TABLE 1.
Summary of main characteristics of 33 studies reporting on associations between hair, salivary, serum, and/or urinary cortisol concentration with adiposity-related outcomes in children.
| First author, Publication year; Country and development context | Study design | Sample size (% Girls) | Age (years, Mean ± SD, Range) | Race/Ethnicity | Cortisol measure | Adiposity outcomes | Effect size/Associations between cortisol and adiposity/Covariates | ||
|
|
|
||||||||
| Measure source | Measurement method | Measures | Method of ascertainment | ||||||
| Hair cortisol | |||||||||
| 1 1 Vehmeijer et al. (55); Netherlands (Developed) | Cohort | 2,042 (52.5%) | 5.90 (5.70–8.00) | European and non-European | Hair, 3 cm | LC-MS/MS | BMI-SDS | BMI was calculated based on measured weight and height. BMI-SDS was generated based on Dutch reference growth charts | Increase of BMI SDS (β = 0.06, 95% CI: 0.02, 0.09) per quintile of hair cortisol Covariates included: (1) Child’s: Sex and age, maternal pre-pregnancy BMI, psychological distress during pregnancy (2) Maternal: Educational level and marital status at 6 years, child’s ethnicity, hair color and television watching time |
| 12 Vehmeijer et al. (55); Netherlands (Developed) | Cohort | 2,042 (52.5%) | 5.90 (5.70–8.00) | European and non-European | Hair, 3 cm | LC-MS/MS | FMI-SDS | FMI was measured by DXA | Increase of FMI-SDS (β = 0.05, 95% CI: 0.02, 0.08) per quintile of hair cortisol Covariates included: (1) Child’s: Sex and age, maternal pre-pregnancy BMI, psychological distress during pregnancy (2) Maternal: Educational level and marital status at 6 years, child’s ethnicity, hair color and television watching time |
| 13 Vehmeijer et al. (55); Netherlands (Developed) | Cohort | 2,042 (52.5%) | 5.90 (5.70–8.00) | European and non-European | Hair, 3 cm | LC-MS/MS | Overweight vs. Non-overweight | BMI was calculated by measuring weight and height. Weight status was defined based on the International Obesity Task Force cut-offs, the age- and sex- specific cut-off points | Increased risk for overweight or obesity of (OR = 1.18, 95% CI: 1.07, 1.29) per quintile of hair cortisol Covariates included: (1) Child’s: Sex and age, maternal pre-pregnancy BMI, psychological distress during pregnancy (2) Maternal: Educational level and marital status at 6 years, child’s ethnicity, hair color and television watching time |
| 21 Bethancourt et al. (56); Bolivia (Developing) | Cross-sectional | 167 (53.2%) | 9.70 (6.00–15.00) | Not reported | Hair, 1.5 cm | ELISA | BMI z-score | BMI was calculated based on measured weight and height. BMI z-score was generated based on the WHO reference values and macros | Increase of −0.02 BMI z-score (SE = 0.02, p = 0.26) per 20% increase of hair cortisol Covariates included: Maternal: Community and household of residence, age, household adult equivalents, household income, self-reported perceived social status |
| 22 Bethancourt et al. (56); Bolivia (Developing) | Cross-sectional | 167 (53.2%) | 9.70 (6.00–15.00) | Not reported | Hair, 1.5 cm | ELISA | Body fat percentage | Body fat percentage was measured using a Tanita BF-680W bioelectric impedance scale | Increase of −0.29 percentage of body fat (SE = 0.12, p = 0.01) per 20% increase of hair cortisol Covariates included: Maternal: Community and household of residence, age, household adult equivalents, household income, self-reported perceived social status |
| #31 Petimar et al. (57); United States (Developed) | Cohort | 491 (NR) | 7.80–13.10 | White | Hair, ≥3 cm | LC-MS/MS | BMI z-score | BMI was calculated based on measured weight and height. BMI z-score was generated based on US CDC growth charts (2000) | Associations between log hair cortisol concentration (HCC) and BMI z-score: β = 0.00, 95% CI: −0.08, 0.07 Covariates included: Child’s: Age, sex, birthweight-for-sex-and-gestational age z score, second-hand smoke exposure, mid-childhood pubertal development score, and early childhood BMI z-score Maternal: enrollment age, maternal education, pre-pregnancy BMI, maternal smoking during pregnancy, household income |
