Rationale: Available prospective studies of obesity and asthma have used only body mass index (BMI) as an indicator for adiposity; studies using detailed obesity measures are lacking, and the role of physical fitness level and sedentary time remains unexplored in the link between obesity and asthma.
Objectives: To compare various anthropometric measures of obesity in relation to childhood asthma, and to further characterize the interrelations among central obesity, physical fitness level, sedentary time, and asthma.
Methods: The nationwide Taiwan Children Health Study followed 2,758 schoolchildren from fourth to sixth grade, annually collecting data regarding physical fitness, sedentary time, obesity measures (comprising body weight and height, abdominal and hip circumference, skin fold thickness, and body composition), asthma, and pulmonary function tests. The generalized estimating equation was used for 3 years of repeated measurements to analyze the interrelation among obesity, sedentary time, physical fitness level, and asthma; a structural equation model was used to explore the pathogenesis among these factors. Asthma incidence was analyzed during a 2-year follow-up among centrally obese and nonobese groups in baseline children without asthma.
Measurements and Main Results: Central obesity most accurately predicts asthma. Low physical fitness levels and high screen time increase the risk of central obesity, which leads to asthma development. Obesity-related reduction in pulmonary function is a possible mechanism in the pathway from central obesity to asthma.
Conclusions: Central obesity measures should be incorporated in childhood asthma risk predictions. Children are encouraged to increase their physical fitness levels and reduce their sedentary time to prevent central obesity–related asthma.
Prospective studies have suggested that obesity antedates development of asthma. The role of physical fitness and sedentary time remain unexplored in the link between obesity and asthma.
Among various obesity measures, central obesity most accurately predicts asthma; physical fitness and sedentary time are leading factors in the pathway from central obesity to asthma and decreased pulmonary function is a possible mechanism in this link. Children are encouraged to increase their physical fitness levels and reduce sedentary time to prevent central obesity–induced asthma.
Obesity and asthma severely deteriorate child health. Epidemiologic studies have consistently demonstrated that obesity is closely linked to asthma, and that obesity typically antedates asthma occurrence. A recent metaanalysis that included six relevant child studies showed that overweight children exhibited increased risk of incident asthma (1). A positive dose-response relationship between body mass index (BMI) and asthma risk was also observed.
Available prospective studies of obesity and asthma have primarily used BMI as an indicator of adiposity. Alternative obesity measures that account for fat distribution include skinfold thickness, waist and hip circumference, and body composition. Two studies explored the relationship between skinfold thickness and asthma, but yielded inconsistent results (2, 3). One study observed no significant relationship between body fat percentage (measured using bioelectrical impedance) and asthma (4). In a pediatric study, Vahlkvist and coworkers (5) noted that children with asthma were less fit and had a higher body fat percentage compared with children without asthma. A cross-sectional study by Musaad and coworkers (6) demonstrated that abdominal obesity was more strongly related to childhood allergic asthma than was BMI. Moreover, another recent cross-sectional study by Forno and coworkers (7) reported that BMI, percent body fat, and waist circumference (WC) were all associated with increased risk of asthma. Besides, fat-free mass was noted to be inversely correlated to adult asthma (8). The results of the above studies were not consistent. Children studies using detailed obesity measures, such as total body fat or fat-free mass, are lacking and may provide additional insight into the obesity-asthma link.
Sedentary time, which is defined as the time spent in sitting behaviors, is associated with deleterious health consequences, such as all-cause mortality (9), obesity (10), and adverse metabolic profiles (11). Screen time, defined as time engaging with screen media, has been correlated with low cardiorespiratory fitness levels in previous longitudinal studies (12, 13). Although studies have confirmed the associations between physical fitness level, sedentary time, and obesity, studies exploring how physical fitness level and sedentary time affect asthma development are scant (14, 15). The role of physical fitness level and sedentary time in the link between obesity and asthma remains largely unexplored. Only one study concerning the obesity-asthma link adjusted physical activity as a confounder; however, the study concluded that the obesity-asthma relationship was independent of physical activity (16). Because physical fitness and sedentary time are correlated with obesity and asthma (14, 15, 17), we hypothesize that they are intermediate or precipitating factors in the pathogenesis from obesity to asthma.
The Taiwan Children Health Study (TCHS) followed 2,758 schoolchildren from the fourth to sixth grades, annually collecting data on physical fitness, sedentary time, obesity measures, and respiratory outcomes. Therefore, the TCHS data provide a unique opportunity to test our hypothesis: exploring the relationship between central obesity and childhood asthma; and investigating the interrelations among central obesity, physical fitness level, sedentary time, and asthma. Moreover, according to previous literature, one possible biologic mechanism from central obesity to asthma could be mediated by poor pulmonary function (18, 19). Therefore, we also aim to examine the role of pulmonary function between central obesity and childhood asthma.
The TCHS is a nationwide prospective multidisciplinary study of a school-based cohort, comprising two parts. The first cohort of seventh- to eighth-grade children was enrolled in 2007, and the second cohort of fourth-grade schoolchildren was enrolled in 2010 (20). The analysis herein involved data from the second cohort because it contained several obesity measures. A questionnaire was distributed to parents to address child respiratory health and associated confounders, and a questionnaire was given to children to assess the time spent on sedentary behaviors (only collected in 2010 and 2012). Between 2010 and 2012, we performed the annual follow-ups approximately from April to June regarding the parent and child questionnaires, obesity measures, physical fitness levels, and pulmonary function tests. Our study followed the guiding principles of the Declaration of Helsinki (21) and the protocol was approved by the Institutional Review Board at National Taiwan University Hospital.
We used parent-reported questionnaire to define the lifetime occurrence of physician-diagnosed asthma. The question was “Has a doctor ever diagnosed your child as having asthma”? Additionally, active asthma cases were distinguished from normal children by asking two questions: “Has a doctor ever diagnosed your child as having asthma”? and “Did your child ever experience difficulty breathing, or did you observe any wheezing or whistling from his or her chest in the past 12 months”? If the answers were “Yes” to both questions, we classified the child as an active asthma case. Moreover, during the follow-up surveys, the questionnaire responses of the parents were used to categorize the incident asthma statuses of the child participants.
