Severe Obesity,Obesity,and Cardiometabolic Risk in Children 0 to 6 Years of Age

Abstract

Background:

The prevalence of obesity in children has been shown to be plateauing or decreasing in some countries. However, the burden of severe obesity is often not assessed. Children with severe obesity may be at an increased risk for cardiovascular disease and diabetes. The objective is to determine the prevalence of severe obesity in young children and to examine the association with cardiometabolic risk factors.

Methods:

A longitudinal study was conducted through The Applied Research Group for Kids! (TARGet Kids!), a practice-based research network in Toronto, Canada. Healthy children from birth to 6 years of age were recruited and followed through middle childhood. The main outcomes of the study were as follows: total cholesterol, high-density lipoprotein (HDL), low-density lipoprotein (LDL), and non-HDL cholesterol, triglycerides, glucose, systolic blood pressure (SBP), and diastolic blood pressure (DBP). BMI was age and sex standardized based on the WHO growth standards. A BMI z-score >3 was used to define obesity for children <5 years of age, and severe obesity for children ≥5 years of age.

Results:

Among 5738 children <5 years, 0.8% had a zBMI >3. In 626 children who were 5 and 6 years old, 2.1% had a zBMI >3. In the multivariable analysis adjusted for age, sex, maternal ethnicity, and family history, using repeated measures, children with a zBMI >3 had significantly higher odds of having abnormal SBP [odds ratio (OR) 6.4, 95% confidence interval (CI) 1.5–27.9; p = 0.01] and DBP (OR, 3.6 95% CI 1.2–10.6; p = 0.02), respectively, as compared with healthy-weight children. Trends demonstrating an association between a zBMI >3 and abnormal lipid levels were also identified.

Conclusion:

Young children with a zBMI >3 have significantly higher blood pressure measures and trends toward worse lipid profiles than children at lower zBMIs.

Introduction

The prevalence of overweight and obesity has increased in all age groups in the United States, Canada, and many countries globally over the past 30 years.^1,2 In 2013, an estimated 42 million children <5 years of age worldwide were overweight or obese.³ However, some developed countries are reporting that child obesity prevalence has reached a plateau or is decreasing, especially in young children.^1,4–6 Although this is encouraging, these studies only report on overweight and obesity, but they do not consider obesity severity. Therefore, the prevalence of obesity may be stable or decreasing overall, but the number of children who are severely obese may be increasing.⁷ Severe obesity has been associated with poor health outcomes in childhood as well as adulthood, particularly for cardiometabolic health^8,9 and quality of life.¹⁰

There are multiple definitions and nomenclatures for weight status categories in children depending on age, country, and growth reference used. The definition in the United States has traditionally used the CDC age- and sex-specific percentiles at the 85th, 95th, and 99th levels to categorize overweight, obese, and severely obese, respectively.¹¹ However, a more precise definition has recently been proposed by the American Heart Association (AHA) due to the lack of sensitivity of the 99th percentile, and to better capture the potential detrimental health effects at the extreme ends of the distribution.⁹

The new definition categorizes obesity severity by using classes, which also aligns with the adult obesity literature.¹² In children ≥2 years of age, class I obesity is defined as ≥95th percentile and class II obesity is ≥120% of the 95th percentile or a BMI of ≥35 kg/m².⁹ These new cut-offs, using the CDC growth charts, may more accurately predict cardiometabolic risk (CMR) but have only been studied in children 5 to 19 years.¹³ In children <2 years of age, the CDC recommends using the World Health Organization (WHO) growth standards, and there are no recommendations for this class definition in this age group.

The WHO growth standards are recommended in Canada for all aged children. According to the WHO, for children ≥5 years of age, obesity is defined as >2 standard deviations (SD) away from the median z-score for BMI (∼97th percentile), and severe obesity is defined as >3 SD (∼99.9th percentile).¹⁴ For children <5 years, WHO defines overweight as >2 SD away from the median z-score for BMI, for weight-for-length and obese is defined as >3 SD, with no cut-offs defined for severe obesity in the WHO nomenclature in this age group.^15,16

Currently, the prevalence of severe obesity in Canadian children is unknown. If there is evidence that the severity of obesity is an important determinant of cardiometabolic health in children, it is important to measure the potential burden of disease in the population. With resource-intensive interventions required to treat obesity in childhood, it is important to characterize who is at greatest risk, to develop interventions that are individualized for intensity, and to measure outcomes. Our primary objective was to determine the prevalence of severe obesity in a population of young children by using the WHO definition. Our secondary objective was to evaluate the association between severity of obesity by using BMI z-scores and CMR factors, including total cholesterol, non-high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, HDL cholesterol, triglycerides, non-fasting glucose, systolic blood pressure (SBP), diastolic blood pressure (DBP), and a CMR score. The AHA definition was also examined to assess any differences in the associations with CMR based on a different definition of severe obesity.

Materials and Methods

Study Design

In this longitudinal cohort study, children aged 0 to 6 years were recruited from The Applied Research Group for Kids (TARGet Kids!), a practice-based research network in Toronto, Canada (www.targetkids.ca).¹⁷ Children were recruited into TARGet Kids! from nine primary care pediatric and family physician practices between 2009 and 2015 at their regularly scheduled well-child visits. Children are followed at their regularly scheduled well-child visits at 0–6, 9–12, 18, and 24 months of age and annually afterward; children may enter the study at any age before 6 years of age, and some children have now been followed until early adolescence. Children born very preterm (<32 weeks gestational age), and children with severe developmental delay or chronic illness (except for asthma), including conditions affecting growth were excluded. Children were also excluded if their parents were unable to complete questionnaires in English.

