Insulin resistance is higher in prepubertal girls but switches to become higher in boys at age 16: A Cohort Study (EarlyBird 57).

BACKGROUND: The risk of type 2 diabetes is increasing in teenage girls, and is associated with their greater insulin resistance (IR). HYPOTHESIS: We hypothesized that the adverse metabolic profile of girls (compared with boys) would persist from childhood through adolescence. PATIENTS AND METHODS: Community-based longitudinal cohort of 292 children (147 boys) studied annually from 9 to 16 years. MEASURES: IR (homeostasis-model-assessment-2), high-density lipoprotein-cholesterol (HDL-C), triglycerides, % body-fat (dual-energy x-ray absorptiometry), pubertal stage (age at peak height velocity), physical activity (accelerometry). Multi-level modelling established the age-related trends in IR and lipids and the influence of covariates. RESULTS: Each year from 9 to 15 years, girls had 21% to 63% higher IR than boys (girls mean IR 0.73-1.33, boys 0.51-0.89, P < .005). At 16 years the gender difference was not significant (girls IR 0.60, boys 0.56, P = .45). Girls had lower HDL-C from 9 to 12 years, higher triglycerides from 9 to 14 years, greater adiposity throughout, and earlier puberty, but boys were more active than girls (all P < .05). After adjustment for %-fat, puberty and activity, the gender difference in IR between girls and boys aged 9 to 15 years became non-significant (IR girls 0.66-1.01, boys 0.65-1.04, P > .07). However, after adjustment at 16 years, girls' IR was 25% lower than boys' (girls 0.44, boys 0.63, P = .001), and they had 22% higher HDL-C (P < .001) and 20% lower triglycerides (P = .003). CONCLUSIONS: The higher IR of prepubertal and early pubertal girls diminishes during late puberty, and boys begin to exhibit greater metabolic risk. Despite being leaner and more active, boys at 16 years have higher IR than girls, suggesting future higher risk for diabetes, thus we reject our hypothesis.

Proinsulin is stable at room temperature for 24 hours in EDTA: A clinical laboratory analysis (adAPT 3).

AIMS: Reference laboratories advise immediate separation and freezing of samples for the assay of proinsulin, which limit its practicability for smaller centres. Following the demonstration that insulin and C-peptide are stable in EDTA at room temperature for at least 24hours, we undertook simple stability studies to establish whether the same might apply to proinsulin. METHODS: Venous blood samples were drawn from six adult women, some fasting, some not, aliquoted and assayed immediately and after storage at either 4°C or ambient temperature for periods from 2h to 24h. RESULTS: There was no significant variation or difference with storage time or storage condition in either individual or group analysis. CONCLUSION: Proinsulin appears to be stable at room temperature in EDTA for at least 24h. Immediate separation and storage on ice of samples for proinsulin assay is not necessary, which will simplify sample transport, particularly for multicentre trials.

Generational change in fasting glucose and insulin among children at ages 5-16y: Modelled on the EarlyBird study (2015) and UK growth standards (1990) (EarlyBird 69).

AIM: Pre-diabetes is a state of beta-cell stress caused by excess demand for insulin. Body mass is an important determinant of insulin demand, and BMI has risen substantially over recent time. We sought to model changes in the parameters of glucose control against rising BMI over the past 25years. METHODS: Using random coefficient mixed models, we established the correlations between HbA1C, fasting glucose, fasting insulin, HOMA2-IR and BMI in contemporary (2015) children (N=307) at ages 5-16y from the EarlyBird study, and modelled their corresponding values 25years ago according to the distribution of BMI in the UK Growth Standards (1990). RESULTS: There was little change in HbA1C or fasting glucose over the 25y period at any age or in either gender. On the other hand, the estimates for fasting insulin and HOMA2-IR were substantially higher in both genders in 2015 compared with 1990. CONCLUSION: Insofar as it is determined by body mass, there has been a substantial rise in beta cell demand among children over the past 25years. The change could be detected by fasting insulin and HOMA2-IR, but not by fasting glucose or HbA1C.

Physical activity attenuates the mid-adolescent peak in insulin resistance but by late adolescence the effect is lost: a longitudinal study with annual measures from 9-16 years (EarlyBird 66).