| #32 Petimar et al. (57); United States (Developed) | Cohort | 493 (NR) | 7.80–13.10 | White | Hair, ≥3 cm | LC-MS/MS | Waist circumference | Waist circumference was measured | Associations between log HCC and WC: β = −0.04, 95% CI: −0.83, 0.74 Covariates included: Child’s: Age, sex, birthweight-for-sex-and-gestational age z score, second-hand smoke exposure, mid-childhood pubertal development score, and early childhood BMI z-score Maternal: enrollment age, maternal education, pre-pregnancy BMI, maternal smoking during pregnancy, household income |
| #33 Petimar et al. (57); United States (Developed) | Cohort | 491 (NR) | 7.80–13.10 | White | Hair, ≥3 cm | LC-MS/MS | Waist to height ratio (WtHR) | WtHR was calculated based on measured waist and height | Associations between log HCC and WtHR: β = 0.002, 95% CI: −0.003, 0.007 Covariates included: Maternal: enrollment age, education, pre-pregnancy BMI, maternal smoking during pregnancy, household income Child’s: Age, sex, birth weight-for-sex-and-gestational age z score, second-hand smoke exposure, mid-childhood pubertal development score, and early childhood waist to height ratio |
| 4 Bryson et al. (58); Australia (Developed) | Cross-sectional | 297 (60.6%) | 3.10 ± 0.10 | Not reported | Hair, 3 cm | ELISA | BMI z-score | BMI was calculated based on measured weight and height. BMI z-score was generated based on Cole’s international criteria (Cole, Bellizzi, 2000) | Associations between HCC and BMI z-score: β = 0.76, 95% CI: 0.51, 1.12, p = 0.16 Covariates included: Age, gender, season of assessment, site of hair collection, randomized controlled trial randomization status |
| 5 Baan et al. (59); Netherlands (Developed) | Cross-sectional | 298 (45.64%) | 12.60–13.20 | Not reported | Hair, 3 cm | LC-MS/MS | BMI z-score | BMI was calculated based on measured weight and height. BMI z-score was generated based on the 1997 Dutch nationwide growth study | Associations between log HCC and BMI z-score: β = 0.13, 95% CI: 0.04, 0.22, p = 0.01 No covariates were reported |
| 61 Smith et al. (36); Australia (Developed) | Cross-sectional | 128 (68.0%) | 8.44 ± 0.34 | Not reported | Hair, 3 cm | ELISA | Waist circumference | Waist circumference was measured | Correlation between log HCC and WC: r = 0.015 No covariates were reported |
| 62 Smith et al. (36); Australia (Developed) | Cross-sectional | 128 (68.0%) | 8.44 ± 0.34 | Not reported | Hair, 3 cm | ELISA | BMI | BMI was calculated based on measured weight and height | Correlation between log HCC and BMI: r = −0.047 No covariates were reported |
| 7 Evans et al. (19); Sweden (Developed) | Cross-sectional | 92 (NR) | 10 | Dutch | Hair, 3 cm | LC-MS/MS | BMI | BMI was calculated based on measured weight and height | Correlation between HCC and BMI: r = 0.01 No covariates were reported |
| #8 Distel et al. (18); United States (Developed) | Cohort | 52 (61%) | 6–10 | Mexican | Hair, NR | ELISA | BMI | BMI was calculated based on measured weight and height | Associations between HCC and BMI: β = 4.62, 95% CI: 1.41, 7.83, p < 0.01 Covariates included: Age and food insecurity |
| 9 Sun et al. (60); China (Developing) | Cross-sectional | 1,000 (57.9%) | 8.97 ± 0.86 | Han Chinese | Hair, ≥10 mg | ELISA | BMI | BMI was calculated based on measured weight and height | Associations between HCC and BMI: β = 0.17, 95% CI: 0.05, 0.29 Covariates included: PRS polygenic risk score |
| 10 Lu et al. (34); China (Developing) | Cross-sectional | 85 (45.9%) | 11.40 ± 0.30 | Chinese, not otherwise specified | Hair, 3 cm | LC-MS/MS | BMI z-score | BMI was calculated based on measured weight and height. BMI z-score was defined as the number of standard deviation units from the mean or reference value | Associations between log HCC and BMI: β = 0.193, 95% CI: 0.19, 0.20, p = 0.004 No covariates were reported |
| 11 Papafotiou et al. (35); Greece (Developed) | Cross-sectional | 50 (100%) | 7.60 ± 1.30 | Greek | Hair, 3 cm | LC-MS/MS | BMI z-score | BMI was calculated based on measured height and weight. BMI z-score was generated based on Cole’ s international criteria (Cole, Bellizzi, 2000) | Correlations between HCC and BMI z-score: r = 0.327, p = 0.025 No covariates were reported |
| 121 Gerber et al. (32); Switzerland (Developed) | Cross-sectional | 318 (53.1%) | 7.26 ± 3.51 | Not reported | Hair, 3 cm | CLIA | BMI | BMI was calculated based on measured weight and height | Correlations between HCC and BMI: r = 0.16, p < 0.01, r = 0.13 for boys, r = 0.16 for girls (p < 0.05) No covariates were reported |