Consent was obtained from the parents of all participating children, and ethical approval was obtained from the Research Ethics Board at The Hospital for Sick Children and St. Michael's Hospital. TARGet Kids! is registered at www.clinicaltrials.gov; NCT01869530.

Data were collected through standardized parent-completed nutrition and health questionnaires, physical measurements, and a nonfasting blood sample. All data were uploaded to the central MediData Rave Platform (MediData Solutions, New York) that was housed at the Applied Health Research Centre of the Li Ka Shing Knowledge Institute, St. Michael's Hospital, Toronto.

Anthropometric Measures

Height and weight were measured by trained research staff. Weight was measured by using a precision digital scale (SECA, Germany), standing height was measured by using a stadiometer (SECA), and length for children <2 years of age was measured by using a length board and standardized methods. Waist circumference for the CMR score was measured by using standardized protocols from National Health and Nutrition Examination Survey (NHANES) using a measuring tape around the waist at the midaxillary line and uppermost lateral border of the ilium.¹⁸ BMI was calculated by dividing weight in kilograms by height (or length) in meters squared. BMI z-scores were standardized by using the WHO growth standards (children <5 years) and growth reference (for children 5 years and older).¹⁶ Although Canada recommends defined cut-offs from the WHO, there is no recommended cut-off for severe obesity in children <5 years of age. We used the WHO z-score definition for all ages, noting that labels for each category change with age. For children <5 years of age, BMI z-scores >1 are labeled as “at risk of overweight,” >2 as “overweight,” and >3 as “obese”; in children 5 years and older, these labels are changed to “overweight,” “obese,” and “severely obese” to comply with the recommendations in Canada. BMI-for-age was used for all ages instead of weight-for-length for children <2 years because studies have shown that BMI-for-age is comparable to weight-for length¹⁹ and may be a better predictor of future weight status.²⁰

In a supplemental analysis, BMI status was also categorized according to the AHA cut-offs as follows: underweight (<5th percentile), healthy weight (5th to <85th percentile), overweight (≥85th to <95th percentile), class I obesity (≥95th to <120% of the 95th percentile), and class II obesity (≥120% of the 95th percentile).

Cardiometabolic Measures

All laboratory analyses were performed by the Mount Sinai Services Laboratory using standard procedures as follows: Glucose was measured by using enzymatic reference methods with hexokinase; lipids (total cholesterol, triglycerides, HDL, and LDL) were measured by using the enzymatic colorimetric method on the Roche Modular Platform. Cut-off points for high risk were determined based on existing standards for children.²¹ SBP and DBP were measured by trained research assistants in children aged 3 years and older, according to the guidelines of the National High Blood Pressure Education Program²³ and using the NHANES cut-offs.²¹

A continuous CMR score was examined as a summary outcome. As previously reported, the CMR score was calculated as the sum of age- and sex-standardized waist circumference, triglycerides, SBP, glucose, and HDL.^22,24 A continuous CMR score was calculated for each subject as follows: CMR score = z-WC + z-triglycerides + z-SBP + z-glucose – z-HDL. Higher HDL is indicative of a healthier metabolic profile; therefore, the inverse of HDL was used in the score. A lower CMR score indicates lower CMR.

Potential Confounders

All potential confounders were selected a priori based on the literature, and data were collected by using a parent-completed, standardized questionnaire. The potential confounders include: age, sex, maternal ethnicity, and family history of cardiometabolic diseases. Maternal ethnicity was categorized into four categories: European, East Asian, Southeast Asian/South Asian, and Other. Other included African, Afro-Caribbean, Latin, Arab, North African, West Asian, Aboriginal, and Mixed.

A positive family history was dichotomized to “any” or “none” if the child's parent reported that any person in the immediate family (mother, father, or sibling) or extended family (aunt, uncle, or grandparent) was diagnosed with the following conditions: heart disease, hypertension, high cholesterol, stroke, diabetes, depression, and overweight or obesity. A family history of cardiovascular disease, diabetes, depression,²⁵ and overweight/obesity has previously been shown to be associated with children's CMR.^26–28

Statistical Analysis

Descriptive analyses of main exposure, outcomes, and covariates were examined at baseline. A comparison of the patients in the TARGet Kids! cohort who provided a blood sample versus those who did not was examined by using student's t-test and chi-squared statistic to assess the differences. Cross-sectional, univariate analyses for all a priori independent variables and outcomes were examined. Residual distributions were examined to evaluate linear and logistic model assumptions, and case-wise diagnostics were performed to evaluate influential observations.

For the primary objective, the prevalence of severe obesity was estimated by using the WHO definition. We evaluated this definition in relation to the eight individual cardiometabolic outcomes and the summary score by using chi-square tests. For children with repeated measures of blood lipids, glucose, and blood pressure, only the first blood sample provided by the individual participant was used to determine the proportion at high risk. To determine statistical significance, 95% confidence intervals (CIs) and p-values at a significance level <0.05 were used.

A multivariable logistic regression was used to observe the association between BMI status and CMR factors adjusting for age, sex, maternal ethnicity, and family history. A generalized estimating equation (GEE) was used to account for the repeated measurements and potential correlation within subjects contributing more than one set of data points. Exponentiation of the coefficients was used to report adjusted odds ratios. A GEE multivariable linear regression was used to estimate the association between BMI status and CMR score. In a supplemental analysis, BMI status was also categorized according to the AHA cut-offs to determine the prevalence of severe obesity in children ≥2 years of age, and we calculated the effect on CMR. All statistical calculations were performed by using SAS version 9.3 (SAS Institute, Cary, NC).

