19th January 2018, Volume 131 Number 1468

Hajar Mazahery, Carolyn Cairncross, Cathryn Conlon, Lisa Houghton, Jane Coad, Carlos Camargo Jr, Cameron Grant, Pamela von Hurst

Cows’ milk, the most commonly consumed dairy product among children, is a rich source of macronutrients (including protein, carbohydrate and fat) and micronutrients (eg, vitamins B2, B12 and A, and minerals calcium and potassium).1 Milk consumption has been associated with better diet quality and with meeting dietary nutrient intake requirements.2–4 Milk consumption has also been shown to favourably affect bone health in children.5 Accrual of good peak bone mass by the end of the growth period has been shown to have a protective effect on low bone mineral density and fracture risk early in life, and bone loss and osteoporosis later in life.6–9 Furthermore, consumption of milk/dairy products during childhood has been shown to be positively related to linear growth10,11 and inversely related to body mass index (BMI),12 adiposity12,13 and the risk of overweight and obesity,13,14 all of which are important indicators of nutritional status and health.

The New Zealand Ministry of Health recommends that children and young people from age two to 18 years consume 2–3 servings of milk or dairy products each day, preferably from low- or reduced-fat options.1 In the light of the rising prevalence of paediatric obesity and the significance of obesity in the aetiology of non-communicable diseases (eg, type 2 diabetes),15 reduced- or low-fat milk and milk products are recommended from the age of two years because they provide less energy and fat (including saturated fat) and more protein and calcium than regular fat varieties.1 Despite the repeated emphasis on the importance of milk and dairy products in children’s diets, consumption is below the recommendations for a large proportion of the population in high-income countries including Australia16,17 and New Zealand.18 Based on the New Zealand National Children’s Nutrition survey conducted in 2008/2009, approximately 22% of children and young people never consume milk.18 Furthermore, low- or reduced-fat milk was consumed by approximately 25% of New Zealand children in the 2002 survey.19

Prior studies have shown that milk consumption is influenced by age, sex, ethnicity,18,19 parental education,17 household size,20 socioeconomic status,21 medical conditions (lactose intolerance and milk allergy)22,23 and weight status.24 Some of these factors (including, but not limited to, ethnicity, socioeconomic status and overweight/obesity) are also associated with the type of milk consumption during childhood.19,25–27 The consumption of low- or reduced-fat milk has been consistently reported to be higher in some ethnic groups and in children of higher socioeconomic status. For instance, the New Zealand 2002 National Children’s Nutrition survey suggests that low- or reduced-fat milk consumption is higher in New Zealand European than Māori and Pacific children.19 Among children and adolescents in the US, low- or reduced-fat milk consumption is higher in children from households of higher versus lower socioeconomic status and in children of non-Hispanic white ethnicity compared with non-Hispanic black and Hispanic ethnicity.25 However, the relationship between overweight/obesity and milk fat content is less clear. While some researchers reported a positive relationship between regular milk fat content and increased risk of overweight/obesity,26 some have found no relationship28 and others reported an inverse relationship.27,29 Given these discrepancies in epidemiological data, lack of randomised controlled trials and the current emphasis of dietary guidelines on promoting low- or reduced-fat milk consumption in children, further investigation is warranted. To the best of our knowledge, no epidemiological research exists addressing both the pattern of type of milk consumption and its determinants in New Zealand preschool-aged children.

Therefore, the aims of this cross-sectional analysis were: (1) to investigate the type of milk usually consumed among preschool children living in New Zealand; and (2) to investigate the sociodemographic characteristics that correlate with milk type consumed by these children.

Methods

We investigated sociodemographic factors and medical conditions associated with the type of milk consumption in preschool children living in New Zealand, and explored the relationship between type of milk consumed and overweight/obesity. Data were drawn from a nationwide cross-sectional study named Te Ra Whakaora (Sunshine and Health). Study design and data collection methods for this study are described briefly below, with further details reported elsewhere.30

Study design

Preschool children aged 2 to <5 years from throughout New Zealand were enrolled in this study from August to October 2012. It is important to note that the data is five years old, though to the best of our knowledge, this study is the latest nationwide investigation of milk consumption pattern in New Zealand preschool children and no such a large study has been published since then. Data were collected via paper or online questionnaires. The 59-item study questionnaire comprised of questions describing the child’s demographics, medical history and eating habits. Questions were directed to the main caregiver/parent. Anthropometric measurements (weight and height) were obtained during the child’s visit at testing centres. The collection of data took place at 54 testing centres (49 community pharmacies, Human Nutrition Unit at Massey University, Auckland, University of Otago and Dunedin, two Auckland health services who carry before school check for four-year old children, and a Wellington general practice clinic) in 17 New Zealand cities and towns with a good geographical coverage achievement. The study participants were recruited using a variety of techniques including sending emails advertising the study, displaying flyers and posters in early childhood education centres and kindergartens in close proximity to testing centres, distributing press releases to media outlets, and interviews on two radio stations and a Sunday evening TV news programme. The primary recruitment method was the purchase of access to an email database of the online services company that contains 160,000 families with newborn and young children (up to seven years old) and provides products and information for parents. Over a 10-week period, 2,267 families expressed their interest, of whom 1,812 families enrolled in the study and 1,329 families provided information regarding sociodemographic characteristics and milk consumption pattern. Ethical approval for this study was obtained from the Health and Disability Ethics Committee, Northern Region, reference number NTX/12/04/036, and a parent/caregiver of each child provided written informed consent for participation of their child in the study.