| 122 Gerber et al. (32); Switzerland (Developed) | Cross-sectional | 318 (53.1%) | 7.26 ± 3.51 | Not reported | Hair, 3 cm | CLIA | Body fat percentage | Percentage body fat was calculated based on measured skinfold | Correlations between HCC and PBF: r = 0.14, p < 0.01, r = 0.12 for boys, r = 0.16 for girls (p < 0.05) No covariates were reported |
| 123 Gerber et al. (32); Switzerland (Developed) | Cross-sectional | 318 (53.1%) | 7.26 ± 3.51 | Not reported | Hair, 3 cm | CLIA | Waist circumference | Waist circumference was measured | Correlations between HCC and WC: r = 0.14, p < 0.01, r = 0.18, p < 0.05 for boys, r = 0.11 for girls No covariates were reported |
| 13 Rippe et al. (25); Netherlands (Developed) | Cross-sectional | 2,484 (51.7%) | 6.20 ± 0.70 | Danish-Caucasian Western and other European | Hair, 3 cm | LC–MS/MS | BMI | BMI was calculated based on measured weight and height | Associations between log HCC and BMI: (95% CI) = 0.025 (0.02, 0.03; p = 0.001) No covariates were reported |
| 14 Olstad et al. (61); Australia (Developed) | Cross-sectional | 30 (43.3%) | 14.30 ± 3.90 | Not reported | Hair, 3 cm | ELISA | BMI z-score | BMI was calculated based on measured weight and height. BMI z-score was generated based on the CDC growth charts of U.S (2000) | Associations between HCC and BMI z-score: β = 0.20, 95% CI: −0.85, 1.25, p = 0.694 Covariates included: (1) Childs’s: Age (2) Maternal: BMI, education |
| 151 Noppe et al. (20); Netherlands (Developed) | Cross-sectional | 2,953 (51.9%) | 6.20 ± 0.60 | European and non-European, not otherwise specified | Hair, 3 cm | LC–MS/MS | BMI | BMI was calculated based on measured weight and height | Associations between HCC and BMI: β = 0.19, 95% CI: 0.12, 0.26 Covariates included: Child’s: Age, sex, ethnicity, and topical glucocorticoid use |
| 152 Noppe et al. (20); Netherlands (Developed) | Cross-sectional | 2,953 (51.9%) | 6.20 ± 0.60 | European and non-European, not otherwise specified | Hair, 3 cm | LC–MS/MS | FMI-SDS | Fat mass index was measured by DXA | Associations between HCC and FMI: β = 0.05, 95% CI: 0.01, 0.09 Covariates included: Child’s: Age, sex, ethnicity, and topical glucocorticoid use |
| 16 Murray et al. (62); Australia (Developed) | Cross-sectional | 95 (52.6%) | 9.50 ± 0.34 | Not reported | Hair, 3 cm | ELISA | BMI | BMI was calculated based on measured weight and height | Correlations between HCC and BMI: r = −0.26 No covariates were reported |
| 171 Larsen et al. (33); Denmark (Developed) | Cross-sectional | 317 (NR) | 5 (4–7) | Danish, not otherwise specified | Hair, 1–2 cm | ELISA | BMI z-score | BMI was calculated based on measured weight and height. BMI z-score was generated using the Lambda-Mu-Sigma method | Associations between HCC and BMI z-score: β = 0.01, 95% CI: −0.04, 0.07, p = 0.70 Covariates included: Child’s: Intervention status, gender, physical activity, maternal education level, and age |
| 172 Larsen et al. (33); Denmark (Developed) | Cross-sectional | 280 (NR) | 5 (4–7) | Danish, not otherwise specified | Hair, 1–2 cm | ELISA | FMI z-score | FMI was measured by BIA-method, and calculated based on an equation described by Goran et al. (1996) in young Children | Associations between HCC and FMI z-score: β = 0.03, 95% CI: −0.03, 0.08, p = 0.32 Covariates included: Child’s: Intervention status, gender, physical activity, maternal education, and age |
| 173$ Larsen et al. (33); Denmark (Developed) | Cross-sectional | 280 (NR) | 5 (4–7) | Danish, not otherwise specified | Hair, 1–2 cm | ELISA | FFMI z-score | FFMI was calculated by subtracting FFM from body weight. FMI was measured by BIA-method | Associations between HCC and FMI z-score: β = −0.01, 95% CI: −0.07, 0.05, p = 0.69 Covariates included: Child’s: Intervention status, gender, physical activity, maternal education, and age |
| 174$ Larsen et al. (33); Denmark (Developed) | Cross-sectional | 309 (NR) | 5 (4–7) | Danish, not otherwise specified | Hair, 1–2 cm | ELISA | Waist circumference | BMI was calculated based on measured weight and height. BMI z-score was generated using the Lambda-Mu-Sigma method | Associations between HCC and WC: β = 0.10, 95% CI: −0.09, 0.30, p = 0.30 Covariates included: Child’s: Intervention status, gender, physical activity, maternal education, and age |