Body weight was determined using a body composition machine that comprised an electronic scale (IOI 353; Jawon Medical, Korea; accuracy, 0.1 kg); the participants were weighed wearing light clothing and no shoes. Body height was measured to the nearest 0.1 cm by using the wall-mounted stadiometer (MAGATA, BW-120, Taiwan) at each school. BMI was calculated as weight divided by (height)2 (kg/m2) and was converted into age- and sex-specific BMI percentiles (22).WC was measured to the nearest 1 mm by using a flexible tape at the natural waist (midpoint between the lower ribcage and the iliac crest) and hip circumference was measured at the level of the maximal posterior extension of the buttocks. Three central obesity measures were taken: (1) WC, (2) waist-to-hip ratio (WHR), and (3) waist-to-height ratio (WHtR). Participants who exhibited sex- and age-specific central obesity measures greater than or equal to 85th percentile were classified as exhibiting central obesity. Skinfold thickness was measured in duplicate to the nearest 0.5 mm by using Lange calipers (Beta Technology, Santa Cruz, CA). Two skinfold sites over the bilateral triceps and gastrocnemius were measured based on standard procedures. The same investigator took all measurements. The skinfold sums were calculated by adding the average skinfold of the bilateral triceps and gastrocnemius. The body fat percentage was calculated based on the skinfolds according to equations from previous literature (23). Resistance and reactance were measured using a tetrapolar multifrequency bioelectrical impedance analysis (BIA) machine (IOI 353; Jawon Medical), which yielded detailed body composition data, such as body fat percentage, total body fat, fat-free mass, total body water, and total body muscle. Children were asked to avoid fluid or food intake and vigorous exercise 2 hours before BIA measurements.
Physical fitness tests were annually performed using a standardized protocol at each school during the follow-up surveys (24). Among these tests, an 800-m sprint was used to determine the cardiorespiratory endurance of each child. Starting from a resting position, the 800-m sprint was measured using electronic time measurement (seconds) at a precision of 1 per 100 seconds. The physical fitness index (z score) was calculated based on the speed of the 800-m sprint, using sex- and age-specific means and standard deviations. We defined a high physical fitness level as a positive z score in the 800-m sprint.
We used the Chinese version of the International Physical Activity Questionnaire (25) to determine child sedentary time. Sedentary time was defined as the hours per day that children spent sitting without doing physical activities. Screen time was the sum of hours per day spent watching television, using a computer, or playing videogames. Sedentary and screen times during school days and holidays were separately determined and the average daily sedentary and screen times were calculated using weighted means. The cut-off point used to establish an excessive sedentary time was an age- and sex-specific median split. Screen time was classified as high if children reported more than 2 hours per day (13).
The pulmonary function tests were performed according to our previous standardized protocol (26). Five pulmonary function indices were analyzed: (1) FVC, (2) FEV1, (3) FEV1/FVC, (4) maximal mid-expiratory flow, and (5) peak expiratory flow. We obtained sex-specific percentages for the predicted pulmonary function indices; details of this method were listed in a previous study (26).
The normality of the continuous variables was tested using a Kolmogorov-Smirnov test. The continuous variables were expressed as mean ± standard deviation and categorical variables were expressed as frequencies and percentages. We first categorized distinct obesity measures by using age- and sex-specific percentiles (<25, ≥25% to <85%, and ≥85%) and the study population as reference. To evaluate the repeated measurements throughout the 3-year follow-up period, we used generalized estimating equations (GEE) (27) to examine the interrelations among central obesity, physical fitness level, sedentary time, and respiratory outcomes (active and physician-diagnosed asthma). The primary advantage of GEE is that it accounts for within-child association. The correlation structure assumed for repeated measurements is a working independence correlation. We considered various confounders, namely, age, sex, parental educational level, family income, family history of atopy, whether the mother breastfed, and whether the child was exposed to in utero maternal smoke. Subjects with missing covariate information were included in the model using missing indicators. All analyses were conducted using SAS software version 9.2 (SAS Institute, Cary, NC). The statistical significance was set at 5%, based on two-sided estimation.
A structural equation model (SEM) was applied to the data from the 3-year follow-up, using Mplus software (Mplus version 5; Muthen and Muthen, Los Angeles, CA). The skewed-distributed continuous variables were log-transformed. We assumed that physical fitness level and sedentary time determined central obesity, which subsequently caused asthma. Furthermore, we hypothesized that the biologic pathway from central obesity to asthma could be mediated by poor pulmonary function. We used a linear combination of the respiratory outcomes and pulmonary function indices (FEV1/FVC) from survey 1 to 3. An estimated coefficient/SE value estimate of 1.96 or larger was considered statistically significant (equivalent to the 0.05 level). The final model was reached when the comparative fit index was greater than 0.90, the Tucker-Lewis index was greater than 0.90, and the root mean square error of approximation approached 0.05.
To address reverse causation, we restricted our analysis to 2,515 fourth graders without asthma and observed the incidence of asthma during a 2-year follow-up among centrally obese and nonobese groups in survey 1. Moreover, we chose fourth graders without central obesity and compared the incidence of central obesity during a 2-year follow-up among disparate physical fitness level and sedentary time groups in survey 1. The PROC LIFEREG procedure of SAS was used to calculate the hazard risk ratio.
The study population comprised a total of 2,758 fourth-grade schoolchildren in 2010. The average annual follow-up rate was 83.9% after three surveys. The BMI, total body fat, fat-free mass, total body muscle, and body fat percentage from BIA gradually increased throughout adolescence (Table 1). The prevalence of overweight children rose from 30.8% in survey 1, to 32.1% in survey 2, and finally to 34.4% in survey 3. The average sedentary time in survey 1 was 6.88 hours per day; it declined to 4.67 hours per day in survey 3. In survey 1, 8.8% of children had physician-diagnosed asthma and 4.6% exhibited active asthma in the previous 12 months. Tables 2–4 included GEE model results using repeated measurement of the 3-year data. Table 5 demonstrated SEM results in the total study population (n = 2,758). In Table 6, we demonstrated the survival analysis using children without asthma (n = 2,515), and in Table 7, non–centrally obese participants (WC group, n = 1,980; WHR group, n = 1,979; WHtR group, n = 1,994) in survey 1.
