Child Care Exposure Influences Childhood Adiposity at 2 Years: Analysis from the ROLO Study

Abstract

Background:

The first 2 years of life are instrumental for childhood physical development. Factors contributing to childhood obesity are difficult to determine; child care exposure is one to consider, by influencing food preference and physical activity development.

Objective:

To investigate the association of child care exposure with adiposity at 2 years.

Methods:

Data were collected as part of the secondary analysis of the prospective ROLO study (randomized control trial of low glycemic index diet) in Dublin, Ireland. Mothers were recruited antenatally and followed up at 2 years postpartum. Maternal and childhood anthropometric data and lifestyle questionnaires, reporting on child care attendance (defined as nonparental care), exposure (weeks), and infant-feeding practices, were collected.

Results:

Anthropometric measures and lifestyle data were collected for 273 mothers and children aged 2 years, 52.7% of whom attended child care. Child care was predominately provided by a nonrelative (83.7%), either in a crèche (57%) or by a childminder (26.7%). More than half (56.2%) of the children attended child care part-time (≤30 hours/week). Central adiposity measures (abdominal circumference, waist:height ratio) and total adiposity (sum of all skin folds) were significantly elevated in children with increasing time in child care. Children provided with “meals and snacks” had elevated adiposity measures versus those given “snacks or no food.” No difference in the infant-feeding practices was identified between the child care groups.

Conclusions:

Children attending child care have higher total and central adiposity, proportional to exposure. More research is required to investigate this link to appropriately design health promotion and obesity prevention programs targeting children at 2 years.

Introduction

An estimated one in four children in Ireland are overweight or obese, and this is expected to increase year on year.¹ Factors contributing to the childhood obesity epidemic are multifarious and difficult to determine.² Intrinsic factors such as socioeconomic status (SES), parental BMI, and breastfeeding practices, combined with obesogenic influences, including greater snack foods availability, a westernized diet high in sugar and fat, and increased exposure to junk-food advertising, have been previously researched.^2,3 Recent societal changes may also be influential, for instance, the increased proportion of women in the workplace, an increase in parents working nonstandard hours, higher proportion of children in child care, increased dependence on technology, and lack of community structure can contribute to the obesogenic environment.^4–6

Child care attendance in informal (provided by grandparent/family) and formal (centre/crèche-based) settings is linked to increased overweight and obesity in children.^7–9 However, the findings are inconsistent.^10–12 In addition, studies conducted among predominately high¹³ and low¹⁴ SES cohorts have shown an increased propensity for obesity depending on the child care environment, with differential outcomes by gender.¹⁵ Early child care attendance is associated with undesirable infant-feeding practices, including shortened breastfeeding duration and early introduction to solid foods, with negative implications for childhood adiposity.^16,17

Meals provided in child care and child care providers' attitudes and behaviors toward food directly impact the child's development of food choice and eating habits.³ Childhood dietary preferences are based on food exposure, with repeated exposures helping to minimize the rejection of unfamiliar foods.¹⁸ As such, child care providers are highly influential in the development of food preferences via the availability and variety of foods provided, meal structure, eating styles, food social practices, food parenting, and food modeling, that is, observational learning.^3,18 The adoption of dietary patterns with high fruit and vegetable frequency is protective against future morbidities such as cardiovascular disease, diabetes, stroke, and obesity.¹⁸ Dietary and physical activity behaviors developed at the pre-school stage affect lifetime habits and future disease risk.¹⁹ Thus, early child care offers an opportunity for diet and physical activity promotion as a preventative measure to tackle childhood adiposity.

To quantify the burden of childhood obesity at an individual and population level, it is necessary to utilize appropriate body compositional assessments. Using multiple methodologies, such as BMI centiles, circumferences, indices, and skinfolds thickness, an accurate measure of adiposity can be achieved. Although these measures are not without their limitations, ongoing research has enabled the validation with other measures of body fat such as dual-energy X-ray absorptiometry (DXA) and developed reference values and standardized growth curves for comparison in children from infancy to adulthood.^20,21

The aims of this study are to determine the association between child care exposure (defined as time spent in weeks in any nonparental care) and meal provision in child care with adiposity, and to investigate variations in anthropometric measures depending on infant-feeding patterns in child care.

