Serum Vitamin Levels and the Risk of Asthma in Children

Transcription

1 American Journal of Epidemiology Copyright 2004 by the Johns Hopkins Bloomberg School of Public Health All rights reserved Vol. 159, No. 4 Printed in U.S.A. DOI: /aje/kwh053 Serum Vitamin Levels and the Risk of Asthma in Children Raida I. Harik-Khan 1, Denis C. Muller 1, and Robert A. Wise 2 1 Clinical Research Branch, National Institute on Aging, National Institutes of Health, Baltimore, MD. 2 Pulmonary and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD. Received for publication April 28, 2003; accepted for publication August 27, Dietary intake, especially of antioxidant vitamins A, C, E, and the carotenoids, has been linked with the presence and severity of asthma. From the Third National Health and Nutrition Examination Survey (NHANES III), conducted in the United States between 1988 and 1994, the authors selected 4,093 children (aged 6 17 years) for whom relevant medical, socioeconomic, and anthropometric data were complete. The children were 50.6% female, and 9.7% reported a diagnosis of asthma. Bivariate analyses showed that asthma diagnosis was associated with lower levels of serum vitamin C, α-carotene, β-carotene, and β-cryptoxanthin. However, antioxidant levels may be surrogate markers for socioeconomic variables such as race, poverty, tobacco exposure, or general nutritional status. In logistic models that included age, body mass index, socioeconomic variables, antioxidant levels, parental asthma, and household smoking, the only antioxidants significantly associated with asthma were vitamin C (odds ratio = 0.72 per mg/dl, 95% confidence interval = 0.55, 0.95) and α-carotene (odds ratio = 0.95 per µg/dl, 95% confidence interval = 0.90, 0.99). The odds ratio for asthma in the highest quintile of serum vitamin C relative to the lowest was 0.65 (p < 0.05), whereas it was 0.74 for α-carotene (p = 0.066). The authors concluded that low vitamin C and α-carotene intakes are associated with asthma risk in children. adolescent; antioxidants; ascorbic acid; asthma; child Abbreviation: NHANES III, Third National Health and Nutrition Examination Survey. In recent years, there has been increasing interest in the role of nutrition in the development of asthma. Because of the increasing prevalence of asthma, particularly in the inner city and developing nations, it has been postulated that changes in diet and nutrition may play a role in promoting its onset. In particular, interest has focused on the intake of antioxidant vitamins, especially vitamins A, C, and E and the carotenoids, in the modulation of oxidative lung or airway injury from inflammatory cells. Vitamin A derivatives of retinol influence the development, maintenance, differentiation, and regeneration of lung epithelial cells and may play a central role in the development of airway diseases (1, 2). To our knowledge, there are no long-term clinical trials on the effect of diet on the development of asthma, and most previous studies linking asthma with vitamin intake have relied on inferences from epidemiologic data. These data consistently showed an association between the presence of asthma and the intake of vitamin C (3). However, many of the previous studies have used food frequency questionnaires to assess vitamin intake, which may not accurately reflect biologic availability of vitamins or take multivitamin ingestion into account (4). Furthermore, vitamin intake may be a marker or epiphenomenon of other risk factors for asthma such as low socioeconomic status, low birth weight, obesity, urban environment, tobacco smoke exposure, regional or cultural dietary habits in minority subgroups, or dietary modification in families with other members who have asthma or allergy. The Third National Health and Nutrition Examination Survey (NHANES III) provides an ideal data set for examining the relation between asthma and vitamin intake. The study population represents a national population sample, and the data set includes information on general health, socioeconomic status, and serum levels of some vitamins. The question that we wanted to answer in this analysis of the NHANES III data set was whether serum vitamin Correspondence to Dr. Robert A. Wise, Pulmonary and Critical Care Medicine, Johns Hopkins University School of Medicine, 5501 Hopkins Bayview Circle, Baltimore, MD ( rwise@jhmi.edu). 351

2 352 Harik-Khan et al. concentrations, specifically vitamins A, C, E, and the carotenoids, were associated with a clinical diagnosis of asthma when other factors such as body weight, socioeconomic status, passive smoke exposure, urban environment, and ethnicity were taken into account. MATERIALS AND METHODS Study population NHANES III examined a random national population sample of the United States with oversampling of minorities and children (5). Participants undertook health questionnaires completed by adult caretakers where appropriate, and a subsample underwent physical examination, were tested for pulmonary function, and had blood samples taken to measure serum vitamins. For the present analysis, we used linked information from the questionnaire, physical examination, and laboratory data for participants aged 6 17 years. The