Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Relationship between temperament, nonresting energy expenditure, body composition, and physical activity in girls

Abstract

Objectives: To assess the extent that predilection for movement, as measured by a temperament questionnaire (activity temperament), contributes to nonresting energy expenditure and body composition in girls.

Design, Setting, and Participants: Baseline data for 196 premenarcheal non-obese girls aged 8–12 y were obtained from a longitudinal study of growth and development. The association of activity temperament with nonresting energy expenditure in girls with low and high levels of physical activity was evaluated, as was the association of activity temperament with body composition.

Measures: Maternal reports of child activity temperament were obtained by questionnaire. Nonresting energy expenditure was calculated as total energy expenditure (measured by doubly labeled water) minus resting energy expenditure (obtained by indirect calorimetry). Body composition was estimated by total body water. Questionnaires and activity diaries were used to assess physical activity and sedentary behavior.

Results: Higher activity temperament was associated with higher nonresting energy expenditure after multivariate control for weight, vigorous activity, walking and light activity, and television viewing, although activity temperament did not account for a large percentage of the variability in nonresting energy expenditure (partial squared correlation coefficient=0.03). In girls with physical activity levels below the median, high activity temperament was associated with a mean±s.d., nonresting energy expenditure of 310±138 kJ (74±33 kcal) above that of girls with a low activity temperament. Girls with a high activity temperament had less body fat than did girls with a low activity temperament (21.6 vs 24.5%, a difference of 2.9 percentage points; 95% confidence interval, 1.3–4.4 percentage points).

Conclusion: Predilection for movement, as measured by a temperament questionnaire, contributes to nonresting energy expenditure and may be useful in capturing an aspect of energy expenditure in population studies. The cross-sectional observation that girls with a high activity temperament were leaner than girls with a low activity temperament suggests that a constitutional predilection for movement may play a role in the development of obesity.

Introduction

Obesity results when energy intake exceeds energy expenditure. Study of the components of energy expenditure is critical to the development of approaches to prevent obesity. Nonresting energy expenditure (NREE), the component of energy expenditure not due to resting energy expenditure (REE), is often thought of as the energy of activity. NREE can be accurately determined by doubly labeled water, but large studies of free-living subjects generally use physical activity questionnaires, motion sensors, or direct observation to estimate energy expended in activity because of the monetary and time costs associated with analysis of doubly labeled water. In such population studies, the energy spent in physical activity is most often conceptualized as the energy spent in volitional activity and ‘exercise’, and energy spent fidgeting is unlikely to be captured.

Nonexercise activity thermogenesis (NEAT) has been characterized as the nonvolitional component of NREE.1 NEAT has been proposed to explain the variability between subjects in their tendency to gain weight in overfeeding experiments where physical activity level is controlled.1 Nonvolitional movements (such as fidgeting) are a substantial contributor to energy expenditure in some individuals. For example, fidgeting while sitting or standing has been estimated, in carefully controlled experiments of adults, to increase energy expenditure by 2.5±1.7 kJ/min (0.6±0.4 kcal/min) and 4.2±2.1 kJ/min (1.0±0.5 kcal/min) above that required for sitting or standing quietly.2

Temperament, a construct of behavioral style, is considered relatively stable in an individual. Of the many scales designed for assessing temperament in children, one of the most widely used was designed by Thomas and Chess,3 and includes nine subscales, with an activity subscale designed to assess predilection for movement. We conceptualized two pathways through which predilection for movement, as measured by a temperament questionnaire (activity temperament), could affect NREE and thereby influence energy balance (Figure 1). Activity temperament may directly increase energy expenditure via fidgeting. Alternatively, activity temperament, may be indirectly associated with energy expenditure by predicting a child's physical activity level; for example, children with a high activity temperament, and thus a greater drive to move, might engage in high levels of physical activity.

Figure 1
figure1

Pathways through which activity temperament could affect NREE.