Results

Between January 2009 and July 2015, 6687 children from 0 to 6 years of age were recruited into the TARGet Kids! cohort. Over this period, 4023 children provided at least one blood sample and 2346 of these children provided more than one (range from 2 to 5), resulting in 6364 total blood samples. A total of 3843 children >3 years of age contributed 8350 blood pressure measurements. All of these children had standardized height and weight measurements. Table 1 shows the socio-demographic characteristics of the study cohort at enrollment into TARGet Kids! stratified by whether or not they provided a blood sample (children who did not provide blood were not included further in this study).

Table 1.

Baseline Characteristics of TARGet Kids! Cohort

Baseline characteristic	N	With blood work	Without blood work	p
N (%)	6687	4023 (60.16)	2664 (39.84)
Age (months)	6670			<0.01
Mean (SD)		35.18 (24.33)	25.49 (19.56)
<1 year		553 (13.75)	688 (25.99)
1 to <3 years		1568 (39.00)	1101 (41.59)
3+ years		1902 (47.28)	858 (32.41)
Sex	6670			0.07
Male		2144 (53.31)	1352 (51.06)
Female		1878 (46.69)	1296 (48.94)
Maternal ethnicity	6002			0.22
European		2365 (65.24)	1528 (64.17)
East Asian		243 (6.70)	168 (7.06)
Southeast/South Asian		390 (10.76)	294 (12.35)
Other		627 (17.30)	391 (16.42)
Neighborhood income	6098			0.10
Less than $20,000		65 (1.75)	45 (1.89)
$20,000 to $39,999		656 (17.64)	441 (18.53)
$40,000 to $59,999		1509 (40.59)	906 (38.07)
$60,000 to $89,999		1214 (32.65)	769 (32.31)
$90,000 to $149,999		245 (6.59)	194 (8.15)
$150,000 or more		29 (0.78)	25 (1.05)
Child BMI z-score^a	6364			0.13
<−2		134 (3.50)	119 (4.69)
≥−2 to ≥1		3005 (78.50)	1985 (78.27)
>1 to ≤2		528 (13.79)	325 (12.82)
>2 to ≤3		122 (3.19)	85 (3.35)
>3		39 (1.02)	22 (0.87)
Positive family history^b
Any	6392	1156 (52.88)	2254 (53.59)	0.59
Heart disease	6381	412 (18.90)	799 (19.02)	0.91
High cholesterol	6385	695 (31.82)	1299 (30.92)	0.46
Hypertension	6382	498 (22.83)	936 (22.28)	0.62
Stroke	5426	232 (12.53)	440 (12.31)	0.82
Diabetes	4678	98 (6.20)	166 (5.36)	0.24
Depression	5425	392 (21.17)	848 (23.73)	0.03
Overweight/obesity	5426	197 (10.64)	307 (8.59)	0.01
CMR factors
N (%) outside normal range
Total cholesterol	4004	245 (6.12)
Non-HDL cholesterol	3988	295 (7.40)
HDL cholesterol	3989	535 (13.41)
LDL cholesterol	3957	172 (4.35)
Triglycerides	3995	885 (22.71)
Glucose	4023	256 (6.52)
SBP	3842	30 (0.78)
DBP	3843	135 (3.51)
CMR score^c
Mean (SD)	1949	0.05 (2.61)

Parent report of any family member (mother, father, sibling, or aunt/uncle/grandparent) at last visit. With blood work, N = 1156; without blood work, N = 2254.

CMR score calculated for children ≥3 years of age with complete measures for waist circumference, non-HDL cholesterol, SBP, triglycerides, and glucose.

Bold values indicate significance <0.05.

CMR, cardiometabolic risk; DBP, diastolic blood pressure; SBP, systolic blood pressure; SD, standard deviation.

As well, Table 1 shows the overall proportion of children outside the normal range for the eight CMR factors. Children who did not provide a blood sample were slightly younger and had parents who reported more family history of depression and overweight or obesity than children with blood work. No other significant differences were seen between the two groups.

The prevalence of severe obesity was assessed by using the WHO definition and reported in Table 2 by age and sex. Overall, among the 6364 children enrolled in TARGet Kids! with a valid BMI z-score, 4.0% had a zBMI <−2, 78.4% had a zBMI ≥−2 and ≤1, 13.4% had a zBMI >1 and ≤2, 3.3% had a zBMI >2 and ≤3, and 1.0% had a zBMI >3. Figure 1 shows the proportion of children classified as high risk for each CMR factor by zBMI status. Based on the chi-square analysis, only SBP was statistically significant between weight categories (data not shown), although children with a zBMI >3 had a much higher prevalence of abnormality for risk factors.

Figure 1.

Proportion of children with cardiometabolic risk factors by zBMI (using WHO definition). Z-scores of −2.0, 1.0, 2.0, and 3.0 correspond approximately to percentiles of the 3rd, 85th, 97th, and 99.9th, respectively. For children <5 years: underweight (<−2.0), healthy weight (≥−2.0 and ≤1.0), at risk for overweight (>1.0 to ≤2.0), overweight (>2.0 to ≤3.0), and obesity (>3.0). For children <5 years: underweight (<−2.0), healthy weight (≥−2.0 and ≤1.0), overweight (>1.0 to ≤2.0), obesity (>2.0 to ≤3.0), and severe obesity (>3.0).

Table 2.