Dependent variables

Milk consumption was described in this study using a multiple response question. Parents were asked to indicate if the child usually consumed cows’ milk and to specify the type of milk the child usually consumed. The latter included the following options: Standard (blue top), Lite (light blue top), Trim (green top), Anchor Calci Plus, Anchor Mega, Anchor Superblue and Meadowfresh Calcitrim or Calcistrong. This question was followed by a free-text option for the recording of brand name or type if it was not included in the list. These options together with options that the parents/caregivers provided were then collapsed into two categories according to their sugar and fat content; 1) fat- and sugar-containing dairy milks that comprised of standard fat cows’ milk, unpasteurised/raw cow’s milk, full-fat flavoured cows’ milk and full-fat goat’s milk, and 2) plain low- or reduced-fat cows’ milks (including both fortified and non-fortified cows’ milk). Throughout this paper, we use the term “full-fat cows’ milk” to refer to all “fat- and sugar-containing milks” because a small number of children consumed other types of milk (eg, goat’s milk, n=6 and flavoured milk, n=13) and their effect on the findings would be negligible. Furthermore, the significance of independent variables on the consumption of each individual milk type could not be determined due to a very small number of children in each milk type category. Please refer to Supplementary Tables 1 and 2 for the nutritional content of different types of dairy milk and type of milk consumed by children. For those children who consumed more than one type of milk, the priority was given to the full-fat cows’ milk (eg, if a parent/caregiver reported child consuming both standard fat and light blue top cows’ milk, the child was assigned to the “full-fat cows’ milk” group). To investigate the effect of including children who drink both low- and high-fat milk as high-fat milk drinkers, sensitivity analyses were performed while excluding those children. The findings are presented in Supplementary Table 3. Exclusion of those children had no effect on findings (comparable ORs), and therefore the results presented in the results section includes those drinking both low- and high-fat milk as high-fat milk drinkers.

Independent variables

The questionnaire content allowed for description of the child’s demographics and medical conditions and mother’s educational level. Parents/caregivers were asked to record their child’s age and sex, and list all ethnicities they identified for their child. For analytical purposes, ethnicity was then assigned using the following prioritisation; Māori > Pacific > non-European other > New Zealand European. Parents/caregivers were asked for information on the number of children in household as a measure of household size, with the number then collapsed into two groups (households with <3 children or ≥3 children). Parents/caregivers also provided information on who cared for their child during the day, and if attending a formal preschool education centre (eg, daycare, preschool or kindergarten) and for how many hours per week. Factors affecting milk consumption in daycare centres have been previously studied,31 but data regarding the relationship between milk consumption and daycare attendance are lacking.

Parents/caregivers described their highest educational qualification obtained: intermediate or secondary school with no qualification; school certificate (New Zealand National Certificate of Education Achievement level 1, NZ NCEA, or similar overseas qualification); completed high school certificate (NCEA level 2 and 3 or similar overseas qualification); full-time study since leaving school; and university degree or diploma. Parental/caregiver education was then classified into three groups; no secondary qualification, secondary qualification or post-secondary qualification.

Socioeconomic status was determined using the New Zealand Index of Social Deprivation, a census-based measure.32 Geo-coding of residential address was performed using ArcGIS geographic information system software (Eris, Redlands, California), along with the subsequent allocation of deprivation index. Each household was allocated to a national household deprivation index of 1 to 10 (with 1 being the least deprived and 10 the most deprived decile of households). These deprivation index scores were then collapsed into five categories; deprivation indices 1 & 2, 3 & 4, 5 & 6, 7 & 8 and 9 & 10 together. Residential address was used to establish latitude, with the country then separated into three regions accordingly, northern (34-38oS), central (39-42oS) and southern (43-46oS) New Zealand.

Parents/caregivers were asked if the child had ever had any food allergies and, if they reported that their child had a food allergy, were provided with a short list of foods commonly associated with allergic reactions in preschool-aged children (eg, milk, egg, peanut, seafood and an open-text option). Finally, parents/caregivers were asked if their child was diagnosed with any specific conditions, including celiac disease, Crohn’s disease, cystic fibrosis and other inflammatory bowel disorders. Lactose intolerance and/or milk sensitivity symptoms overlap with symptoms of these diseases.33–36

Weight and height were measured without shoes by the trained testers, using household non-calibrated scales and wall-mounted measuring devices. Body mass index (BMI) percentile for age and sex was calculated, and then classified into three groups; normal, overweight or obese (based on the cut-offs provided by Cole et al (2000)37).

Statistical analyses

Statistical analyses were performed using IBM SPSS version 21.0 (IBM Corp; Armonk, NY, USA). Descriptive statistics were used to report the proportion of preschool children consuming milk and type of milk and other categorical independent variables. We investigated the relationship between potential variables and milk and type of milk consumption using cross-tabulation. Predictors of cows’ milk consumption and cows’ milk type consumption were assessed employing binary logistic regression analysis (univariable and multivariable). Household size and number of children in household were not included in the regression analyses because they were not associated with type of cows’ milk consumption (preliminary cross-tabulation analysis). To avoid the violation of multicollinearity and incomplete information from the predictors (due to many variables with many categories) and because there was a strong relationship between maternal education and ethnicity (a larger proportion of New Zealand European versus Pacific children (82% vs 58% P<0.0001) had mothers with post-secondary education, we ran two regression analyses including either of the mentioned independent variables and report the results of each in separate tables. Imbalanced data with binary outcome variables are associated with biases in the estimated probability of an event.38 We investigated the models to determine if all the assumptions were met and which model had a better model fit (assessing -2 log likelihood). We also added interaction terms into the models to investigate for interaction effects between variables but no significant results were observed. Associations were described using adjusted odds ratios (OR) and 95% confidence intervals (CI).

Results

Participants’ socio-demographic characteristics by cows’ milk consumption

Table 1 shows the characteristics of the study population by cows’ milk consumption. A larger proportion of children were within the 2–3 year-old category, were of New Zealand European ethnicity, had mothers with post-secondary qualification, lived in households with <3 children, lived in the northern region or spent 11–20 hours/week rather than less or more time in daycare/preschool/kindergarten. A larger proportion were of lower deprivation indices than higher deprivation indices (the highest deprivation index is the most deprived area).