The objectives of these analyses were to assess whether activity temperament is associated with NREE, whether activity temperament and physical activity interrelate as components of NREE, and to explore whether the effect of activity temperament on NREE is associated with body composition. Since activity temperament and physical activity may be interrelated in girls who report high levels of physical activity (indirect pathway), we expected to see a greater contribution of activity temperament to NREE in girls with low levels of physical activity (direct pathway) (Figure 1). To understand these associations, we analyzed baseline data from a longitudinal study of growth and development in girls.

Methods

Study design and population

A total of 196 girls were enrolled in the Massachusetts Institute of Technology (MIT) Growth and Development Study between September 1990 and June 1993. A description of the study has been previously published.4 Premenarcheal girls (aged 8–12 y) with a triceps skinfold thickness below the 85th percentile based on a First National Health and Nutrition Examination Survey (NHANES I) reference standard5 were recruited from Cambridge and Somerville public schools in Massachusetts, the MIT summer day camp, and friends and siblings of enrollees. The cohort is 71% white subjects, 14% black subjects, and 15% other race/ethnicity. The study protocol was approved by both the Committee on the Use of Humans as Experimental Subjects at MIT, and the Institutional Review Board of the Tufts-New England Medical Center in Boston, Massachusetts.

Measures

The participants were admitted to the Clinical Research Center at MIT for an overnight visit. REE was measured by indirect calorimetry.4,6 Total energy expenditure (TEE) and total body water were measured using doubly labeled water over a 2-week period as previously described.4 TEE was calculated from the mean daily rate of carbon dioxide (rCO2) production,7 the calculated oxygen consumption from rCO2, and the food quotient from a 7-day food record kept during the second week of the 2-week period, using Weir's equation.4,8 NREE was computed by subtracting REE from TEE. Height was measured to 0.1 cm with a wall-mounted stadiometer, and weight was measured in a hospital gown using a Seca scale accurate to 0.1 kg. We calculated body mass index (BMI) as weight in kilograms divided by the square of height in meters. We determined BMI z-score using the Centers for Disease Control and Prevention growth reference.9 Fat–free mass was determined from total body water according to the method of Halliday and Miller,10 with the assumptions that fat-free mass was 73% water, the 18O dilution space was 1% higher than total body water, and the deuterium dilution space was 3% higher than the 18O dilution space.4,11 We calculated percentage body fat as the difference of body weight and fat-free mass (to yield fat mass), divided by body weight.

We estimated the girls' physical activity level from questionnaires and 7-day activity diaries. Participants were asked by questionnaire to estimate time spent, on a typical weekend day and on a typical school day, sleeping, sitting, standing, walking (including light activity such as doing chores around the house or the yard), and engaged in vigorous activity (exercising, ‘playing or working very hard where your heart beats fast, you breathe hard, and you are sweating,’ or being involved in sports).12 A 24-h grid in which girls could select up to four activities per hour was used. Estimates of time sitting, standing, walking and light activity, and in vigorous activity were also measured with a 7-day activity diary, which corresponded with the first week of the measured energy expenditure period. Girls were asked to record hourly their activity level (sleeping, sitting, standing, walking and light activity, vigorous activity), with a possibility for two activity levels within an hour block. A standardized protocol was used to explain the activity diary and activity questionnaire to the girls. We calculated weekly time spent walking (including light activity) and in vigorous activity from the questionnaire and the activity diary, and then averaged questionnaire and diary estimates to arrive at typical weekly hours of walking and light activity and hours of vigorous activity. To create a physical activity score, we summed the estimate of walking and light activity time and two times the hours of vigorous activity (to reflect relative metabolic cost).13 We assessed time typically spent watching television on weekdays and weekend days by questionnaire and calculated average weekly hours.

Temperament was characterized by the girls' mothers with an instrument adapted from one developed by Thomas and Chess.3 Four subscales were measured: activity, persistence, intensity, and distractibility. We used only the activity subscale in these analyses. Mothers were asked to rate their daughter on nine items related to predilection for movement (Table 1); each item consisted of a four-level scale (hardly ever, sometimes, often, almost always). Responses to all nine items were summed to obtain the activity temperament score (possible range: 9–36, with higher scores indicating greater predilection for movement). To test reliability, 50 mothers repeated the ratings 2 weeks after the first administration. To test stability, 41 mothers completed the ratings again 1 year after the first administration.