Parameter | Survey 1 (n = 2,758) | Survey 2 (n = 2,150) | Survey 3 (n = 1,933) |
---|---|---|---|
Age | 9.7 (0.5) | 10.6 (0.5) | 11.7 (0.5) |
Male sex | 1,409 (51.1) | 1,075 (50.0) | 965 (49.9) |
BMI, kg/m2 | 18.6 (3.8) | 19.3 (3.8) | 20.4 (4.1) |
Body fat percentage from skinfold, % | 23.2 (8.1) | 28.1 (8.5) | 25.7 (7.6) |
Sum of skinfolds, mm | 29.9 (11.7) | 37.3 (12.3) | 33.7 (11.1) |
WC, cm | 62.1 (9.9) | 65.5 (10.0) | 64.9 (10.4) |
WHR | 0.82 (0.06) | 0.82 (0.06) | 0.78 (0.24) |
WHtR | 0.45 (0.06) | 0.45 (0.06) | 0.44 (0.06) |
Body fat percentage from BIA, % | 17.3 (7.5) | 17.7 (7.7) | 19.6 (7.8) |
Total body fat, kg | 6.8 (4.5) | 7.9 (5.3) | 9.6 (5.9) |
Fat-free mass, kg | 29.8 (5.7) | 33.5 (5.9) | 36.4 (6.3) |
Total body water, kg | 21.5 (4.1) | 24.2 (4.3) | 26.6 (4.2) |
Total body muscle, kg | 27.7 (5.2) | 31.1 (5.4) | 33.7 (5.8) |
Physical fitness index | 0.00 (1.00) | −0.04 (0.99) | 0.16 (1.07) |
Sedentary time, h/d | 6.88 (5.02) | — | 4.67 (2.40) |
Screen time, h/d | 1.94 (1.99) | — | 2.94 (2.49) |
Pulmonary function FEV1/FVC, % | 90.8 (6.7) | 90.8 (6.7) | 92.2 (6.7) |
Active asthma | 127 (4.6) | 101 (4.7) | 68 (3.5) |
Physician-diagnosed asthma | 243 (8.8) | 198 (9.2) | 188 (9.7) |
Obesity Measures | Active Asthma | Physician-diagnosed Asthma | ||||
---|---|---|---|---|---|---|
OR1 (95% CI) | OR2 (95% CI) | OR3 (95% CI) | OR1 (95% CI) | OR2 (95% CI) | OR3 (95% CI) | |
BMI percentiles | ||||||
<25% | 1 | 1 | 1 | 1 | 1 | 1 |
≥25% to <85% | 1.67 (1.12–2.51) | 1.51 (0.89–2.55) | 1.45 (0.93–2.27) | 1.15 (0.88–1.50) | 1.14 (0.87–1.49) | 1.20 (0.90–1.59) |
≥85% | 2.16 (1.31–3.57) | 2.48 (1.35–4.55) | 2.33 (1.39–3.92) | 1.42 (1.01–1.99) | 1.35 (0.96–1.91) | 1.59 (1.12–2.27) |
P for trend | 0.002 | 0.003 | 0.002 | 0.05 | 0.10 | 0.01 |
Body fat percentage from skinfold | ||||||
<25% | 1 | 1 | 1 | 1 | 1 | 1 |
≥25% to <85% | 1.23 (0.86–1.76) | 1.35 (0.84–2.16) | 1.17 (0.78–1.73) | 1.02 (0.80–1.31) | 1.01 (0.78–1.30) | 1.05 (0.81–1.37) |
≥85% | 1.76 (1.12–2.77) | 2.25 (1.26–4.04) | 1.99 (1.20–3.99) | 1.12 (0.80–1.58) | 1.19 (0.84–1.69) | 1.34 (0.94–1.92) |
P for trend | 0.02 | 0.001 | 0.02 | 0.54 | 0.42 | 0.16 |
Sum of skinfolds | ||||||
<25% | 1 | 1 | 1 | 1 | 1 | 1 |
≥25% to <85% | 1.19 (0.84–1.70) | 1.30 (0.81–2.08) | 1.12 (0.76–1.68) | 1.04 (0.81–1.33) | 1.03 (0.80–1.33) | 1.07 (0.82–1.40) |
≥85% | 1.74 (1.11–2.73) | 2.23 (1.26–3.98) | 1.98 (1.20–3.25) | 1.15 (0.82–1.62) | 1.21 (0.85–1.71) | 1.38 (0.97–1.98) |
P for trend | 0.02 | 0.01 | 0.02 | 0.44 | 0.35 | 0.11 |
WC | ||||||
<25% | 1 | 1 | 1 | 1 | 1 | 1 |
≥25% to <85% | 2.12 (1.39–3.24) | 2.03 (1.18–3.48) | 1.84 (1.16–2.92) | 1.13 (0.88–1.46) | 1.13 (0.87–1.46) | 1.20 (0.91–1.59) |
≥85% | 2.61 (1.53–4.43) | 3.18 (1.69–5.99) | 2.70 (1.57–4.66) | 1.29 (0.92–1.81) | 1.27 (0.90–1.78) | 1.39 (0.97–1.99) |
P for trend | <0.001 | <0.001 | <0.001 | 0.14 | 0.17 | 0.07 |
WHR | ||||||
<25% | 1 | 1 | 1 | 1 | 1 | 1 |
≥25% to <85% | 1.65 (1.18–2.30) | 1.70 (1.06–2.71) | 1.51 (1.01–2.25) | 1.05 (0.83–1.82) | 1.08 (0.86–1.37) | 1.09 (0.84–1.41) |
≥85% | 2.23 (1.44–3.46) | 3.06 (1.75–5.34) | 2.42 (1.49–3.93) | 1.36 (0.99–1.87) | 1.38 (0.99–1.90) | 1.53 (1.09–2.17) |
P for trend | <0.001 | <0.001 | <0.001 | 0.09 | 0.08 | 0.03 |
WHtR | ||||||
<25% | 1 | 1 | 1 | 1 | 1 | 1 |
≥25% to <85% | 1.50 (1.03–2.18) | 1.44 (0.89–2.32) | 1.20 (0.80–1.80) | 1.07 (0.84–1.34) | 1.06 (0.84–1.35) | 1.02 (0.78–1.33) |
≥85% | 2.09 (1.29–3.39) | 2.76 (1.54–4.93) | 2.30 (1.40–3.77) | 1.44 (1.04–2.00) | 1.45 (1.04–2.05) | 1.56 (1.11–2.20) |
P for trend | 0.002 | 0.001 | 0.003 | 0.05 | 0.05 | 0.03 |
Body fat percentage from BIA | ||||||
<25% | 1 | 1 | 1 | 1 | 1 | 1 |
≥25% to <85% | 1.02 (0.72–1.45) | 0.95 (0.59–1.54) | 1.00 (0.65–1.54) | 1.06 (0.82–1.32) | 1.08 (0.84–1.40) | 1.03 (0.77–1.38) |
≥85% | 1.30 (0.80–2.09) | 1.38 (0.75–2.55) | 1.55 (0.91–2.65) | 1.23 (0.87–1.75) | 1.26 (0.88–1.81) | 1.22 (0.83–1.80) |
P for trend | 0.34 | 0.38 | 0.15 | 0.27 | 0.23 | 0.35 |
Total body fat | ||||||
<25% | 1 | 1 | 1 | 1 | 1 | 1 |
