second cycle of the National Health and Nutrition Examination Survey (NHANES II) in 1976–1980, obesity was defined as a body mass index (BMI) >95th percentile.9 The reference data chosen to represent the 95th percentile of BMI was collected in 1976–1980, before a significant increase in childhood obesity in the United States. Recently, percentiles standards for BMI in United States children were published using data from more than 66 000 children.10These values are in close agreement with the BMI values used in this analysis. The BMI is the weight in kilograms divided by the square of the height in meters. There is a strong positive linear correlation between BMI and the percentage of body weight that is fat.11,12 There is growing international consensus that a BMI >95th percentile is a valid and clinically useful definition of obesity.13
One thousand five hundred twenty-eight children were enrolled in the NCICAS. We screened out errors in anthropometric measurements, or data entry, by examining outliers using the following criteria: 1) weight for age z score less than −6 or >6; 2) height for age z score less than −6 or >6; and 3) weight for heightz score less than −4 or >6. Two of the authors (P.F.B. and E.L.) independently reviewed the anthropometric data of the 58 children who met the criteria for outliers and using standard National Center for Health Statistics (NCHS) growth charts agreed to exclude 15 of these children whose measurements seemed biologically implausible. After these exclusions, full anthropometric data, including weight for height z scores, were available for 1380 children. Fifty-eight children with a BMI <5th percentile were excluded from analysis so that cachectic children would not be included in the analysis of nonobese children.
After enrollment, a structured interview was conducted with the child's primary caretaker concerning the demographic characteristics of the household, including self-reported race/ethnicity, the child's access to medical care, adherence to medical therapy, and history of medication use in the 3 months before the baseline interview. Information on the home environment and exposure to tobacco smoke were also collected. Skin-testing to indoor allergens (cat, dog, rat, mouse, roach, mite, Alternaria, and Penicillium) was performed on the child, and urine was collected to assay for cotinine, a metabolite of nicotine.
The child's psychological health was measured using a modified version of the Child Behavior Checklist (CBCL). The CBCL is a 113-item questionnaire that generates a score for behavioral problems and symptoms. A modification of the CBCL was developed for the NCICAS that eliminates 13 items which could be confounded by asthma symptoms (eg, difficulty sleeping).14 Raw scores on the modified test were subsequently converted to T-scores by comparison with a normative population. A score of 64 or greater was considered indicative of substantial psychological problems in the child.
The mental health of the child's primary caretaker was evaluated using the Brief Symptom Inventory, a standardized 53-item questionnaire that generates scores for 3 global dimensions.14 For the current study, summary scores were converted to T-scores by comparison with a normative population that was matched to the population recruited into the study in terms of ethnic composition and socioeconomic status. A score above 63 was considered to indicate psychological problems in the caretaker.
At intervals of 3, 6, and 9 months from the baseline visit, measurements of health service use and asthma symptoms were obtained from the primary caretaker by a trained interviewer using a standard questionnaire. These assessments were completed by telephone in 93% of cases and by in-person interview in 7% of cases. Follow-up interviews at the 3, 6, and 9-month time points were completed for 90%, 92%, and 94% of the sample, respectively.
Measures of Health Service Use and Asthma Symptoms
The measurements of health service use examined for this study included the following: 1) hospitalizations for asthma, 2) unscheduled doctor or clinic visits for asthma (including ED visits), and 3) ED visits for asthma. Hospitalizations, unscheduled visits, and ED visits were dichotomized into the following categories: 1) none, or 2) any, because the vast majority of participants reported either none or 1 event. Results are expressed as the proportion of individuals reporting one or more events during the 9-month period after the baseline assessment.
Questions assessing asthma symptoms included the number of days the child wheezed, the number of nights the child was awakened by asthma, and the number of days the child's play was slowed because of asthma symptoms in the previous 2 weeks. Results of the symptom frequency reports are expressed as a mean per 2-week period.
Medication usage was not assessed during the follow-up period.