Prevalence of Obesity and Severe Obesity Using BMI z-Scores Standardized to the WHO Growth Standards and Reference by Age Group and Sex, N (%)

Age group	Total, N	<−2.0	−2.0 to ≤1.0	>1.0 to ≤2.0	>2.0 to ≤3.0	>3.0
Overall	6364	253 (4.0)	4990 (78.4)	853 (13.4)	207 (3.3)	61 (1.0)
0 to <2 years^a	3281	210 (6.4)	2631 (80.2)	340 (10.4)	83 (2.5)	17 (0.5)
2 to <5 years^a	2457	36 (1.5)	1860 (75.7)	427 (17.4)	103 (4.2)	31 (1.3)
5 to 6 years^b	626	7 (1.1)	499 (79.7)	86 (13.7)	21 (3.4)	13 (2.1)
Boys
0 to <2 years	1741	122 (7.0)	1369 (78.6)	184 (10.6)	55 (3.2)	11 (0.6)
2 to <5 years	1261	18 (1.4)	919 (72.9)	243 (19.3)	59 (4.7)	22 (1.7)
5 to 6 years	334	4 (1.2)	264 (79.0)	45 (13.5)	13 (3.9)	8 (2.4)
Girls
0 to <2 years	1540	88 (5.7)	1262 (82.0)	156 (10.1)	28 (1.8)	6 (0.4)
2 to <5 years	1196	18 (1.5)	941 (78.7)	184 (15.4)	44 (3.7)	9 (0.8)
5 to 6 years	292	3 (1.0)	235 (80.5)	41 (14.0)	8 (2.7)	5 (1.7)

Z-scores of −2.0, 1.0, 2.0, and 3.0 correspond approximately to percentiles of the 3rd, 85th, 97th, and 99.9th, respectively.

For children <5 years: underweight (<−2.0), healthy weight (≥−2.0 and ≤1.0), at risk for overweight (>1.0 to ≤2.0), overweight (>2.0 to ≤3.0), and obesity (>3.0).

For children ≥5 years: underweight (<−2.0), healthy weight (≥−2.0 and ≤1.0), overweight (>1.0 to ≤2.0), obesity (>2.0 to ≤3.0), and severe obesity (>3.0).

In the multivariable analysis adjusted for age, sex, maternal ethnicity, and family history, using repeated measures, children with a zBMI >3 had significantly higher odds of having abnormal SBP and DBP (Table 3). Although not significant at a 0.05 p-value, children with a zBMI >3 showed increased odds for all other risk factors, ranging from 1.18 (95% CI 0.47–2.96; p = 0.72) for total cholesterol to 2.00 (95% CI 0.77–5.20; p = 0.15) for LDL cholesterol, except for glucose. The linear regression model for BMI status and the CMR score is presented in Table 4. All levels of overweight and obesity were significantly associated with an increased CMR standardized risk score compared with healthy weight children; for example, children with a zBMI >3 had an increased risk score of 4.32 units (95% CI 3.14–5.50; p < 0.01) compared with healthy-weight children.

Table 3.

Generalized Estimating Equation Multivariable Logistic Regressions of BMI Status (Using WHO z-Score Definition) and Cardiometabolic Risk Factors Using Repeated Measures

	Total cholesterol			Non-HDL cholesterol			HDL cholesterol			LDL cholesterol
	aOR	95% CI	p	aOR	95% CI	p	aOR	95% CI	p	aOR	95% CI	p
BMI z-score
<−2	0.77	0.34–1.77	0.54	1.00	0.51–1.97	0.99	0.71	0.40–1.27	0.25	1.45	0.69–3.04	0.33
≥−2 to ≥1	REF			REF			REF			REF
>1 to ≤2	0.94	0.69–1.29	0.71	1.22	0.89–1.67	0.22	1.13	0.88–1.45	0.33	1.14	0.76–1.71	0.54
>2 to ≤3	0.39	0.14–1.05	0.06	1.03	0.52–2.03	0.94	0.91	0.55–1.51	0.70	0.52	0.16–1.67	0.27
>3	1.18	0.47–2.96	0.72	1.56	0.61–3.98	0.36	2.09	0.93–4.67	0.07	2.00	0.77–5.20	0.15
Age (months)	0.99	0.99–1.00	0.0009	0.98	0.97–0.98	<0.0001	0.97	0.97–0.98	<0.01	0.98	0.98–0.99	<0.01
Sex^a	1.58	1.22–2.04	0.0005	1.75	1.37–2.24	<0.0001	1.31	1.09–1.57	0.004	1.87	1.35–2.59	0.0002
Maternal ethnicity^a
East Asian	1.29	0.81–2.07	0.28	1.04	0.63–1.69	0.89	0.56	0.36–0.87	0.01	1.78	1.00–3.15	0.05
Southeast/South Asian	1.24	0.83–1.84	0.29	1.11	0.75–1.63	0.61	0.72	0.52–1.00	0.05	1.81	1.14–2.86	0.01
Other	1.16	0.82–1.63	0.41	0.96	0.69–1.33	0.80	0.81	0.63–1.04	0.10	1.45	0.96–2.18	0.07
Family History^a	1.69	1.34–2.12	<0.0001	1.55	1.24–1.94	0.0001	0.72	0.60–0.86	0.0004	1.61	1.20–2.17	0.002