Table 1: Socio-demographic characteristics of preschool children by cows’ milk consumption.

Variables

Total (n=1,329)

Cows’ milk consumption

P value*

Yes (n=1,166)

No (n=163)

Age

 

 

 

0.80

2-year-old

496 (37)

432 (87)

64 (13)

 

3-year-old

448 (34)

393 (88)

55 (12)

 

4-year-old

385 (29)

341 (89)

44 (11)

 

Sex

 

 

 

0.36

Female

684 (49)

574 (89)

74 (11)

 

Male

681 (51)

592 (87)

89 (13)

 

Ethnicity

 

 

 

0.06

New Zealand European

926 (70)

801 (87)

125 (14)

 

Māori

174 (13)

163 (94)

11 (6)

 

Pacific

44 (3)

40 (91)

4 (9)

 

Other

185 (14)

162 (88)

23 (12)

 

Maternal education

 

 

 

0.002

Post-secondary qualification

1,072 (81)

925 (86)

147 (14)

 

Secondary qualification

124 (9)

113 (91)

11 (9)

 

No secondary qualification

133 (10)

128 (96)

5 (4)

 

Socioeconomic statusa

 

 

 

0.30

Deprivation index 1 & 2

343 (26)

290 (85)

53 (16)

 

Deprivation index 3 & 4

331 (25)

297 (90)

34 (10)

 

Deprivation index 5 & 6

287 (22)

253 (88)

34 (12)

 

Deprivation index 7 & 8

227 (17)

202 (89)

25 (11)

 

Deprivation index 9 & 10

141 (11)

124 (88)

17 (12)

 

Residential region

 

 

 

0.11

Northern

651 (49)

559 (86)

92 (14)

 

Central

378 (28)

341 (90)

37 (9)

 

Southern

300 (23)

266 (89)

34 (11)

 

Household size (number of children)

 

 

 

0.52

<3

942 (71)

823 (87)

119 (13)

 

≥3

387 (29)

343 (89)

44 (11)

 

Daycare attendance (hours)

 

 

 

0.53

0

192 (14)

172 (90)

20 (10)

 

1–10

257 (19)

220 (86)

37 (14)

 

11–20

460 (35)

400 (87)

60 (13)

 

21–30

234 (18)

206 (88)

28 (12)

 

30+

186 (14)

168 (90)

18 (10)

 

Cows’ milk allergy

 

 

 

<0.0001

No

1,225 (92)

1,120 (91)

105 (9)

 

Yes

104 (8)

46 (44)b

58 (56)

 

Medical conditions (that might affect milk consumption)c

 

 

 

0.23

No

1,322 (99)

1,161 (88)

161 (12)

 

Yes

7 (1)

5 (71)

2 (29)

 

BMI (calculated for age and sex) (n=1,315)

 

 

 

0.13

Normal

1,094 (83)

952 (87)

142 (13)

 

Overweight

173 (13)

159 (92)

14 (8)

 

Obese

48 (4)

44 (92)

4 (8)

 

BMI, body mass index; CI, confidence intervals; OR, odds ratio.
All values are reported as n (%).
*Cross-tabulation, P<0.05 considered significant.
aThe highest deprivation index is the most deprived area.
bOut of the 46 cows’ milk drinkers who reported having cows’ milk allergy, 16 (34.8%) children drank lactose-free cows’ milk. 
cThe focus was on medical conditions with symptoms overlapping those of milk allergy/lactose intolerance. Medical conditions identified in these children were celiac disease (n=4), cystic fibrosis (n=1) and inflammatory bowel disorder (n=1).

Of the total study population, 104 (8%) parents/caregivers reported their children had a milk allergy, and seven (0.5%) reported their children had a medical condition that may affect cows’ milk consumption. Seventeen percent (221/1,315) of children were classified as either overweight or obese.

For a detailed breakdown of types of milk consumed by children, refer to Supplementary Table 2. Approximately 88% (1,166/1,329) of parents/caregivers reported their child consumed cows’ milk, 8% (163/1,329) not consuming cows’ milk, and 4% (53/1,329) not consuming any type of milk. The proportion of children consuming cows’ milk varied by parental report of the child having a milk allergy (x2=198.5, P<0.0001); parental education (x2=12.4, P<0.01) and ethnicity (x2=7.4, P=0.06). The odds of drinking cows’ milk were decreased for children with versus those without cows’ milk allergy (44% vs 91%, OR=0.06, 95% CI 0.04–0.10) and increased for children of mothers with no secondary versus post-secondary educational qualifications (96% vs 86%, OR=4.83, 95% CI 1.68–13.89) and for Māori versus New Zealand European children (94% vs 87%, OR=2.46, 95% CI 1.20–5.03).

Table 2: Factors predicting type of milk consumption (plain low- or reduced-fat milk vs full-fat cows’ milk) in the 1,166 preschool children who consumed milk.

Variables

Plain low- or reduced-fat cows’ milk

307 (26.3)

Full-fat cows’ milk

859 (73.7)

Association with cows’ milk consumption type*

Univariable

OR (95% CI)

Multivariable

OR (95% CI)

Age

2-year old

84 (19)

348 (81)

Reference category

3-year old

116 (30)

277 (70)

1.74 (1.26–2.40)

1.29 (1.80–2.50)

4-year old

107 (31)

234 (69)

1.90 (1.36–2.64)

1.93 (1.38–2.72)

Ethnicity

New Zealand European

228 (28)

573 (72)

Reference category

Māori

32 (20)

132 (80)

0.59 (0.39–0.90)

0.56 (0.36–0.88)

Pacific Island

5 (13)

35 (87)

0.36 (0.14–0.93)

0.32 (0.12–0.86)

Other

43 (27)

119 (73)

0.91 (0.62–1.33)

0.83 (0.56–1.24)

Residential region

Northern

177 (32)

382 (68)