Methods

This is a secondary analysis of a prospective cohort of mothers and children who were enrolled in the randomized control trial of the low glycemic index diet (ROLO) study at the National Maternity Hospital, Dublin, Ireland. Pregnant women who previously delivered a macrosomic infant (birth weight >4 kg) and met the eligibility criteria were recruited between January 2007 and January 2011.²² Participants were randomized to receive low glycemic dietary advice, or usual antenatal care without dietary advice. Both groups were prospectively followed up. The primary aim of this randomized control trial was to prevent recurrence of macrosomia, which was not reduced, and the secondary outcomes were gestational weight gain and glucose intolerance, both of which were reduced in the intervention group.²²

At 2 years postpartum, participants were invited to attend a follow-up appointment with their toddler. From the total cohort of 800 women, 336 mother–child pairs were followed up at 2 years. Ninety-six participants did not complete the questionnaire, whereas an additional 33 participants who did not attend the measurements completed the questionnaire. For the purpose of this analysis, participants who did not complete questionnaires were excluded. Of the total cohort of 273 children, 144 were in the intervention group and 129 were in the control group.

Anthropometry

Maternal weight and height were recorded by medical staff at the first antenatal appointment (10–18 weeks' gestation), and BMI was calculated. At delivery, neonate birthweight was recorded to the nearest 0.1 g. Using the Gestation Networks Bulk Calculator (version 6.2.3), birthweight centile was calculated and corrected for gestational age and gender. Maternal age, weight, height, parity, ethnicity, and neonate gestational age were extracted from medical records. Neonate skinfold thickness measures were added to the study protocol during the progression of the study, as such skinfold measures are not available for all infants at delivery.

At the 2 year follow-up, maternal and child anthropometric and skinfold measures were taken by the research dietitian. Maternal mid-upper arm circumference, height, and weight were measured, and maternal BMI was calculated. Circumferences were measured by using an Seca ergonomic circumference measuring tape, and they were recorded to the nearest 0.1 cm.

Children's height, weight, and circumferences, including mid-upper arm, abdominal, chest, thigh, head, and hip, were recorded for all at the follow-up. Skinfold thickness measurements were later added to the body composition assessments at the 2 year follow-up (n = 204), including biceps, triceps, subscapular, and thigh thicknesses. A Holtain skinfold calliper was used to measure skinfold thickness to the nearest 0.2 mm by the same research dietitian to minimize inter-observational error.

To correct for gender and child age at exam, measurements were converted to Z-scores by using WHO growth standards. Children were classified as “normal” (BMI-for-age Z-score −2 to 1), “at risk of overweight” (BMI-for-age Z-score 1.01–2), “overweight” (BMI-for-age Z-score 2.1–3), and “obese” (BMI-for-age Z-score >3).

Proxy measures for central adiposity included waist:hip ratio, waist:height ratio, and subscapular skinfold:triceps skinfold ratio. Markers of total adiposity included sum of triceps and subscapular, and sum of all skinfolds.

Lifestyle Information

Questions on general health, eating habits, lifestyle, physical activity, infant-feeding practices, and child care arrangements were taken from the Survey of Lifestyle, Attitudes, and Nutrition in Ireland survey.²³ Infant-feeding practices were assessed by using questions on the Lifeways study questionnaires, including: whether or not the child was breastfed, breastfeeding duration, age that drinks other than breast milk were introduced, and age of solid foods introduction.²⁴ Mothers were also asked an open-ended question about their child care arrangements, for example, “Does your child attend child care (crèche, nursery, childminder, etc.)” and if they answered yes, to note the age the child started and describe further. In the present analysis, child care attendance was defined as nonparental care, which may have been in the home or outside the home. Child care exposure was calculated by using age of child care commencement and age of the child at the 2 year exam. Responses were categorized into child care provider (relative vs. nonrelative) and child care setting (crèche, child-minder, relative) for analysis.

Ethics

The ROLO study was conducted according to the guidelines outlined in the Declaration of Helsinki. All procedures involving human subjects were approved by the National Maternity Hospital Research Ethics Committee. Written informed consent was obtained at recruitment antenatally and at the 2-year follow-up appointments.

Statistical Analysis

All statistical analyses were performed by using IBM SPSS Statistics software version 20.0 (SPSS, Inc., Chicago, IL). Continuous descriptive data were tested for normality by the Kolmogorov–Smirnov test and by examining plotting distribution graphs. Parametric variables were analyzed by the independent-sample t-test to detect mean differences between groups at baseline. As birth weight centile was not normally distributed, it was analyzed by the Mann–Whitney U test. The nonparametric variable triceps skinfold to subscapular skinfold ratio was log-transformed for analysis by the independent-sample t-test. p-values are two tailed, with significance defined as p < 0.05. Z-scores were calculated, where possible, to correct for age and gender of the child.

Chi-square tests with Pearson's correlation were used to exam the relationship between categorical variables at baseline, intervention group, gender, maternal education level, SES, and attendance to child care.