Table 1 Activity temperament questions

Statistical analyses

NREE adjusted for the participant's body weight was calculated by regressing weight on NREE and using the residual plus the mean of NREE as the outcome variable (adjusted NREE) in our analyses. We used linear regression to determine whether activity temperament independently predicted adjusted NREE after controlling for physical activity score and television viewing. Participants with complete data for adjusted NREE, activity temperament, physical activity score, and television viewing were included in the model. We log-transformed activity temperament score and square-root-transformed television viewing to better meet the assumptions of linear regression analysis. Regression models were estimated in the full sample and also for two subsamples (‘low physical activity’ and ‘high physical activity’) constructed by dividing the data at the median of physical activity score. In addition, we dichotomized activity temperament score at the median to categorize girls as having a high or a low activity temperament. To estimate the difference in adjusted NREE between girls with a high activity temperament and girls with a low activity temperament in the high and low physical activity subsamples, adjusted means for high and low activity temperament were estimated from the regression models. We assessed the association between activity temperament score and percentage body fat by the Pearson correlation coefficient. The difference in percentage body fat between girls with a high activity temperament and girls with a low activity temperament was assessed by t-test after splitting the sample at the median level of physical activity.

Results

A total of 172 girls (average age, 10.1 y) had complete data for REE, TEE, and activity temperament. Their mean±s.d. NREE was 2954±771 kJ (706±184 kcal), REE averaged 5148±618 kJ (1230±148 kcal), and TEE averaged 8102±1163 kJ (1935±278 kcal) (Table 2). Percentage body fat averaged 23% (range, 10–36%). The median hours of television watched per week was 21.5 h (interquartile range, 13–35 h); data were missing for two girls. In all 13 girls were excluded from analyses using physical activity because they had incomplete activity diaries. The girls walked or engaged in light activity an average of 18.6±7.5 h per week, and engaged in vigorous activity an average of 9.5±5.7 h per week. The means reported in Table 2 did not change substantively when the data were restricted to the 159 girls with complete data on physical activity. The physical activity score ranged from a minimum of 8 to a maximum of 76 with a mean of 37.6. Activity temperament score ranged from a minimum of 12 to a maximum of 34 with a median of 21. The Pearson correlation between activity temperament score measured 2 weeks apart was r=0.90 (n=50). The Pearson correlation between activity temperament score measured one year apart was r=0.62 (n=41).

Table 2 Participant characteristicsa

Physical activity score was positively associated with adjusted NREE (R2=0.11, P<0.0001). Television viewing was negatively associated with adjusted NREE (R2=0.03, P=0.027) and remained an independent predictor of adjusted NREE (P=0.033) when added to the regression of adjusted NREE on the physical activity score. Higher activity temperament independently predicted higher adjusted NREE (P=0.030) in a model containing physical activity score, television viewing, and activity temperament (R2 for full model=0.16) (Table 3), but the association was attenuated when activity temperament was the single predictor in the model (R2=0.02, P=0.056; data not shown). The partial squared correlation coefficient for activity temperament in the full model containing physical activity score, television viewing, and activity temperament was 0.03 (Table 3).

Table 3 Regression equations predicting adjusted NREEa

As we hypothesized, activity temperament independently predicted adjusted NREE after controlling for physical activity score (continuous scale) and television viewing among girls with low physical activity, but not among girls with high physical activity (Table 3). The physical activity score was positively associated with adjusted NREE in the high physical activity subsample, but in the low physical activity subsample the slope of the score with adjusted NREE was shallower and the association was not statistically significant. In both subsamples, television viewing and the physical activity score were not correlated (r=0.03, P=0.81 low activity; r=−0.12, P=0.29 high activity). Activity temperament score was weakly correlated with the physical activity score in both subsamples (r=0.23, P=0.04, low physical activity; r=0.20, P=0.07, high physical activity).