| 175$ Larsen et al. (33); Denmark (Developed) | Cross-sectional | 308 (NR) | 5 (4–7) | Danish, not otherwise specified | Hair, 1–2 cm | ELISA | WtHR | WtHR was calculated based on measured waist circumference and height | Associations between HCC and WtHR: β = −0.001, 95% CI: −0.003, 0.002, p = 0.52 Covariates included: Child’s: Intervention status, gender, physical activity, maternal education, and age |
| 181 Veldhorst et al. (37); Netherlands (Developed) | Cross-sectional | 40 (75%) | 8–12 | Caucasian, no-Caucasian | Hair, 1 cm | ELISA | BMI-SDS | BMI was calculated based on measured weight and height. BMI SDS was generated based on the 2010 Dutch nationwide growth study | Correlations between log HCC and BMI-SDS: r = 0.407, p < 0.01 No covariates were reported |
| 182 Veldhorst et al. (37); Netherlands (Developed) | Cross-sectional | 40 (75%) | 8–12 | Caucasian, no-Caucasian | Hair, 1 cm | ELISA | Waist circumference | Waist circumference was measured | Correlations between log HCC and WC: r = 0.43, p < 0.01 No covariates were reported |
| 191 Noppe et al. (63); Netherlands (Developed) | Cross-sectional | 128 (50.8%) | 8.40 (4.25–14.13) | Not reported | Hair, 3 cm | ELISA | Waist circumference | Waist circumference was measured | Correlations between log HCC and WC: r = 0.19, p = 0.04 No covariates were reported |
| 192 Noppe et al. (63); Netherlands (Developed) | Cross-sectional | 128 (50.8%) | 8.40 (4.25–14.13) | Not reported | Hair, 3 cm | ELISA | WtHR | WtHR was calculated based on measured waist circumference and height | Correlations between log HCC and WtHR: r = 0.19, p = 0.04 No covariates were reported |
| 193* Noppe et al. (63); Netherlands (Developed) | Cross-sectional | 128 (50.8%) | 8.40 (4.25–14.13) | Caucasian | Hair, 3 cm | ELISA | BMI | BMI was calculated based on measured weight and height | NR |
| Salivary cortisol | |||||||||
| 11 Pruszkowska-Przybylska et al. (64); Poland (Developing) | Cross-sectional | 73 (100%) | 8.92 (7–11) | Not reported | Saliva (8 a.m.–2 p.m.) | ELISA | Body fat percentage | Body fat percentage was measured using the BIA-method | Association between salivary cortisol and FM% was β = −0.089, SE = 0.12, p = 0.462 Covariates included: Child’s: Vitamin D concentration Maternal: Education, 2D:4D digit ratio, socio-economic status |
| 12 Pruszkowska-Przybylska et al. (64); Poland (Developing) | Cross-sectional | 73 (100%) | 8.92 (7–11) | Not reported | Saliva (8 a.m.–2 p.m.) | ELISA | BMI z-score | BMI was calculated based on measured weight and height. The calculated method of BMI z-score was not reported | Association between salivary cortisol and BMI z-score was β = −0.027, SE = 0.117, p = 0.818 Covariates included: Child’s: Vitamin D concentration Maternal: Education, 2D:4D digit ratio, socio-economic status |
| 13 Pruszkowska-Przybylska et al. (64); Poland (Developing) | Cross-sectional | 60 (0%) | 8.92 (7–11) | Not reported | Saliva (8 a.m.–2 p.m.) | ELISA | Body fat percentage | Body fat percentage was measured using the BIA-method | Association between salivary cortisol and FM% was β = −0.091, SE = 0.137, p = 0.511 Covariates included: Child’s: Vitamin D concentration Maternal: Education, 2D:4D digit ratio, socio-economic status |
| 14 Pruszkowska-Przybylska et al. (64); Poland (Developing) | Cross-sectional | 60 (0%) | 8.92 (7–11) | Not reported | Saliva (8 a.m.–2 p.m.) | ELISA | BMI z-score | BMI was calculated based on measured weight and height. The calculated method of BMI z-score was not reported | Association between salivary cortisol and BMI z-score was β = −0.148, SE = 0.134, p = 0.273 Covariates included: Child’s: Vitamin Concentration Maternal: Education, 2D:4D digit ratio, socio-economic status |
| 21 Dai et al. (65); United States (Developed) | Cross-sectional | 689 (53.0%) | 9.20 (SD = 0.41) | Caucasian, not otherwise specified | Saliva (waking, 30 mins post-waking) | ELISA | Body composition | Body composition was indexed by BMI and waist-to-hip ratio. BMI was calculated based on measured weight and height | Salivary cortisol was associated with body composition: β = −0.20, SE = 0.05, p < 0.01 Covariates included: Child’s: Sex, age, race, socioeconomic, and medication use |