To investigate the dose-dependent effect of child care exposure, we completed a Pearson's correlation between child care exposure and 20 anthropometric measures. A cut-off point of p < 0.5 was used to choose those correlations that would undergo further analysis using multiple linear regression. This cut-off point was chosen to prevent false negative results. Child care exposure (predictor) was entered in the model as a continuous variable to assess its effect on the anthropometric measures (outcomes) after being controlled with known cofounders (birth weight, child gender, study group, duration of breastfeeding, child's age at the 2-year exam, maternal BMI at 2-year exam, and maternal education level).

Those anthropometric measures found to be predicted by child care exposure were further analyzed by using an independent-sample t-test to take meal patterns into account. Meals provided in child care were categorized into two groups, “meals and snacks” and “snacks or no food,” which were compared to detect variances in mean anthropometric values.

Results

Two hundred and seventy-three mothers returned for the 2-year follow-up and completed the lifestyle questionnaire; 52.7% of mothers had a child in child care at age 2 years. Of the children attending child care, 83.7% were provided by a nonrelative, 57% attended a formal crèche setting, with a further 26.7% minded by a child-minder and the remaining 16.3% by a relative. The majority (56.2%) of children attended child care part-time, defined as less than a cumulative 30 hours spent in child care per week. Demographics of the total study population, and dichotomized by child care groups, are shown in Table 1.

Table 1.

Baseline Demographics

	Total population			Not attending child care			Attending child care
	n	Mean	SD	n	Mean	SD	n	Mean	SD	p-value
Mother's age at delivery (y)	273	33.2	3.7	129	32.5	4.2	144	33.9	3.1	0.01^*
Mother's weight at first appointment (kg)	273	71.6	12.3	129	72.4	10.8	144	70.8	13.5	0.28
BMI at first appointment (kg/m²)	273	25.9	4.2	129	26.2	3.7	144	25.5	4.5	0.19
Gestation days	247	282.6	7.6	118	282.1	7.8	129	283.2	7.5	0.28
Birth sum of skinfolds (mm)	122	28.1	5.1	54	28.4	4.9	68	27.9	5.2	0.49
Birth weight centile^†	255	80.5	33.7	120	82.4	29.9	135	79.3	32.7	0.36
	n	%		n	%		n	%		p -value
Total cohort	273			129	47.3		144	52.7
ROLO intervention group	144	52.7		66	45.8		78	51.2		0.62
Mother achieved third-level education	150	59.5		63	42.0		87	58.0		0.03^*
Mid-high SES	165	60.4		81	49.1	84		50.9		0.60
Male child	132	48.4		63		47.7	69	52.3		0.88

Continuous data were tested for normality by the Kolmogorov–Smirnov test. Parametric variables were analyzed by the independent-sample t-test to detect between group mean differences.

p-values are two tailed, with significance defined as p < 0.05. Nonparametric variables (†), were analyzed by the Mann–Whitney U test, with significance defined as p < 0.05. Results are presented as median and IQR in place of mean and SD, respectively. Categorical data were analyzed with chi-square tests with Pearson's correlation, considered significant at p < 0.05 (^*).

IQR, interquartile range; ROLO, randomized control trial of low glycemic index diet; SES, socioeconomic status.

Mean maternal BMI at the first antenatal appointment (10–18 weeks' gestation) was 25.9 kg/m² (median 25.02 kg/m², interquartile range 5.24), indicating an overweight population. Mean neonate birthweight centile was 80.5, illustrating that this cohort of children was heavier than the average population matched for age and sex. Mothers with children in child care at 2 years were more likely to be older (p < 0.001) and to have achieved a third-level education (p = 0.03). There were no differences in maternal anthropometry taken at the first antenatal appointment (10–18 weeks' gestation) and neonate measures (recorded at delivery) between child care groups. No difference was found between the ROLO intervention group and the usual care group; they were combined for further analysis.

Child body-composition measures at 2 years are shown in Table 2, presented for the total cohort, and dichotomized by child care attendance. Abdominal circumference was significantly increased in children attending child care (51.2 ± 3.5 cm vs. 50.1 cm ± 3.3, p = 0.01). Waist:height ratio, an indicator of central adiposity, was greater in the group attending child care (0.57 ± 0.04 vs. 0.56 ± 0.04, p = 0.01), as was the sum of all skinfolds (41.5 ± 7.3 cm vs. 39.3 ± 6.7 cm, p = 0.03), indicating an elevated adiposity. Regarding infant feeding, no significant difference was detected for breastfeeding duration, age that other drinks were given, or age of solid introduction (Table 3).

Table 2.