AIMS/HYPOTHESIS: The aim of this work was to test whether the mid-adolescent peak in insulin resistance (IR) and trends in other metabolic markers are influenced by long-term exposure to physical activity. METHODS: Physical activity (7 day ActiGraph accelerometry), HOMA-IR and other metabolic markers (glucose, fasting insulin, HbA1c, lipids and BP) were measured annually from age 9 years to 16 years in 300 children (151 boys) from the EarlyBird study in Plymouth, UK. The activity level of each child was characterised, with 95% reliability, by averaging their eight annual physical activity measures. Age-related trends in IR and metabolic health were analysed by multi-level modelling, with physical activity as the exposure measure (categorical and continuous) and body fat percentage (assessed by dual-energy X-ray absorptiometry) and pubertal status (according to age at peak height velocity and Tanner stage) as covariates. RESULTS: The peak in IR at age 12-13 years was 17% lower (p < 0.001) in the more active adolescents independently of body fat percentage and pubertal status. However, this difference diminished progressively over the next 3 years and had disappeared completely by the age of 16 years (e.g. difference was -14% at 14 years, -8% at 15 years and +1% at 16 years; 'physical activity × age(2), interaction, p < 0.01). Triacylglycerol levels in girls (-9.7%, p = 0.05) and diastolic blood pressure in boys (-1.20 mmHg, p = 0.03) tended to be lower throughout adolescence in the more active group. CONCLUSIONS/INTERPRETATION: Our finding that physical activity attenuates IR during mid-adolescence may be clinically important. It remains to be established whether the temporary attenuation in IR during this period has implications for the development of diabetes in adolescence and for future metabolic health generally.

Anti-Müllerian hormone in children: a ten-year prospective longitudinal study (EarlyBird 39).

BACKGROUND: Anti-Müllerian hormone (AMH) is produced by Sertoli cells of the testes and granulosa cells of the ovary. There are limited prospective longitudinal data assessing AMH concentrations throughout childhood in both sexes. OBJECTIVE: This study aimed to examine AMH throughout childhood with particular reference to the relationship of AMH to pubertal development in both sexes. DESIGN: This is a prospective longitudinal non-intervention cohort study with annual sampling for participants aged 5-14 years. SETTING: Community cohort study. PARTICIPANTS: A total of 307 healthy children (170 boys) recruited at 5 years from randomly selected schools in Plymouth, UK, participated in this study. Data sets are complete in 76% of the children at 14 years of age. MAIN OUTCOME MEASURE(S): Annual measures of serum AMH, follicle stimulating hormone (FSH) and luteinising hormone (LH), Tanner stage (TS). RESULTS: Boys: AMH was stable from 5 to 7 years, increased slightly from 8 to 10 years, then declined at TS2. This decline was preceded by rising FSH and the appearance of LH. AMH correlated inversely with gonadotrophic hormones during puberty. Girls: AMH increased slightly between 6 and 10 years, peaking during the final prepubertal year before returning to near baseline levels at TS3. Inverse correlations between AMH and FSH were apparent during the prepubertal years. CONCLUSIONS: Our longitudinal data clarified the development of individual AMH levels over a 10-year period. We described modest late prepubertal peaks in both boys and girls, and confirmed the pubertal decline in boys. The inverse association of AMH with gonadotrophins in young females supports its role as a marker of ovarian function, while the precise role for AMH in relation to testicular function in young males remains unclear.

Exploring the Adolescent Fall in Physical Activity: A 10-yr Cohort Study (EarlyBird 41).

INTRODUCTION: Contemporary adolescents are deemed inactive, especially girls, but whether for biological reasons associated with their maturation, changes in their behavior or because of environmental constraints, is uncertain. We examined the trends in physical activity (PA) in relation to both biological and environmental factors in an attempt to establish what drives activity patterns from childhood through adolescence. METHODS: Physical activity (7-d Actigraph accelerometry) was measured annually from 5 to 15 yr in a single cohort of some 300 UK children. Total PA (TPA; in-school and out-of-school separately and combined as whole day) and intensity-specific PA (sedentary, light, and moderate-and-vigorous [MVPA]) were analyzed. Biological age (years before/after measured peak height velocity) and pubertal stage (self-reported pubic hair development-Tanner staging) were also measured as was socioeconomic status (postcode-derived index of multiple deprivation [IMD]). RESULTS: Total PA was stable from 5 to 8 yr (trend P = 0.10) but fell progressively from 9 to 15 yr (by approximately 30% in girls and approximately 20% in boys, both P < 0.001; sex interaction, P < 0.01). Half of this fall was attributable to light intensity PA and only a quarter to MVPA. The decline in PA was related similarly to chronological and biological age, whereas pubertal stage explained the more rapid PA decline in girls (puberty-adjusted sex interaction, P = 0.51). Total PA fell to the same extent for in-school and out-of-school settings (both P < 0.001), and for lower and higher IMD areas (both P < 0.001). Total PA tracked moderately to strongly from childhood into adolescence (r = 0.58; P < 0.001). CONCLUSIONS: The adolescent decline in PA is consistent across different environmental settings, attributable to falls in light-intensity/habitual activity and influenced by puberty, suggesting that the inactivity of adolescence may, in part, be under biological control.

