• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Am J Prev Med. Author manuscript; available in PMC Jul 1, 2010.
Published in final edited form as:
PMCID: PMC2760597

Sustained Effect of Early Physical Activity on Body Fat Mass in Older Children



Physical activity is assumed to reduce excessive fatness in children. This study examined whether the benefits of early childhood moderate-to-vigorous physical activity (MVPA) on fatness are sustained throughout childhood.


MVPA minutes per day (min/d) and fat mass (kilograms; kg) were measured using accelerometry and dual-energy x-ray absorptiometry in 333 children aged 5, 8, and 11 years who were participating in the Iowa Bone Development Study. Mixed regression models were used to test whether MVPA at age 5 years had an effect on fat mass at age 8 years and age 11 years, after adjustment for concurrent height, weight, age, maturity, and MVPA. The analysis was repeated to control for fat mass at age 5 years. Using mixed-model least-squares means, adjusted means of fat mass at age 8 years and age 11 years were compared between the highest and lowest quartiles of MVPA at age 5 years. Data were collected between 1998 and 2006 and analyzed in 2008.


For boys and girls, MVPA at age 5 years was a predictor of adjusted fat mass at age 8 years and age 11 years (p<0.05). In girls, the effect of MVPA at age 5 years was not significant when fat mass at age 5 years was included. Boys and girls in the highest quartile of MVPA at age 5 years had a lower fat mass at age 8 years and age 11 years than children in the lowest MVPA quartile at age 5 years (p<0.05; mean difference 0.85 kg at age 8 years and 1.55 kg at age 11 years).


Some effects of early-childhood MVPA on fatness appear to persist throughout childhood. Results indicate the potential importance of increasing MVPA in young children as a strategy to reduce later fat gains.


Childhood obesity is associated with increased cardiovascular risks such as hypertension, hyperlipidemia, type 2 diabetes mellitus, and early development of atherosclerotic lesions.1 Lack of physical activity during childhood is widely assumed to contribute to obesity. Many studies have investigated the relationship between physical activity and obesity; however, the results have been inconsistent.2 This inconsistency has raised the issue of the measurement accuracy of physical activity, fat mass, or both. In response, investigators have turned to the use of objective measures of children’s physical activity and fatness to better quantify relationships.

Ness et al.3 reported significant associations between physical activity measured using accelerometry and fat mass measured using dual-energy x-ray absorptiometry (DXA) in a large cohort of children aged 12 years (n=5500). The results suggested the beneficial effect of activity on fatness, although this presumption is not definitive because study design was cross-sectional.

Using a longitudinal design, Janz et al.4 studied the relationship between physical activity and fatness in 379 young children (baseline age 5 years). Physical activity was measured using accelerometry and fatness was measured using DXA. The study found that children maintaining a high level of physical activity were less likely than peers to be in the upper quartile for DXA-measured fatness at follow-up and were less likely to gain fatness during the study period. Also using a longitudinal study design, Johnson et al.5 studied whether physical activity energy expenditure influenced fat-mass change during a 3-to-5-year follow-up (baseline age 4 years to 11 years). This study measured physical activity energy expenditure using doubly labeled water and fat-mass change using DXA. The authors reported that physical activity energy expenditure at baseline did not predict fat-mass change.

Moore et al.6 measured physical activity using accelerometry and estimated body fatness using skinfolds and BMI. These researchers demonstrated that accumulated physical activity over 7 years (from age 4 years to age 11 years) was associated with fatness at age 11 years. However, that study did not find a relationship between physical activity at age 4 years and fatness at age 11 years. Its findings suggested that the protective benefits of physical activity at an early age are not sustained unless the activity level is maintained. However, in a 3-year follow-up study, Stevens et al.7 investigated associations between accelerometry-determined physical activity and percentage of body fat estimated with bio-electric impedance in 454 2nd-grade American Indian children. That study demonstrated that baseline physical activity was associated with later percentage of body fat in normal-weight children but not overweight children. The effort suggested the potential for sustained effects of early physical activity on later fatness; however, the authors did not adjust their final analysis for concurrent physical activity, and therefore the results are inconclusive.

Understanding if early physical activity influences fat mass later in life because of a sustained effect would help to inform early-intervention decisions aimed at preventing obesity. The current study expands on the previous work of Janz et al.4 by examining in a cohort (n=333) of children the associations between accelerometry-measured physical activity of children aged 5 years and the DXA-derived fat mass of children aged 8 and 11 years. It was hypothesized that physical activity during early childhood would result in less fat mass later in childhood via sustained effects.