Anthropometric Measures at 2 Years

	Total population					Not attending child care			Attending child care
	n	Min	Max	Mean	SD	n	Mean	SD	n	Mean	SD	p-value
Anthropometry
Weight at 2 years (kg)	243	9.9	18.7	13.1	1.5	113	13.0	1.4	130	13.1	1.6	0.56
Abdominal circumference (mm)	240	41.3	62.0	50.6	3.5	114	50.1	3.3	126	51.2	3.5	0.01^*
Thigh circumference (mm)	240	18.2	37.6	27.6	2.7	113	27.4	2.6	127	27.8	2.7	0.22
Chest circumference (mm)	233	42.0	57.5	50.5	2.5	112	50.5	2.5	121	50.6	2.5	0.71
Hip circumference (mm)	198	43.0	59.8	49.6	2.9	94	49.5	2.6	104	49.7	3.2	0.49
Head circumference for age (Z-score)	236	−1.5	5.2	1.3	1.1	109	1.3	1.2	127	1.4	0.9	0.50
MUAC for age (Z-score)	239	−0.8	3.4	1.3	0.8	113	1.3	0.7	126	1.3	0.8	0.70
Ratio
Waist:height ratio	240	0.45	0.68	0.56	0.04	114	0.56	0.04	126	0.57	0.04	0.01^*
Waist:hip ratio	198	0.8	1.2	1.1	0.1	94	1.1	0.1	104	1.0	0.1	0.06
Weight for height (Z-score)	240	−1.9	3.2	0.4	1.0	111	0.4	0.9	129	0.5	1.1	0.49
BMI
BMI for age (Z-score)	241	−2.2	3.1	0.4	1.0	112	0.4	0.9	129	0.4	1.1	0.64
Height for age (Z-score)	240	−2.7	4.4	0.7	1.1	113	0.6	0.9	127	0.7	1.2	0.49
Weight for age (Z-score)	240	−1.7	3.3	0.7	0.9	111	0.7	0.9	129	0.8	0.9	0.47
Skinfold measures
Triceps skinfold thickness (mm)	209	4.0	17.6	10.1	2.4	97	9.8	2.3	112	10.4	2.5	0.08
Biceps skinfold thickness (mm)	210	3.4	15.4	7.6	2.2	97	7.5	2.1	113	7.8	2.2	0.45
Thigh skinfold thickness (mm)	207	6.0	28.0	15.6	4.8	96	15.0	4.6	111	16.2	4.9	0.08
Subscapular skinfold thickness (mm)	206	4.0	12.0	7.2	1.7	95	7.2	1.8	112	7.3	1.7	0.59
TSF to SSF ratio^†	204	−0.4	0.2	−0.2^†	0.2^†	93	−0.1^†	0.2^†	111	−0.1^†	0.2^†	0.79
Sum of all skinfolds (mm)	204	23.8	60.6	40.5	7.1	94	39.3	6.7	110	41.5	7.3	0.03^*
Sum of TSF and SSF (mm)	206	8.0	28.0	17.3	3.4	95	16.9	3.3	111	17.7	3.5	0.15

Results are presented as median and IQR in place of mean and SD, respectively.

Continuous data were tested for normality by the Kolmogorov–Smirnov test. Parametric variables were analyzed by the independent-sample t-test to detect between group mean differences. p-values are two-tailed, with significance defined as p < 0.05 (^*). The nonparametric variable, TSF to SSF ratio, was log transformed for analysis (†).

MUAC, mid-upper arm circumference; SSF, subscapular skinfolds; TSF, triscapular skinfolds.

Table 3.

Breastfeeding Practices Recorded at 2 Years

	Total population					Not attending child care			Attending child care
	n	Min	Max	Median	IQR	n	Median	IQR	n	Median	IQR	p-value
Breastfeeding duration	171	0.3	117.4	21.7	31.1	72	21.7	34.1	99	21.7	31.1	0.98
Age stopped breastfeeding	172	0.3	117.4	21.7	30.4	73	21.7	30.6	99	21.7	30.4	0.86
Age drinks other than breast milk	171	0.0	84.8	15.2	22.1	71	17.4	20.1	100	12.0	23.1	0.70
Age given solid foods	264	10.0	52.2	21.7	8.7	125	21.7	8.7	139	21.7	6.5	0.27

All measures expressed in weeks.

The results from the series of adjusted linear regressions exploring the association between child care exposure and adiposity are shown in Table 4. Abdominal circumference, thigh skinfolds, sum of all skinfolds, and waist:height ratio were all significantly associated with child care exposure after adjusting for confounders (p = 0.010, 0.025, 0.019, and 0.036, respectively). Child care exposure had the strongest standardized beta coefficient in each of the models, indicating that child care exposure had the greatest unique contribution toward the anthropometric outcomes among the independent variables. After applying the conservative Bonferroni correction to our 20 anthropometric outcomes of interest, none of them would remain significant at the multiplicity-adjusted p-value threshold of 0.0025 (0.05/20). As our outcomes are correlated with each other, this adjustment is expected to over-correct.

