Conflict of interest: None declared.
Physical activity and respiratory symptoms in children: The generation R study†
Article first published online: 10 JUL 2013
Copyright © 2013 Wiley Periodicals, Inc.
Volume 49, Issue 1, pages 36–42, January 2014
How to Cite
Driessen, L. M., Kiefte-de Jong, J. C., Jaddoe, V. W.V., Hofman, A., Raat, H., de Jongste, J. C. and Moll, H. A. (2014), Physical activity and respiratory symptoms in children: The generation R study. Pediatr. Pulmonol., 49: 36–42. doi: 10.1002/ppul.22839
- Issue published online: 16 DEC 2013
- Article first published online: 10 JUL 2013
- Manuscript Revised: 2 MAY 2013
- Manuscript Accepted: 2 MAY 2013
- Manuscript Received: 30 NOV 2012
- Erasmus Medical Center, the Erasmus University Rotterdam, the Netherlands Organization for Health Research and Development (Zon Mw)
- Europe Container Terminals (ECT)
- physical activity;
- respiratory symptoms;
- shortness of breath;
- young children;
To assess the relationship between physical activity in second year of life and respiratory symptoms during the pre-school period.
This study was embedded in the Generation R Study, a large prospective birth-cohort study in Rotterdam, the Netherlands. Physical activity was measured in the second year of life by an Actigraph accelerometer in a subgroup of 347 children (182 boys, 165 girls; mean age 25.1 months) and data were expressed as counts per 15 sec in categories: light activity (302–614 counts/15 sec), moderate activity (615–1,230 counts/15 sec), and vigorous activity (≥1,231 counts/15 sec). Respiratory symptoms were assessed by the International Study of Asthma and Allergies in Childhood Questionnaire in the third and fourth year of life.
Physical activity levels were not associated with wheezing symptoms in the third and fourth year of life (OR: 0.98; 95% CI: 0.92–1.05 and OR: 0.99; 95% CI: 0.92–1.07 for total activity, respectively), nor associated with shortness of breath symptoms (OR: 0.98; 95% CI: 0.92–1.05 and OR 1.03; 95% CI: 0.96–1.11 for total activity, respectively).
These results suggest that physical activity may not play an important role in the development of respiratory symptoms in pre-school children. Pediatr Pulmonol. 2014; 49:36–42. © 2013 Wiley Periodicals, Inc.
Asthma remains the most common chronic disease in childhood and the prevalence is increasing in Western countries. The causes for this rising prevalence are not fully understood, but life style factors seem likely to be of importance. We previously found that a “Western-like” dietary pattern was associated with asthma-like symptoms at pre-school age. Besides dietary patterns, physical activity could possibly also influence the development of asthma. A recent study showed that regular activity in children is insufficient. Increasing activity levels in children may protect them against asthma.[5-9] But not all literature supports this hypothesis.[10, 11] A recent review by Eijkemans et al. showed that high physical activity may lower the risk of asthma but results are heterogeneous across different populations (e.g., children vs. adults). In contrast, according to Ownby et al., active children may even be more susceptible to develop asthma. However, most previous studies assessing the relationship between physical activity and asthmatic symptoms have been conducted with questionnaires to measure activity patterns in children. In this study, we used accelerometry, a more valid method for physical activity assessment.[14-16] Studies on asthma that used accelerometry were mostly performed in children older than 2 years,[17-20] but physical activity and asthma in younger children is an understudied area. Especially in very young children it might be important to develop prevention strategies, because the first 2 years of life are critical for lung development. In this period, the lungs are more susceptible for changes. If prevention strategies could already be applied in these young children, prevalence of asthma might be much lower later in life.
Therefore, in this study, we assessed the association between accelerometry-measured physical activity at the age of 2 years and asthma-like symptoms during the pre-school period.
The present study was embedded in the Generation R Study, a population-based prospective cohort study from fetal life until young adulthood. In total, 9,778 mothers with a delivery date between April 2002 and January 2006 were enrolled. Among a randomly selected subgroup of Dutch children and their parents, more detailed assessments were conducted. Of the 1,246 participants enrolled in this subcohort, 1,106 mothers and their children continued to participate in postnatal follow up. The exact cohorts, design and measurements has been described previously.
Between December 2005 and February 2008, 617 two-year-old children participating in this subcohort visited the Generation R Study clinic (mean age 25.1 ± 1.1 months). For the purpose of the present study, 500 of these 617 participants were asked to wear an accelerometer as described previously. Due to the limited number of accelerometers, not all children were offered an accelerometer. Of these initial 500 children, 30 children did not agree to wear it, 72 children agreed to wear the accelerometer but eventually did not wear it, and data of 51 children were excluded because of technical problems or insufficient wearing time. The remaining study population consisted of 347 children (Fig. 1).