	Triglycerides			Glucose			SBP			DBP
	aOR	95% CI	p	aOR	95% CI	p	aOR	95% CI	p	aOR	95% CI	p
BMI z-score
<−2	0.78	0.50–1.22	0.28	0.82	0.38–1.77	0.61	4.19	0.97–18.11	0.06	2.48	0.88–6.94	0.08
≥−2 to ≥1	REF			REF			REF			REF
>1 to ≤2	1.12	0.92–1.36	0.27	0.92	0.68–1.24	0.58	1.89	0.88–4.08	0.10	1.31	0.84–2.05	0.24
>2 to ≤3	1.03	0.68–1.56	0.88	0.82	0.45–1.48	0.51	3.41	1.16–10.05	0.03	1.94	0.94–3.98	0.07
>3	1.86	0.97–3.60	0.06	0.94	0.37–2.37	0.89	6.41	1.47–27.92	0.01	3.59	1.24–10.35	0.02
Age (months)	0.98	0.98–0.98	<0.0001	1.01	1.00–1.01	0.001	0.99	0.97–1.00	0.16	0.99	0.98–1.00	0.01
Sex^a	1.21	1.05–1.40	0.01	0.80	0.65–0.99	0.04	0.92	0.51–1.67	0.79	0.88	0.63–1.22	0.44
Maternal ethnicity^a
East Asian	1.28	0.96–1.71	0.09	1.61	1.11–2.33	0.01	1.96	0.75–5.09	0.17	1.09	0.61–1.94	0.77
Southeast/South Asian	0.81	0.62–1.04	0.10	1.37	0.99–1.90	0.06	2.00	0.84–4.80	0.12	0.64	0.30–1.35	0.24
Other	0.69	0.56–0.85	0.0005	1.23	0.93–1.63	0.14	1.52	0.66–3.53	0.33	1.16	0.73–1.84	0.53
Family history^a	0.95	0.82–1.10	0.50	1.08	0.88–1.33	0.44	1.82	0.99–3.34	0.05	0.65	0.45–0.94	0.02

Z-scores of −2.0, 1.0, 2.0, and 3.0 correspond approximately to percentiles of the 3rd, 85th, 97th, and 99.9th, respectively. For children <5 years: underweight (<−2.0), healthy weight (≥−2.0 and ≤1.0), at risk for overweight (>1.0 to ≤2.0), overweight (>2.0 to ≤3.0), and obesity (>3.0). For children ≥5 years: underweight (<−2.0), healthy weight (≥−2.0 and ≤1.0), overweight (>1.0 to ≤2.0), obesity (>2.0 to ≤3.0), and severe obesity (>3.0). Cut-offs for abnormal cardiovascular risk factors can be found at Pediatrics Supplement. Repeated measure: N = 6364 observations on N = 4023 unique subjects used in the analysis of lipid and glucose outcomes; N = 8350 observations on N = 3843 unique subjects in the analysis of blood pressure.

Referent: Male, European; no positive family history.

Bold values indicate significance <0.05.

aOR, adjusted odds ratio; CI, confidence interval.

Table 4.

Generalized Estimating Equation Multivariable Linear Regression of BMI Status and Cardiometabolic Risk Score Using WHO z-Score Definition

	CMR score^a
Variable	Estimate (95% CI)	p
BMI z-score
<−2	−0.51 (−1.30 to 0.28)	0.20
≥−2 to ≥1	REF
>1 to ≤2	1.33 (1.05 to1.60)	<0.01
>2 to ≤3	2.96 (2.28 to 3.64)	<0.01
>3	4.32 (3.14 to 5.50)	<0.01
Age (months)	−0.00 (−0.01 to 0.00)	0.80
Sex	0.13 (−0.08 to 0.34)	0.23
Maternal ethnicity^b
East Asian	−0.29 (−0.69 to 0.11)	0.16
Southeast/South Asian	−0.05 (−0.40 to 0.31)	0.80
Other	−0.44 (−0.74 to −0.13)	0.01
Family history^b	−0.17 (−0.36 to 0.02)	0.09

Z-scores of −2.0, 1.0, 2.0, and 3.0, correspond approximately to percentiles of the 3rd, 85th, 97th, and 99.9th, respectively. For children <5 years: underweight (<−2.0), healthy weight (≥−2.0 and ≤1.0), at risk for overweight (>1.0 to ≤2.0), overweight (>2.0 to ≤3.0), and obesity (>3.0). For children ≥5 years: underweight (<−2.0), healthy weight (≥−2.0 and ≤1.0), overweight (>1.0 to ≤2.0), obesity (>2.0 to ≤3.0), and severe obesity (>3.0).

N = 2879 observations on N = 1949 unique subjects.

Referent: Male, European; no positive family history.

Bold values indicate significance <0.05.

Results of the analysis using the AHA definition for children ≥2 years are presented in Supplementary Tables S1 and S2 (Supplementary Data are available online at www.liebertpub.com/chi). Using the AHA definition, 7.0% of children ≥2 years were underweight, 76.0% were healthy weight, 9.9% were overweight, 6.6% had class I obesity, and 0.6% had class II obesity.

Discussion

To our knowledge, this study is the first to show the prevalence of severe obesity in a young pediatric population in urban Canada and the associated CMR. Overall, 1.3% of children 2 to <5 years and 2.1% of children 5–6 years had a zBMI >3. Despite different definitions, these estimates are in line with the most recent NHANES data that estimated the prevalence of severe obesity to be between 1.7%²⁹ and 2.1%⁷ in children 2–5 years of age. Similarly, in a cohort of U.S. preschool-aged children (3–5 years) in California, the prevalence of severe obesity was 1.6%³⁰; in a national cohort of Australian children, the reported prevalence was 2.0%.³¹ However, the Australian study analyzed data predominately on older children and all four studies used the CDC 2000 growth charts. Our results in older children aged 5 to 6 years support the international severe obesity estimates of ∼2%.

