Research ArticleDEVELOPMENTAL BIOLOGY

Constraint and trade-offs regulate energy expenditure during childhood

See allHide authors and affiliations

Science Advances  18 Dec 2019:
Vol. 5, no. 12, eaax1065
DOI: 10.1126/sciadv.aax1065
  • Fig. 1 Predicted, observed, and modeled patterns of children’s energy expenditure in industrial (US/UK) and subsistence-based (Shuar) populations.

    Predicted additive TEE and constrained TEE models of human energy use (A), with observed (B) and modeled (C) components of TEE among Shuar and industrialized children. (A) Children from subsistence-based populations with relatively heavy burdens of infectious disease and active lifestyles are predicted to either increase TEE as a result of increased REE and AEE (additive TEE model) or maintain TEE at industrialized levels as a result of REE and AEE trade-offs (constrained TEE model). (B) Observed energy expenditures in Shuar and U.S./U.K. cohorts follow the constrained TEE model with greater Shuar REE but lower AEE and no overall TEE difference. (C) Modeled energy expenditures accounting for immune activity and growth, as well as possible population differences in exercise efficiency and REE circadian variation, produce TEE estimates for Shuar and industrial populations that vary by only 8% (Materials and Methods). TEE, total energy expenditure; REE, resting energy expenditure; AEE, activity energy expenditure; TEF, thermic effect of food.

  • Fig. 2 Energy expenditure and daily physical activity measures for Shuar children (red) and industrialized cohorts (blue).

    Scatterplot solid lines (shaded 95% confidence intervals) indicate regressions of energetic measures on log–fat-free mass (FFM) adjusting for age, sex, and log–fat mass (FM), with dotted lines denoting population estimated marginal means from final energetic models. Plots of accelerometry activity counts and MVPA display unadjusted population means (95% confidence intervals). No population difference was observed for TEE. However, Shuar children had greater REE, lower AEE and PAL, and greater activity counts and MVPA than industrialized references. MVPA, moderate-vigorous physical activity.

  • Fig. 3 Shuar total immunoglobulin G (IgG) concentration predicts child REE measures.

    Shuar total immunoglobulin G (IgG) concentration versus measured REE (top) and REE elevation above predicted U.S./U.K. values (bottom). Solid lines (shaded 95% confidence intervals) indicate regression of log-IgG adjusting for age, sex, log-FM, log-FFM, and time of REE data collection (log-REE analysis) or time of REE data collection (REE elevation analysis). REE elevation was calculated as the difference between measured REE values and REE predicted from the best-fit model of the U.S./U.K. sample (R2 = 0.588, P < 0.001; adjusting for age, sex, log-FM, and log-FFM).

  • Table 1 Measures of interest for Shuar and industrialized cohorts.

    Population
    Shuar (n = 44)Industrial (n = 40)
    Descriptive, mean (95% CI)
    Sex (% male)50%56%
    Age (years)8.1 (7.5 to 8.7)*7.1 (6.7 to 7.5)
    Anthropometry, adjusted mean (95% CI)
    Stature (cm)117.0 (115.2 to 118.8)***124.6 (122.7 to 126.4)
    Body mass (kg)22.8 (21.6 to 24.0)***26.0 (24.7 to 27.3)
    Body mass index (kg/m2)16.5 (16.1 to 17.0)16.5 (16.1 to 17.0)
    Fat mass (kg)2.7 (2.4 to 3.0)***5.7 (5.0 to 6.4)
    Fat-free mass (kg)19.7 (18.9 to 20.3)19.5 (18.7 to 20.3)
    Body fat percentage (%)11.9% (11.0 to 12.9)***22.4% (20.5 to 24.5)
    Energetics, adjusted mean (95% CI)
    TEE (kcal/day)1738 (1670 to 1809)1811 (1733 to 1892)
    REE (kcal/day)1255 (1215 to 1296)***1042 (1006 to 1080)
    Activity energy
    expenditure
    (kcal/day)
    276 (225 to 338)***558 (447 to 698)
    Physical activity level1.38 (1.31 to 1.46)***1.76 (1.68 to 1.84)
    Shuar (n = 30)Industrial (n = 2855)§
    Daily physical activity, mean (95% CI)
    Activity wear time
    (hours/day)
    12.2 (11.6 to 12.8)***13.5 (13.4 to 13.6)
    Activity counts (CPM)474 (433 to 516)***379 (365 to 392)
    Sedentary activity
    (min/day)
    240 (228 to 252)***478 (473 to 483)
    Light activity (min/day)405 (382 to 428)***273 (268 to 278)
    MVPA (min/day)89 (76 to 102)***58 (55 to 61)||

    †U.S./U.K. cohort (16, 17).