Peak Flow Measurements
Peak flow measurements were obtained using a mini-Wright peak flow meter (Clement Clarke, Columbus, OH). The research staff taught the child how to use the meter and observed the child using it correctly. The recorded measurement was the highest of 3 maximal expiratory maneuvers obtained during the baseline visit. For children recruited during an ED visit, measurements of baseline peak flow measurements were deferred for 4 weeks. Raw peak flow measurements were converted to percentages of predicted peak expiratory flow rates using gender- and race-specific nomograms.15 Mean PEFR at baseline were calculated for the obese and nonobese groups. In addition, PEFR measurements were plotted against weight for heightz score to seek a trend of decreasing PEFR with increasing obesity that would not be apparent in a dichotomous comparison of obese and nonobese individuals. Results of this analysis are presented in Fig 1.
The EPINUT module of EPI Info (EPI Info Version 6.04, Centers for Disease Control and Prevention, Atlanta, GA) program was used to calculate z scores for weight for age, height for age, and weight for height. Continuous morbidity measures such as the number of days of wheezing, the number of nights child was awakened, and the number of days the child's play was slowed were square root transformed to make the data comply better with the normality requirement. The t tests and χ2tests were used to compare the obese and nonobese children for continuous and dichotomous variables respectively, and multiple regression and logistic regression were used to compare the obese and nonobese groups in terms of asthma morbidity after controlling for race/ethnicity. Baseline differences in medication use were believed to be related to the other outcomes that were measured (asthma symptoms and health care use), and therefore were not controlled.
The NCICAS was approved by the institutional review boards of each of the participating medical centers. Written informed consent was obtained from each caretaker as well as assent from the child according to local institutional review board guidelines.
Of the 1322 children enrolled in the NCICAS with BMI values above the 5th percentile, 249 children (19%) had BMI values >95th percentile and were labeled obese, whereas 1073 (81%) had BMI values between the 5th and 95th percentiles and were labeled nonobese. The sample was composed of 967 non-Hispanic black and 250 Hispanic children. The remaining 91 children, designated other, were of mixed racial/ethnic origin, white, or Asian.
The demographic and baseline characteristics by obesity are noted inTable 1. The groups were not different in terms of age, gender composition, income, education, smoke exposure or in the number of positive skin tests to indoor allergens. A relatively greater proportion of the obese children were Latino (28% vs 17%;P < .01) and in the 3 months before the baseline interview the obese children were more likely to have used oral steroids (30% vs 24%; P < .05) or a combination of 2 or more asthma medications (63% vs 55%; P < .05)
Figure 1 presents a scatterplot of PEFR values versus weight for heightz score for the 749 individuals with baseline PEFR measurements among our sample. The trend line for PEFR (a cubic spline best fit) does not show appreciable decrease in participants with extreme obesity.
Table 2 presents the results for mean percent-predicted PEFR, health service use, and asthma symptoms by obesity. A greater proportion of the obese children had ED visits (39% vs 31%; P < .04) and this difference was significant after adjusting for baseline differences in the racial/ethnic composition of the groups. The obese children had a higher mean number of days of wheezing per 2-week period (4.0 vs 3.4; P < .02). After adjusting for differences in the racial/ethnic composition of the groups, this finding had a P value of .05. Over the course of 1 year, this represents 16 more days of wheezing among the obese group. There were no differences between the obese and nonobese groups in terms of PEFR, hospitalizations, unscheduled doctor or clinic visits, nocturnal awakening, or slowed play.
In this sample of inner-city children with asthma, obese children used more asthma medication, had more reported days of wheezing, and were more likely to visit an ED. These differences were of a modest degree. The obese group had the equivalent of 2 additional weeks (worth of days) of reported wheezing annually than the nonobese group. Over the course of 9 months, the obese group had an 8% increase in ED use, and over the course of the 3 months during which medication use was assessed this group was 6% more likely to be prescribed oral steroids. Although the burden of additional morbidity associated with obesity in this study is mild, the findings are consistent in so far as it would be expected that increased symptoms (wheezing) would prompt more health care use (ED visits) and more medication use. The fact that the obese group were reported to have been prescribed oral steroids more often lends some validity to caregiver reports of greater symptoms. Our finding of greater medication use among obese asthmatics is consistent with the previously published study of Luder et al4 which showed that obese asthmatic children were twice as likely to be prescribed ≥3 concurrent medications than the nonobese group. Important negative findings in our study were that obese and nonobese children did not differ in terms of baseline PEFR, or in the other measures of asthma symptomatology and health care use that we measured. Why the obese children in our study wheeze more and have more ED visits, yet do not have more nocturnal awakening, limitation of play, or more hospitalization is not clear. In an analysis of the NHANES II data, composed of 5672 children <11 years old, Schwartz et al16 found that parental reports of wheezing but not physician-diagnosed asthma were associated with obesity.