We replaced the activity temperament score with a dichotomous variable (high and low activity temperament) in the high and low physical activity regression models to estimate the difference in energy expenditure between high activity temperament girls and low activity temperament girls while controlling for television viewing and physical activity level as continuous variables. In the low physical activity model, the mean adjusted NREE in girls with high activity temperament was 310 kJ (74 kcal) higher than in the girls with low activity temperament (3045 kJ (727 kcal) vs 2735 kJ (653 kcal), 95% confidence interval (CI) for the difference, 42–582 kJ (10–139 kcal)). In the high physical activity model, the overall mean-adjusted NREE levels were higher, as expected, than in the low physical activity model, and the difference in adjusted NREE between girls with high activity temperament and girls with low activity temperament was not statistically significant (3188 kJ (762 kcal) vs 3005 kJ (718 kcal), 95% CI for the difference, −109 to 473 kJ (−26 to +113 kcal)) (Figure 2).

Figure 2
figure2

Adjusted NREE in girls with low activity temperament and girls with high activity temperament stratified by physical activity. Models were adjusted for physical activity score and television viewing (see Methods). Error bars represent one standard error. NS=not statistically significant. NREE (kcal) adjusted for body weight.

Activity temperament was negatively associated with percentage body fat (r=−0.20). Those girls with high activity temperament had a lower mean percentage body fat than girls with low activity temperament overall (21.6 vs 24.5%, 95% CI for the difference in means 1.3–4.4 percentage points) and in both the low (21.1 vs 24.7%, 95% CI for the difference in means 1.3–5.9 percentage points) and the high (21.4 vs 23.7%, 95% CI for the difference in means 0.2–4.6 percentage points) physical activity subsamples.

Discussion

We observed that activity temperament was positively associated with NREE cross-sectionally in premenarcheal girls. Our hypothesis that higher activity temperament is associated with higher NREE in girls who have low levels of physical activity was supported. As we conceptualized in Figure 1, high activity temperament could directly increase NREE through increases in movements not generally thought of as physical activity, or high activity temperament could predispose to increased participation in physical activity, thereby leading to greater NREE (indirect pathway). Although within an individual, both pathways would likely be operating, it follows from this conceptual model that the effect of activity temperament on NREE would be most readily detected in children who do not engage in high levels of physical activity.

Physical activity is the most evident component of NREE, but measurement of energy expended in physical activity has proven difficult in population studies.14,15 The nonvolitional components of NREE, such as fidgeting, are not well captured by physical activity questionnaires or activity monitors. Even using instruments designed to measure body movement and position (inclinometers and accelerometers) in free-living participants, Levine et al16 were not able to detect the energy expended fidgeting. In experiments conducted in metabolic chambers, substantial differences in energy expenditure have been demonstrated where postures (sitting, standing) while still and while fidgeting were compared.2 Further, several studies suggest that energy expended in fidgeting or in activities that would not be considered ‘exercise’ may explain some of the variability in weight gain in response to overfeeding.1,17 Even light activities such as crossing and uncrossing one's legs, shuffling small objects across a table, and rising repeatedly from a chair, increase energy expenditure by 40–113% above resting metabolic rate.18 In a carefully controlled experiment of seven nonobese adults, gum chewing raised energy expenditure by 19%.19 Energy expended in fidgeting could be one mechanism through which some people are less prone to weight gain. The results of our analyses indicate that measurement of activity temperament may help to capture aspects of energy expenditure in population studies, such as fidgeting, that are not captured by physical activity questionnaires.

Temperament is a psychological construct that is posited to be inborn and relatively stable.3,20 In a study of overfeeding among identical adult male twins, variability in weight gain at a controlled level of exercise was substantially greater between unrelated participants than within identical twin pairs; this result suggests a genetic component to the factor or factors that account for the differences in weight gain in response to overfeeding.17 In a study of Pima Indian siblings, 57% of the variance in spontaneous physical activity measured in a metabolic chamber was attributable to family membership.21 In another study, NREE estimated from TEE and REE correlated with spontaneous physical activity measured in a metabolic chamber for adult men and women.22 Thus, predilection for movement may be a heritable and reproducible determinant of nonexercise NREE.