Adiposity, chronic inflammation, and the prepubertal decline of sex hormone binding globulin in children: evidence for associations with the timing of puberty (Earlybird 58).

BACKGROUND: The regulation and role of SHBG in children are poorly defined. Here we investigated whether adiposity-related mechanisms regulate SHBG and whether SHBG levels are associated with the age of puberty. METHODS: Longitudinal modelling of annual physiological and endocrine measurements from age 5 to 15 years in a cohort of 347 Plymouth schoolchildren. RESULTS: SHBG levels were highest at age 5 years and then declined. Mean (SE) SHBG levels were higher in boys than girls at age 5 years [mean (SE) difference 7.68 (3.80) nmol/L; P = .045] but lower in boys by age 15 years [difference 12.19 (3.4) nmol/L; P < .001]. SHBG correlated inversely with adiposity [body mass index SD score (BMI SDS)], insulin, IGF-I, C-reactive protein (CRP), and leptin and positively with adiponectin but not with dehydroepiandrosterone sulphate, androstenedione, or T. In linear mixed models, five adiposity-related covariates (insulin, leptin, adiponectin, IGF-I, and CRP) all exerted significant main effects on SHBG (boys P = .04 to < .001; girls P = .007 to < .001). However, the further addition of BMI SDS rendered the effects of leptin, insulin, and adiponectin nonsignificant, whereas CRP and IGF-I remained significant. In separate models, the individual effects on SHBG of insulin, leptin, IGF-I, and adiponectin, but not CRP, were displaced by BMI SDS. Finally, in linear regression, BMI SDS little affected R(2) resulting from the five adiposity-related signals. Girls with lower SHBG levels at age 5 years reached Tanner stage 2 earlier, tended to have earlier LH secretion, and earlier age at peak height velocity and menarche. In contrast, boys with lower SHBG levels at age 5 years reached Tanner stage 2 earlier, but there were no relationships between SHBG and earlier onset of LH secretion or age at peak height velocity. CONCLUSIONS: Adiposity-related endocrine mechanisms and chronic inflammation were associated with the prepubertal decline of SHBG, and lower SHBG levels anticipated earlier puberty. These findings may be relevant to the occurrence of earlier puberty in recent decades.

Metabolic risk in contemporary children is unrelated to socio-economic status: longitudinal study of a UK urban population (EarlyBird 42).

Lifestyle interventions to improve health in young children tend to target areas of relative deprivation, but the evidence for so doing is largely historical. Accordingly, we have re-examined the link between deprivation, obesity and metabolic risk in contemporary UK children. Using a postcode-based index of multiple deprivation (IMD), we assessed 269 children from the community-based EarlyBird Study, attending 53 schools representing a wide socio-economic range. Annual measures of fatness from 5 to 8 yr included body mass index (BMI), waist circumference (WC), and sum of five skinfolds (SSF). A metabolic risk score, based on blood pressure, lipids and insulin resistance, was derived from annual fasting blood samples. There were no significant associations between deprivation and any measure of adiposity in girls (all p > 0.37). In boys, there was a weak but consistently inverse relationship between deprivation and WC (r = -0.19, p = 0.03) and BMI (r = -0.14, p = 0.09) at 8 yr. Changes in adiposity over 3 yr were unrelated to deprivation in boys. In girls there was a slight but significant increase in SSF only (1 mm/yr per 20 IMD units, p = 0.001). Importantly, in both genders, metabolic risk score was unrelated to deprivation throughout (r values -0.05 to -0.13, all p > 0.12), as was change in metabolic risk (all p > 0.30). Our data do not support the assumption that obesity, metabolic disturbance and thus risk of type 2 diabetes are more prevalent among poorer children. In today's increasingly obesogenic environment, youngsters from all backgrounds appear to be vulnerable, with population-wide implications for public health spending, and the prevention of diabetes in contemporary youth.