Table 4.

Child Care Exposure Effect on Anthropometric Measures Entered in MLR

			Confidence intervals
	Beta	p-value	Lower	Upper	Adjusted R²	F-value	p-value
Abdominal circumference (mm)	0.215	0.010^*	0.005	0.038	0.078	2.473	0.016^*
Thigh skinfolds (mm)	0.194	0.025^*	0.003	0.050	0.061	2.081	0.042^*
Sum of all skinfolds (mm)	0.203	0.019^*	0.007	0.077	0.052	1.921	0.062
Waist:height ratio	0.178	0.036^*	0.000	0.000	0.040	1.721	0.099

Analyzed by a series of linear regression models on the association of child care exposure, with each row variable as the outcome, and adjusted for duration of breastfeeding, age at the 2-year exam, birth weight, mother's BMI at 2-year exam, child gender, study group, and educational status.

Significance at p < 0.05 (^*).

MLR, multiple linear regression.

Data from lifestyle questionnaires gathered information on the provision of meals to children in child care. Of the total attending child care group (n = 139), 76.3% of children received all their meals and snacks while attending child care and 23.7% received either snacks or no food. The t-test analysis comparing meals provided found a significant difference in anthropometric measures between those given “meals and snack” and those given “snacks or no food” (Table 5). Chest circumference, abdominal circumference, waist:height ratio, thigh skinfolds, and sum of all skinfolds were found to be significantly increased in those provided with more meals (p = 0.001, 0.006, 0.012, 0.022, and 0.037, respectively).

Table 5.

Mean Anthropometric Data Based on Meals Provided in Child Care

	Meals & snacks		Snacks & no food
	Mean	SD	Mean	SD	p-value
Weight at 2 years (kg)	13.27	1.44	12.72	2.01	0.097
Weight for age	0.84	0.88	0.52	1.22	0.119
Weight for height	0.54	1.03	0.16	1.06	0.082
Chest circumference (mm)	51.05	2.38	49.36	2.50	0.001^*
Abdominal circumference (mm)	51.75	3.25	49.73	3.89	0.006^*
Thigh circumference (mm)	28.06	2.79	27.05	2.50	0.081
Hip circumference (mm)	50.04	3.17	48.99	3.12	0.149
Thigh skinfolds (mm)	16.84	4.78	14.37	4.69	0.022^*
Sum of skinfolds (mm)	42.33	6.97	38.97	7.53	0.037^*
Waist:height ratio	0.58	0.04	0.55	0.03	0.012^*

Independent-sample t-test analysis of meals provided in child care. Meals provided were categorized into two groups: children provided with “meals and snacks”; whereas in child care, they were compared with those given “snacks” or “no food given” to detect variances in mean anthropometric values.

Significance at p < 0.05 (^*).

Discussion

This study comparatively examines adiposity levels in children at 2 years and their association with child care exposure, while considering meals provided and infant-feeding practices. Correlation analysis between child care exposure and anthropometric measures, adjusted for known confounders, found that child care exposure was positively associated with central adiposity, measured by abdominal circumference and waist:height ratio, and total adiposity, indicated by sum of all skinfolds. Children of higher SES and maternal education usually have lower adiposity.²⁵ Given that this cohort was an older, highly educated socioeconomic group, it would be expected that the children attending child care would tend to be a leaner population. However, this cohort showed raised central and total adiposity measures, which could indicate an association with child care exposure. Other evidence also indicates that children in child care have higher anthropometric and body compositional measures. An investigation by Mackle et al. found that early child care attendance was correlated with increased waist circumference and BMI at school entry age in girls; however, this was not shown in boys.²⁶ Furthermore, Benjamin found increased weight for length at 1 year and elevated BMI at 3 years in children who had attended early child care,¹⁶ whereas Gubbels et al. found that child care attendance at age 1–2 years was predictive of BMI at 2 years and BMI variation between 1 and 2 years.²⁷ A subsequent study by Benjamin et al. discovered that child care attendance was associated with infant total adiposity, in terms of increased BMI and a greater odds ratio of overweight/obesity at 12 months.⁸ These findings are in agreement with our results, as increased total skinfolds thickness may explain the variation in BMI.