≥25% to <85% | 1.28 (0.87–1.87) | 1.22 (0.72–2.07) | 1.27 (0.80–2.03) | 1.22 (0.93–1.59) | 1.23 (0.94–1.62) | 1.22 (0.90–1.65) |
≥85% | 1.60 (0.97–2.63) | 1.90 (1.00–3.60) | 1.77 (1.01–3.09) | 1.24 (0.86–1.78) | 1.21 (0.83–1.75) | 1.19 (0.80–1.78) |
P for trend | 0.06 | 0.06 | 0.05 | 0.21 | 0.24 | 0.32 |
Fat-free mass | ||||||
<25% | 1 | 1 | 1 | 1 | 1 | 1 |
≥25% to <85% | 1.38 (0.91–2.10) | 1.41 (0.80–2.49) | 1.42 (0.87–2.33) | 0.98 (0.75–1.29) | 0.98 (0.74–1.30) | 1.10 (0.81–1.48) |
≥85% | 1.62 (0.97–2.70) | 1.85 (0.95–3.60) | 1.79 (1.00–3.21) | 0.86 (0.59–1.25) | 0.82 (0.56–1.18) | 0.94 (0.63–1.39) |
P for trend | 0.06 | 0.07 | 0.04 | 0.46 | 0.33 | 0.87 |
Total body water | ||||||
<25% | 1 | 1 | 1 | 1 | 1 | 1 |
≥25% to <85% | 1.38 (0.91–2.10) | 1.41 (0.80–2.48) | 1.41 (0.87–2.31) | 0.98 (0.75–1.29) | 0.98 (0.74–1.30) | 1.09 (0.81–1.47) |
≥85% | 1.59 (0.95–2.67) | 1.83 (0.94–3.56) | 1.76 (0.99–3.16) | 0.85 (0.59–1.23) | 0.81 (0.55–1.17) | 0.92 (0.62–1.38) |
P for trend | 0.07 | 0.07 | 0.05 | 0.42 | 0.30 | 0.82 |
Total body muscle | ||||||
<25% | 1 | 1 | 1 | 1 | 1 | 1 |
≥25% to <85% | 1.39 (0.91–2.11) | 1.40 (0.79–2.46) | 1.41 (0.86–2.31) | 0.98 (0.75–1.28) | 0.97 (0.74–1.28) | 1.10 (0.81–1.47) |
≥85% | 1.58 (0.94–2.66) | 1.86 (0.95–3.62) | 1.81 (1.01–3.21) | 0.83 (0.57–1.20) | 0.78 (0.54–1.14) | 0.92 (0.62–1.37) |
P for trend | 0.07 | 0.06 | 0.04 | 0.36 | 0.24 | 0.82 |
Physical Fitness/Sedentary Time | Active Asthma | Physician-diagnosed Asthma | ||||
---|---|---|---|---|---|---|
OR1 (95% CI) | OR2 (95% CI) | OR3 (95% CI) | OR1 (95% CI) | OR2 (95% CI) | OR3 (95% CI) | |
Physical fitness | ||||||
High | 1 | 1 | 1 | 1 | 1 | 1 |
Low | 1.16 (0.89–1.50) | 1.15 (0.87–1.51) | 1.03 (0.77–1.36) | 0.95 (0.79–1.14) | 0.93 (0.77–1.52) | 0.90 (0.79–1.09) |
P value | 0.28 | 0.33 | 0.86 | 0.56 | 0.44 | 0.29 |
Sedentary time | ||||||
Low | 1 | 1 | 1 | 1 | 1 | 1 |
High | 1.26 (0.91–1.76) | 1.32 (0.95–1.85) | 1.27 (0.91–1.78) | 0.96 (0.78–1.19) | 0.93 (0.75–1.15) | 0.91 (0.74–1.26) |
P value | 0.16 | 0.10 | 0.16 | 0.73 | 0.50 | 0.39 |
Screen time | ||||||
Low | 1 | 1 | 1 | 1 | 1 | 1 |
High | 1.06 (0.78–1.43) | 1.09 (0.79–1.50) | 1.04 (0.76–1.42) | 0.93 (0.76–1.15) | 0.86 (0.69– 1.07) | 0.85 (0.68–1.05) |
P value | 0.72 | 0.61 | 0.81 | 0.52 | 0.73 | 0.14 |
Physical Fitness/Sedentary Time | Central Obesity | ||
---|---|---|---|
WC (≥85%) OR (95% CI) | WHR (≥85%) OR (95% CI) | WHtR (≥85%) OR (95% CI) | |
Physical fitness | |||
High | 1 | 1 | 1 |
Low | 1.14 (1.12–1.16) | 1.11 (1.09–1.14) | 1.13 (1.11–1.16) |
P value | <0.001 | <0.001 | <0.001 |
Sedentary time | |||
Low | 1 | 1 | 1 |
High | 1.02 (1.00–1.05) | 1.01 (0.98–1.03) | 1.01 (0.99–1.04) |
P value | 0.06 | 0.59 | 0.22 |
Screen time | |||
Low | 1 | 1 | 1 |
High | 1.06 (1.03–1.08) | 1.03 (1.00–1.05) | 1.05 (1.02–1.07) |
P value | <0.001 | 0.04 | <0.001 |
Model for Active Asthma | Model for Physician- diagnosed Asthma | |
---|---|---|
Pathway | ||
Positive link from central obesity to asthma | 0.10* | 0.08* |
Negative link from physical fitness to central obesity | −0.42† | −0.42† |
Positive link from sedentary time to central obesity | 0.12† | 0.12† |
Negative link from physical fitness to sedentary time | −0.20† | −0.21† |
Link from physical fitness to asthma | 0.01 | 0.05 |
Link from sedentary time to asthma | 0.02 | −0.02 |
Negative link from central obesity to pulmonary function | −0.25† | −0.26† |
Negative link from pulmonary function to asthma | −0.09† | −0.12† |
Link from sedentary time to pulmonary function | 0.06 | 0.05 |
Indirect Effect | ||
From physical fitness to asthma | −0.05‡ | −0.03* |
From sedentary time to asthma | 0.01* | 0.01 |
From central obesity to asthma | 0.02‡ | 0.03† |
Model fitness | ||
Chi-square test of model fit (P value) | <0.001 | <0.001 |
CFI | 0.98 | 0.98 |
TLI | 0.97 | 0.97 |
RMSEA | 0.02 | 0.02 |
Central Obesity Measures in Survey 1 | Active Asthma | Physician-diagnosed Asthma | ||
---|---|---|---|---|
Incidence | HR* (95% CI) | Incidence | HR* (95% CI) | |
WC | ||||
<85% | 1.91 | 1 | 3.20 | 1 |