Reference category

Central

65 (19)

276 (81)

0.51 (0.37–0.71)

0.50 (0.36–0.70)

Southern

65 (24)

201 (76)

0.70 (0.50–0.97)

0.65 (0.46–0.92)

BMI (calculated for age and sex)

Normal

238 (25)

714 (75)

Reference category

Overweight

53 (33)

106 (67)

1.50 (1.05–2.15)

1.74 (1.20–2.54)

Obese

12 (27)

32 (73)

1.13 (0.57–2.22)

1.48 (0.73–3.01)

†Variables included in the multivariate analysis were age, sex, ethnicity, deprivation index, residential region, milk allergy and body mass index (BMI, calculated for age and sex). Model x2 (14)=61.08, P<0.0001.
When ethnicity was replaced with maternal education, the same variables were found to be significant predictors of type of cows’ milk consumed (with comparable ORs).
*Cows’ milk type consumption was coded as 1 = full–fat milk and 2 = plain reduced- or low-fat milk.

Pattern and predictors of different types of cows’ milk consumption

Of 1,166 cows’ milk consumers, 307 (26%) consumed plain low- or reduced-fat milk. For a detailed breakdown of type of milk included in each category and their consumption in these children, refer to Supplementary Table 2. Type of cows’ milk consumption was associated with age, ethnicity, maternal education and residential region (Table 2). The odds of consuming plain low- or reduced-fat milk was increased for three-year old (OR=1.80, 95% CI 1.29–2.50) and four-year old children (OR=1.93, 95% CI 1.38–2.72) compared with two-year old. The odds of consuming plain low- or reduced-fat milk were lower in Māori (OR=0.56, 95% CI 0.36–0.88) and Pacific Island children (OR=0.32, 95% CI 0.12–0.86) than children of New Zealand European ethnicity and in children living in central (OR=0.50, 95% CI 0.36–0.70) and southern regions (OR=0.65, 95% CI 0.46–0.92) than in northern region. When maternal education was replaced with ethnicity, the results for all variables were comparable (data not reported here), and the odds of consuming plain low- or reduced-fat milk were lower in children of mothers with no secondary qualification (OR=0.57, 95% CI 0.34–0.95) and secondary qualification (OR=0.59, 0.36–0.97) than those of mothers with post-secondary qualification (17% and 19% vs 28%, respectively).

Type of milk consumption was associated with BMI percentiles (calculated for age and sex); with overweight children having an increased odds of consuming plain low- or reduced-fat cows’ milk (OR=1.74, 95% CI 1.20–2.54) compared with normal weight children. Medical conditions (known to affect milk consumption) were not entered in the logistic regression analysis because there were only five children with one of these.

Discussion

In the study, the majority of children were reported as consuming cows’ milk, however only 26% of these children consumed plain low- or reduced-fat cows’ milk despite guidelines recommending low-fat milk for all children >2 years of age.1 The odds of consuming plain low- or reduced-fat cows’ milk were increased in older children, those of New Zealand European ethnicity and among children of mothers who have a post-secondary education and living in the Northern regions. We also identified an association between type of milk and overweight, with the odds of consuming plain low- or reduced-fat cows’ milk increased in overweight but not obese children compared with children with a normal BMI.

The smaller proportion of children consuming plain low- or reduced-fat cows’ milk (than full-fat milk) in our study mirrors the findings of other reports from New Zealand and the US, showing that approximately 26% and 13–21% of children, respectively, consumed lower-fat milk varieties.19,27 These data highlight the fact that most parents/caregivers do not adhere to the dietary guidelines which recommend reduced-fat and low-fat milk be given to children.1

The odds of consuming plain low or reduced milk were increased for three- and four-year old compared with two-year old children (an average of 9% increase in the proportion). These results confirm the findings of Scharf et al (2013) and Fox et al (2010) who reported an approximately 8% difference in the proportion of four-year old children who consumed lower-fat milk alternatives than two- and three-year old children.27,39 However, the 2002 New Zealand National Children’s Nutrition survey did not find any relationship between age and type of milk consumed.19 It is worth noting that preschool children were not included in the 2002 survey, and different variables other than age (including, but not limited to, peer and environmental influences)40 could be involved in shaping food choices in older children. It is also important to note that there are two sets of guidelines that are applicable to the two-year old age group in New Zealand; the 0–241 and 2–18-year old1 guidelines. While the 0–2 guidelines recommend no more than 500ml of milk, all of which should be whole milk,41 the 2–18 dietary guidelines recommend 2–3 servings of dairy products that should be of low-fat alternatives.1 Therefore, some parents/caregivers of 2-year old children given that their children potentially (depending on interpretation of age) fall into both sets of recommendations might have uncertainty about choosing the appropriate set of guidelines.

Ethnicity and maternal education, but not socioeconomic status, were significantly associated with type of cows’ milk consumed by children. Children of New Zealand European origin or of mothers with post-secondary qualification had increased odds of consuming plain low- or reduced-fat milk compared with Māori or Pacific children or children of mothers with no secondary qualification. Ethnicity and maternal education have consistently been reported to be significant determinants of milk and type of milk consumption, with non-white or less educated populations being more likely to consume full-fat varieties.19,25,42 The reasons for the differences in the type of milk consumed in some ethnic groups likely involves culture, socioeconomic status, environment and education (and possibly interactions among these variables).43 Among these characteristics, maternal education seems to be the strongest predictor, though other ethnicity-related factors not explored in this study (like taste preferences and acceptability) might also be involved. Evidence suggests that parental knowledge of nutrition guidelines about milk (as well as parental dietary pattern) influences the amount of milk served to 2–8-year old children.44 Thus, educating parents/caregivers who have a high degree of control over the environment and experiences of their preschool child is of particular importance in delivering information to young children.