Hara et al. identified waist:height ratio as the strongest predictor for an unfavorable lipid profile and cardiovascular risk in school children.²⁸ Furthermore, Freedman et al. discovered that central fat distribution is a strong predictor of deranged lipid profile in children from age 5, independent of weight, height, sex, age, and ethnicity.²⁹ The negative implications for children who become overweight/obese are well established; they experience greater illness frequency, higher blood pressure, increased metabolic dysfunction, and poorer health status.³⁰ Children who gain weight quickly in the first few years of life have greater risk of early childhood adiposity.¹⁶ This adiposity continues into later life, causing a cascade of future morbidity risk.³⁰ In fact, it has been hypothesized that the rising childhood/adolescent obesity prevalence and its clinical implications may result in parents outliving their children for the first time in modern times.³¹ Thus, our finding of higher central adiposity and total adiposity with increased time spent in child care is of public health importance.

Child care has been found to affect infant-feeding practices, as typically mothers stop breastfeeding once the infant has commenced child care, thus increasing their obesity risk.¹⁰ However, infant-feeding practices, including breastfeeding duration, age at which drinks (other than water and milk) were given, and age of solid foods introduction, were comparable between child care attending and nonattending groups in this study. This is in keeping with similar results found by Pearce et al., who found that breastfeeding was not protective against increased adiposity in children attending child care at 3 years.⁷ This contradicts other findings that determined that child care attendance was associated with a reduced likelihood of being breastfed and an increased likelihood of premature introduction to solid foods.¹⁷ However, breastfeeding data gathered in this study were not specified between exclusive, expressed milk and combined formula feeding, which may have illustrated some distinction between the child care groups.

In this study, 52.7% of children were attending some form of child care at 2 years. This is significantly higher than figures from the OECD, which reported 28.8% of children under 3 years attending child care.³² Qualitative data gathered found that within our sample the majority of child care was provided by a nonrelative, in a formal crèche setting where all meals and snacks were provided. Data analysis suggests that children provided with meals and snacks have increased adiposity measures compared with those given snacks only or no food while in child care. This illustrates a direct relationship between the meals provided in care and increased adiposity and suggests that meals provided in child care are implicated in the role of child care in obesity. This is in keeping with a study that found that almost 60% of children in center-based care received a sweet snack (e.g., cookie, donut) and did not meet the recommendations for vegetables, fruits, grains, and meat.³³ The availability of meals high in saturated fats, and lacking essential vitamins and minerals is another reported concern.¹⁴ Other studies have also found that informal child care (provided by grandparent/family) has a tendency for increased rates of obesity.^7,13,34 The influence of caregivers on children's diet and physical activity has previously been investigated, with the significance of this strengthened by reduced time spent with parents.³ As an increasing number of children spend a significant proportion of time in child care, the more pivotal the influence of child care attendance becomes.⁵ Literature suggests that child care providers are increasingly becoming as influential as parents on forming the child's food preferences.¹⁸ As such, “other people minding their children” was identified by Irish parents of children aged 1–4 years as a barrier to providing a balanced diet for their children.³⁵

Diet preference is formed in children from age 2; as such, the pre-school age represents the formative years during which diet preference is developed.^18,19 Takahashi et al.³⁶ have identified hours spent undertaking physical activity in child care at 3 years as a strong influence on child adiposity and development. Therefore, a window of opportunity exists to initiate healthy diet and physical activity behaviors at pre-school age.³⁷ Previous health interventions in pre-school children have seen positive reductions in child weight.³⁸ An obesity prevention program specially designed for the pre-school setting, Nutrition and Physical Activity Self-assessment for Child Care (NAP SACC), has been found to be effective in improving nutrition and physical activity.³⁹ Our data combined with the existing research on young children in child care highlight the need for public health investments to enhance nutritional and physical activity standards in formal child care settings. Currently, in Ireland, “Food and Nutrition Guidelines for Pre-School Services” for dietary guidelines for child care providers exist, and “Get active Ireland,” coordinated by the Health Service Executive, has published physical activity recommendations for child care providers.^40,41 These documents act only as guidelines, as adherence to these recommendations are not routinely monitored or assessed. It should be recommendable that policy makers consider implementing a program such as the NAP SACC to improve nutrition and physical activity in the child care environment.

A limitation of this study is that the participants belong to a relatively higher socioeconomic group, and more than half of the original ROLO cohort achieved a third-level education and had a higher BMI score. Although there was no difference between those children attending and not attending child care, study group and birth weight were controlled in the further analysis to account for this limitation. The open-ended question to obtain information on child care arrangement is also considered a limitation of this study. The child care arrangement listed at 2 years was taken as the predominant child care arrangement, as there were insufficient data to analyze child care arrangements before the 2 year time point. Information on breastfeeding exclusivity could not differentiate between “any” and “exclusive” breastfeeding in the questionnaire.

The primary strength of this study is the combination of various anthropometric assessments and skinfold measurements to gain a reliable indication of child adiposity and weight status. This study is the first to use these body compositional measures to explore the influence of child care settings in children aged 2 years.