| 22 Dai et al. (65); United States (Developed) | Longitudinal | 647 (55.0%) | 10.53 (SD = 0.52) | Caucasian, not otherwise specified | Saliva (waking, 30 mins post-waking) | ELISA | Body composition | Body composition was indexed by BMI and waist-to-hip ratio. BMI was calculated based on measured weight and height | Salivary cortisol at baseline was associated with body composition at follow-up: β = 0.00, SE = 0.02, p > 0.05 Covariates included: Child’s: Sex, age, race, socioeconomic, and medication use |
| 3 Pruszkowska-Przybylska et al. (66); Poland (Developing) | Cross-sectional | 132 (56.8%) | 6–13 | Not reported | Saliva (8 a.m.–2 p.m.) | ELISA | Fat mass percentage | Fat mass was measured by BIA-method | Salivary cortisol was associated with fat mass percentage β = −0.17, SE = 0.076, p = 0.026 No covariates were reported |
| 4X Marceau et al. (52); United States (Developed) | Cohort | 361 (43%) | 4.50–9 | White, African, American, Hispanic, Latino, Multiethnic, other | Saliva (morning) | DELFIA | BMI | BMI was calculated based on measured weight and height | Associations between morning salivary cortisol and BMI: β = −1.34, 95% CI: −2.28, −0.4, p < 0.05 No covariates were reported |
| 4z Marceau et al. (52); United States (Developed) | Cohort | 361 (43%) | 4.50–9 | White, African, American, Hispanic, Latino, Multiethnic, other | Saliva (evening) | DELFIA | BMI | BMI was calculated based on measured weight and height | Associations between evening salivary cortisol and BMI: β = −0.52, 95% CI: −3.52, 2.48 No covariates were reported |
| 51*Lynch et al. (51); United States (Developed) | Cross-sectional | 147 (57.1%) | 10–12 | African American, Asian, Caucasian, Hispanic | Saliva (mid-morning) (9:30 A.M.–11:00 A.M.) |
ELISA | Waist Circumference | Waist circumference was measured | NR |
| 52* Lynch et al. (51); United States (Developed) | Cross-sectional | 147 (57.1%) | 10–12 | African American, Asian, Caucasian, Hispanic | Saliva (mid-morning) (9:30 A.M.–11:00 A.M.) | ELISA | BMI | BMI was calculated based on measured weight and height | NR |
| 6 Lu et al. (34); China (Developing) | Cross-sectional | 85 (45.9%) | 11.4 0 ± 0.30 | Chinese, not otherwise specified | Saliva cortisol (lnAUCi, TSST-C) | LC-MS/MS | BMI z-score | BMI was calculated based on measured weight and height. BMI z-score is defined as the number of standard deviation units from the mean or reference value | Associations between salivary cortisol lnAUCi and BMI: β = 0.051, 95% CI: −1.74, 1.84 No covariates were reported |
| 71 Papafotiou et al. (35); Greece (Developed) | Cross-sectional | 50 (100%) | 7.60 ± 1.30 | Greek | Saliva (AUCg) | ECLIA | BMI z-score | BMI was calculated based on measured height and weight. BMI z-score was generated based on Cole’ s international criteria (Cole, Bellizzi, 2000) | Correlations between salivary cortisol (AUCg) and BMI z-score: r = 0.352, p = 0.016 No covariates were reported |
| 72x Papafotiou et al. (35); Greece (Developed) | Cross-sectional | 50 (100%) | 7.60 ± 1.30 | Greek | Saliva (morning) | ECLIA | BMI z-score | BMI was calculated based on measured height and weight. BMI z-score was generated based on Cole’ s international criteria (Cole, Bellizzi, 2000) | Correlations between morning salivary cortisol and BMI-Z score: r = 0.321, p = 0.031 No covariates were reported |
| 73z Papafotiou et al. (35); Greece (Developed) | Cross-sectional | 50 (100%) | 7.60 ± 1.30 | Greek | Saliva (evening) | ECLIA | BMI z-score | BMI was calculated based on measured height and weight. BMI z-score was generated based on Cole’ s international criteria (Cole, Bellizzi, 2000) | Correlations between salivary cortisol and BMI z-score: r = 0.413, p = 0.006 No covariates were reported |
| 8 Chu et al. (49); China (Developing) | Cross-sectional | 110 (50.9%) | 4–5 | Chinese, not otherwise specified | Saliva (morning) | LC-MS/MS | BMI | BMI was calculated based on measured weight and height | Correlations between salivary cortisol and BMI: r = 0.001, p = 0.426 No covariates were reported |
| 91 Lu et al. (45); China (Developing) | Cross-sectional | 87 (44.8%) | 12–13 | Chinese, not otherwise specified | Saliva (AUCi, after TSST-C) | ELISA | BMI | BMI was based on measured weight and height | Correlations between salivary cortisol (AUCi) and BMI: r = 0.15 No covariates were reported |