As expected, the proportion of children with abnormal CMR factors in this young population was generally low. Similar to previous studies, the prevalence in the highest obesity category increased sharply for most risk factors,^32,33 except for glucose. In contrast, our study showed no observable difference in the prevalence of abnormal CMR factors between a zBMI >1, or a zBMI >2 except for SBP and DBP. Compared with healthy-weight children, children with a zBMI >3 had higher odds of having each of the CMR factors examined [odds ratios (ORs) ranged from 1.18 to 6.41] and statistically higher odds of having high blood pressure and a worse CMR standardized score.

Similarly, Skinner et al. found that older children and adolescents with class II and class III obesity were at a significantly higher risk for low levels of HDL, high SBP, high DBP, and high triglycerides compared with children with class I obesity, although blood pressure was only measured in children 8 to 19 years of age, and their sample size in children 3 to 5 years was less than 300 participants.³³ Severe obesity was also significantly associated with the cluster CMR score at all levels of obesity compared with healthy-weight children. Valerio et al. examined clustered risk factors in a cohort of patients from pediatric obesity clinics and found that 67% of patients had one or more cardiometabolic abnormalities.¹³ It is worrisome to note that even in this young age group, our results suggest that the severity of obesity is associated with abnormal CMR profiles.

Age and sex were significantly associated with almost all CMR factors. Younger children had slightly lower odds of having abnormal risk factors and girls had higher odds of having abnormal risk factors compared with boys, expect for glucose and blood pressure. Li et al. found that boys were more likely to have low HDL and high fasting glucose, and girls were more likely to have higher triglycerides.³² Notable sex differences were also seen in the study by Skinner et al.; boys with severe obesity had a higher prevalence of abnormal SBP and triglycerides.³³

In Canada and internationally, multiple agencies have endorsed the WHO growth standards and references for children. To the best of our knowledge, no one has compared the use of the >3.0 z-score with 120% of the 95th percentile applied to the WHO standardized growth in young children. Different definitions will affect the number of children categorized as severely obese and may affect diagnosis and appropriate identification of adverse CMR. Canadian guidelines characterize weight categories differently than in the United States and do not provide any guidelines for severe obesity in children <5 years of age.¹⁴ These Canadian guidelines contrast recommendations from the American Academy of Pediatrics (AAP), which changed the nomenclature of categorizing weight status to eliminate “at risk” categories and only have overweight, obese, and classes of severe obesity. This reason for this change was to more precisely identify individuals with high adiposity and the associated negative health risks.³⁴ Based on the emerging evidence, including this study, where severity of obesity is an important determinant of cardiometabolic health risk, it may be prudent to define severe obesity in this young pediatric population where appropriate early clinical assessment and intervention may impact outcomes.

The strengths of this study are the large cohort of young children providing blood samples, measured heights/lengths and weights and blood pressure. No other studies examined non-HDL cholesterol in their blood measures. The AAP expert panel on cardiovascular risk in children has listed non-HDL as a potential screening blood lipid for future disease.²¹ Parents also provided known family history of cardiovascular disease and maternal ethnicity that were included in the models as potential confounders. Finally, we used the most conservative cut-off points for CMR (values considered “high” risk). The cut-off points obtained from the Expert Panel on Integrated Guidelines used data mostly from older children; therefore, the use of the highest level of risk was considered prudent for the younger age group in this study.

There are some limitations to this study. Although the overall sample of children was large in this study, the prevalence of severe obesity was very small as was the prevalence of certain CMR factors such as SBP and DBP. Therefore, the CIs for the prevalence of abnormal risk factors were quite large. To improve the estimates and address this potential limitation, we used repeated measures of children with longitudinal data and a GEE to account for the correlation within subjects. A second limitation is that the measurement of blood pressure in a pediatric population can have low reliability. Although our research assistants are trained to measure blood pressure with appropriate pediatric cuffs, in five clinics, manual measurement is used and can be subject to measurement error.³⁵

The blood sample taken during regularly scheduled well-child visits is a nonfasting blood sample; however, previous work within this cohort showed that glucose was only minimally impacted.³⁶ We decided not to adjust for the multiple hypotheses testing because lowering the risk of Type I error by adjusting significance levels would inherently increase the risk of a Type II error, which may mask some associations.³⁷ In addition, the inclusion of the CMR score is one method to address multiple outcomes testing by using this composite endpoint. However, the validation on CMR scores in children is still emerging, and we, therefore, decided to include all individual outcomes. Finally, our study sample comprises mainly urban children of European ethnicity with a slightly higher-than-average median household income. The results in this population may not be generalizable to the wider Canadian population.

Despite these limitations, this is the first study to examine the prevalence of severe obesity in a cohort of young children and the associated CMR. Due to the low prevalence of severe obesity and abnormal CMR factors, these results should only be considered exploratory, and future research should characterize the growth of children with severe obesity over time. Nevertheless, it was shown that even in this young age group, children with a zBMI >3 (defined by the WHO as obesity in children <5 years, and severe obesity in children 5 years and older) have significantly higher DBP and SBP, consistent with the results of other studies of CMR in children. Growth monitoring for young children is recommended to assess a child's overall health and as a tool to screen for overweight and obesity. Monitoring strategies may need to include a focus on children with severe obesity to ensure adequate identification and intervention in this population with potentially important cardiometabolic health outcomes.