detailed methods for the questionnaires, examination, and laboratory measurements are described elsewhere (5, 6). Medical, environmental, and sociodemographic information was provided by an adult caretaker. The presence of physician-diagnosed asthma ( ever asthma ) was ascertained by a positive response to the question, Has a doctor ever said that your child has asthma? Body mass index was calculated as weight in kilograms divided by height in meters squared. Serum vitamin concentrations were measured by isocratic high-performance liquid chromatography at the laboratories of the Centers for Disease Control and Prevention (6). Ascorbic acid intake was obtained from the dietary recall interview covering the last 24 hours. Information on supplemental dietary intake of vitamins and minerals in the last month was also reviewed. Statistical analyses Data analysis was conducted by using SAS software (SAS Institute, Inc., Cary, North Carolina). In this paper, data are presented as mean (standard deviation) unless otherwise specified. Student s t test was used to check for the significance of the difference of the variables listed in table 1 between asthmatics and nonasthmatics. Categorical variables in table 1 were compared by using chi-square tests. Logistic regression analysis was used to examine the relation between various explanatory variables and the probability of having an asthma diagnosis. The full model included age, gender, household size, body mass index, household head educational status, household head gender, household head employment status, diagnosis of asthma or hay fever in at least one parent, race, the presence of at least one smoker in the household, and serum levels of the antioxidant vitamins A, C, and E and the carotenoids: α-carotene, β-carotene, β- cryptoxanthin, lutein/zeaxanthin, and lycopene. The final model was obtained by backward selection (p = 0.05 significance level for inclusion) with the stipulation that age and all racial/ethnic categories remain in the final model. Inclusion of region did not alter the inferences; hence, it was removed from the model. Computation of variance inflation factors showed that multicollinearity among the explanatory variables was not significant. The odds ratios and their upper and lower 95 percent confidence limits were computed for each explanatory risk factor. A two-tailed p value of <0.05 was required for statistical significance. RESULTS Overall, for 5,433 participants between the ages of 6 and 17 years, the question, Has a doctor ever said that your child has asthma? was answered. For these children, technically reportable serum vitamin measurements were available for 4,427 (81 percent). For these 4,427 participants, one or more items of the physical examination or sociodemographic information were missing for 334, leaving a final study population of 4,093 children for whom measurements were complete. Of these children, 2,072 (50.6 percent) were girls, and 397 (9.7 percent) had a physician diagnosis of asthma. Of the latter group, 278 children (70 percent) had current asthma. Table 1 shows that for 46 percent of the children with asthma, their first asthma diagnosis was made by age 3 years, while 22 percent and 26 percent were diagnosed with asthma between the ages of 3 and 6 and of 6 and 12 years, respectively. Only 5 percent were diagnosed between the ages of 12 and 17 years. Bivariate analyses comparing children with asthma with children without asthma (table 1) confirmed previous findings that environmental and socioeconomic factors were associated with asthma. These asthma risk factors include male gender, family history of asthma or hay fever, smaller household size, unemployed head of household, lower educational status of head of household, female head of household, and urban environment. Household income as evidenced by the poverty index (defined as the ratio of family income in the last 12 months to the federal poverty line; thus, a higher poverty index indicates a higher socioeconomic status) and the presence of household smokers was not significantly associated with higher asthma risk. Among the asthmatics, there were significantly fewer Mexican Americans and more children classified as being in the other ethnic group (which included Asian Americans and other categories of Hispanics). Furthermore, these data confirmed previous findings (7) that body mass index was higher in asthmatics than nonasthmatics (p < ). Serum concentrations of vitamins A and E were not different between asthmatics and nonasthmatics. However, significant differences were present for serum concentrations of vitamin C, α-carotene, β-carotene, and β-cryptoxanthin. No differences were found between asthmatics and nonasthmatics with respect to lutein/zeaxanthin, lycopene, cholesterol, and triglyceride serum concentrations. We also did not find any difference in ascorbic acid dietary intake, calculated from the recall interview of the last 24 hours, or in the intake of supplemental minerals and vitamins between asthmatics and nonasthmatics. Univariate analyses showed that, in this population, moderate statistically significant correlations existed between some of the asthma risk factors and vitamin concentrations (table 2). The parsimonious logistic regression model enabled us to examine the predictive strength of serum vitamin concentrations for childhood asthma while adjusting statistically for