| 92$ Lu et al. (45); China (Developing) | Cross-sectional | 87 (44.8%) | 12–13 | Chinese, not otherwise specified | Saliva (AUCi, after TSST-C) | ELISA | Body fat percentage | Percentage body fat was measured by BIA-method | Correlations between salivary cortisol (AUCi) and PBF: r = 0.15 No covariates were reported |
| 10x Ruttle et al. (23); United States (Developed) | Cohort | 323 (NR) | 11–18 | Largely Caucasian | Saliva (morning) | ELISA | BMI | BMI was calculated based on measured weight and height | Associations between morning salivary cortisol and BMI: β = −0.17, 95% CI: −0.29, −0.05, p < 0.01 Covariates included: Child sex and pubertal status, maternal BMI |
| 10y Ruttle et al. (23); United States (Developed) | Cohort | 323 (NR) | 11–18 | Largely Caucasian | Saliva (afternoon) | ELISA | BMI | BMI was calculated based on measured weight and height | Associations between afternoon salivary cortisol and BMI: β = −0.15, 95% CI: −0.27, −0.03, p < 0.01 Covariates included: Child sex and pubertal status, maternal BMI |
| 10z Ruttle et al. (23); United States (Developed) | Cohort | 323 (NR) | 11–18 | Largely Caucasian | Saliva (evening) | ELISA | BMI | BMI was calculated based on measured weight and height | Associations between evening salivary cortisol and BMI: β = −0.12, 95% CI: −0.24, −0.002, p < 0.01 Covariates included: Child’s sex and pubertal status, maternal BMI |
| 111$ Miller et al. (46); United States (Developed) | Cross-sectional | 218 (50.9%) | 4.40 ± 0.58 | White, Black, Biracial, Hispanic/Latino | Saliva (AUC, Stress-elicitation challenge tasks) | ELISA | BMI z-score | BMI was calculated based on measured weight and height. BMI z-score was calculated based on US Centers for Disease Control reference growth curves for age and sex | Associations between salivary cortisol (AUC) and BMI z-score: β = −0.17, 95% CI: −0.31, −0.03, p = 0.018 Covariates included: Child’s: Age, sex, ethnicity Maternal: Overweight and family income-to-needs ratio |
| 112$ Miller et al. (62); United States (Developed) | Cohort | 115 (NR) | 4.40 ± 0.58 | White, Black, Biracial, Hispanic/Latino | Saliva (AUC, Stress-elicitation challenge tasks) | ELISA | Change of BMI z-score | BMI was calculated based on measured weight and height. BMI z-score was calculated based on US Centers for Disease Control reference growth curves for age and sex | Associations between salivary cortisol (AUC) and change of BMI-Z score: β = 0.002, 95% CI: −0.004, 0.008, p = 0.410 Covariates included: Child’s: Age, sex, ethnicity Maternal: Overweight and family income-to-needs ratio |
| 12c Francis et al. (31); United States (Developed) | Cross-sectional | 32 (NR) | 5–7 | White, Black, others | Saliva (AUCi, TSST-C) | ELISA | BMI z-score | BMI was extracted from medical record | Associations between salivary cortisol (AUCi) and BMI z-score: β = 0.07, 95% CI: −0.32, 0.46 Covariates included: Child’s: Eating in the absence of hunger Parental: Combined education |
| 12d Francis et al. (31); United States (Developed) | Cross-sectional | 11 (NR) | 8–9 | White, Black, others | Saliva (AUCi, TSST-C) | ELISA | BMI z-score | BMI was extracted from medical record | Associations between salivary cortisol (AUCi) and BMI z-score: β = 1.38, 95% CI: 0.46, 2.30, p < 0.01 Covariates included: Child’s: Eating in the absence of hunger Parental: Combined education |
| 13a$ Hill et al. (50); United States (Developed) | Cohort | 153 (0%) | 9.60 ± 0.90 | Caucasian, African American | Saliva (morning) | ELISA | Change in BMI z-score | BMI was calculated based on measured weight and height | Correlations between morning salivary cortisol and change in BMI z-score: r = 0.15, p = 0.009 No covariates were reported |
| 13b$ Hill et al. (50); United States (Developed) | Cohort | 163 (100%) | 9.60 ± 0.90 | Caucasian, African American | Saliva (morning) | ELISA | Change in BMI z-score | BMI was calculated based on measured weight and height | Correlations between morning salivary cortisol and change in BMI z-score: r = –0.015 |
| 14a Dockray et al. (30); United States (Developed) | Cross-sectional | 56 (0%) | 11.44 | Hispanic, Hispanic, African American, Asian American | Saliva cortisol (lnAUCi, TSST-C) | ELISA | BMI | BMI was calculated based on measured weight and height | Correlations between saliva cortisol lnAUCi and BMI: r = 0.29, p < 0.05 No covariates were reported |