≥85% | 2.54 | 1.21 (0.56–2.60) | 3.70 | 1.22 (0.60–2.50) |
WHR | ||||
<85% | 1.84 | 1 | 3.28 | 1 |
≥85% | 3.19 | 1.20 (0.56–2.56) | 3.35 | 1.53 (0.64–3.66) |
WHtR | ||||
<85% | 1.92 | 1 | 3.31 | 1 |
≥85% | 2.69 | 1.56 (0.76–3.19) | 3.10 | 1.17 (0.57–2.39) |
Physical Fitness/Sedentary Time in Survey 1 | WC (≥85%) | WHR (≥85%) | WHtR (≥85%) | |||
---|---|---|---|---|---|---|
Incidence | HR* (95% CI) | Incidence | HR* (95% CI) | Incidence | HR* (95% CI) | |
Physical fitness | ||||||
High | 2.62 | 1 | 5.24 | 1 | 2.01 | 1 |
Low | 5.11 | 1.87 (1.23–2.83) | 9.25 | 1.58 (1.14–2.18) | 4.74 | 1.66 (1.18–2.34) |
Sedentary time | ||||||
Low | 3.93 | 1 | 6.80 | 1 | 2.97 | 1 |
High | 3.66 | 1.12 (0.82–1.54) | 7.37 | 0.94 (0.73–1.23) | 3.77 | 1.24 (0.93–1.67) |
Screen time | ||||||
Low | 3.30 | 1 | 6.50 | 1 | 2.70 | 1 |
High | 5.37 | 1.26 (0.92–1.72) | 8.39 | 1.01 (0.76–1.33) | 5.05 | 1.15 (0.87–1.53) |
In Table 2, although BMI percentiles reflect significant dose responsiveness in relation to active asthma, three central obesity indicators (WC, WHR, and WHtR) consistently revealed more significant dose-responsive relationships. Comparing two asthma outcomes, central obesity indicators significantly related to active asthma rather than physician-diagnosed asthma. This likely resulted because active asthma reflected a recent asthma status (within 12 mo), which should be more closely correlated to recent obesity measures compared with a lifetime diagnosis of asthma. OR1 was produced by GEE models adjusting for age, sex, parental education, family income, family history of atopy, breastfeeding, and in utero smoking. To examine the role of physical fitness and sedentary time in the relationships among central obesity and asthma, we additionally adjusted physical fitness to produce OR2 and sedentary time to yield OR3. Comparing OR1 with OR2 showed that estimated effect of active asthma changed by more than 10% when physical fitness was introduced to the base model (the risk of active asthma in WC ≥85% vs. <25%, OR1 = 2.61; OR2 = 3.18); this suggests that physical fitness is a crucial confounder, instead of intermediate factor, in the relationship between central obesity and asthma.
A SEM was used to confirm the association between central obesity and asthma (Table 5, Figure 1). The standardized path coefficients for the pathways from central obesity to asthma were significantly positive (0.1 in the model for active asthma; 0.08 in the model for physician-diagnosed asthma). Furthermore, to prove the temporal association, we calculated the incidence of asthma among distinct central obesity groups for a 2-year follow-up period in children without asthma of survey 1 (Table 6; see Figure E1 in the online supplement). The incidence of active asthma in the greater than or equal to 85% group was consistently higher than that of the less than 25% group, based on all central obesity indicators (the incidences of active asthma: WC ≥85% group vs. <25% group was 2.54 vs. 1.91 cases per 100 person-years; WHR ≥85% group vs. <25% group was 3.19 vs. 1.84 cases per 100 person-years; WHtR ≥85% group vs. <25% group was 2.69 vs. 1.92 cases per 100 person-years). The findings were nearly identical regarding the outcome of physician-diagnosed asthma.
The physical fitness index was negatively associated with all obesity measures, whereas sedentary and screen times were positively associated with obesity measures throughout the 3 years of repeated measurements (Table 4). Regarding the effect on central obesity, each unit increase in the physical fitness index related to a 0.05-cm reduction in WC. An additional hour of screen time per day correlated to a 0.35-cm increase in WC. The SEM (Table 5, Figure 1) results showed a negative link between physical fitness and central obesity and positive link between sedentary time and central obesity (standardized path coefficients, −0.42 and 0.12, respectively). We determined that physical fitness and sedentary time were negatively correlated (standardized path coefficients, −0.20).