3 Serum Vitamin C Levels and Asthma in Children 353 TABLE 1. Subject characteristics,* Third National Health and Nutrition Examination Survey, United States, Characteristic Children with no asthma diagnosis (n = 3,696) Children with an asthma diagnosis (n = 397) p value Age (years) 11.3 (3.1) 11.5 (3.1) 0.2 FEV 1 /FVC 0.88 (0.06) 0.83 (0.08) Body mass index (kg/m 2 ) (4.63) (5.79) Serum vitamin A (mg/dl) (8.85) (10.94) 0.10 Serum vitamin C (mg/dl) (0.428) (0.40) Serum vitamin E (mg/dl) (171.2) (189.0) 0.7 Serum α-carotene (mg/dl) 3.42 (2.91) 2.97 (2.09) Serum β-carotene (mg/dl) (9.15) (8.99) Serum cryptoxanthin (mg/dl) (6.90) (5.45) Serum lutein/zeaxanthin (mg/dl) 19.5 (8.07) 18.9 (8.4) 0.13 Serum lycopene (mg/dl) 24.2 (10.30) 24.4 (10.7) 0.6 Poverty index 1.65 (1.25) 1.76 (1.30) 0.12 Gender (% female) Household size (no. of persons in the dwelling) 5.1 (1.9) 4.8 (1.7) Household head unemployment status (%) # Household head education > high school (%) # Household head gender (% female) # Parent with asthma or hay fever (%) # Presence of at least one smoker in the household (% ) # Metropolitan area population >1 million (%) # Subjects with current asthma (%) 71.7 Serum cholesterol (mg/dl) (29.3) (30.8) 0.4 Serum triglycerides (mg/dl) 94.9 (52.6) 97.0 (55.1) 0.5 Ascorbic acid intake (last 24-hour dietary recall data)** (41.8) (100.8) 0.6 Used vitamin and mineral supplements in the last month (%) Age (years) at onset of asthma (%) < < < Race (%) White # African American # Mexican American # Other # * Values are expressed as mean (standard deviation) unless otherwise noted. Test of significance was performed by using Student s t test. Because spirometry was conducted on subjects aged 8 years or older, forced expiratory volume in 1 second (liters) (FEV 1 )/forced vital capacity (liters) (FVC) measurements were available for only 2,925 children with no asthma and 312 children with asthma. Defined as the ratio of family income in the last 12 months to the federal poverty line; thus, a higher poverty index indicates a higher socioeconomic status. Sample sizes for poverty index for children without asthma and for those with asthma were 3,403 and 376, respectively. Refers to whether the family head worked in the past 2 weeks and, if not, whether he or she had a job or business. # Test of significance was performed by using the chi-square test. ** The sample size for ascorbic acid intake from the last 24-hour recall data was 3,951 children. Includes Hispanics who are not Mexican Americans as well as Asian Americans.

4 354 Harik-Khan et al. TABLE 2. Correlation* of key measures from a bivariate analysis comparing children with asthma with children without asthma, Third National Health and Nutrition Examination Survey, United States, Variable Vitamin A Vitamin C Vitamin E α-carotene β-carotene Body mass index Poverty index Vitamin A (µg/dl) Vitamin C (mg/dl) (0.0004) Vitamin E (µg/dl) (0.0001) (0.0001) α-carotene (µg/dl) (0.42) (0.0001) (0.0001) β-carotene (µg/dl) (0.0340) (0.0001) (0.0001) (0.0001) Body mass index (kg/ m 2 ) (0.0001) (0.0001) (0.0001) (0.0001) (0.0001) Poverty index (0.0001) (0.0001) (0.0001) (0.0001) (0.0002) (0.0822) Educational status of family head (0.0001) (0.0017) (0.0001) (0.0059) (0.0702) (0.0400) (0.0001) * n = 4,093 for all correlations except poverty index, where n = 3,779 because of incomplete data. Values are expressed as Pearson s r (p value). important socioeconomic factors. We found that low levels of vitamin C and α-carotene were predictive of asthma (table 3). None of the other antioxidant vitamins added significantly to the predictive value of the model once these two vitamins were included. In addition to increased body mass index, the socioeconomic and demographic characteristics that were significant predictors of childhood asthma included male gender, urban residence, female head of TABLE 3. Risk of a diagnosis of asthma for children participating in Third National Health and Nutrition Examination Survey, United States, Risk factor Odds ratio Full model 95% confidence interval Odds ratio Final model 95% confidence interval Age , , Body mass index (kg/m 2 ) 1.041* 1.016, * 1.021, Gender 0.594* 0.477, * 0.479, Metropolitan area population >1 million 1.382* 1.109, * 1.109, Household size , Gender (0 = male, 1 = female) 1.375* 1.073, * 1.179, Household head unemployment status , Household head education , Presence of at least one smoker in the household , Serum vitamin A (mg/dl) , Serum vitamin E (mg/dl) , Serum vitamin C (mg/dl) 0.735* 0.552, * 0.546, Serum α-carotene (µg/dl) , * 0.899, Serum β-carotene (mg/dl) , Serum cryptoxanthin (mg/dl) , Serum lutein/zeaxanthin (µg/dl) , Serum lycopene (µg/dl) , Parent with asthma or hay fever , * 2.484, Race Mexican American (reference) White , , African American , , Other , * 1.172, * p < The final model was obtained by backward selection (p = 0.05), with the stipulation that age and race categories remain in the final model. Includes Hispanics who are not Mexican Americans as well as Asian Americans.