| 14b Dockray et al. (30); United States (Developed) | Cross-sectional | 55 (100%) | 10.49 | Hispanic, Hispanic, African American, Asian American | Saliva cortisol (logAUCi, TSST-C) | ELISA | BMI | BMI was calculated based on measured weight and height | Correlations between saliva cortisol logAUCi and BMI: r = 0.52, p < 0.01 No covariates were reported |
| 15 Barat et al. (48); France (Developed) | Cross-sectional | 19 (63.2%) | 6–13 | Not reported | Saliva (morning) | RIA | Truncal distribution of fat mass (TDFM) | TDFM was assessed with dual energy X-ray absorptiometry | Correlations between morning salivary cortisol and TDFM: r = 0.38 for total children, r = −0.33 for boys, r = 0.53 for girls No covariates were reported |
| 16 Rosmalen et al. (47); Netherlands (Developed) | Cross-sectional | 894 (100%) | 10–12 | Not reported | Saliva (AUCi, normal condition) | TRFIA | BMI | BMI was calculated based on measured weight and height | Correlation between salivary cortisol (AUCi) and BMI: r = 0.072, p = 0.042 No covariates were reported |
| Serum cortisol | |||||||||
| 11 Gallagher et al. (67); Greek (Developed) | Cross-sectional | 2,665 (49.5%) | 9–13 | Not reported | Serum | ELISA | Visceral fat | Visceral fat was measured by BIA method | Serum cortisol was associated with visceral fat: β = −0.04, 95% CI: −0.1, −0.07, p = 0.01 Covariates included: Child’s: Sex, tanner stage, total daily energy intake and total steps per day Maternal: Education |
| 12 Gallagher et al. (67); Greek (Developed) | Cross-sectional | 2,665 (49.5%) | 9–13 | Not reported | Serum | ELISA | BMI | BMI was calculated based on measured weight and height | Serum cortisol was associated with visceral fat: β = −0.03, 95% CI: −0.1, 0.0, p = 0.06 Covariates included: Child’s: Sex, tanner stage, total daily energy intake and total steps per day Maternal: Education |
| 13 Gallagher et al. (67); Greek (Developed) | Cross-sectional | 2,665 (49.5%) | 9–13 | Not reported | Serum | ELISA | BMI z-score | BMI z-score was calculated based on WHO 2007 growth reference for age | Serum cortisol was associated with BMI z-score: β = −0.01, 95% CI: −0.0, 0.0, p = 0.11) Covariates included: Child’s: Sex, tanner stage, total daily energy intake and total steps per day Maternal: Education |
| 2* Koester-Weber et al. (68); Multi-Centre in Europe (Developed) | Cross-sectional | 927 (55%) |
14.90 ± 1.20 | Not reported | Serum | ELISA | Overweight vs. Non-overweight | BMI was calculated by measuring weight and height. Overweight: BMI > 25 Kg/m2, Obesity: BMI > 30 kg/m2 | NR |
| 31$ Hillman et al. (53); United States (Developed) | Cross-sectional | 218 (100%) | 14.90 ± 2.20 | White, Black, Other | Serum (afternoon and AUCi) | RIA | BMI z-score | BMI was calculated based on measured weight and height. BMI z-score was generated based on the CDC growth charts of U.S (2000) | Associations between serum cortisol (AUCi) and BMI z-score: β = −0.02, 95% CI: −0.04, −0.003, p = 0.02 No covariates were reported |
| 32$ Hillman et al. (53); United States (Developed) | Cross-sectional | 218 (100%) | 14.90 ± 2.20 | White, Black, Other | Serum (afternoon and AUCi) | RIA | BMI | BMI was calculated based on measured weight and height | Associations between serum cortisol (AUCi) and BMI: β = −0.06, 95% CI: −0.156, 0.03, p = 0.20 Covariates included: Child’s: Age, race, Tanner stage, and socio-economic status |
| 33$ Hillman et al. (53); United States (Developed) | Cross-sectional | 218 (100%) | 14.90 ± 2.20 | White, Black, Other | Serum (afternoon and AUCi) | RIA | Body fat percentage | Percentage body fat was measured by DXA | Associations between serum cortisol (AUCi) and PBF: β = −0.05, 95% CI: −0.17, 0.08, p = 0.49 Covariates included: Child’s: Age, race, and socio-economic status |
| 41 Adam et al. (69); United States (Developed) | Cross-sectional | 211 (43.6%) | 10.80–11.10 | Latino | Serum | RIA | BMI | BMI was calculated based on measured weight and height | Correlations between serum cortisol and BMI: r = 0.06 No covariates were reported |