The Iowa Bone Development Study is a longitudinal study of bone health during childhood. The study participants are a subset of a larger cohort of Midwestern children (n=890) recruited during 1998–2001 and then participating in the Iowa Fluoride Study. A detailed description of the demographic characteristics of participants can be found elsewhere.4,8 Physical activity and body-fat data were obtained three times per child during a 6-year span. Four hundred thirty-three cohort children aged approximately 5 years (baseline) participated in physical activity and fat-mass measurements, along with 495 children aged approximately 8 years and 406 children aged approximately 11 years. A total of 333 children (148 boys and 185 girls) who had physical activity and fat-mass data for all three visits were included in the data analysis.

There were no statistical differences in mean baseline heights and weights between children with physical activity and fat-mass data at all three visits and those without. There was a small mean baseline-age difference between children aged 5.3 years with data at all three visits versus children aged 5.2 years without them. The study was approved by The University of Iowa IRB. Written, informed consent was provided by the parents of the children and assent was obtained from the children. Data were collected between 1998 and 2006 and analyzed in 2008.

Fatness Measured by DXA

For children aged 5 years and 8 years, whole-body scans using a Hologic QDR 2000 DXA were conducted with software version 7.20B and fan-beam mode. For children aged 11 years, the Hologic QDR 4500 DXA (Delphi upgrade) with software version 12.3 and fan-beam mode was used for scan acquisition. Using criterion-carcass analysis of pigs, Pintauro et al.9 have shown DXA to be an accurate and precise measure of fat mass in children. Quality-control scans were performed daily using the Hologic phantom. To minimize operator-related variability, all measurements were conducted by one of three experienced technicians. To adjust for the differences between the two DXA machines, translational equations were used from 4500 DXA measures to 2000 DXA measures for the records of children aged 11 years. The translational equations were developed specifically for the two scanners in a pilot study where 60 of the children (32 boys, 28 girls) aged 9.9–12.4 years (M=11.4, SD=0.4) were scanned on each machine in random order during one clinic visit (TLB, unpublished observations, 2007). Total body-fat mass (kilograms; kg) was derived from the scan images. Percentage of body fat was calculated based on body weight and total fat mass (total fat mass ÷ body weight × 100). The coefficient for determination (R2) for the 4500 DXA data regressed onto the 2000 DXA data was 0.9979, and the actual observations were very tight around the regression line.

Physical Activity Measured by Accelerometry

ActiGraph uniaxial accelerometers (model 7164) were used to measure physical activity levels (movement counts) of children aged 5, 8, and 11 years. This monitor has been validated for measuring physical activity in children.1013 The procedure for physical activity measurement has been described elsewhere.14 In brief, children aged 5 years and 8 years were asked to wear the monitor all day during waking hours for 4 consecutive days, including 1 weekend day, during one of the autumn months. Children aged 11 years were asked to wear the monitor all day during waking hours for 5 consecutive days, including both weekend days, during one of the autumn months. The number of wear days was increased for children aged 11 years because previous research11 demonstrates less stability in accelerometry-measured physical activity in older children compared to younger children. Parents were instructed to fasten the belt at their child’s waist (on the mid-axillary line). Monitors and data-recording sheets were sent to parents and returned via prepaid U.S. mail. Only children who wore the accelerometer at least 8 hours per day for at least 3 days and within 15 months of the DXA scan were regarded as completing the physical activity measurement. Movement-count values were accumulated and summed over 1-minute intervals.

In this study, a summary variable of daily minutes spent in moderate-to-vigorous physical activity (MVPA) was used. The variable was derived using the cut-point threshold of 3000 accelerometer movement counts per minute (ct·min−1). In laboratory- and field-based studies, this cut point has been associated with MVPA at normal walking speeds and is predictive of fat mass and heart-disease risk factors in children and adolescents.8,12,1517

Anthropometry and Maturity Assessment Methods

At each DXA visit, research nurses trained in anthropometry measured the child’s height (in centimeters; cm) using a Harpenden stadiometer and body mass (kg) using a Healthometer physician’s scale. Both devices were routinely calibrated. Sitting height was also measured at age 11 years. Maturity offset (year from peak height velocity) was calculated using predictive equations established by Mirwald and colleagues.18 The equations included height, weight, age, gender, sitting height, and leg length as predictors of years from peak height velocity (or somatic maturity). This equation has been validated in white Canadian children and adolescents (R2=0.91, 0.92, SE of the estimate=0.49, 0.50). The maturity-offset variable was dichotomized as 0 (prior to peak height velocity, or pre-mature) or 1 (≥ peak height velocity, or mature).