Footnotes

Acknowledgments

The authors thank all of the participating families for their time and involvement in TARGet Kids! and are grateful to all practitioners who are currently involved in the TARGet Kids! practice-based research network. Advisory Committee: Eddy Lau, Andreas Laupacis, Patricia C. Parkin, Michael Salter, Peter Szatmari, and Shannon Weir. Research Team: Charmaine Camacho, Arthana Chandraraj, Dharma Dalwadi, Ayesha Islam, Thivia Jegathesan, Tarandeep Malhi, Megan Smith, and Laurie Thompson. Applied Health Research Centre: Christopher Allen, Bryan Boodhoo, Judith Hall, Peter Juni, Gerald Lebovic, Karen Pope, Jodi Shim, and Kevin Thorpe; Mount Sinai Services Laboratory: Azar Azad.

Funding to support TARGet Kids! was provided by multiple sources, including the Canadian Institutes for Health Research (CIHR), namely the Institute of Human Development, Child and Youth Health; the Institute of Nutrition, Metabolism and Diabetes; and the St. Michael's Hospital Foundation. The Paediatric Outcomes Research Team is supported by a grant from The Hospital for Sick Children Foundation. Funding agencies had no role in the design, collection, analyses, or interpretation of the results of this study or in the preparation, review, or approval of this article.

TARGet Kids! Collaboration: Co-Leads: Catherine S. Birken, Jonathon L. Maguire; Scientific Committee: Kawsari Abdullah, Yamna Ali, Laura N. Anderson, Imaan Bayoumi, Cornelia M. Borkhoff, Sarah Carsley, Shiyi Chen, Yang Chen, Denise Darmawikarta, Cindy-Lee Dennis, Karen Eny, Stephanie Erdle, Kayla Furlong, Kanthi Kavikondala, Christine Koroshegyi, Christine Kowal, Grace Jieun Lee, Dalah Mason, Jessica Omand, Patricia C. Parkin, Navindra Persaud, Lesley Plumptre, Meta van den Heuvel, Shelley Vanderhout, Peter Wong, and Weeda Zabih; Site Investigators: Murtala Abdurrahman, Barbara Anderson, Kelly Anderson, Gordon Arbess, Jillian Baker, Tony Barozzino, Sylvie Bergeron, Dimple Bhagat, Nicholas Blanchette, Gary Bloch, Joey Bonifacio, Ashna Bowry, Anne Brown, Jennifer Bugera, Douglas Campbell, Sohail Cheema, Elaine Cheng, Brian Chisamore, Ellen Culbert, Karoon Danayan, Paul Das, Mary Beth Derocher, Anh Do, Michael Dorey, Kathleen Doukas, Anne Egger, Allison Farber, Amy Freedman, Sloane Freeman, Keewai Fung, Sharon Gazeley, Charlie Guiang, Dan Ha, Curtis Handford, Laura Hanson, Hailey Hatch, Teresa Hughes, Sheila Jacobson, Lukasz Jagiello, Gwen Jansz, Paul Kadar, Tara Kiran, Lauren Kitney, Holly Knowles, Bruce Kwok, Sheila Lakhoo, Margarita Lam-Antoniades, Eddy Lau, Fok-Han Leung, Alan Li, Jennifer Loo, Joanne Louis, Sarah Mahmoud, Roy Male, Vashti Mascoll, Rosemary Moodie, Julia Morinis, Maya Nader, Sharon Naymark, Patricia Neelands, James Owen, Jane Parry, Michael Peer, Kifi Pena, Marty Perlmutar, Navindra Persaud, Andrew Pinto, Tracy Pitt, Michelle Porepa, Vikky Qi, Nasreen Ramji, Noor Ramji, Jesleen Rana, Alana Rosenthal, Katherine Rouleau, Janet Saunderson, Rahul Saxena, Vanna Schiralli, Michael Sgro, Hafiz Shuja, Susan Shepherd, Hafiz Shuja, Barbara Smiltnieks, Cinntha Srikanthan, Carolyn Taylor, Suzanne Turner, Fatima Uddin, Joanne Vaughan, Thea Weisdorf, Sheila Wijayasinghe, Peter Wong, Anne Wormsbecker, Ethel Ying, Elizabeth Young, Michael Zajdman, and Ian Zenlea.

Author Disclosure Statement

No competing financial interests exist.

References

Ogden

, Carroll

, Kit

, Flegal

. Prevalence of childhood and adult obesity in the United States, 2011–2012. JAMA, 2014; 311:806–814.

Roberts

, Shields

, de Groh

, et al. Overweight and obesity in children and adolescents: Results from the 2009 to 2011 Canadian Health Measures Survey. Health Rep, 2012; 23:37–41.

World Health Organization. Report of the Commission on Ending Childhood Obesity. World Health Organization, Geneva, Swizterland. 2016.

Olds

, Maher

, Zumin

, et al. Evidence that the prevalence of childhood overweight is plateauing: Data from nine countries. Int J Pediatr Obes, 2011; 6:342–360.

Rodd

, Sharma

. Recent trends in the prevalence of overweight and obesity among Canadian children. CMAJ, 2016; 188:E313–E20.

Wen

, Gillman

, Rifas-Shiman

, et al. Decreasing prevalence of obesity among young children in Massachusetts from 2004 to 2008. Pediatrics, 2012; 129:823–831.

Skinner

, Perrin

, Skelton

. Prevalence of obesity and severe obesity in US children, 1999–2014. Obesity (Silver Spring), 2016; 24:1116–1123.