Interestingly, the odds of consuming plain low- or reduced-fat cows’ milk was lower in children living in Central and Southern regions of New Zealand than those living in the Northern region. The reason for this geographical relationship is unclear (because the association persisted beyond adjustment for other sociodemographic factors and overweight/obesity), though an explanation for such finding could be that those living in central and southern regions of New Zealand who participated in the study might have been from rural areas where dairy farmers are more likely to feed their children full-fat milk. Further research is warranted to better elucidate such relationship.

Overweight but not obese children had higher adjusted odds of drinking plain low- or reduced-fat cows’ milk than normal weight children. A lack of statistical significance for the association of obesity with milk type could potentially be due to small numbers of children who were obese (n=44, 4% of total sample). Our findings are inconsistent with the findings of LaRowe et al (2007) and O’Connor et al (2006) who reported no statistically significant relationship between milk fat content and overweight/obesity.26,28 It should be noted that the former study group reported higher BMI in 2–5-year old children consuming 2% fat milk (reduced fat) than those consuming lower-fat milk (1% fat).26

Our findings, however, confirm those of Vanderhout et al (2016) who also reported a significant inverse relationship between milk fat percentage and BMI z-score (adjusted for milk volume consumed), and children consuming standard fat milk having a 0.72 [95% CI, 0.68–0.76] lower BMI z-score than those consuming lower-fat milk alternatives.29 A similar relationship has also been reported from a longitudinal study of preschool-aged children born in 2001 in the US; Scharf et al (2013)27 reported increased odds of overweight/obesity in preschool children consuming low- or reduced-fat milk than those consuming higher-fat milk alternatives. The former study group also found that children consuming lower-fat milk alternatives (1% vs 2% fat) were more likely to become overweight between two and four years,27 a finding confirmed by some evidence from adults. Data from the US-based Women’s Health Study showed that women aged 45 years and over, with normal weight to begin with, gained significantly less weight with higher dairy intakes.45 Interestingly, only high-fat dairy products, and not low-fat dairy products, were associated with significantly less weight gain.45 Those in the top fifth of the high-fat dairy product intake range had a significant 8% lower risk of becoming overweight or obese.

It is important to note that the cause and effect relationship cannot be determined from these observational studies. Our findings may reflect the fact that parents/caregivers of overweight children or in families where overweight/obesity is prevalent may intentionally give their children lower-fat milk to lower the risk of overweight/obesity (per dietary guidelines). Some authors, however, hypothesise that lower-fat milk alternatives may increase children’s caloric intake from other foods because fat has a satiating effect.46 It is important to note that the cause of obesity is multifactorial, with a full range of sociodemographic factors that affect physical activity and dietary practices (with dietary fat being just one of the many components involved), and therefore, randomised controlled trials are required to disentangle the direction of causality (the relationship between obesity and milk fat content).

In the light of high and increasing prevalence of paediatric overweight/obesity47 and serious overweight/obesity-related comorbidities,48 the dietary guidelines around low- or reduced-fat milk alternatives seem prudent. Individuals are advised to reduce their energy intake through decreasing dietary fat intake as a weight gain management strategy. Theoretically, replacing one serving of standard fat milk with low- or reduced-fat milk alternatives may reduce daily fat intake by approximately 5–8g which equates to a reduction in daily energy intake by 45–72 kcal/serving of milk. However, the downside of promoting a low-fat diet for weight management has been acknowledged by experts. It has been suggested that a singular focus on reduction of total and saturated fat can be counterproductive, because dietary fat is typically replaced by refined carbohydrates and refined carbohydrates are likely to be more detrimental to health compared to total and saturated fat.49

Our study has several limitations. The type of cows’ milk consumed by children in the study was reported by parents/caregivers and was not assessed directly. Furthermore, information regarding the type of milk parents/caregivers drink was not collected. Parents/caregivers dietary habit may play a major role in what they provide their children. The enrolled sample of children was not representative of the New Zealand preschool population. We enrolled a larger proportion of parents/caregivers with higher socioeconomic status and higher education and of children with lower BMI. Compared to the data from New Zealand Health Survey (2015/16),47 the rate of overweight/obesity was lower in our study population (32% vs 17%, respectively). With regard to the obesity-type of milk consumption relationship, we lacked data on other sources of dairy products, frequency and volume of milk consumed, total energy intake, types of food consumed and physical activity. Therefore, we could not adjust for these well-established confounders. Finally, given the cross-sectional nature of this study, causal inferences cannot be made. However, our study is the first in New Zealand to investigate the relationship between sociodemographic factors and type of milk consumption, and to explore the association between type of milk consumed and overweight in preschool children.

In conclusion, we found that most children drank cows’ milk (88%), the prevalence of consuming plain low- or reduced-fat milk was low in preschool children (26%), and was influenced by several sociodemographic characteristics, including age, maternal education, ethnicity and residential region. There was no relationship between consumption of full-fat cow’s milk and overweight/obesity. Identification of the determinants of milk type consumption in children provides guidance for targeted interventions to improve milk consumption pattern in children. It is also of high significance to investigate whether the dietary guidelines are clear for parents/caregivers or people advising them. We found that consumption of plain low- or reduced-fat milk was associated with an increased odds of overweight but not obesity. This finding may suggest that parents/caregivers of overweight children are more likely to adhere to dietary guidelines to manage weight gain in children. However, this hypothesis is assumed to be true only if parents/caregivers have an accurate perception of their young children’s normal weight status. Further research is warranted to better elucidate the reasons for obesity-type of milk consumption relationship.