Statistical Analysis

Data were analyzed by gender using SAS version 9.1.3. Gender-specific descriptive analyses including t-tests were conducted for measures at ages 5, 8, and 11 years. Mixed regression models for correlated data were used to examine whether physical activity at age 5 years predicted fat mass at age 8 years and age 11 years. The residual observations within-children were correlated through the within-person variance–covariance matrix. Matrix structure type was determined based on Akaike’s Information Criterion (AIC) for goodness of fit. An unstructured variance–covariance matrix was chosen because it allowed for an assumption of higher variance for measures at age 11 years together with the within-person covariance. This analysis controlled for concurrent (at age 8 years or age 11 years) height, weight, age, maturity, and MVPA. Residual and studentized residual graphs were used to confirm the models’ assumptions and fit. The analysis was repeated to control also for fat mass at age 5 years. This latter approach tested whether there was an additional effect on fat mass over and above the sustained effect between MVPA at age 5 years and fat mass at age 5 years.

To show the impact of high versus low early-age physical activity, the highest-quartile group and the lowest-quartile group based on MVPA at age 5 years were identified for the whole sample, and the analysis was stratified by gender. The highest-quartile group included 37 girls and 47 boys. The lowest-quartile group included 25 boys and 59 girls. Mixed-model least-squares means calculated at person-level of covariates (concurrent age, height, weight, maturity, and MVPA) for age group were used to compare the fat mass in children aged 8 years and 11 years in the highest and lowest quartiles of MVPA at age 5 years. Least-squares mean fat mass for children at age 5 years was calculated in a separate cross-sectional model. The level of significance was set at 0.05 for all analyses.


Characteristics of Participants

Table 1 presents the characteristics of the children at the time of each examination (age 5 years, age 8 years, and age 11 years), including age, height, weight, MVPA, fat mass, and percentage of body fat. Mean height and weight were similar between boys and girls. At all three time points, boys engaged in more MVPA than girls and had lower percentages of body fat. At age 5 years and age 8 years, boys also had lower fat mass. The average time between DXA and MVPA measurement was <4 months at age 5 years and age 11 years and <5 months at age 8 years. As shown in Figure 1, the majority of boys and girls engaged in MVPA <35 minutes per day at age 5 years. The distribution of MVPA levels for girls was more negatively skewed than for boys. All boys and girls at age 5 years and age 8 years were classified as pre-mature. All boys and 81% of girls at age 11 years were pre-mature. Therefore, the maturity-offset variable was included as a possible predictor only in the models for girls.

Figure 1
The distribution of MVPA among boys and girls at age 5 years
Table 1
Characteristics of children (n=148 boys; n=185 girls)

The Effect of Early Physical Activity on Later Fat Mass

Gender-specific regression models for fat mass are presented in Table 2 and Table 3. The unstructured within-person covariance-error matrix improved the model fit when compared to compound symmetry structure. For example, in Table 2, the AIC for boys decreased from 1181.9 to 1126.8; for girls, it decreased from 1441.2 to 1396.2. After adjustment for concurrent (age 8 years or age 11 years) age, height, weight, and MVPA, MVPA at age 5 years was a significant predictor of later fat mass in both boys and girls (p<0.05; Table 2). However, in girls, when the model included fat mass at age 5 years, MVPA at age 5 years did not reach significance (Table 3). In all models, concurrent (age 8 years or age 11 years) MVPA was significantly associated with fat mass (p<0.05).

Table 2
Mixed regression model analysis of fat mass at age 8 and age 11 as predicted by MVPA at age 5 yearsa
Table 3
Mixed regression model analysis of fat mass at age 8 years and age 11 years as predicted by MVPA at age 5 yearsa

Fat-mass differences between the highest and the lowest quartiles of MVPA at age 5 years are presented in Table 4. The MVPA means at age 5 years for the highest and lowest quartiles were 51 minutes and 9 minutes for boys and 45 minutes and 11 minutes for girls, respectively. Therefore, the mean difference in MVPA at age 5 years between the highest and lowest quartiles was 42 minutes for boys and 34 minutes for girls. After adjustment for concurrent age, height, weight, and MVPA, boys and girls in the highest MVPA quartile at age 5 years had significantly lower fat mass at age 8 years and age 11 years than those in the lowest quartile (p<0.05). Fat-mass means for the highest and lowest quartiles of MVPA at any age gradually increased over time regardless of gender. The magnitude of the difference of least-squares mean fat mass also increased with age (1.07 kg for boys and 0.62 kg for girls at age 8 years and 1.36 kg for boys and 1.74 kg for girls at age 11 years).