A biological, latent variable model of health (EarlyBird 68).

OBJECTIVE: To determine whether factor analysis of a set of health-related biomarkers provides evidence of an underlying common dimension of variation, and to explore the relationship between this dimension of variation with positive and negative affect. METHOD: Twelve health-related metabolic, immune and body-composition biomarkers at ages 5, 7, 9, 11, 14 and 16years were obtained from the EarlyBird longitudinal cohort of 347 children and supplemented by positive affect (PA) and negative affect (NA) measured at age 16years. RESULTS: At each age, principal factor analysis revealed that nine of the 12 biomarkers consistently loaded on the first extracted factor, accounting for 25% of the variance at age 5, and 37-44% of the variance at 7-16years. High loading biomarkers included physical indicators of adiposity, insulin resistance, C-reactive protein, triglycerides, and cholesterol. Factor scores at different ages correlated between .48 and .85. Correlations between the first factor scores and mood measured at age 16 were r=-.17 (p=.02) for PA and r=.13 (p=.07) for NA. CONCLUSIONS: There is a latent variable, h, that accounts for about a third of the variance of a set of health related physical and biochemical biomarkers. h is comparatively stable during childhood and is a weak predictor of mood. These data provide a rationale for aggregating biomarkers in psychoneuroimmunological research. The concept of h provides a possible biological rationale for the role of common factors in disease onset and progression, mental illness, and functional disorders.

PGC1α promoter methylation in blood at 5-7 years predicts adiposity from 9 to 14 years (EarlyBird 50).

The early environment, acting via epigenetic processes, is associated with differential risk of cardiometabolic disease (CMD), which can be predicted by epigenetic marks in proxy tissues. However, such measurements at time points disparate from the health outcome or the environmental exposure may be confounded by intervening stochastic and environmental variation. To address this, we analyzed DNA methylation in the peroxisome proliferator-activated receptor γ coactivator 1α promoter in blood from 40 children (20 boys) collected annually between 5 and 14 years of age by pyrosequencing. Body composition was measured annually by dual X-ray absorptiometry, physical activity by accelerometry, and pubertal timing by age at peak high velocity. The effect of methylation on transcription factor binding was investigated by electrophoretic mobility shift assays. Seven cytosine guanine dinucleotide (CpG) loci were identified that showed no significant temporal change or association with leukocyte populations. Modeling using generalized estimating equations showed that methylation of four loci predicted adiposity up to 14 years independent of sex, age, pubertal timing, and activity. Methylation of one predictive locus modified binding of the proadipogenic pre-B-cell leukemia homeobox-1/homeobox 9 complex. These findings suggest that temporally stable CpG loci measured in childhood may have utility in predicting CMD risk.

Mood and its association with metabolic health in adolescents: a longitudinal study, EarlyBird 65.

BACKGROUND: Mood comprises two main traits - positive and negative affect, both associated with depression and anxiety. Studies in children have linked depression with obesity, but the association with metabolic health is unclear. OBJECTIVE: To explore the relationship between mood and metabolic health in adolescents. METHODS: We studied 208 healthy children (115 boys) enrolled in the longitudinal EarlyBird Diabetes Study, and reviewed at 7 and 16 yr. Participants completed the Positive Affect and Negative Affect Schedule - Child Form (PANAS-C) at 16yr to assess positive and negative affect, together representing mood. Measures at 7 and 16 yr: body mass index (BMI), fat (%; dual energy X-ray absorptiometry), physical activity (accelerometer), metabolic risk z-score comprising homeostasis model assessment-insulin resistance (HOMA-IR), triglycerides, total cholesterol/high density lipoprotein (HDL) ratio and blood pressure. Pubertal development was determined by age at peak height velocity. RESULTS: Positive affect was higher in boys than girls, (50 vs. 46, p = 0.001), negative affect higher in girls than boys (26 vs. 22, p < 0.001). Those with lower mood were fatter (r = -0.24, p < 0.001), had higher HOMA-IR (r = -0.12, p = 0.05), higher cholesterol:HDL ratio (r = -0.14, p = 0.02), were less active (r = 0.20, p = 0.003) and had earlier pubertal development (r = 0.19, p = 0.004). Inverse associations between mood and metabolic risk z-score and change in metabolic risk z-score 7-16yr (ß = -0.26, p = 0.006, and -0.40, p = 0.004, respectively) were independent of adiposity, physical activity and puberty and sex. CONCLUSIONS: Low mood in healthy children is associated with poorer metabolic health independently of adiposity. These findings may have implications for the physical and mental health of contemporary youngsters, given their increasing obesity and cardiometabolic risk.

