Meat consumption among 18-month-old children participating in the Childhood Asthma Prevention Study.

Abstract

Objective: To document meat consumption among 18-month-old children, for use in refining population dietary assessment methods and dietary guidance for young children.

Design: A secondary analysis of data collected in 1998-2000 from the 18 months follow up of the Childhood Asthma Prevention Study: an intervention trial of omega-3 oil supplementation and house-dust mite reduction, from birth to five years.

Subjects and setting: Pregnant women whose unborn children had an atopic family history were recruited from antenatal clinics of six hospitals in western Sydney. Carers of 429 18-month-old children (80% response) satisfactorily completed three-day weighed food records.

Main outcome measures: Mean and median intakes per day and portion sizes of various meats and meat products.

Statistical analyses: T-tests for comparing gender differences; Pearson correlation and one-way analysis of variance for relationships between meat and nutrient intake.

Results: During the recording period 94% of the children ate meat. McDonald's Chicken McNuggets[TM], beef mince, and beef sausages were the most frequently consumed meats. Median portion sizes ranged from 20-50 g, and were considerably smaller than 'usual' portions specified on food frequency questionnaires in common use. Higher meat diets in this age group were not associated with higher intakes of iron or zinc per MJ.

Conclusion: The marginally low intakes of iron and zinc in this age group could be improved by greater use of cuts of red meat appropriately prepared for toddlers.

Key words: dietary intake, nutrient intakes, children's diet, iron, zinc

(Nutr Diet 2005;62:12-20)

**********

Introduction

Information about dietary intakes of infants and very young children is relatively scarce, particularly regarding food consumption patterns. The 1995 National Nutrition Survey was the first national survey of diets to include pre-school children but the lower age limit for the sample frame was two years (1). We could find no published information from population-based studies of children aged one to two years in Australia. In the US, data on a subsample of children under two years has been collected over a number of years in national nutrition surveys including the National Health and Nutrition Examination Survey (NHANES), the Continuing Survey of Food Intake of Individuals (CSFII), and the Nationwide Food Consumption Survey (2-4). In the UK, at least two large purpose-specific surveys of children under two years have been conducted in the last decade, one a national survey and the other, a large population cohort (5,6).

The importance of obtaining dietary information in this age group in Australia relates to at least two areas of public health requiring improvement: methods to measure and monitor the food and nutrition situation in sub-groups of the Australian population (7), and tailoring or refining population dietary guidance appropriate for various age groups (8).

While detailed population dietary surveys are the cornerstone of national food and nutrition monitoring, such surveys are expensive and challenging to conduct, are likely to be repeated infrequently and often provide inadequate information for population sub-groups of particular interest. Less detailed measurement methods can be used to monitor food habits (9), including food frequency questionnaires and short survey questions about food habits, both of which have been used to measure aspects of diet among adults in Australian Bureau of Statistics National Health Surveys, and the state Computer Assisted Telephone Interview (CATI) survey programs (10,11). Interest in monitoring aspects of children's diets is increasing, with recent state surveys in Australia having used or adapted questions and/or questionnaires previously used with adults (12). The relative validity of selected short diet questions has been investigated among adults, but has not been reported for children (7). Key aspects of short diet questions, and food frequency questionnaires that are likely to influence their validity in children are: whether the foods, quantities, frequencies, prompts and response categories used in the survey questions, are relevant to the current eating habits of young children and the specific nutritional concerns for this age group. Thus, relatively precise information about types of foods commonly consumed, portion sizes, the contribution of relevant foods to nutrient intakes, and the diet quality of those who consume more or less of particular foods is needed to begin to tailor population survey questions and tools for this age group.

The lack of official national dietary guidance for children in the transition from infancy to a varied diet, specifically from one to two years, has been highlighted as a significant problem facing the nutrition community (8). The Australian Guide to Healthy Eating (AGHE) is the official national food selection guide, including quantitative recommendations for 'core' food groups (meats, dairy, fruit and vegetables, and cereals) and is used widely for population-based dietary education and planning (13). In its current form, the AGHE does not include recommendations for children under the age of four years. Other dietary policies, including the NHMRC Dietary Guidelines for Children and Adolescents (14) provide guidance about infant feeding, and general principles for planning diets of older children, but these are not quantitative, nor intended for use in detailed diet planning.