Supplementary Table 1: Nutritional information of different types of dairy milk.*

Milk types

 

Nutrients/100 ml

Energy (kj)

Protein (g)

Fat total (g)

Carbohydrate (g)

Calcium (mg)

Anchor

Standard blue top

260

3.3

3.3

4.8

117

Lite

194

3.3

1.5

4.9

117

Trim

156

4.0

0.1

5.0

133

Superblue

272

5.0

2.8

4.8

164

Mega

243

5.0

2.0

5.0

160

Calci+

193

5.8

0.2

5.2

190

Zero lacto blue

260

3.3

3.3

4.0

117

Zero lacto trim

156

4.0

0.1

5.0

133

Meadow Fresh

Standard blue top

270

3.9

3.3

4.7

135

Lite

210

4.0

1.5

4.8

135

Trim

170

4.1

0.4

4.9

140

Calci Trim

190

5.9

0.2

4.8

190

Calci Strong

Chocolate flavoured milk

270

4.4

1.5

8.4

145

Strawberry flavoured milk

260

4.4

1.5

7.9

145

Banana flavoured milk

260

4.4

1.5

7.5

145

Goat’s milk*

243

2.9

3.3

4.2

NR

NR, not reported
*All nutritional information is based on product Nutrition Information Panel, unless otherwise stated.

**Derived from http://www.meadowcroftfarm.co.nz/products/goats-milk 

Supplementary Table 2: Detailed description of milk consumption (by type) among preschool children (n=1,329).

Milk and type of milk consumption

n (%)

Description

Does not drink dairy milk

163 (12)

  • Milk added on cereals only otherwise the child doesn’t drink milk (n=7)
  • Breast milk (n=5)
  • Toddler formula (n=37)
  • Plant milk (n=81, including soy, rice, almond and oat milk)
  • No milk at all (n=54)

Drinks dairy milk*

1,166 (88)

 

Full-fat cows’ milk**

859 (74)

  • Standard blue top milk (n=852)
  • Goat’s milk (n=6)
  • Flavoured milk (n=13)
  • Standard fat lactose-free milk (n=15)
  • Toddler formula in combination with other milks from this group (n=34)
  • Breast milk in combination with other milks from this group (n=9)

Plain low- or reduced-fat cows’ milk

307 (26)

 

Light

198 (17)

  • Standard light milk (n=198)
  • Toddler formula in combination with other milks from this group (n=5)

Trim

49 (4)

  • Standard trim milk (n=48)
  • Lactose free trim milk (n=1)
  • Breast milk in combination with other milks from this category (n=1)

Fortified

63 (5)

  • Anchor calciplus (n=24)
  • Anchor Mega (n=10)
  • Anchor Superblue (n=6)
  • Meadow fresh calcitrim or calcistrong (n=23)
  • Breast milk in combination with other milks from this category (n=2)
  • Toddler formula in combination with other milks from this category (n=4)

*Milk type was collapsed into two categories; fat- and sugar-containing milk (that comprises of standard-fat milk (blue top), flavoured milk, goat’s milk and raw/unpasteurised milk) and plain low- or reduced-fat milk (that comprises of light milk, trim milk and fortified low/reduced-fat milk).

**The term “full-fat cows’ milk” is used to refer to “fat- and sugar-containing dairy milk” because only 13 (1%) parents reported their children consuming flavoured milk and six parents reported their children consuming goats’ milk. Some children consumed more than one type of milk. When two or more types of milk were consumed, the priority was given to the higher fat content (eg, if a parent/caregiver reported child consuming both standard and low- or reduced-fat milk, the child was assigned to full-fat cows’ milk group, n=118).

†Percentage of children who consumed dairy milk (n=1,166).

Supplementary Table 3: Sensitivity analysis excluding children who consumed both full-fat and low-fat cow’s milk (n=118). Factors predicting type of milk consumption (plain low- or reduced-fat milk vs full-fat cows’ milk) in the 1,048 preschool children who consumed milk.

Variables

Plain low- or reduced-fat cows’ milk

307 (29)

Full-fat cows’ milk

741 (71)

Association with cows’ milk consumption type*

Multivariable

OR (95% CI)

Age

2-year old

84 (22)

299 (78)

Reference category

3-year old

116 (33)

235 (67)

1.84 (1.31–2.59)

4-year old

107 (34)

207 (66)

1.88 (1.33–2.66)

Ethnicity

New Zealand European

228 (32)

485 (68)

Reference category

Māori

32 (22)

116 (78)

0.54 (0.34–0.84)

Pacific Island

5 (14)

32 (86)

0.29 (0.11–0.77)

Other

43 (29)

107 (71)

0.76 (0.51–1.15)

Residential region

Northern

177 (36)

317 (64)

Reference category

Central

65 (20)

253 (80)

0.45 (0.32–0.64)

Southern

65 (28)

171 (72)

0.63 (0.44–0.89)

BMI (calculated for age and sex)

 

Normal

238 (28)

619 (72)

Reference category

Overweight

53 (22)

187 (78)

1.84 (1.24–2.73)

Obese

12 (29)

29 (71)

1.51 (0.73–3.12)

Variables included in the multivariate analysis were age, sex, ethnicity, deprivation index, residential region, milk allergy and body mass index (BMI, calculated for age and sex). Model x2 (15)=68.57, P<0.0001.

When ethnicity was replaced with maternal education, the same variables were found to be significant predictors of type of cows’ milk consumed (with comparable ORs).

*Cows’ milk type consumption was coded as 1 = full-fat milk and 2 = plain reduced- or low-fat milk. 

Abstract

Aim

New Zealand dietary guidelines recommend children from two years of age consume low- or reduced-fat milk. We aimed to investigate the predictors of type of milk consumption in preschool children.

Method

Data were drawn from a cross-sectional study which enrolled preschool children (2–<5 years, n=1,329) from throughout New Zealand.

Results

Cows’ milk was consumed regularly by 88% of children. Of these, 26% consumed plain low- or reduced-fat milk, while 74% consumed full-fat milk. The adjusted odds of consuming plain low- or reduced-fat milk were increased in older children: three-year old (OR=1.80, 95% CI 1.29–2.50); four-year old (OR=1.93, 95% CI 1.38–2.72) versus two-year old children, and were decreased in Māori (OR=0.56, 95% CI 0.36–0.88) and Pacific children (OR=0.32, 95% CI 0.12–0.86) compared with New Zealand European children. Approximately 18% of children were overweight/obese. The odds (adjusted for socio-demographic characteristics) of consuming plain low- or reduced-fat milk were increased in overweight children (OR=1.74, 95% CI 1.20–2.54) than normal weight children.