Divergence between HbA1c and fasting glucose through childhood: implications for diagnosis of impaired fasting glucose (Early Bird 52).

OBJECTIVE: An HbA1c threshold of ≥ 6.5% has recently been adopted for the diagnosis of diabetes in adults, and of ≥ 5.7% to identify adults at risk. Little,however, is known of HbA1c's behaviour or diagnostic value in youth. Our aim was to describe the course of HbA1c during childhood, and its association with fasting glucose. RESEARCH DESIGN AND METHODS: HbA1c and glucose were measured every year in a cohort of 326 healthy children (162 boys) from 5 to 15 years. Mixed effects modelling was used to establish the determinants of HbA1c and its development over time. ROC analysis was used to determine the diagnostic value of HbA1c in the 55 individuals who showed impaired fasting glucose(IFG ­ glucose ≥ 5.6 mmol/L). RESULTS: Glucose rose progressively from 4.3 mmol/L at 5 years to 5.1 mmol/Lat 15 years, and although there were positive associations between HbA1c and glucose, from 10 to 13 years, HbA1c fell while glucose continued to rise. IFG developed in 55 children, but HbA1c exceeded 5.7% in only 16 of them. The maximum area under the ROC curve was 0.71 at the age of 14 (p<0.001), and the sensitivity and specificity were optimal at 50 and 80% respectively,corresponding to HbA1c of 5.4%. CONCLUSIONS: Although HbA1c retains a positive association with glucose throughout childhood, it is weak, and their trends diverge from 10 years,suggesting that factors other than glycaemia systematically influence the variance of HbA1c in youth. These findings therefore limit the interpretation of HbA1c for the diagnosis of IFG during childhood.

Is autoimmunity or insulin resistance the primary driver of type 1 diabetes?

Diabetes is usually classified as autoimmune or metabolic but, as difficulties have arisen with the taxonomy of diabetes, it may help to forego the conventional classification for a more inclusive model. Thus, all diabetes can be ascribed to beta cell insufficiency-hyperglycemia occurs only when the insulin supply fails to meet demand. Humans enter the world with a reserve of beta cells, which is eroded variably by apoptosis over the course of a lifetime. For most, the loss is slow and inconsequential but, for others fast enough to be critical within a lifetime. The challenge now is to define the factors that vary the tempo of beta cell loss, because tempo, not type, seems likely to determine whether diabetes occurs at all, in adulthood or in childhood. Insulin resistance is generally believed to underpin T2D, but has been a feature of insulin-dependent diabetes as well for nearly 80 years, though largely ignored until immunotherapy trials to test the autoimmunity hypothesis persistently failed to bring patient benefit. It seems possible that insulin resistance accelerates beta cell loss generally, its impact modulated by an immune response (autoimmunity) to the beta-cell stress whose intensity varies with immunogenotype. If so, the target for prevention of T1D might more logically lie with insulin sensitivity than with immunoregulation.

Evidence of early beta-cell deficiency among children who show impaired fasting glucose: 10-yr cohort study (EarlyBird 56).

OBJECTIVE: Impaired fasting glucose (IFG) is a predictor of future diabetes and is increasingly common in children, but the extent to which it results from excess insulin demand or failure of supply is unclear. Our aim was to compare the behaviour of insulin sensitivity and beta-cell function in children who developed IFG with those whose glucose levels remained within the normal range. METHODS: We examined trends in fasting glucose, insulin sensitivity (HOMA-S) and beta-cell function (HOMA-B) in 327 healthy children annually from 5 to 15 yr, and the parents at baseline. RESULTS: Fifty-five children showed IFG, mostly after age 11 yr. Fasting glucose rose progressively and was higher throughout in those who developed IFG compared with those who did not (p < 0.001). Beta-cell function was lower from the age of 5 yr in those who developed IFG (p = 0.006), but there was no difference in BMI (p = 0.71). A difference in insulin sensitivity was revealed on adjustment for covariates (p = 0.03). Glucose was higher (p < 0.001), beta-cell function lower (p = 0.01), and insulin sensitivity the same (p = 0.86) in the mothers of children who showed IFG, compared with those who did not. CONCLUSIONS: IFG is common in contemporary children, and appears to be related to a defect in beta-cell function already present at 5 yr. Similar findings in the mothers of IFG children suggest that the beta-cell defect may be transmissible.