    ‡Values back-converted from analysis of log-transformed measures (table S1); *P < 0.05; ***P < 0.001; population-level differences in two-tailed t tests (daily physical activity measures) or multivariable models controlling for age and sex (anthropometry measures) or age, sex, log-FM, and log-FFM (energetics measures).

    §Canadian cohort (18).

    ||U.K. sample reported 64 min/day of moderate-vigorous activity (17).

    Supplementary Materials

    • Supplementary material for this article is available at http://advances.sciencemag.org/cgi/content/full/5/12/eaax1065/DC1

      Fig. S1. TEE measures for Shuar children (red) and a larger and more diverse sample of industrialized cohorts (blue).

      Fig. S2. Shuar arm muscle area (AMA) measures as percentiles of US age- and sex-matched references (NHANES III).

      Table S1. Parameter estimates [β (SE)] for final energetics GLM models.

      Table S2. Parameter estimates [β (SE)] for energetics GLM models that do not include FM as a predictor.

      Table S3. Household-level lifestyle, economic, and dietary information for the Shuar study sample (n = 18 households).

      Table S4. Parameter estimates [β (SE)] for GLM models evaluating conservative values of Shuar REE that excluded initial (REEi) or single highest (REEh) repeated weekly measures.

      Table S5. Parameter estimates [β (SE)] for GLM models using an alternative hydration constant of 0.75 for Shuar and US cohort FM and FFM calculation.

      Table S6. Measured TEE and FFM using CRDS and duplicate measures (TEEirms; FFMirms) obtained for six participants using isotope ratio mass spectrometry.

      Table S7. Measured REE and TEE for US/UK children and predicted values calculated by common prediction equations (that were developed using predominantly industrialized samples).

      Data file S1. Primary study data with variable list.

      Data file S2. Daily physical activity summary data for the Canadian cohort.

      Data file S3. Expanded industrialized sample data.

      References (4450)

    • Supplementary Materials

      The PDFset includes:

      • Fig. S1. TEE measures for Shuar children (red) and a larger and more diverse sample of industrialized cohorts (blue).
      • Fig. S2. Shuar arm muscle area (AMA) measures as percentiles of US age- and sex-matched references (NHANES III).
      • Table S1. Parameter estimates β (SE) for final energetics GLM models.
      • Table S2. Parameter estimates β (SE) for energetics GLM models that do not include FM as a predictor.
      • Table S3. Household-level lifestyle, economic, and dietary information for the Shuar study sample (n = 18 households).
      • Table S4. Parameter estimates β (SE) for GLM models evaluating conservative values of Shuar REE that excluded initial (REEi) or single highest (REEh) repeated weekly measures.
      • Table S5. Parameter estimates β (SE) for GLM models using an alternative hydration constant of 0.75 for Shuar and US cohort FM and FFM calculation.
      • Table S6. Measured TEE and FFM using CRDS and duplicate measures (TEEirms; FFMirms) obtained for six participants using isotope ratio mass spectrometry.
      • Table S7. Measured REE and TEE for US/UK children and predicted values calculated by common prediction equations (that were developed using predominantly industrialized samples).
      • Legends for data files S1 to S3
      • References (4450)

      Download PDF

      Other Supplementary Material for this manuscript includes the following:

      • Data file S1 (Microsoft Excel format). Primary study data with variable list.
      • Data file S2 (Microsoft Excel format). Daily physical activity summary data for the Canadian cohort.
      • Data file S3 (Microsoft Excel format). Expanded industrialized sample data.

      Files in this Data Supplement:

    Stay Connected to Science Advances

    Navigate This Article