The groups of obese and nonobese children in our study were not different in a large number of commonly measured social and demographic characteristics. These include age, gender composition, family income, maternal education, and measures of caregiver and child mental health. Of particular note is the fact that exposure to household smoke and dermal reactivity to a panel of common indoor allergens did not differ between groups. This last finding is significant insofar as it has been hypothesized that obese individuals may have a relatively greater exposure to indoor allergens associated with asthma, and this hypothesis has not been previously tested.
In this study, the obese and nonobese group differed in racial/ethnic composition. The obese group was composed of 28% Hispanic (predominantly Puerto Rican) children versus 17% in the nonobese group. We do not believe that the observed increased morbidity among the obese group was attributable to the difference in racial/ethnic composition because the association of obesity to an increase in reported wheezing and unscheduled ED visits remained consistent after statistical adjustment for race/ethnicity.
Although we hypothesized that PEFR would be decreased in the obese group, we did not find measurable decreases of PEFR despite greater symptoms and health care use. There are several potential explanations for this finding. First, the baseline PEFR values used in our study were obtained at least 4 weeks from an acute exacerbation of asthma symptoms, and therefore, would not reflect the greater frequency of asthma symptoms we observed in the obese group. More frequent measurement of PEFR may be needed to detect the episodic decreases in PEFR among asthmatics with bronchospasm. Second, PEFR may be an inadequate test. There is a growing body of literature that documents that normal values of PEFR can be found in children who have significant airflow obstruction as demonstrated by markedly decreased levels of FEV1, and forced expiratory flow between 25% and 75% of vital capacity.17
We questioned whether a threshold effect of obesity on PEFR might occur at a greater level of obesity than >95th percentile of BMI (which corresponds to a weight for height z score of >1.65) and we constructed a continuous graphic plot (examination) of PEFR versus weight for height z score. The results of this analysis suggest that in obese asthmatic children free from acute symptoms, impairment of airflow (detectable by PEFR tests) does not occur even in relatively severe obesity. Ray et al18 showed that in severely obese but otherwise healthy young adults, vital capacity was not significantly reduced until the weight/height ratio in kg/cm was >1.0 This corresponds to a weight of >335 pounds for a 5-foot tall individual. A study of 13 very obese children who weighed a mean of 212% of the average body weight for their height also showed significant reductions in FEV1 and FVC.7 Other studies of obese children that have used less severe definitions of obesity have often found inconsistent relationships between obesity and pulmonary function in otherwise normal individuals. These inconsistent results likely reflect varying degrees of severe obesity in the study groups, as well as imprecision in the definitions of obesity, which rely on height and weight, rather than an assessment of total body fat. Lazarus et al19 have clarified this issue by showing that height-adjusted FEV1 and FVC increase as a function of body weight in a population-based sample of children, but that both FEV1 and FVC decrease as a function of increasing percentage of body fat at any given height and weight. Taken together, these existing studies of obesity and pulmonary function show that increasing lean body mass has a strong positive effect on pulmonary function tests whereas increasing fat mass has a negative effect. This effect of body fat is small when obesity is mild to moderate but becomes measurable when obesity is in the morbid range. Although our study sample included PEFR for 37 children who were >150% of the median weight for height (weight for height z score: 4.5 to 5.6) it included only 1 PEFR value for a child who was 200% of the median weight for height (z score: 6.9) and therefore, we were unable to test whether a threshold for decreased PEFR occurred at this level of obesity.