Conclusion

In this study, an increase in child care exposure is associated with central and total adiposity. This effect was not mediated by infant-feeding practices, such as breastfeeding duration. Meal provision in child care settings was also found to be significant, with those given “meals and snacks” illustrating increased adiposity. This study adds to the growing body of evidence that indicates that public health investments are required in child care settings to enhance the provision of healthy food and physical activity for young children. As such, early childhood health promotion presents an opportunity for future obesity prevention and reduction of the ever-increasing burden of obesity-related comorbidities and mortalities.

Footnotes

Acknowledgments

The ROLO study was funded by the Health Research Board Ireland, the Health Research Centre for Health and Diet Research, the National Maternity Hospital Medical Fund, and the European Union's Seventh Framework Programme (FP7/2007–2013), project 59 EarlyNutrition under grant agreement no. 289346. The Current Controlled Trials Registration Number was ISRCTN54392969.

Author Disclosure Statement

No competing financial interests exist.

References

Heinen

, Murrin

, Daly

, et al. The Childhood Obesity Surveillance Initiative (COSI) in the Republic of Ireland: Findings from 2008, 2010 and 2012. 2014; 1–52.

Ang

, Wee

, Poh

, Ismail

. Multifactorial influences of childhood obesity. Curr Obes Rep, 2012; 2:10–22.

Roblin

. Childhood obesity: Food, nutrient, and eating-habit trends and influences. Appl Physiol Nutr Metab, 2007; 32:635–645.

Champion

, Rumbold

, Steele

, et al. Parental work schedules and child overweight and obesity. Int J Obes, 2012; 36:573–580.

Horgan

. Childcare in Ireland: Themes and issues. Irish J Appl Soc Stud, 2001; 2:8.

McDonnell

, Doyle

. Maternal employment, childcare and childhood overweight during infancy. 2014. www.ucd.ie/t4cms/WP14_16.pdf (Last accessed November 11, 2016).

Pearce

, Li

, Abbas

, et al. Is childcare associated with the risk of overweight and obesity in the early years findings from the UK millennium cohort study. Int J Obes, 2010; 34:1160–1168.

Neelon Benjamin

, Andersen Schou

, Morgen Schmidt

, et al. A. Early child care and obesity at 12 months of age in the Danish National Birth Cohort. Int J Obes, 2015; 39:33–38.

Geoffroy

M-C

, Power

, Touchette

, et al. Childcare and overw eight or obesity over 10 years of follow-up. J Pediatr, 2013; 62:753–758.

10.

Costa

, Adams

, Phillips

, Benjamin Neelon

. The relationship between childcare and adiposity, body mass and obesity-related risk factors: Protocol for a systematic review of longitudinal studies. Syst Rev, 2016; 5:141.

11.

Lumeng

, Gannon

, Appugliese

, et al. Preschool child care and risk of overweight in 6- to 12-year-old children. Int J Obes Relat Metab Disord, 2004; 29:60–66.

12.

Zahir

, Heyman

, Wojcicki

. No association between childcare and obesity at age 4 in low‐income Latino children. Pediatr Obes, 2013; 8:e24–e28.

13.

Lin

, Leung

, Hui

, et al. Is informal child care associated with childhood obesity? Evidence from Hong Kong's “Children of 1997” birth cohort. Int J Epidemiol, 2011; 40:1238–1246.

14.

McGrady

, Mitchell

, Theodore

, et al. Preschool participation and BMI at kindergarten entry: The case for early behavioral intervention. J Obes, 2010; 2010.

15.

McLaren

, Zarrabi

, Dutton

, et al. Child care: Implications for overweight/obesity in Canadian children?. Chronic Dis Inj Can, 2012; 33:1–11.

16.

Benjamin

, Rifas-Shiman

, Taveras

, et al. Early child care and adiposity at ages 1 and 3 years. Pediatrics, 2009; 124:555–562.

17.

Kim

, Peterson

. Association of infant child care with infant feeding practices and weight gain among us infants. Arch Pediatr Adolesc Med, 2008; 162:627–633.

18.

Nicklas

, Baranowski

, et al. Family and child-care provider influences on preschool children's fruit, juice, and vegetable consumption. Nutr Rev, 2001; 59:224–235.

19.

Skinner

, Carruth

, Bounds

, et al. Do food-related experiences in the first 2 years of life predict dietary variety in school-aged children?. J Nutr Educ Behav, 2002; 34:310–350.

20.

Wells

JCK

, Williams

, Chomtho

, et al. Body-composition reference data for simple and reference techniques and a 4-component model: A new UK reference child. Am J Clin Nutr, 2012; 96:1316–1326.