The Generation R Study has been approved by the Medical Ethical Committee of the Erasmus Medical Center, Rotterdam, the Netherlands.
Children were asked to wear an ActiGraph accelerometer (Model AM-7164, ActiGraph, Pensacola, FL) during 1 week day and 1 weekend day at the age of 2 years (median: 2 days; range 1–2 days. In total, 71% of the children had worn the accelerometer for 1 week day and 1 weekend day. This uni-axial accelerometer measures normal human movement and has proven to be accurate and valid for physical activity measurement of 3- to 18-year-old children.[14-16] The accelerometer was attached to the child's right hip by an elastic waist belt during waking hours. During sleeping and exposure to water the accelerometer was removed. All accelerometer data were processed using Actisoft 3.2 (ActiGraph, Pensacola, FL) and MAHUffe 126.96.36.199 (Institute of Metabolic Science, Medical Research Council Epidemiology Unit, Cambridge, UK) software programs. These data were included in analyses when the accelerometer was worn for at least 400 min/day Because of the sporadic and intermittent nature of children's physical activity, short epochs were used. In the present study, movement values (counts per minute) were collected every 15 sec. Substantial periods of zero activity counts (>10 min) and sleeping periods were excluded. Data were expressed as mean counts per minute and as time spent in activities of different intensity using age-specific count cut points. As there were no count cut points available for 2-year-old children, count cut points of 3-year-old children were used, as validated by Sirard. Intervals were categorized as light (302–614 counts/15 sec), moderate (615–1,230 counts/15 sec), or vigorous activity (≥1,231 counts/15 sec). Physical activity data were adjusted for the time of wearing the accelerometer by calculating the mean percentage of physical activity per day relative to the amount of minutes per day of wearing the accelerometer.
The primary outcome was the presence or absence of parental reported respiratory symptoms in third and fourth year of life.
Information regarding wheezing and shortness of breath in the past year was obtained on the basis of International Study of Asthma and Allergies in Childhood (ISAAC) questionnaire at the age of 24, 36, and 48 months. Wheezing was defined as at least one episode of wheezing in the third and fourth year, respectively. Shortness of breath was defined as at least one episode of shortness of breath in the third and fourth year, respectively. Answers were given by the parents. Because of potential reverse causality, data at 24 months (i.e., wheezing and shortness of breath in the second year of life) were not included in the outcome analyses since it was too close to the physical activity measurements. Answers were analyzed dichotomously (wheezing or shortness of breath yes vs. no).
Anthropometric measurements at 36 and 48 months were collected at regular visits to the child health centers. Body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters. Data on gender and gestational age at birth were obtained from midwife and hospital registries. At the age of 3 months, neuromotor development was assessed using an adapted version of Touwen's Neurodevelopmental examination.[27-29] Additional characteristics (maternal age and BMI, maternal smoking during pregnancy, household income, maternal educational level, attendance of childcare in second year of life and food consumption in second year of life) were obtained from pre- and post-natal questionnaires completed by the parents. Maternal educational level was divided into low–moderate (no education finished, primary school or secondary school) and high (vocational training, bachelor's degree or academic education). For household income, a cut-off point of 2,000 euro/month was used.
By means of Chi-square statistics for categorical variables and independent sample t-tests for continuous variables, differences in baseline characteristics were assessed between children with and without wheezing or shortness of breath. To compare physical activity between the groups, Mann–Whitney U-tests were performed.
Logistic regression models were used to examine the association between physical activity and respiratory symptoms and adjusted for relevant confounding factors when the regression coefficient changed by more than 10%. All measures of association are presented in odds ratios (OR) with their 95% confidence intervals (CIs).
To improve the validity of our results, the missing values were multiple-imputed. Five imputed datasets were created using a fully conditional specified model. Imputations in each dataset were based on the relationships between missing data and all the other variables included in the study. In this paper, we presented the pooled results from the five imputed datasets. A P-value of ≤0.05 was considered statistically significant.
Statistical analyses were performed using the Statistical Package of Social Sciences version 17.0 for Windows (SPSS, Inc., IBM, Chicago, IL).
Table 1 shows the baseline characteristics of the study population. The prevalence of wheezing was stable in the third and fourth year of life (for both ages 14%, P < 0.01) and the prevalence of shortness of breath slightly decreased from 18% in the third year of life to 13% in the fourth year of life (P < 0.01). Both in the third and in the fourth year of life maternal age and maternal educational levels were slightly lower in the groups with respiratory symptoms.