Effectiveness of intervention on physical activity of children: systematic review and meta-analysis of controlled trials with objectively measured outcomes (EarlyBird 54).

OBJECTIVE: To determine whether, and to what extent, physical activity interventions affect the overall activity levels of children. DESIGN: Systematic review and meta-analysis. DATA SOURCES: Electronic databases (Embase, Medline, PsycINFO, SPORTDiscus) and reference lists of included studies and of relevant review articles. STUDY SELECTION: Design: randomised controlled trials or controlled clinical trials (cluster and individual) published in peer reviewed journals. INTERVENTION: incorporated a component designed to increase the physical activity of children/adolescents and was at least four weeks in duration. OUTCOMES: measured whole day physical activity objectively with accelerometers either before or immediately after the end of the intervention period. DATA ANALYSIS: INTERVENTION effects (standardised mean differences) were calculated for total physical activity, time spent in moderate or vigorous physical activity, or both for each study and pooled using a weighted random effects model. Meta-regression explored the heterogeneity of intervention effects in relation to study participants, design, intervention type, and methodological quality. RESULTS: Thirty studies (involving 14,326 participants; 6153 with accelerometer measured physical activity) met the inclusion criteria and all were eligible for meta-analysis/meta-regression. The pooled intervention effect across all studies was small to negligible for total physical activity (standardised mean difference 0.12, 95% confidence interval 0.04 to 0.20; P<0.01) and small for moderate or vigorous activity (0.16, 0.08 to 0.24; P<0.001). Meta-regression indicated that the pooled intervention effect did not differ significantly between any of the subgroups (for example, for total physical activity, standardised mean differences were 0.07 for age <10 years and 0.16 for ≥ 10 years, P=0.19; 0.07 for body mass index across the entire range and 0.22 for exclusively overweight/obese children, P=0.07; 0.12 for study duration ≤ 6 months and 0.09 for >6 months, P=0.71; 0.15 for home/family based intervention and 0.10 for school based intervention, P=0.53; and 0.09 for higher quality studies and 0.14 for lower quality studies, P=0.52). CONCLUSIONS: This review provides strong evidence that physical activity interventions have had only a small effect (approximately 4 minutes more walking or running per day) on children's overall activity levels. This finding may explain, in part, why such interventions have had limited success in reducing the body mass index or body fat of children.

Age before stage: insulin resistance rises before the onset of puberty: a 9-year longitudinal study (EarlyBird 26).

OBJECTIVE: Insulin resistance (IR) is associated with diabetes. IR is higher during puberty in both sexes, with some studies showing the increase to be independent of changes in adiposity. Few longitudinal studies have reported on children, and it remains unclear when the rise in IR that is often attributed to puberty really begins. We sought to establish from longitudinal data its relationship to pubertal onset, and interactions with age, sex, adiposity, and IGF-1. RESEARCH DESIGN AND METHODS: The EarlyBird Diabetes study is a longitudinal prospective cohort study of healthy children aged 5-14 years. Homeostasis model assessment (HOMA-IR), skinfolds (SSF), adiposity (percent fat, measured by dual-energy X-ray absorptiometry), serum leptin, and IGF-1 were measured annually in 235 children (134 boys). Pubertal onset was adduced from Tanner stage (TS) and from the age at which luteinizing hormone (LH) first became serially detectable (≥0.2 international units/L). RESULTS: IR rose progressively from age 7 years, 3-4 years before TS2 was reached or LH became detectable. Rising adiposity and IGF-1 together explained 34% of the variance in IR in boys and 35% in girls (both P < 0.001) over the 3 years preceding pubertal onset. The contribution of IGF-1 to IR was greater in boys, despite their comparatively lower IGF-1 levels. CONCLUSIONS: IR starts to rise in mid-childhood, some years before puberty. Its emergence relates more to the age of the child than to pubertal onset. More than 60% of the variation in IR prior to puberty was unexplained. The demography of childhood diabetes is changing, and prepubertal IR may be important.