5 Serum Vitamin C Levels and Asthma in Children 355 FIGURE 1. Age-adjusted odds ratios and 95% confidence intervals (vertical bars) for quintiles 2 5 relative to quintile 1 of serum vitamin C (A) or α-carotene (B) level for 4,093 children, aged 6 17 years, from the Third National Health and Nutrition Examination Survey population, United States, * p < household, and family history of asthma or hay fever. The only racial group that had a higher risk of asthma relative to the Mexican Americans was the other category. The results of table 3 enabled us to estimate the asthma risk associated with changes in serum vitamin C and α- carotene concentrations. Thus, an increase of 0.43 mg/dl, equivalent to one standard deviation, was associated with a decrease of 13 percent in the risk of having asthma. Similarly, an increase of one standard deviation in serum α- carotene (2.85 µg/dl) was associated with a 15 percent decreased risk of asthma. In separate analyses, we computed the effect of quintiles of serum vitamin C or α-carotene on the age-adjusted risk of asthma. The three highest quintiles of vitamin C were protective relative to the lowest quintile (figure 1, A). On the other hand, only the highest quintile of serum α-carotene offered borderline protection (p = 0.067) relative to the lowest quintile (figure 1, B). DISCUSSION The major finding of this study is that the risk of a childhood asthma diagnosis is increased by lower levels of serum vitamin C and the carotenoids, including α-carotene, β- carotene, and β-cryptoxanthin. We did not find any association of lower serum concentrations of vitamin A or E with the risk of having asthma. After statistical adjustment for potentially confounding socioeconomic factors, only low vitamin C and α-carotene serum levels continued to be predictive of asthma, whereas β-carotene dropped out of the model. Because of the relatively high correlation of the two variables (r = 0.61), we are unable to confidently ascribe the relation of asthma to one or the other form of carotenes. This study strengthens a growing body of data indicating that vitamin C plays a role in the development or expression of asthma (8 12). To our knowledge, this study is the first large childhood study that has relied on serum vitamin levels rather than food frequency questionnaires and has also controlled for several other potentially confounding characteristics. We also investigated the temporal relation between vitamin C and asthma by examining the effect of serum vitamin C on the presence of current asthma. Similar logistic regression analyses showed essentially similar results regarding vitamin C and several other risk factors for asthma diagnosis. The difference between the current asthma and the ever asthma results was that in the current asthma group, α-carotene was not a risk factor, whereas employment status and educational status of the household head were the most predictive socioeconomic characteristics. Vitamin C as ascorbic acid is a water-soluble antioxidant present in physiologic concentrations in the airway and alveolar lining fluid. As an antioxidant, vitamin C can modify oxidative insults from inhaled agents, infectious agents, or cellular inflammation (13, 14). In several studies, large acute doses of vitamin C have been found to have a beneficial effect on airways reactivity (15 18). Consistent with this conclusion is the finding by Romieu et al. (19) that supplementation with vitamins C and E in asthmatic children had a protective effect against high levels of ozone. Although the evidence of benefit of vitamin C supplementation as a treatment for chronic asthma was inconclusive in two recent literature reviews (20, 21), other reviews have concluded that vitamin C supplementation provides a short-term benefit in treating asthma (22, 23). We also analyzed the relation between asthma and antioxidant dietary intake and vitamin supplementation by using the available NHANES III intake of dietary ascorbic acid and of supplemental vitamins and minerals. We found no difference between children with and without asthma regarding these two measures. We do not know the reason for the discrepancy between the interview data and serum vitamin C level. One possibility is that the recall data are flawed and do not reflect true vitamin C intake. Alternatively, it is possible that in asthma patients, metabolism or excretion of vitamin C is different. For example, it may be that increases in the level of oxidative stress due to inflammation lead to lower levels of serum vitamin C (24).