|Maternal age at intake (mean ± SD)||32 ± 4|
|Maternal BMI at intake (mean ± SD)||24.5 ± 4.2|
|Smoking during pregnancy, N (%)||79 (23%)|
|Family history of asthma or atopy, N (%)||180 (52%)|
|Household income, N (%)|
|<2,000 euro/month||34 (10%)|
|≥2,000 euro/month||313 (90%)|
|Educational level mother, N (%)|
|Male, N (%)||182 (52%)|
|Gestational age at birth (mean ± SD)||40.1 ± 1.5|
|Birth weight z-score (mean ± SD)||0.04 ± 0.97|
|BMI 29–35 months (mean ± SD)||16.2 ± 1.2|
|Motor function (mean ± SD)||6.6 ± 2.5|
|Attendance of childcare in second year of life, N (%)||263 (76%)|
|Fruit intake in second year of life (g/day) (mean ± SD)||171 ± 100|
|Vegetables in second year of life (g/day) (mean ± SD)||51 ± 43|
|Salty snacks in second year of life (g/day) (mean ± SD)||3 ± 0.3|
|Wheezing in third year of life, N (%)||50 (14%)|
|Wheezing in fourth year of life, N (%)||48 (14%)|
|Shortness of breath in third year of life, N (%)||63 (18%)|
|Shortness of breath in fourth year of life, N (%)||43 (13%)|
Attendance at childcare in second year of life was significantly lower in children with shortness of breath in the fourth year of life (P < 0.01).
Median (range) total activity was 71 (130) versus 70 (137) min/day and 68 (123) versus 71 (150) min/day in children with wheezing relative to those without wheezing in the third and fourth year of life, respectively (P = 0.53, and P = 0.39, respectively). In addition, activity levels were similar in children with and without shortness of breath. Median (range) total activity was 71 (122) versus 70 (143) min/day and 75 (123) versus 70 (150) min/day in children with shortness of breath compared to children without shortness of breath at the ages of 3 and 4 years, respectively (P = 0.64 and P = 0.74 for 3 and 4 years, respectively). Wearing time was not different between children with and without respiratory symptoms (mean difference less than 15 min/day; P > 0.20).
We found no significant associations between physical activity and the risk of respiratory symptoms at the age of 3 and 4 years (Tables 2 and 3). Results were similar after stratification of physical activity levels into tertiles and after additional adjustment for any respiratory symptoms at the age of 2 years (data not shown).
|Physical activity||Univariate model OR (95% CI)||P-value||Multivariate model OR (95% CI)a||P-value|
|Wheezing in third year of life|
|Light activity||0.97 (0.86–1.08)||0.56||0.94 (0.83–1.06)||0.32|
|Moderate—vigorous activity||1.01 (0.89–1.15)||0.86||1.00 (0.88–1.15)||0.70|
|Wheezing in fourth year of life|
|Light activity||0.99 (0.89–1.11)||0.87||0.95 (0.84–1.08)||0.43|
|Moderate—vigorous activity||1.01 (0.88–1.16)||0.92||1.00 (0.85–1.18)||0.96|
|Physical activity||Univariate model OR (95% CI)||P-value||Multivariate model OR (95% CI)a||P-value|
|Shortness of breath in third year of life|
|Light activity||0.97 (0.87–1.08)||0.54||0.95 (0.85–1.06)||0.35|
|Moderate—vigorous activity||1.00 (0.89–1.13)||0.99||1.00 (0.88–1.13)||0.96|
|Shortness of breath in fourth year of life|
|Light activity||1.09 (0.96–1.22)||0.19||1.04 (0.91–1.19)||0.54|
|Moderate—vigorous activity||1.09 (0.95–1.26)||0.23||1.08 (0.93–1.25)||0.33|
In this prospective study, no differences in activity pattern at 2 years were found between children with and without wheezing or shortness of breath at pre-school age.
These results do not confirm our hypothesis that reduced physical activity is associated with the development of respiratory symptoms. There is evidence that regular training could raise the threshold for respiratory symptoms. Recently, Kosti et al. showed that leisure time activity was inversely associated with life-time asthma in 10–12-year-old children. A study among Greek schoolchildren aged 10–12 years found an association between decreased physical activity measured by questionnaire and exercise-induced bronchoconstriction. Also, in a population of children aged 3–5 years, decreased physical activity measured by accelerometry was associated with a history of asthma whereas Rundle et al. did not find any association in 4-year olds. In contrast, the ISAAC found that both vigorous activity and sedentary behavior, measured by questionnaire, was associated with a higher prevalence of wheezing symptoms and asthma.