21.

Wohlfahrt-Veje

, Tinggaard

, Winther

, et al. Body fat throughout childhood in 2647 healthy Danish children: Agreement of BMI, waist circumference, skinfolds with dual X-ray absorptiometry. Eur J Clin Nutr, 2014; 68:664–670.

22.

Walsh

, McGowan

, Mahony

, et al. Low glycaemic index diet in pregnancy to prevent macrosomia (ROLO study): Randomised control trial. Br Med J, 2012; 345:e5605.

23.

Harrington

, Perry

, Lutomski

, et al. SLÁN 2007: Survey of lifestyle, attitudes and nutrition in Ireland. Dietary Habits of the Irish Population. Psychol Rep, 2008; 6.

24.

O'Mahony

, Fallon

, Hannon

, et al. The Lifeways Cross-Generation Study: Design, recruitment and data management considerations. Ir Med J, 2007; 100:suppl:3–6.

25.

Cameron

, Spence

, Laws

, et al. A review of the relationship between socioeconomic position and the early-life predictors of obesity. Curr Obes Rep, 2015; 4:350–362.

26.

Mackle

. Early child care and adiposity at kindergarten entry, Electronic Thesis and Dissertation Repository, 2012, Vol. Paper 660. http://ir.lib.uwo.ca/etd/660 (Last accessed November 11, 2016).

27.

Gubbels

, Kremers

SPJ

, Stafleu

, et al. Child-care use and the association with body mass index and overweight in children from 7 months to 2 years of age. Int J Obes, 2010; 34:1480–1486.

28.

Hara

, Saitou

, Iwata

, et al. Waist-to-height ratio is the best predictor of cardiovascular disease risk factors in Japanese schoolchildren. J Atheroscler Thromb, 2002; 9:127–132.

29.

Freedman

, Serdula

, Srinivasan

, Berenson

. Relation of circumferences and skinfold thicknesses to lipid and insulin concentrations in children and adolescents: The Bogalusa Heart Study. Am J Clin Nutr, 1999; 69:308–317.

30.

Must

, Strauss

. Risks and consequences of childhood and adolescent obesity. Int J Obes Relat Metab Disord, 1999; 23:S2–S11.

31.

Olshansky

, Passaro

, Hershow

, et al. A potential decline in life expectancy in the United States in the 21st century. N Engl J Med, 2005; 352:1138–1145.

32.

OECD Social Policy Division Directorate of Employment Enrolment in childcare and pre-schools.. 2014; 1–6. www.oecd.org/els/soc/PF3_2_Enrolment_childcare_preschool.pdf (Last accessed November 11, 2016).

33.

Ball

, Benjamin

, Ward

. Dietary intakes in North Carolina child-care centers: Are children meeting current recommendations?. J Am Diet Assoc, 2008; 108:718–721.

34.

Maher

, Li

, Carter

, Johnson

. Preschool child care participation and obesity at the start of kindergarten. Pediatrics, 2008; 122:322–330.

35.

Walton

. National Pre-School Nutrition Survey. Summary Report on: Food and Nutrient Intakes, Physical Measurements and Healthy Eating. 2012; 1–36. www.iuna.net/wp-content/uploads/2012/06/Summary_Report_National_PreSchool_Nutrition_Survey_June_2012.pdf (Last accessed November 11, 2016).

36.

Takahashi

, Yoshida

, Sugimori

, et al. Influence of behavioral and environmental factors on the development of obesity in three-year-old children—A case-control study based on Toyama Study. Environ Health Prev Med, 1997; 2:74–78.

37.

Ward

, Bélanger

, Donovan

, et al. Correlates, determinants, and effectiveness of childcare educators' practices and behaviours on preschoolers' physical activity and eating behaviours: A systematic review protocol. Syst Rev, 2015; 4:18.

38.

Bluford

DAA

, Sherry

, Scanlon

. Interventions to prevent or treat obesity in preschool children: A review of evaluated programs. Obesity, 2007; 15:1356–1372.

39.

Ward

, Benjamin

, Ammerman

, et al. Nutrition and physical activity in child care: Results from an environmental intervention. Am J Prev Med, 2008; 35:352–356.

40.

Department of Health and Children Fact sheet for Childcare Providers 2006, 2006. www.getirelandactive.ie/Preschoolers/Resources/Fact-sheet-for-Childcare-Providers.pdf (Last accessed November 11, 2016).

41.

Department of Health and Children Food and Nutrition Guidelines for Pre-School Services. 2004. http://health.gov.ie/wp-content/uploads/2014/03/Food-and-Nutrition-Guidelines-for-Pre-School-Services.pdf (Last accessed November 11, 2016).