Table 7 and Figure E2 show that the incidence of central obesity was consistently higher in the low physical fitness group than in the high physical fitness group for all central obesity indicators. The incidence of central obesity, as demonstrated using WC, was 5.11 and 2.62 cases per 100 person-years in the low- and high-level physical fitness groups, respectively; high sedentary and screen times also yielded high incidences of central obesity.
Table 3 shows the associations among physical fitness, sedentary time, and asthma based on the GEE model. Regardless of the confounders adjusted in the model, no significant relationships were observed. The SEM results (Table 5, Figure 1) show that the links between physical fitness level or sedentary time and asthma were nonsignificant. However, the physical fitness levels and sedentary time indirectly and significantly affected asthma. Although physical fitness levels and sedentary time did not directly influence asthma risk, they were mediated through central obesity.
Among the five pulmonary function indices, FEV1/FVC displayed the most significant associations with the obesity measures (see Table E1). Therefore, we chose FEV1/FVC as the representative index of pulmonary function in the SEM; the SEM results show that a high degree of central obesity causes correspondingly poor pulmonary function, increasing the risk of asthma (Table 5, Figure 1).
By comparing various obesity measures among children, we revealed that central obesity is the critical obesity measure for predicting asthma. Exploring the pathogenesis from central obesity to asthma showed that low physical fitness levels and high sedentary time are precipitating factors, rather than intermediate factors, in the pathway from central obesity to asthma. Decreased pulmonary function is a possible mechanism in the link between central obesity and asthma.
The finding that central obesity relates to asthma and incident asthma supports the reports of the few previous cross-sectional studies (6, 28, 29). However, these studies reported that the correlations were observed only in restricted asthma phenotypes, such as nonatopic asthma (29), children with asthma that exhibited allergic rhinitis (6), or only in women (28). These discrepancies likely resulted from the differing races of the study participants and the misclassification of asthma phenotypes. Only one longitudinal adult study in Sweden attained findings similar to those in the current study (30). Adolescence is the critical period for developing central obesity (31); however, this study was the first children study to compare the associations among various obesity measures and asthma, using 3-year repeated measurements. The possible mechanisms of central obesity–related asthma include the mechanical compression of abdominal obesity on the thoracic cage and the influence of proinflammatory mediators (produced by visceral fat) on airway remodeling.
Consistent with previous cross-sectional studies (32, 33), the SEM results yielded a significant inverse correlation between physical fitness levels and sedentary time. A recent longitudinal study demonstrated that excessive screen time during childhood predicts low cardiorespiratory fitness levels in adolescence (13). Additionally, previous studies regarding the relationship between physical fitness level and sedentary time and central obesity have been scarce and primarily cross-sectional (17, 34, 35). One adult study concluded that sedentary behavior is not associated with abdominal obesity (36). According to a review of the literature, this is the first study to show that low physical fitness levels and high sedentary time predict the development of central obesity, based on a 2-year follow-up study. Adolescence is a critical time for changes in body composition, sedentary lifestyle, and physical fitness levels. To prevent central obesity–related cardiometabolic and respiratory diseases in the teenage years and later life, improvements in physical fitness levels and decreases in sedentary behaviors must be advocated.
Because physical fitness levels and sedentary time were proposed as possible pathways linking obesity and asthma, we used SEM to examine their roles. This is the first study to show that low physical fitness levels and high sedentary time indirectly affect the development of asthma and are mediated by central obesity. The GEE and SEM analyses yielded no direct association between physical fitness levels or sedentary time and asthma. The SEM data (not shown) could not prove whether intermediate factors of physical fitness levels or sedentary time affected the pathway from central obesity to asthma. Furthermore, we investigated the direction of the pathophysiology by calculating the incidence from low physical fitness levels and high sedentary time to central obesity, and from central obesity to asthma. The reverse causality that children with asthma are inactive, leading to further risk of overweight, was not supported by a previous pediatric cohort study (37). In clinical practice, many physicians emphasize the effect of physical fitness training to lessen symptoms and improve quality of life in children with asthma. This might explain why we did not observe lower levels of physical fitness in children with asthma. Previous studies exploring the relationship among physical fitness levels, sedentary time, and asthma have used diverse methodologies and yielded inconsistent results. Two 10-year follow-up studies revealed that low physical fitness levels in childhood were associated with developing asthma in young adulthood. Rasmussen and coworkers (15) determined that physical fitness levels (measured by bicycle ergometer) were weakly correlated with incident asthma during adolescence. Ortega and coworkers (14) subsequently discovered that the protective effect of high cardiorespiratory fitness levels on incident asthma was particularly significant among obese individuals. This study supported our finding that the association between physical fitness levels and asthma is mediated by adiposity. However, these studies evaluated physical fitness levels only at the beginning of the study rather than repeatedly recording these levels throughout the study period. In addition, Sherriff and coworkers (38) demonstrated that excessive television viewing duration in children who exhibited no symptoms of wheezing at 3.5 years of age was associated with incident asthma in later childhood; however, no direct association was indicated between sedentary behaviors and bronchial hyperresponsiveness, which should be similar to asthma. The exact mechanism warrants exploration in subsequent studies.
We discovered that central obesity–related decline in pulmonary function (measured using FEV1/FVC) is a pathway that links to asthma. Similar findings have been reported in the literature in adult and child studies (39, 40). Abdominal obesity measures are more accurate predictors of pulmonary function than are body weight or BMI (41, 42). A longitudinal study that used a 7-year follow-up period indicated that changes in WC were associated with changes in FEV1 (43). Abdominal obesity may mechanically restrict the diaphragm and limit lung expansion, triggering asthma attacks.
The major strengths of our study lie in the prospective design and the large population-based child study that involved various obesity and physical fitness measures and questionnaires, which were annually repeated for 3 years. Furthermore, we used multiple exposure and outcome measures to address possible misclassifications. The results were similar regardless of the exposures and outcomes used. We discovered novel interrelations among central obesity, physical fitness levels, sedentary time, and asthma that were independent of various confounders and robust in the three distinct statistical analyses.