| 42 Adam et al. (69); United States (Developed) | Cross-sectional | 211 (43.6%) | 10.80–11.10 | Latino | Serum | RIA | Waist circumference | Waist circumference was measured | Correlations between serum cortisol and WC: r = −0.03 No covariates were reported |
| 5 Weigensberg et al. (70); United States (Developed) | Cross-sectional | 205 (42.4%) | 11.10 ± 1.70 | Latino | Serum | RIA | Waist circumference | Waist circumference was measured | Correlations between serum cortisol and WC: r = 0.02 No covariates were reported |
| 6 Barat et al. (48); France (Developed) | Cross-sectional | 39 (43.6%) | 6–13 | Not reported | Serum | RIA | Truncal distribution of fat mass (TDFM) | TDFM was assessed with dual energy X-ray absorptiometry | Correlations between morning salivary cortisol and TDFM: r = 0.17 for total children, r = 0.33 for boys, r = 0.40 for girls No covariates were reported |
| Urine cortisol | |||||||||
| 11$ Hillman et al. (53); United States (Developed) | Cross-sectional | 218 (100%) | 14.90 ± 2.20 | White, Black, Other | Urine free cortisol (afternoon) | RIA | BMI | BMI was calculated based on measured weight and height. BMI z-score was generated based on the CDC growth charts of U.S (2000) | Associations between urine cortisol and BMI: β = 3.54, 95% CI: 1.12, 5.97, p = 0.005 Covariates included: Child’s: Age, race, Tanner stage, and socio-economic status |
| 12$ Hillman et al. (53); United States (Developed) | Cross-sectional | 218 (100%) | 14.90 ± 2.20 | White, Black, Other | Urine free cortisol (afternoon) | RIA | BMI z-score | BMI was calculated based on measured weight and height | Associations between urine cortisol and BMI z-score: β = 0.56, 95% CI: 0.16, 0.96, p = 0.007 Covariates included: Child’s: Race, Tanner stage, and socio-economic status |
| 13$ Hillman et al. (53); United States (Developed) | Cross-sectional | 218 (100%) | 14.90 ± 2.20 | White, Black, Other | Urine free cortisol (afternoon) | RIA | Body fat percentage | Percentage body fat was measured by DXA | Associations between urine cortisol and PBF: β = 2.60, 95% CI: −0.65, 5.85, p = 0.12 Covariates included: Child’s: Age, race, and socio-economic status |
| 2$ Barat et al. (48); France (Developed) | Cross-sectional | 28 (50%) | 6–13 | Not reported | Urine free cortisol morning | RIA | Truncal distribution of fat mass (TDFM) | TDFM was assessed with dual energy X-ray absorptiometry | Correlation between urine cortisol and TDFM was: r = −0.28 for total children, r = 0.09 for boys and r = 0.25 for girls No covariates were reported |
BMI, body mass index; WC, waist circumference; PBF, percentage body fat; BMI-SDS, BMI standard deviation score; SDS, standard deviation score; FMI, fat mass index; FFMI, free fat mass index; WtHR, waist to height ratio; CDC, Centers for Disease Control and Prevention; CI, confidence interval; M, mean; NR, not reported;β, beta coefficient; ELISA, enzyme-linked immunosorbent assay; CLIA, chemiluminescence immunoassay; HPLC-MS/MS, high-performance liquid chromatography-tandem mass spectrometry; LC-MS/MS, liquid chromatography tandem mass spectrometry; ECLIA, electrochemiluminescence immunoassay; RIA, radioimmunoassay; DELFIA, dissociation-enhanced lanthanide fluorescence immunoassay; TRFIA, a time-resolved fluorescence immunoassay; TSST-C, Trier Social Stress Test for Children; AUC, area under the curve with respect to ground; AUCi, saliva cortisol area-under-the-curve-increase.
#The two cohort studies that showed associations between hair cortisol and weight status were not included in the meta-analysis because the explanation of effect size (β) is different.
#These studies were not included in the meta-analysis because the meaning of effect size (β) was different as those of the other studies.
*These studies were not included in the meta-analysis because the effect sizes were not reported.
$These studies were not included in the meta-analysis because the studies reported the homologous association < 2.
a and b: The studies data was extracted from one publication by gender, a for boys and b for girls, respectively.
c and d: The studies data was extracted from one publication in the age groups.
x, y, and z: Study data were extracted from one publication according to different measurement times – x for morning, y for afternoon and z for evening. In meta-analysis, we only included the association between morning cortisol and weight status.