Our results confirm our hypothesis that activity temperament accounts for a measurable, albeit small, proportion of NREE in girls with lower levels of reported physical activity. We found that girls in the lower half of the distribution of physical activity expended significantly more energy per day (310 kJ (74 kcal) on average) if they had a high activity temperament compared to girls with a low activity temperament. The effect of activity temperament may be more detectable in girls who engage in low levels of physical activity because, as we conceptualized in Figure 1, the ‘direct pathway’ is more dominant under that scenario. Although girls with higher levels of physical activity who had a high activity temperament expended more energy than girls with a low activity temperament, this difference was smaller and did not reach statistical significance. Some girls who have a high activity temperament may discharge their energy through participation in sports and other forms of exercise, activities that would be recognized as physical activity and would therefore be more likely than nonvolitional activity to be recorded in a physical activity questionnaire. Thus, some girls with a high activity temperament may get more exercise as a ‘secondary’ factor relating to their temperament (‘indirect pathway’). Girls with a low activity temperament who do not regularly engage in physical activity may be most at risk for weight gain (see Figure 2, first bar from the left). Further, girls with a low activity temperament who do engage in physical activity may find maintaining their weight difficult if they lose interest in physical activity (see Figure 2, third bar from the left). For example, girls who play sports in elementary school may reduce or discontinue their participation upon reaching puberty, when social priorities shift. Declining levels of physical activity in girls over the transition from childhood to adolescence have been well documented.23,24 We speculate that girls with a high activity temperament may be better protected from weight gain because their tendency to expend energy is greater. Our cross-sectional observation that girls with a high activity temperament have lower percentage body fat than girls with a low activity temperament provides preliminarily support for this hypothesis.

In the only study of activity temperament and obesity we found in the literature, Carey et al25 found a weak positive association between activity temperament measured between ages 8 and 11 y (one of the nine temperament subscales they measured, but not the focus of their investigation) and relative weight increase in young girls and boys studied at ages 4–5 y and again at 8–9 y. Activity temperament was not related cross-sectionally to relative weight at age 8–9 y or to obesity in a separate sample of 6- to 12-y-old children.25 In their discussion of this counterintuitive result, the authors note that the temperament questionnaires they employed may capture vigor rather than total activity or caloric expenditure.25

The major strength of our cross-sectional analysis is the quality of our measures. The measurement of TEE by doubly labeled water and REE by indirect calorimetry provides an accurate determination of NREE; and percentage body fat estimated by total body water provides an accurate assessment of body composition. The reliability and stability of the activity temperament construct measured in two subsets of girls was quite good: over a 2-week period test–retest reliability was r=0.90, and over a 1-y period stability was r=0.62.

Several limitations of this analysis are noteworthy. Since the girls were not obese, the association between percentage body fat and activity temperament we observed did not include the full range of relative weight in childhood; nonetheless, we had enough variability in percentage body fat (range 10–36%) to detect an effect of activity temperament. In addition, physical activity was assessed by self-report. A previously published report of 2-week test–retest reliability for a subset of 69 girls from this cohort found correlations of r=0.84 and r=0.81 for weekday and weekend day television viewing, respectively, r=0.48 and r=0.32 for weekday and weekend day walking and light activity, and r=0.36 and r=0.32 for report of vigorous activity.12 Television viewing aside, these correlations are not high. However, because we used physical activity level to divide the sample into high and low physical activity and did not focus on the actual amounts of physical activity engaged in by each girl, this concern is reduced. We did not measure the thermic effect of food, but this effect contributes only 6–10% of NREE and varies little between individual children.26 Furthermore, any variability in the thermic effect of food is likely to be unrelated to activity temperament or physical activity level. Lastly, it is important to recognize that although activity temperament contributes to NREE, particularly in girls with a low level of physical activity, much of the variability in NREE was unexplained by our models. Improved methodologies for assessing all components of NREE in population studies are needed; methods for assessment of nonexercise energy expenditure in population studies are particularly underdeveloped.