Freedman

, Mei

, Srinivasan

, et al. Cardiovascular risk factors and excess adiposity among overweight children and adolescents: The Bogalusa Heart Study. J Pediatr, 2007; 150:12–17.e12.

Kelly

, Barlow

, Rao

, et al. Severe obesity in children and adolescents: Identification, associated health risks, and treatment approaches: A scientific statement from the American Heart Association. Circulation, 2013; 128:1689–1712.

10.

Makkes

, Renders

, Bosmans

, van der Baan-Slootweg

, Seidell

. Cardiometabolic risk factors and quality of life in severely obese children and adolescents in The Netherlands. BMC Pediatr, 2013; 13:62.

11.

Kuczmarski

, Ogden

, Guo

, et al. 2000 CDC growth charts for the United States: Methods and development. Vital Health Stat, 2002:1–190.

12.

Katzmarzyk

, Mason

. Prevalence of class I, II and III obesity in Canada. CMAJ, 2006; 174:156–157.

13.

Valerio

, Maffeis

, Balsamo

, et al. Severe obesity and cardiometabolic risk in children: Comparison from two international classification systems. PLoS One, 2013; 8:e83793.

14.

Dietitians of C, Canadian Paediatric S, College of Family Physicians of C, Community Health Nurses of C, Secker

. Promoting optimal monitoring of child growth in Canada: Using the new WHO growth charts. Can J Diet Pract Res, 2010; 71:e1–e3.

15.

WHO AnthroPlus for personal computers Manual: Software for assessing growth of the world's children and adolescents. Geneva: WHO. 2009. Available at www.who.int/growthref/tools/readme_sas.pdf?ua=1

16.

WHO Child Growth Standards based on length/height, weight and age. Acta Paediatr Suppl, 2006; 450:76–85.

17.

Carsley

, Borkhoff

, Maguire

, et al. Cohort profile: The Applied Research Group for Kids (TARGet Kids!). Int J Epidemiol, 2015; 44:776–788.

18.

NHANES. Third national health and nutrition examination anthropometric procedures video. Prevention, 2003.

19.

Furlong

KRAL

, Kang

, Lebovic

, et al. BMI-for-age and weight-for-length in children 0 to 2 years. Pediatrics, 2016; 138:e20153809.

20.

Roy

, Spivack

, Faith

, et al. Infant BMI or weight-for-length and obesity risk in early childhood. Pediatrics, 2016; 137:pii: e20153492.

21.

Expert panel on integrated guidelines for cardiovascular health and risk reduction in children and adolescents: Summary report. Pediatrics, 2011; 128 Suppl 5:S213–S256.

22.

Anderson

, Lebovic

, Hamilton

, et al. Body mass index, waist circumference, and the clustering of cardiometabolic risk factors in early childhood. Paediatr Perinat Epidemiol, 2016; 30:160–170.

23.

Adolescents NHBPEPWGoHBPiCa. The Fourth Report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents. Pediatrics, 2004; 114(Suppl 2):555–576.

24.

Eisenmann

. On the use of a continuous metabolic syndrome score in pediatric research. Cardiovasc Diabetol, 2008; 7:17.

25.

Martin

, Ul-Haq

, Nicholl

, et al. Cardiometabolic disease and features of depression and bipolar disorder: Population-based, cross-sectional study. Br J Psychiatry, 2016; 208:343–351.

26.

Brogan

, Danford

, Yeh

, Jen

. Cardiovascular disease risk factors are elevated in urban minority children enrolled in head start. Child Obes, 2014; 10:207–213.

27.

Khoury

, Manlhiot

, Gibson

, et al. Universal screening for cardiovascular disease risk factors in adolescents to identify high-risk families: A population-based cross-sectional study. BMC Pediatr, 2016; 16:11.

28.

Ritchie

, Murphy

, Ice

, et al. Universal versus targeted blood cholesterol screening among youth: The CARDIAC project. Pediatrics, 2010; 126:260–265.

29.

Ogden

, Carroll

, Lawman

, et al. Trends in obesity prevalence among children and adolescents in the united states, 1988–1994 through 2013–2014. JAMA, 2016; 315:2292–2299.

30.

, Chandra

, Sinaiko

, et al. Severe obesity in children: Prevalence, persistence and relation to hypertension. Int J Pediatr Endocrinol, 2014; 2014:3.

31.

Garnett

, Baur

, Jones

, Hardy

. Trends in the prevalence of morbid and severe obesity in Australian children aged 7–15 years, 1985–2012. PLoS One, 2016; 11:e0154879.

32.

, Perez

, Wu

, et al. Cardiometabolic risk factors among severely obese children and adolescents in the United States, 1999–2012. Child Obes, 2016; 12:12–19.

33.

Skinner

, Perrin

, Moss

, Skelton

. Cardiometabolic risks and severity of obesity in children and young adults. N Engl J Med, 2015; 373:1307–1317.

34.

Barlow

. Expert committee recommendations regarding the prevention, assessment, and treatment of child and adolescent overweight and obesity: Summary report. Pediatrics, 2007; 120 Suppl 4:S164–S192.

35.

Moyer

. Screening for primary hypertension in children and adolescents: U.S. Preventive Services Task Force recommendation statement. Pediatrics, 2013; 132:907–914.

36.

Anderson

LNMJ

, Lebovic

, Hanley

, et al. Fasting serum lipids and metabolic profile in early childhood. J Pediatr, 2016.

37.

Feise

. Do multiple outcome measures require p-value adjustment?. BMC Med Res Methodol, 2002; 2:8.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.02 MB

0.03 MB