Conclusion

The type of milk consumed by preschool children varies with child demographics and anthropometry. Further research is warranted to investigate caregivers/parents’ knowledge about dietary guidelines and to determine the causal relationship between obesity and milk type consumption. The findings of the current study may have important implications for developing and shaping interventions and in helping shape public health policy and practice to promote cows’ milk consumption in preschool children.

Author Information

Hajar Mazahery, Institute of Food Science and Technology—School of Food and Nutrition, Massey University, Auckland; Carlos A Camargo Jr, Department of Emergency Medicine, Massachusetts General Hospital, Boston, USA; Carolyn Cairncross, Faculty of Health and Environmental Sciences, AUT, Auckland; Lisa A Houghton, Department of Human Nutrition, University of Otago, Dunedin; Cameron C Grant, Department of Paediatrics: Child and Youth Health, Auckland City Hospital, University of Auckland, The Centre for Longitudinal Research—He Ara ki Mua, School of Population Health, University of Auckland, Auckland, Starship Children’s Hospital, Auckland District Health Board, Auckland; Jane Coad, Institute of Food Science and Technology, School of Food and Nutrition, Massey University, Palmerston North; Cathryn A Conlon, Institute of Food Science and Technology—School of Food and Nutrition, Massey University, Auckland; Pamela R von Hurst, Institute of Food Science and Technology—School of Food and Nutrition, Massey University, Auckland.

Correspondence

Dr Pamela R von Hurst, Institute of Food Science and Technology—School of Food and Nutrition, Massey University, Auckland 0745.

Correspondence Email

p.r.vonhurst@massey.ac.nz

Competing Interests

HM was funded to write this manuscript with a summer studentship from Fonterra Cooperative Group (Auckland, New Zealand). The funder had no input into the collection and analysis of the data, and provided peer review of the writing of the manuscript. Te Ra Whakaora study was funded by the New Zealand Health Research Council (Auckland, New Zealand).