Activity temperament may represent an important construct in the study of obesity, for two reasons: first, it may be a predictor of weight gain. Second, activity temperament, as a proxy for nonvolitional or spontaneous physical activity, may help to account for energy expenditure that is not captured by physical activity questionnaires and activity diaries. Measurement of activity temperament in population studies, by questionnaire, appears to be valuable for capturing a portion of the component of NREE that is not thought of as exercise, physical activity, recreational activity, or sport. Our observation that girls with a high activity temperament had a leaner body composition irrespective of reported physical activity level suggests a possible protective effect of a constitutionally high activity temperament against weight gain. Longitudinal studies are needed to determine whether activity temperament can help identify girls at risk for weight gain. Although participation in physical activity tends to decline as girls mature, girls with a high activity temperament may be ‘more protected’ from concomitant drops in energy expenditure, either because their inherent predilection for movement will burn calories (fidgeting), or because they will be more likely to maintain participation in sports and physical activities. These investigations would improve our understanding of the etiology of weight gain during the critical period of adolescence.

References

  1. 1

    Levine JA, Eberhardt NL, Jensen MD . Role of nonexercise activity thermogenesis in resistance to fat gain in humans. Science 1999; 283: 212–214.

    CAS  Article  Google Scholar 

  2. 2

    Levine JA, Schleusner SJ, Jensen MD . Energy expenditure of nonexercise activity. Am J Clin Nutr 2000; 72: 1451–1454.

    CAS  Article  Google Scholar 

  3. 3

    Thomas A, Chess S . Temperament and development. Brunner/Mazel Publishers: New York; 1977.

    Google Scholar 

  4. 4

    Bandini LG, Must A, Spadano JL, Dietz WH . Relation of body composition, parental overweight, pubertal stage, and ethnicity to energy expenditure among premenarcheal girls. Am J Clin Nutr 2002; 76: 1040–1047.

    CAS  Article  Google Scholar 

  5. 5

    Must A, Dallal GE, Dietz WH . Reference data for obesity: 85th and 95th percentiles of body mass index (wt/ht2) and triceps skinfold thickness. Am J Clin Nutr 1991; 53: 839–846.

    CAS  Article  Google Scholar 

  6. 6

    Bandini LG, Morelli JA, Must A, Dietz WH . Accuracy of standardized equations for predicting metabolic rate in premenarcheal girls. Am J Clin Nutr 1995; 62: 711–714.

    CAS  Article  Google Scholar 

  7. 7

    Lifson N . Theory of use of the turnover rates of body water for measuring energy and material balance. J Theor Biol 1966; 12: 46–74.

    CAS  Article  Google Scholar 

  8. 8

    Weir JB . New method of calculating metabolic rate with special reference to protein metabolism. J Physiol 1949; 109: 1–9.

    Article  Google Scholar 

  9. 9

    Ogden CL, Kuczmarski RJ, Flegal KM, Mei Z, Guo S, Wei R, Grummer-Strawn LM, Curtin LR, Roche AF, Johnson CL . Centers for disease control and prevention 2000 growth charts for the United States: improvements to the 1977 National Center for Health Statistics version. Pediatrics 2002; 109: 45–60.

    Article  Google Scholar 

  10. 10

    Halliday D, Miller AG . Precise measurement of total body water using trace quantities of deuterium oxide. Biomed Mass Spectrom 1977; 4: 82–87.

    CAS  Article  Google Scholar 

  11. 11

    Schoeller DA, van Santen DW, Peterson DW, Dietz W, Jaspan J, Klein PD . Total body water measurement in humans with 18O and 2H labeled water. Am J Clin Nutr 1980; 33: 2686–2693.

    CAS  Article  Google Scholar 

  12. 12

    Ching PLYH, Dietz WH . Reliability and validity of activity measures in preadolescent girls. Pediatr Exerc Sci 1995; 7: 389–399.

    Article  Google Scholar 

  13. 13

    Ainsworth BE, Haskell WL, Leon AS, Jacobs DR Jr, Montoye HJ, Sallis JF, Paffenbarger, RS . Compendium of physical activities: classification of energy costs of human physical activities. Med Sci Sports Exerc 1993; 25: 71–80.

    CAS  Article  Google Scholar 

  14. 14

    Sallis JF, Saelens B . Assessment of physical activity by self-report: status, limitations, and future directions. Res Q Exerc Sport 2000; 71 (2 Suppl): S1–S14.