Table 4
Mixed-model least-squares means for fat mass in children in the highest and lowest quartiles of MVPA at age 5 years


Using objective measures, this study examined the association between early physical activity and later fatness during childhood. It provided evidence that early physical activity affects later fat mass. The effect was somewhat stronger in boys, given that significant associations persisted after adjusting for fatness at age 5 years. Similarly, Ness and colleagues3 reported a stronger relationship in boys compared to girls with respect to objective measures of MVPA and fat mass. Importantly, this report’s regression analysis indicated that early physical activity predicted later fat mass even after adjustment for concurrent physical activity. This finding lends support to the hypothesis that there is a pathway between early physical activity and later fat mass that is independent of the effect of accumulated physical activity. Physical activity at an early age may influence the physiologic mechanism of fat accumulation during growth so that early physical activity may have a sustained effect on the fatness phenotype later in life. These findings also suggest that children who are less physically active at an early age may be more susceptible to fat accumulation later in childhood.

In this cohort, it has previously been shown that concurrent physical activity at age 5 years is associated with fat mass at age 5 years.8 This article establishes that concurrent physical activity is also a significant predictor of fat mass at age 8 years and age 11 years. These findings suggest that engaging in physical activity at an early age (age 5 years) has an immediate effect on body-fat level and that maintaining physical activity throughout childhood has a preventive effect on both body-fat accumulation during childhood and, presumably, the development of obesity. This interpretation is supported by the study by Moore et al.,6 which showed that higher levels of accumulated physical activity during a 7-year period were associated with less body fat later (age 11 years).

How much early physical activity is needed for a potentially protective effect? The measurement error inherent in accelerometry methods and discrepancies in approaches to calibrating movement-count data to minutes of MVPA preclude a precise recommendation of protective minutes. However, the β coefficients of the regression model (Table 2) indicate that if all other variables were held constant, 10 minutes of MVPA at age 5 years would result, on average, in 0.2 kg less fat mass at age 8 years and age 11 years, whereas 10 minutes of concurrent MVPA result in 0.1 kg less fat mass. The preventive advantages of early MVPA for the most-active participants appeared to increase with age. The mean fat-mass difference between these highest and lowest quartiles for boys was 1.07 kg at age 8 years and 1.37 kg at age 11 years. For girls, the mean difference in fat mass was 0.63 kg at age 8 years and 1.73 kg at age 11 years.

Limitations of the present research include the use of a Midwestern convenience sample with low minority representation and relatively high SES. In addition, the analyses did not consider other factors associated with fat accumulation such as energy intake, fat intake, sedentary time, and genetic factors. Other (unknown) confounders may have influenced the results of this study. Epochs of 1 minute may underreport the amount of physical activity accumulated by young children.19 Physical activity outcomes differ depending on accelerometry cut-point values used to define intensity levels.20,21 This study used the same accelerometry cut point throughout the study period to define MVPA. There are no universally agreed-on, age-related cut-point values, and there is limited guidance for interpreting accelerometry data in longitudinal research. However, using a receiver operating characteristic (ROC) curve approach, Evenson and colleagues17 have recently shown no difference in MVPA cut points between children aged 5–8 years. Additional studies are needed to establish evidence-based cut points for accelerometer movement counts throughout childhood.

On the other hand, this is one of the few studies investigating longitudinal associations between objectively measured physical activity and fatness in a relatively large sample of children. The use of objective measures helps to clarify associations between physical activity and obesity. This study is also useful in that it investigated the association between physical activity and fatness in children from approximately kindergarten age to junior high age. In the last 2 decades, the prevalence of childhood obesity has tripled in children and adolescents.22,23 Results from this work support the importance of physical activity engagement at an early age for reducing later fat accumulation. The current public health emphasis of continuous physical activity promotion throughout childhood and adolescence appears warranted.


The authors thank the staff of the Iowa Fluoride Study for their organizational efforts, especially exercise specialist Ms. Kelli O’Neil. They gratefully acknowledge and thank the children and parents of the Iowa Fluoride Study and the Iowa Bone Development Study, because without their contributions this work would not have been possible.


This study was supported by the National Institute of Dental and Craniofacial Research R01-DE12101 and R01-DE09551 and the General Clinical Research Centers Program from the National Center for Research Resources, M01-RR00059.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

No financial disclosures were reported by the authors of this paper.