The study was limited by the questionnaire-derived sedentary time that the children self-reported; however, the relationships among sedentary time, physical fitness levels, and obesity were consistent with those reported in the literature. We used the same validated questionnaire, the International Physical Activity Questionnaire, in 2010 and 2012 surveys. The participants were likely sufficiently mature to validly report their sedentary time and parent-reported asthma was used to assess the outcomes. The SEM and incident rate calculations yielded similar findings for active and physician-diagnosed asthma. Furthermore, self-reported asthma questionnaires are widely used in epidemiologic studies and have been proven a valid measure of asthma. Another limitation of the study is that we used an 800-m sprint performed at a school as an indicator of physical fitness level rather than using accelerometers or bicycle ergometers; however, an 800-m sprint performed using standardized protocol is similar to the vigorous exercise attained using bicycle or accelerometers to represent cardiorespiratory fitness. Repeated 3-year measurements also improved the validity and accuracy of the study. Besides, BIA measurements vary among different ethnic groups and are influenced by consumption of food and vigorous exercise. Our BIA machine used predictive equations adjusted for Asian population and we tried to perform the measurement under standardized protocols. BIA is still considered a reliable and practical tool to estimate percent body fat in children (44).
Given the burdens of childhood obesity and asthma, these findings may have critical implications for physicians and in future disease-prevention policies. Children are encouraged to increase their physical fitness levels and reduce their sedentary time to prevent central obesity–induced asthma. Reducing central obesity enhances pulmonary function, preventing the occurrence of asthma.
1. | Chen YC, Dong GH, Lin KC, Lee YL. Gender difference of childhood overweight and obesity in predicting the risk of incident asthma: a systematic review and meta-analysis. Obes Rev 2013;14:222–231. |
2. | Berntsen S, Lødrup Carlsen KC, Hageberg R, Aandstad A, Mowinckel P, Anderssen SA, Carlsen KH. Asthma symptoms in rural living Tanzanian children; prevalence and the relation to aerobic fitness and body fat. Allergy 2009;64:1166–1171. |
3. | 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. Association with social class, perinatal events, and race. Am Rev Respir Dis 1990;142:555–562. |
4. | Irei AV, Takahashi K, Le DSNT, Ha PTN, Hung NTK, Kunii D, Sakai T, Matoba T, Yamamoto S. Obesity is associated with increased risk of allergy in Vietnamese adolescents. Eur J Clin Nutr 2005;59:571–577. |
5. | Vahlkvist S, Inman MD, Pedersen S. Effect of asthma treatment on fitness, daily activity and body composition in children with asthma. Allergy 2010;65:1464–1471. |
6. | Musaad SMA, Patterson T, Ericksen M, Lindsey M, Dietrich K, Succop P, Khurana Hershey GK. Comparison of anthropometric measures of obesity in childhood allergic asthma: central obesity is most relevant. J Allergy Clin Immunol 2009;123:1321–1327. |
7. | Forno E, Acosta-Pérez E, Brehm JM, Han YY, Alvarez M, Colón-Semidey A, Canino G, Celedón JC. Obesity and adiposity indicators, asthma, and atopy in Puerto Rican children. J Allergy Clin Immunol (In press) |
8. | Song W-J, Kim S-H, Lim S, Park Y-J, Kim M-H, Lee S-M, Lee S-B, Kim K-W, Jang H-C, Cho S-H, et al. Association between obesity and asthma in the elderly population: potential roles of abdominal subcutaneous adiposity and sarcopenia. Ann Allergy Asthma Immunol 2012;109:243–248. |
9. | Dunstan DW, Barr EL, Healy GN, Salmon J, Shaw JE, Balkau B, Magliano DJ, Cameron AJ, Zimmet PZ, Owen N. Television viewing time and mortality: the Australian Diabetes, Obesity and Lifestyle Study (AusDiab). Circulation 2010;121:384–391. |
10. | Liou YM, Liou T-H, Chang L-C. Obesity among adolescents: sedentary leisure time and sleeping as determinants. J Adv Nurs 2010;66:1246–1256. |
11. | Dunstan DW, Salmon J, Healy GN, Shaw JE, Jolley D, Zimmet PZ, Owen N; AusDiab Steering Committee. Association of television viewing with fasting and 2-h postchallenge plasma glucose levels in adults without diagnosed diabetes. Diabetes Care 2007;30:516–522. |
12. | Mitchell JA, Pate RR, Blair SN. Screen-based sedentary behavior and cardiorespiratory fitness from age 11 to 13. Med Sci Sports Exerc 2012;44:1302–1309. |
13. | Aggio D, Ogunleye AA, Voss C, Sandercock GRH. Temporal relationships between screen-time and physical activity with cardiorespiratory fitness in English schoolchildren: a 2-year longitudinal study. Prev Med 2012;55:37–39. |
14. | Ortega FB, Lee DC, Sui X, Ruiz JR, Cheng YJ, Church TJ, Miller CC, Blair SN. Cardiorespiratory fitness, adiposity, and incident asthma in adults. J Allergy Clin Immunol 2010;125:271–273. |
15. | Rasmussen F, Lambrechtsen J, Siersted HC, Hansen HS, Hansen NC. Low physical fitness in childhood is associated with the development of asthma in young adulthood: the Odense schoolchild study. Eur Respir J 2000;16:866–870. |
16. | Camargo CA Jr, Weiss ST, Zhang S, Willett WC, 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. |
17. | Rey-López JP, Ruiz JR, Vicente-Rodríguez G, Gracia-Marco L, Manios Y, Sjöström M, De Bourdeaudhuij I, Moreno LA; HELENA Study Group. Physical activity does not attenuate the obesity risk of TV viewing in youth. Pediatr Obes 2012;7:240–250. |
18. | Beuther DA, Weiss ST, Sutherland ER. Obesity and asthma. Am J Respir Crit Care Med 2006;174:112–119. |
19. | Eneli IU, Skybo T, Camargo CA Jr. Weight loss and asthma: a systematic review. Thorax 2008;63:671–676. |