    CAS  Article  Google Scholar 

  15. 15

    Sirard JR, Pate RR . Physical activity assessment in children and adolescents. Sports Med 2001; 31: 439–454.

    CAS  Article  Google Scholar 

  16. 16

    Levine J, Melanson EL, Westerterp KR, Hill JO . Measurement of the components of nonexercise activity thermogenesis. Am J Physiol Endocrinol Metab 2001; 281: E670–E675.

    CAS  Article  Google Scholar 

  17. 17

    Bouchard C, Tremblay A, Despres J-P, Nadeau A, Lupien PJ, Theriault G, Dussault J, Moorjani S, Pinault S, Fournier G . The response to long-term overfeeding in identical twins. N Engl J Med 1990; 322: 1477–1482.

    CAS  Article  Google Scholar 

  18. 18

    Schoeller DA, Jefford G . Determinants of the energy costs of light activities: inferences for interpreting doubly labeled water data. Int J Obes Relat Metab Disord 2002; 26: 97–101.

    CAS  Article  Google Scholar 

  19. 19

    Levine J, Baukol P, Pavlidis I . The energy expended in chewing gum. N Engl J Med 1999; 341: 2100.

    CAS  Article  Google Scholar 

  20. 20

    Buss AH, Plomin R . Temperament: early developing personality traits. Lawrence Erlbaum Associates: Hillsdale, NJ; 1984.

  21. 21

    Zurlo F, Ferraro RT, Fontvielle AM, Rising R, Bogardus C, Ravussin E . Spontaneous physical activity and obesity: cross-sectional and longitudinal studies in Pima Indians. Am J Physiol 1992; 263 (2 Part 1): E296–E300.

    CAS  PubMed  Google Scholar 

  22. 22

    Snitker S, Tataranni PA, Ravussin E . Spontaneous physical activity in a respiratory chamber is correlated to habitual physical activity. Int J Obes Relat Metab Disord 2001; 25: 1481–1486.

    CAS  Article  Google Scholar 

  23. 23

    Kimm SYS, Glynn NW, Kriska AM, Barton BA, Kronsberg SR, Daniels SR, Crawford PB, Sabry ZI, Liu K . Decline in physical activity in black girls and white girls during adolescence. N Engl J Med 2002; 347: 709–715.

    Article  Google Scholar 

  24. 24

    Aaron DJ, Storti KL, Robertson RJ, Kriska AM, LaPorte RW . Longitudinal study of the number and choice of leisure time physical activities from mid to late adolescence: implications for school curricula and community recreation programs. Arch Pediatr Adolesc Med 2002; 156: 1075–1080.

    Article  Google Scholar 

  25. 25

    Carey WB, Hegvik RL, McDevitt SC . Temperamental factors associated with rapid weight gain and obesity in middle childhood. J Dev Behav Pediatr 1988; 9: 194–198.

    CAS  Article  Google Scholar 

  26. 26

    Bandini LG, Schoeller DA, Edwards J, Young VR, Oh SH, Dietz WH . Energy expenditure during carbohydrate overfeeding in obese and nonobese adolescents. Am J Physiol 1989; 256 (3 Part 1): E357–E367.

    CAS  PubMed  Google Scholar 

Download references

Acknowledgements

We gratefully acknowledge Pamela Ching, Jennifer Spadano, and the staff at the Clinical Research Center for their assistance with the study, as well as the girls who enrolled for their participation and commitment. This study was supported by NIH Grants DK-50537, M01-RR-00088, 5P30 DK46200, T32-DK62032-11.

Author information

Affiliations

Authors

Corresponding author

Correspondence to S E Anderson.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Anderson, S., Bandini, L., Dietz, W. et al. Relationship between temperament, nonresting energy expenditure, body composition, and physical activity in girls. Int J Obes 28, 300–306 (2004). https://doi.org/10.1038/sj.ijo.0802543

Download citation

Keywords

  • child temperament
  • energy expenditure
  • nonexercise activity thermogenesis
  • weight

Further reading

Search

Quick links