Luder et al4 found a significant reduction of PEFR among the obese asthmatic participants in their study using a more liberal definition of obesity (BMI >85th percentile) than we did. The discrepancy between our 2 studies likely reflects the fact that patients in our study were required to wait 4 weeks after an exacerbation of asthma symptoms whereas in the study of Luder et al, peak flow values were obtained both during well and sick visits.
Our study design has several important limitations. Our study is limited by the fact that asthma symptoms and health care use measures in the NCICAS are based on self-reported data. Although self-reported data forms a large part of several important pediatric data sets such as the NHANES II, we cannot rule out the possibility that a caregiver bias to overreport symptoms and a physician bias to overprescribe asthma medication to obese children has influenced our data. We are unaware of any existing literature that documents such a bias.
Although we found an association of obesity with greater asthma morbidity, the nature of our study design can not establish a causal relationship, nor does it suggest a directionality to the causation if, in fact, one exists.
Some authors have posited that asthma may be a risk factor for obesity.6,20 Indirect evidence for this hypothesis exists in a case-control study of asthmatic and nonasthmatic children done in an urban health center by Gennuso et al20 that found a greater prevalence of obesity in the asthmatic participants. However, in this study a dose-effect relationship between asthma severity and degree of obesity was not found. Several longitudinal studies have examined the growth of asthmatic children. In a large cohort of Israeli children examined at age 17, boys with a history of mild asthma in childhood had higher BMI values than controls. However, boys with moderate or severe asthma and girls with all 3 classes of asthma had BMI values that were no different from controls.21 A recently published prospective 4-year study of 3347 Scottish children with asthma showed no increase in weight for age z score over the course of the study irrespective of asthma severity.22 Available literature indicating that asthma can promote the development of obesity is evidently conflicting.
Recently, evidence that obesity may increase one's risk to acquire asthma has been presented by Camargo et al.23 Using prospectively obtained data from the Nurses Health Study II they found that women who gained weight after the age of 18 were at an increased risk to develop asthma; furthermore, the relative risk of developing asthma increased with increasing levels of BMI. A pathophysiological explanation for this phenomenon has not been established. Further evidence that obesity directly affects asthma morbidity has been recently demonstrated in morbidly obese adult asthmatics who showed marked improvement in asthma morbidity after losing weight following laparoscopic gastric banding.24 However, as stated previously, morbidly obese individuals are at risk for significant decreases in pulmonary function, and whether these results would be replicated among asthmatics with lesser degrees of obesity is unknown.
In summary, we found that obese asthmatic children wheezed more, had more unscheduled ED visits, and received more medications. These differences were modest and were not associated with a decrease in baseline PEFR or an increase in the rate of hospitalization. Prospective studies of the effect of weight loss on asthma morbidity may be required to provide a better understanding of the relationship between obesity and asthma morbidity, impairment of pulmonary function, and health service use.
The contract grant sponsor of this work was the National Institute of Allergy and Infectious Disease (National Institute of Health, Bethesda, MD) Grants U01 A1-30751, A1-30752, A1-30756, A1-30772, A1-30773-01, A1-30777, A1-30779, A1-30780, and N01 A1-15105.
We wish to thank Andrew Racine, MD, PhD, Alex Okun, MD, and Ruth Stein, MD, for their review of the manuscript.
- Received April 17, 2000.
- Accepted May 31, 2000.
Reprint requests to (P.F.B.) Pediatric Academic Associates, 1621 Eastchester Rd, Bronx, NY 10461.
↵FNa Drs Mitchell, Islam, and Lynn were formerly with New England Research Institutes, Watertown, Massachusetts.
Presented in part at the Annual Meeting of the Ambulatory Pediatric Association on May 2, 1997.