References

  1. Ministry of Health. Food and Nutrition Guidelines for Healthy Children and Young People (Aged 2 –18 years): A background paper, 2015, Ministry of Health: Wellington.
  2. Ballew C, Kuester S, Gillespie C. Beverage choices affect adequacy of children’s nutrient intakes. Arch Pediatr Adolesc Med, 2000; 154(11):1148–52.
  3. Marshall TA, Eichenberger Gilmore JM, Broffitt B, et al. Diet quality in young children is influenced by beverage consumption. J Am Coll Nutr, 2005; 24(1):65–75.
  4. Rangan AM, Flood VM, Denyer G, et al. Dairy consumption and diet quality in a sample of Australian children. J Am Coll Nutr, 2012; 31(3):185–93.
  5. Rockell JE, Williams SM, Taylor RW, et al. Two-year changes in bone and body composition in young children with a history of prolonged milk avoidance. Osteoporos Int, 2005; 16(9):1016–23.
  6. Bonjour JP, Chevalley T, Ferrari S, et al. The importance and relevance of peak bone mass in the prevalence of osteoporosis. Salud Publica Mex, 2009; 51 Suppl 1:S5–17.
  7. Goulding A, Rockell JE, Black RE, et al. Children who avoid drinking cow’s milk are at increased risk for prepubertal bone fractures. J Am Diet Assoc, 2004; 104(2): 250–3.
  8. Black RE, Williams SM, Jones IE, et al. Children who avoid drinking cow milk have low dietary calcium intakes and poor bone health. Am J Clin Nutr, 2002; 76(3):675–80.
  9. Fiorito LM, Mitchell DC, Smiciklas-Wright H, et al. Girls’ calcium intake is associated with bone mineral content during middle childhood. J Nutr, 2006; 136(5):1281–1286.
  10. Wiley AS. Cow milk consumption, insulin-like growth factor-I, and human biology: A life history approach. Am J Hum Biol, 2012; 24(2):130–138.
  11. Wiley AS. Dairy and milk consumption and child growth: Is BMI involved? An analysis of NHANES 1999–2004. Am J Hum Biol, 2010; 22(4):517–25.
  12. Abreu S, Santos R, Moreira C, Santos PC, et al. Milk intake is inversely related to body mass index and body fat in girls. Eur J Pediatr, 2012; 171(10):1467–74.
  13. Lu L, Xun P, Wan Y, et al. Long-term association between dairy consumption and risk of childhood obesity: a systematic review and meta-analysis of prospective cohort studies. Eur J Clin Nutr, 2016; 70(4):414–423.
  14. Wang W, Wu Y, Zhang D. Association of dairy products consumption with risk of obesity in children and adults: a meta-analysis of mainly cross-sectional studies. Annals Epidemiol, 2016; 26(12):870–882.e2.
  15. Hagman E, Danielsson P, Brandt L, et al. Association between impaired fasting glycaemia in pediatric obesity and type 2 diabetes in young adulthood. Nutr Diabetes, 2016; 6:e227.
  16. Ridoutt B, Baird D, Bastiaans K, et al. Changes in Food Intake in Australia: Comparing the 1995 and 2011 National Nutrition Survey Results Disaggregated into Basic Foods. Foods, 2016; 5(2):40.
  17. Gopinath B, Flood VM, Burlutsky G, et al. Pattern and predictors of dairy consumption during adolescence. Asia Pac J Clin Nutr, 2014; 23(4):612–8.
  18. Clinical Trials Research Unit. A National Survey of Children and Young People’s Physical Activity and Dietary Behaviours in New Zealand: 2008/09 - Key Findings, 2010, Ministry of Health: Wellington.
  19. Ministry of Health. NZ food NZ children: Key results of the 2002 national children’s nutrition survey, 2003, Ministry of Health: Wellington.
  20. Esfarjani F, Mohammadi-nasrabadi F, Roustaee R, et al. Household Milk consumption and Its Socio-economic Associates in West Azarbayejan Province, North-west Iran. Nutr Food Sci Res, 2015; 2(3): 21–27.
  21. Lin SL, Tarrant M, Hui LL, et al. The role of dairy products and milk in adolescent obesity: Evidence from Hong Kong’s “children of 1997” birth cohort. PLOS ONE, 2012; 7(12):e52575.
  22. Kurajdová K, Táborecká-Petrovičová J, Kaščáková A. Factors influencing milk consumption and purchase behavior – Evidence from Slovakia. Procedia Econ Financ, 2015; 34:573–580.
  23. De Alwis A, Edirisinghe J, Athauda A. Analysis of factors affecting fresh milk consumption among the mid-country consumers. Trop Agric Res Exten, 2011; 12(2):103–109.
  24. Parker CE, Vivian WJ, Oddy WH, et al. Changes in dairy food and nutrient intakes in Australian adolescents. Nutrients. 2012; 4(2):1794–811.
  25. Kit BK, Carroll MD, Ogden CL. Low-fat milk consumption among children and adolescents in the United States, 2007–2008. NCHS Data Brief, 2011(75):1–8.
  26. LaRowe TL, Moeller SM, Adams AK. Beverage patterns, diet quality, and body mass index of US preschool and school-aged children. J Am Diet Assoc, 2007; 107(7):1124–1133.
  27. Scharf RJ, Demmer RT, DeBoer MD. Longitudinal evaluation of milk type consumed and weight status in preschoolers. Arch Dis Child, 2013; 98(5):335–340.
  28. O’Connor TM, Yang SJ, Nicklas TA. Beverage intake among preschool children and its effect on weight status. Pediatr, 2006; 118(4):e1010–8.
  29. Vanderhout SM, Birken CS, Parkin PC, et al. Relation between milk-fat percentage, vitamin D, and BMI z score in early childhood. Am J Clin Nutr, 2016; 104 (6):1657–64.
  30. Cairncross CT. Vitamin D and preschool children - predictors of status and relationship with allergic and respiratory diseases in New Zealand. Auckland, New Zealand: Massey University; 2015.
  31. Thomas LF, Keim KS, Long EM, Zaske JM. Factors related to low milk intake of 3- to 5-year-old children in child care settings. J Am Diet Assoc, 1996; 96(9):911–2.
  32. Salmond CE, Crampton P, Atkinson J. NZDep2006 Index of Deprivation. In: health Dop, editor. Wellington: University of Otago; 2007.
  33. von Tirpitz C, Kohn C, Steinkamp M, et al. L Lactose intolerance in active Crohn’s disease: clinical value of duodenal lactase analysis. J Clin Gastroenterol, 2002; 34(1):49–53.
  34. Vernia P, Marinaro V, Argnani F, et al. Self-reported milk intolerance in irritable bowel syndrome: what should we believe? Clin Nutr, 2004; 23(5):996–1000.
  35. Mäkelä M. Milk and wheat allergy, and celiac disease. Clin Transl Allergy, 2011; 1(Suppl 1):S37.
  36. Li L, Somerset S. D Digestive system dysfunction in cystic fibrosis: Challenges for nutrition therapy. Dig Liver Dis, 2014; 46(10):865–874.
  37. Cole TJ, Bellizzi MC, Flegal KM, et al. Establishing a standard definition for child overweight and obesity worldwide: international survey. Bio Med J, 2000; 320(7244):1240.
  38. King G, Zeng L. Logistic regression in rare events data. Polit Anal, 2001; 9(2):137–163.
  39. Fox MK, Condon E, Briefel RR, et al. Food consumption patterns of young preschoolers: are they starting off on the right path? J Am Diet Assoc, 2010; 110(12 Suppl):S52–9.
  40. Birch LL, Fisher JO. Development of eating behaviors among children and adolescents. Pediatr, 1998; 101(Supplement 2):539–549.
  41. Ministry of Health. Food and nutrition guidelines for healthy infants and toddlers (Aged 0–2): A background paper - Partially revised December 2012, 2015: Ministry of Health, Wellington.
  42. Thornton LE, Pearce JR, Ball K. S Sociodemographic factors associated with healthy eating and food security in socio-economically disadvantaged groups in the UK and Victoria, Australia. Public Health Nutr, 2014; 17(1):20–30.
  43. Williams DR. Race and health: basic questions, emerging directions. Ann Epidemiol, 1997; 7(5):322–33.
  44. Brett NR, Vanstone CA, Maguire JL, et al. Parental knowledge, perceptions and consumption of milk and alternatives relates to intakes of young children 2–8 y of age. FASEB J, 2016; 30(1 Supplement):295.3.
  45. Rautiainen S, Wang L, Lee I-M, et al. Dairy consumption in association with weight change and risk of becoming overweight or obese in middle-aged and older women: a prospective cohort study. Am J Clin Nutr, 2016; 103(4):979–988.
  46. Beglinger C, Degen L. Fat in the intestine as a regulator of appetite--role of CCK. Physiol Behav. 2004; 83:617–21.
  47. Ministry of Health. Annual update of key results 2015/16: New Zealand Health Survey, 2016, Ministry of Health. Wellington.
  48. Emmanuel JJ, Coppack SW. Health consequences–obesity associated comorbidities. In: Agrawal S, editor. Obesity, Bariatric and Metabolic Surgery: A Practical Guide. Cham: Springer International Publishing; 2016:29–38.
  49. Hu FB. Are refined carbohydrates worse than saturated fat? Am J Clin Nutr. 2010; 91:1541–2.