1. Jolliffe CJ, Janssen I. Vascular risks and management of obesity in children and adolescents. Vasc Health Risk Manag. 2006;2:171–187. [PMC free article] [PubMed]
2. Wareham NJ, van Sluijs EM, Ekelund U. Physical activity and obesity prevention: a review of the current evidence. Proc Nutr Soc. 2005;64:229–247. [PubMed]
3. Ness AR, Leary SD, Mattocks C, et al. Objectively measured physical activity and fat mass in a large cohort of children. PLoS Med. 2007;4:e97. [PMC free article] [PubMed]
4. Janz KF, Burns TL, Levy SM, et al. Tracking of activity and sedentary behaviors in childhood: the Iowa Bone Development Study. Am J Prev Med. 2005;29:171–178. [PubMed]
5. Johnson MS, Figueroa-Colon R, Herd SL, et al. Aerobic fitness, not energy expenditure, influences subsequent increase in adiposity in black and white children. Pediatrics. 2000;106:E50. [PubMed]
6. Moore LL, Gao D, Bradlee ML, et al. Does early physical activity predict body fat change throughout childhood? Prev Med. 2003;37:10–17. [PubMed]
7. Stevens J, Suchindran C, Ring K, et al. Physical activity as a predictor of body composition in American Indian children. Obes Res. 2004;12:1974–1980. [PubMed]
8. Janz KF, Levy SM, Burns TL, et al. Fatness, physical activity, and television viewing in children during the adiposity rebound period: the Iowa Bone Development Study. Prev Med. 2002;35:563–571. [PubMed]
9. Pintauro SJ, Nagy TR, Duthie CM, et al. Cross-calibration of fat and lean measurements by dual-energy x-ray absorptiometry to pig carcass analysis in the pediatric body weight range. Am J Clin Nutr. 1996;63:293–298. [PubMed]
10. Janz KF. Validation of the CSA accelerometer for assessing children’s physical activity. Med Sci Sports Exerc. 1994;26:369–375. [PubMed]
11. Janz KF, Witt J, Mahoney LT. The stability of children's physical activity as measured by accelerometry and self-report. Med Sci Sports Exerc. 1995;27:1326–1332. [PubMed]
12. Trost SG, Ward DS, Moorehead SM, et al. Validity of the computer science and applications (CSA) activity monitor in children. Med Sci Sports Exerc. 1998;30:629–633. [PubMed]
13. Ekelund U, Sjöström M, Yngve A, et al. Physical activity assessed by activity monitor and doubly labeled water in children. Med Sci Sports Exerc. 2001;33:275–281. [PubMed]
14. Janz KF, Gilmore JM, Levy SM, et al. Physical activity and femoral neck bone strength during childhood: the Iowa Bone Development Study. Bone. 2007;41:216–222. [PMC free article] [PubMed]
15. Trost SG, Way R, Okely AD. Predictive validity of three ActiGraph energy expenditure equations for children. Med Sci Sports Exerc. 2006;38:380–387. [PubMed]
16. Jago R, Wedderkopp N, Kristensen PL, et al. Six-year change in youth physical activity and effect on fasting insulin and HOMA-IR. Am J Prev Med. 2008;35:554–560. [PubMed]
17. Evenson KR, Catellier DJ, Gill K, Ondrak KS, McMurray RG. Calibration of two objective measures of physical activity for children. J Sports Sci. 2008;26:1557–1565. [PubMed]
18. Mirwald RL, Baxter-Jones AD, Bailey DA, et al. An assessment of maturity from anthropometric measurements. Med Sci Sports Exerc. 2002;34:689–694. [PubMed]
19. Welk GJ, Corbin CB, Dale D. Measurement issues in the assessment of physical activity in children. Res Q Exerc Sport. 2000;71:59–73. [PubMed]
20. Schmitz KJ, Treuth M, Hannan P, et al. Predicting energy expenditure from accelerometry counts in adolescent girls. Med Sci Sports Exerc. 2005;37:155–161. [PMC free article] [PubMed]
21. Guinhouya CB, Hubert H, Soubrier S, et al. Moderate-to-vigorous physical activity among children: discrepancies in accelerometry-based cut-off points. Obesity (Silver Spring) 2006;14:774–777. [PubMed]
22. Strauss RS, Pollack HA. Epidemic increase in childhood overweight, 1986–1998. JAMA. 2001;286:2845–2848. [PubMed]
23. Ogden CL, Flegal KM, Carroll MD, et al. Prevalence and trends in overweight among U.S. children and adolescents, 1999–2000. JAMA. 2002;288:1728–1732. [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...