- PEFR =
- peak expiratory flow rates •
- FVC =
- forced vital capacity •
- FEV1 =
- forced expiratory volume in 1 second •
- NCICAS =
- National Cooperative Inner-City Asthma Study •
- ED =
- emergency department •
- NHANES II =
- National Health and Nutrition Examination Survey •
- BMI =
- body mass index •
- NCHS =
- National Center for Health Statistics •
- CBCL =
- Child Behavior Checklist
- Asthma—United States, 1980–1990. MMWR Morb Mortal Wkly Rep. 1992:41:733–735
- Troiano RP, Flegal KM, Kuczmarski RJ, Campbell SM, Johnson CL. Overweight prevalence and trends for children and adolescents: The National Health and Nutrition Examination Surveys, 1963 to 1991. Arch Pediatr Adolesc Med. 1995:149:1085–1091
- Luder E, Melnik TA, DiMaio M. Association of being overweight with greater asthma: symptoms in inner city black and Hispanic children. J Pediatr. 1998:132:699–703
- Fung KP, Lau SP, Chow OKW, Lee J, Wong TW. Effects of overweight on lung function. Arch Dis Child. 1990:65:512–515
- Kaplan TA, Montana E. Exercise-induced bronchospasm in non-asthmatic obese children. Clin Pediatr. 1993:32:220–225
- Inselman LS, Milanese A, Deurloo A. Effect of obesity on pulmonary function in children. Pediatr Pulmonol. 1993:16:130–137
- Mitchell H, Senturia Y, Gergen P, et al. Design and methods of the National Cooperative Inner-City Asthma Study. Pediatr Pulmonol. 1997:24:237–252
- Najjar MF, Rowland M. Anthropometric reference data and prevalence of overweight, United States, 1976–1980. Vital Health Stat. 1987;11(No. 238)
- Rosner B, Prineas R, Loggie J, Daniels SR. Percentiles for body mass index in US children 5 to 17 years of age. J Pediatr. 1998:132:211–222
- Roche AF, Siervogel RM, Chumlea C, Webb P. Grading body fatness from limited anthropometric data. Am J Clin Nutr. 1981:34:2831–2838
- Pietrobelli A, Faith MS, Allison DB, Gallagher D, Chiumello G, Heymsfield SB. Body mass index as a measure of adiposity among children and adolescents: a validation study. J Pediatr. 1998:132:204–210
- Dietz WH, Robinson TN. The use of the body mass index (BMI) as a measure of overweight in children and adolescents. J Pediatr. 1998:132:191–193
- Wade S, Weil C, Holden G, et al. Psychosocial characteristics of inner-city children with asthma. A description of the NCICAS Psychosocial protocol. Pediatr Pulmonol. 1997:24:263–276
- Hsu KHK, Jenkins DE, Hsi BP, et al. Ventilatory functions of normal children and young adults. Mexican American, White and Black II. Wright peak flow meters. J Pediatr. 1979:95:192–196
- Schwartz J, Gold D, Dockery DW, Weiss ST, Speizer FE. Predictors of asthma and persistent wheeze in a national sample of children in the United States. Am Rev Respir Dis. 1990:142:555–562
- Eid N, Yandell B, Howell L, Eddy M, Sheikh S. Can peak expiratory flow predict airflow obstruction in children with asthma? Pediatrics. 2000:105:354–358
- Ray CS, Sue DY, Bray G, Hansen JE, Wasserman K. Effects of obesity on respiratory function. Am Rev Respir Dis. 1983:128:501–506
- Lazarus R, Colditz G, Berkey CS, Speizer FE. Effect of body fat on ventilatory function in children and adolescents: cross-sectional findings from a random population sample of school children. Pediatr Pulmonol. 1997:24:187–194
- Gennuso J, Epstein LH, Paluch RA, Cerny F. The relationship between asthma and obesity in urban minority children and adolescents. Arch Pediatr Adolesc Med. 1998:152:1197–1200
- Shohat M, Shohat, T, Kedem R, Mimouni M, Danon YL. Childhood asthma and growth outcome. Arch Dis Child. 1987:62:63–65
- McCowan C,
- Neville RG,
- Thomas GE,
- et al.
- Camargo CA Jr, Weiss ST, Zhana S, Willet W, Speizer FE. Prospective study of body mass index, weight change and risk of adult-onset asthma in women. Arch Intern Med. 1999:159:2582–2588
- Dixon JB, Chapman L, O'Brien P. Marked improvement in asthma after lap-band surgery for morbid obesity. Obesity Surg. 1999:9:385–389
- Copyright © 2000 American Academy of Pediatrics