6 356 Harik-Khan et al. Since these study findings are descriptive rather than experimental, we must be cautious about inferring causal relations. The consistency of the relation of low vitamin C levels across different studies with varied population samples, age groups, methods of ascertaining disease state, and nutritional intake strengthens the validity of the association. However, the strength of the association is not constant across all studies. A previous study of 51 African asthmatic children and matched controls found that asthmatics had lower plasma ascorbic acid levels and that there was a positive correlation between vitamin C concentrations and socioeconomic class among the children with asthma (9). This finding was attributed to higher intake of fruits by children in the higher socioeconomic groups. Another study of 14 asthmatic children from Turkey showed decreased serum levels of vitamin C as well as of β-carotene and vitamin E compared with 12 normal controls (12). Forastiere et al. (25) found, in a large Italian population sample of children aged 6 7 years, that a history of frequent winter consumption of citrus and kiwi fruits, rich in vitamin C, was associated with protection from respiratory symptoms over the following 12 months. Additionally, Cook et al. reported that, in a group of 2,650 children in England and Wales, fresh fruit intake was associated with better lung function. However, analysis of the results from a subset of 276 children did not confirm a relation between plasma vitamin C levels and forced expiratory volume in 1 second (FEV 1 ) (26). In the adult populations enrolled in the NHANES III surveys, low vitamin C intake or serum concentration has been associated with lower pulmonary function and more respiratory symptoms (10, 27, 28). However, our results are not consistent with those of Powell et al. (29), who reported no significant difference in the concentration of red blood cell vitamin C between a group of 37 asthmatic children and 35 controls. Our conclusions also differ from those of Cook et al. (26), who found that vitamin C plasma levels were not different in 34 children with wheeze compared with 201 children without wheeze (26) and from the findings of Picado et al. (30) that there was no significant difference in vitamin C serum/ plasma levels between 118 adult asthmatics and 121 healthy subjects. Moreover, our results differ from those of Mainous et al. (31), who, in a study of the NHANES III adult population, found no significant relation between serum vitamin C levels and use of health care services for wheezing. The discrepancy between different studies may reflect differences in the ascertainment of vitamin intake. In our analysis, for example, we found that neither a 24-hour food recall questionnaire nor a history of vitamin/mineral supplementation differentiated asthmatics from nonasthmatics. Alternatively, it is possible that nutrition plays different roles in the induction of asthma than in modifying the severity or symptoms of the disease. Finally, there may be discrepant methods of diagnosing asthma, using either questionnaires, functional measures, or health care utilization as indicators of the asthma diagnosis (32). In summary, we concluded from this analysis of a large national population sample of children that low levels of serum vitamin C are associated with an increased risk of asthma. This association was independent of confounding socioeconomic, anthropometric, smoking, and nutritional variables. REFERENCES 1. Chytil F. The lungs and vitamin A. Am J Physiol 1992;262: L517 L Gray T, Koo JS, Nettesheim P. Regulation of mucous differentiation and mucin gene expression in the tracheobronchial epithelium. Toxicology 2001;160: Weiss ST. Diet as a risk factor for asthma. Ciba Found Symp 1997;206: Knutsen SF, Fraser GE, Linsted KD, et al. Comparing biological measurements of vitamin C, folate, alpha-tocopherol and carotene with 24-hour dietary recall information in non- Hispanic blacks and whites. Ann Epidemiol 2001;11: Plan and operation of the Third