Within the Avon Longitudinal Study of Parents and Children (ALSPAC), it was found that longer TV watching at pre-school age was associated with development of asthma in later childhood. Similarly, Rasmussen et al. showed that children with low physical fitness (measured by bicycle ergometer test) were at a higher risk for the development of asthma later in life. These latter results could be further explained by the phenomenon of tracking. A recent review showed that activity levels at young ages probably track during early childhood. We therefore hypothesized that children with a high physical activity level in second year of life, will be more likely to have a high activity level in subsequent years that may result in less respiratory symptoms. However, Rasmussen et al. also showed that when adjustment in the analyses were performed for other risk factors of asthma, physical fitness was only associated with the development of physician diagnosed asthma and not with airway responsiveness. Thus, positive associations between physical fitness and asthma in previous studies may be easily related to other confounding factors. Moreover, most studies did not measure physical activity levels objectively. Nevertheless, our findings are in line with other studies who objectively measured physical activity levels by accelerometry. A Dutch birth cohort from Eijkemans et al. also found no differences in accelerometry-measured physical activity levels between 4- to 5-year-old children with and without wheezing. Likewise, two other case–control studies did not find any difference in physical activity levels between 9 and 11 years old children with and without asthma.[37, 38]
Physical activity may increase the sensation of respiratory symptoms due to triggering of bronchial contraction during exercise. It has been found that asthmatic children may be afraid to develop these symptoms and therefore exercise less. Also, children with asthma report more activity limitations due to breathing problems than children without asthma,[17, 40] implying that physical activity itself may not influence the development of asthma but impaired physical activity levels may be a consequence of the disease itself.
This potential reverse causation might cancel out any positive influence of physical activity on respiratory symptoms. Nonetheless, we assessed the outcome after physical activity measurements and there appeared to be no association between physical activity and respiratory symptoms at the age of 24 months (data not shown). For that reason, this potential reverse causality may not be of major concern in our study group.
Important strengths of this study are the prospective data collection and use of accelerometry to measure physical activity objectively. In addition, we were able to adjust for potential confounders, which were collected in the Generation R Study. Furthermore, we applied a multiple imputation technique to deal with missing data.
A limitation of our study is that disease outcome was based on questionnaires, which might have led to information bias. However, only if parents were less likely to report respiratory symptoms when the child was active, this may have overestimated our results, and this seems unlikely.
Accelerometry has proven to be a valid method to measure physical activity,[14-16] but it remains difficult to assess the actual pattern of activity, as the accelerometer on the hip was not worn all the time (e.g., during sleeping or swimming). We tried to minimize this bias by adjusting for the time of wearing the accelerometer. Also, measuring physical activity levels in children aged 2 years of age is challenging since children just developed the ability to walk and clear physical activity patterns may not be developed sufficiently yet to assess any effect with health outcomes.
We used self-reported wheezing and shortness of breath which is an accepted method in epidemiological studies reflecting the incidence of asthma-related symptoms in pre-school children. Nevertheless, comprehensive lung function measurements are difficult to perform in young children. Therefore, final conclusions related to asthma diagnosis should be made with caution.
Last, our study was not able to detect a significant association between physical activity and wheezing which may be due to the small study group. Therefore, further large-scale studies may be needed to elucidate the potential role of physical activity on the development of respiratory symptoms in children.
Physical activity is not related to respiratory symptoms in subsequent years. Further large-scale studies are required to clarify the effects of objectively measured physical activity in young children on respiratory symptoms later in life.
The Generation R Study is conducted by the Erasmus Medical Center in close collaboration with the School of Law and Faculty of Social Sciences of the Erasmus University Rotterdam, the Municipal Health Service–Rotterdam Metropolitan Area, the Rotterdam Homecare Foundation, and the Stichting Trombosedienst and Artsenlaboratorium Rijnmond. We acknowledge the contributions of children and parents, general practitioners, hospitals, and midwives in Rotterdam. The funders had no role in the design of the study, the data collection and analyses, the interpretation of data, or the preparation of, review of, and decision to submit the manuscript.
- 1Time trends in the prevalence of asthma and allergy among 6–7-year old children. Results from ISAAC phase I and III studies in Kaunas, Lithuana. Medicina (Kaunas) 2008; 44:944–952., , .
- 18Physical activity and asthma symptoms among New York City Head Start Children. J Asthma 2009; 46:803–809., , , , , .
- 19Intensity of physical activity, cardiorespiratory fitness, and body mass index in youth. J Phys Act Health 2010; 7:54–59., , , , , .
- 20Assessing physical activity in children with asthma: convergent validity between accelerometer and electronic diary data. Res Q Exerc Sport 2009; 80:153–163., , .
- 25Calibration and evaluation of an objective measure of physical activity in preschool children. J Phys Act Health 2005; 2:324–336..
- 28Maternal symptoms of anxiety during pregnancy affect infant neuromotor development: the Generation R Study. Dev Neuropsychol 2009; 34:476–493., , , , , , , .
- 29Fetal programming of infant neuromotor development: the Generation R Study. Pediatr Res 2010; 67:132–137., , , , , , .
- 34ISAAC Phase Three Study Group. The association between BMI, vigorous physical activity and television viewing and the risk of symptoms of asthma, rhinoconjunctivitis and eczema in children and adolescents: ISAAC Phase Three. Clin Exp Allergy 2013; 43:73–84., , , , , ,