20. | Chen YC, Chen PC, Hsieh WS, Portnov BA, Chen YA, Lee YL. Environmental factors associated with overweight and obesity in taiwanese children. Paediatr Perinat Epidemiol 2012;26:561–571. |
21. | World Medical Association declaration of Helsinki. Recommendations guiding physicians in biomedical research involving human subjects. JAMA 1997;277:925–926. |
22. | Chen W, Chang MH. New growth charts for Taiwanese children and adolescents based on World Health Organization standards and health-related physical fitness. Pediatr Neonatol 2010;51:69–79. |
23. | Slaughter MH, Lohman TG, Boileau RA, Horswill CA, Stillman RJ, Van Loan MD, Bemben DA. Skinfold equations for estimation of body fatness in children and youth. Hum Biol 1988;60:709–723. |
24. | Taiwan Ministry of Education. Nationwide physical fitness survey of school children and youth. Taipei, Taiwan: Ministry of Education; 2008. |
25. | Macfarlane DJ, Lee CC, Ho EY, Chan KL, Chan DT. Reliability and validity of the Chinese version of IPAQ (short, last 7 days). J Sci Med Sport 2007;10:45–51. |
26. | Lee YL, Hwang BF, Chen YA, Chen JM, Wu YF. Pulmonary function and incident bronchitis and asthma in children: a community-based prospective cohort study. PLoS ONE 2012;7:e32477. |
27. | Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics 1986;42:121–130. |
28. | Chen Y, Rennie D, Cormier Y, Dosman J. Sex specificity of asthma associated with objectively measured body mass index and waist circumference: the Humboldt study. Chest 2005;128:3048–3054. |
29. | Appleton SL, Adams RJ, Wilson DH, Taylor AW, Ruffin RE; North West Adelaide Health Study Team. Central obesity is associated with nonatopic but not atopic asthma in a representative population sample. J Allergy Clin Immunol 2006;118:1284–1291. |
30. | Kronander UN, Falkenberg M, Zetterström O. Prevalence and incidence of asthma related to waist circumference and BMI in a Swedish community sample. Respir Med 2004;98:1108–1116. |
31. | Chrzanowska M, Suder A, Kruszelnicki P. Tracking and risk of abdominal obesity in the adolescence period in children aged 7-15. The Cracow Longitudinal Growth Study. Am J Hum Biol 2012;24:62–67. |
32. | Sandercock GRH, Ogunleye AA. Screen time and passive school travel as independent predictors of cardiorespiratory fitness in youth. Prev Med 2012;54:319–322. |
33. | Tremblay MS, LeBlanc AG, Kho ME, Saunders TJ, Larouche R, Colley RC, Goldfield G, Connor Gorber S. Systematic review of sedentary behaviour and health indicators in school-aged children and youth. Int J Behav Nutr Phys Act 2011;8:98. |
34. | Vale S, Soares-Miranda L, Santos R, Moreira C, Marques A, Santos P, Teixeira L, Mota J. Influence of cardiorespiratory fitness and parental lifestyle on adolescents’ abdominal obesity. Ann Hum Biol 2011;38:531–536. |
35. | Moliner-Urdiales D, Ruiz JR, Ortega FB, Rey-Lopez JP, Vicente-Rodriguez G, España-Romero V, Munguía-Izquierdo D, Castillo MJ, Sjöström M, Moreno LA; HELENA Study Group. Association of objectively assessed physical activity with total and central body fat in Spanish adolescents; the HELENA Study. Int J Obes (Lond) 2009;33:1126–1135. |
36. | McGuire KA, Ross R. Incidental physical activity and sedentary behavior are not associated with abdominal adipose tissue in inactive adults. Obesity (Silver Spring) 2012;20:576–582. |
37. | Eijkemans M, Mommers M, de Vries SI, van Buuren S, Stafleu A, Bakker I, Thijs C. Asthmatic symptoms, physical activity, and overweight in young children: a cohort study. Pediatrics 2008;121:e666–e672. |
38. | Sherriff A, Maitra A, Ness AR, Mattocks C, Riddoch C, Reilly JJ, Paton JY, Henderson AJ. Association of duration of television viewing in early childhood with the subsequent development of asthma. Thorax 2009;64:321–325. |
39. | Pekkarinen E, Vanninen E, Länsimies E, Kokkarinen J, Timonen KL. Relation between body composition, abdominal obesity, and lung function. Clin Physiol Funct Imaging 2012;32:83–88. |
40. | Chen Y, Rennie D, Cormier Y, Dosman JA. Waist circumference associated with pulmonary function in children. Pediatr Pulmonol 2009;44:216–221. |
41. | Chen Y, Rennie D, Cormier YF, Dosman J. Waist circumference is associated with pulmonary function in normal-weight, overweight, and obese subjects. Am J Clin Nutr 2007;85:35–39. |
42. | Ochs-Balcom HM, Grant BJ, Muti P, Sempos CT, Freudenheim JL, Trevisan M, Cassano PA, Iacoviello L, Schünemann HJ. Pulmonary function and abdominal adiposity in the general population. Chest 2006;129:853–862. |
43. | Carey IM, Cook DG, Strachan DP. The effects of adiposity and weight change on forced expiratory volume decline in a longitudinal study of adults. Int J Obes Relat Metab Disord 1999;23:979–985. |
44. | Talma H, Chinapaw MJ, Bakker B, HiraSing RA, Terwee CB, Altenburg TM. Bioelectrical impedance analysis to estimate body composition in children and adolescents: a systematic review and evidence appraisal of validity, responsiveness, reliability and measurement error. Obes Rev 2013;14:895–905. |
Supported by grants #98-2314-B-002-138-MY3 and #101-2314-B-532-002-MY3 from Taiwan National Science Council.
Author Contributions: Y.-C.C. contributed the cohort data collection, statistical analysis, interpretation of data, and writing. Y.-K.T. assisted in the critical part of the statistical analysis and data interpretation. K.-C.H., P.-C.C., and D.-C.C. contributed to critically revising this manuscript for intellectual content. Y.L.L. reviewed the study design and supervised the study.
Originally Published in Press as DOI: 10.1164/rccm.201401-0097OC on March 26, 2014
This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org
Author disclosures are available with the text of this article at www.atsjournals.org.