National Health and Nutrition Examination Survey, Series 1: programs and collection procedures. Vital Health Stat ;32: Gunter EW, Lewis BG, Koncikowski SM. Laboratory procedures used for the Third Health and Nutrition Examination Survey (NHANES III), Hyattsville, MD: Centers for Disease Control and Prevention, von Mutius E, Schwartz J, Neas LM, et al. Relation of body mass index to asthma and atopy in children: the National Health and Nutrition Examination Study III. Thorax 2001;56: Seaton A, Devereux G. Diet, infection and wheezy illness: lessons from adults. Pediatr Allergy Immunol 2000;(suppl 13): Aderle WI, Ette SI, Oduwole O, et al. Plasma vitamin C (ascorbic acid) levels in asthmatic children. Afr J Med Med Sci 1985; 14: Schwartz J, Weiss ST. Dietary factors and their relation to respiratory symptoms. The Second National Health and Nutrition Examination Survey. Am J Epidemiol 1990;132: Soutar A, Seaton A, Brown K. Bronchial reactivity and dietary antioxidants. Thorax 1997;52: Kalayci O, Besler T, Kilinc K, et al. Serum levels of antioxidant vitamins (alpha tocopherol, beta carotene, and ascorbic acid) in children with bronchial asthma. Turk J Pediatr 2000;42: Schock BC, Young IS, Brown V, et al. Antioxidants and protein carbonyls in bronchoalveolar lavage fluid of children: normal data. Pediatr Res 2001;49: Mudway IS, Stenfors N, Blomberg A, et al. Differences in basal airway antioxidant concentrations are not predictive of individual responsiveness to ozone: a comparison of healthy and mild asthmatic subjects. Free Radic Biol Med 2001;31: Mohsenin V, Dubois AB, Douglas JS. Effect of ascorbic acid on response to methacholine challenge in asthmatic subjects. Am Rev Respir Dis 1983;127: Ogilvy CS, Dubois AB, Douglas JS. Effects of ascorbic acid and indomethacin on the airways of healthy male subjects with and without induced bronchoconstriction. J Allergy Clin Immunol 1981;67: Bucca C, Rolla G, Caria E, et al. Effects of vitamin C on airway responsiveness to inhaled histamine in heavy smokers. Eur Respir J 1989;2: Bucca C, Rolla G, Oliva A, et al. Effect of vitamin C on histamine bronchial responsiveness of patients with allergic rhinitis. Ann Allergy 1990;65: Romieu I, Sienra-Monge JJ, Ramirez-Aguilar M, et al. Antioxidant supplementation and lung functions among children with

7 Serum Vitamin C Levels and Asthma in Children 357 asthma exposed to high levels of air pollutants. Am J Respir Crit Care Med 2002;166: Bielory L, Gandhi R. Asthma and vitamin C. Ann Allergy 1994;73: Kaur B, Rowe BH, Ram FS. Vitamin C supplementation for asthma. Cochrane Database Syst Rev 2001;(4):CDC Hatch GE. Asthma, inhaled oxidants and dietary antioxidants. Am J Clin Nutr 1995;61(suppl):625S 30S. 23. Monteleone CA, Sherman AR. Nutrition and asthma. Arch Intern Med 1997;157: Winklhofer-Roob BM, Ellemunter H, Fruhwirth M, et al. Plasma vitamin C concentrations in patients with cystic fibrosis: evidence of associations with lung inflammation. Am J Clin Nutr 1997;65: Forastiere F, Pistelli R, Sestini P, et al. Consumption of fresh fruit rich in vitamin C and wheezing symptoms in children. Thorax 2000;55: Cook G, Carey I, Whincup P, et al. Effect of fresh fruit consumption on lung function and wheeze in children. Thorax 1997;52: Schwartz J, Weiss ST. Relationship between dietary vitamin C intake and pulmonary function in the First National Health and Nutrition Examination Survey (NHANES I). Am J Clin Nutr 1994;59: Hu G, Cassano PA. Antioxidant nutrients and pulmonary function: the Third National Health and Nutrition Examination Survey (NHANES III). Am J Epidemiol 2000;151: Powell CV, Nash AA, Powers HJ, et al. Antioxidant status in asthma. Pediatr Pulmonol 1994;18: Picado C, Deulofeu R, Lleonart R, et al. Dietary micronutrients/ antioxidants and their relationship with bronchial asthma severity. Allergy 2001;56: Mainous AG 3rd, Hueston WG, Connor MK. Serum vitamin C levels and use of health care resources for wheezing episodes. Arch Fam Med 2000;9: Pekkanen J, Pearce N. Defining asthma in epidemiological studies. Eur Respir J 1999;14:951 7.