Meat consumption habits of young children are also of interest because studies have shown that intakes tend to be low, and low intakes of meat are associated with low intakes of iron and zinc, and with poor iron status. In US and UK surveys and a New Zealand study of dietary intakes of very young children, meat intakes have been typically low, and substantial proportions of children did not reach the recommended dietary allowances for iron and zinc (3-5,15). Munoz and others (16) adapted US recommendations for older children and compared the diets of two- to five-year-olds from CSFII data; only 13.9% of males and 12.2% of females met the recommendations for meat intake for three consecutive days. Iron and zinc status in children under 2 years appears to be related to meat intake. Cowin et al. (17), Thane et al. (18), and Soh et al. (19) have found that the prevalence of low iron status was higher among children who consumed little meat or poultry. In the latter study toddlers were also found to be at greater risk of sub-optimal iron intakes than infants. In Australia, Oti-Boetang and others found that iron deficiency was common in Australian children: 28% of children aged six to 24 months in their sample had either nonanaemic iron deficiency or iron deficiency anaemia, and deficiency was related to high milk intake which replaced high iron foods (20).

The Childhood Asthma Prevention Study (CAPS) provides a rare opportunity to investigate nutrient and food intake data and correlates of over 400 Australian children aged 18 months, on whom three-day weighed food records were kept by carers. The focus of this paper, is meat consumption habits in the diets of young Australian children.

Methods

CAPS is a randomised controlled trial based in western Sydney, investigating the effects of both house dust mite reduction and omega-3 fatty acid supplementation, separately and together, on the prevention of asthma during the first five years of life. Over 600 infants were randomised into groups receiving either one, both, or no interventions (21). The dietary intervention for the trial did not attempt to influence children's food habits other than the types of polyunsaturated fat spreads and cooking oils used and would thus be unlikely to affect intakes of most food and nutrients, including intakes of meat.

Subjects

Pregnant women whose unborn children were at a high risk of developing asthma because of a parent or sibling with a current diagnosis of asthma or with frequent wheeze were recruited from antenatal clinics of six hospitals in western Sydney.

Of 7171 pregnant women screened, 2095 women satisfied the selection criteria. At completion of recruitment, 616 agreed to participate. Sixty-two women withdrew from the study after randomisation, which resulted in a total of 554 study participants, of which 538 attended the 18-month assessment. The extent to which the study population differed from the sample frame of women who satisfied the selection criteria was assessed and reported elsewhere (22). In brief, a higher proportion of both fathers and mothers of CAPS children had tertiary education and were Australian-born, compared with those who did not participate in the study, and the population of western Sydney.

Ethical approval for the study was obtained from the ethics committees of each of the participating hospitals, the Area Health Services in which they were located, and the University of Sydney.

Dietary assessment

Dietary assessments were carried out in association with medical assessments conducted when children reached 18 months of age. Data on food consumption were collected using three-day weighed food records including one weekend day. Although the study is longitudinal, this is the first and only weighed food record conducted on the CAPS sample. A research dietitian instructed mothers on how to keep records and issued a food record booklet and a set of Tanita digital kitchen scales (2.0 kg X 1.0 g). At the end of the recording period, the dietitian visited their homes to collect and check the completeness and accuracy of records, including brand names of foods and supplements recorded.

Of 538 participants approached, 483 (90%) agreed to keep food records. Of 55 remaining, 14 (2.6%) declined to keep records because of family disruption around the time of the 18-month assessment (moving house, illness in the family, overseas trip, etc), and 41 (7.5%) were non-responders, that is were not contactable by the research dietitian on several attempts or repeatedly failed to keep appointments for record keeping instructions. Of the 483 who agreed to keep food records, 465 (86.4%) actually did so. Of these, 36 were excluded either because all three days were not completed, the quality of the data supplied was poor, the child's food intake on these days was atypical due to illness affecting food intake, or because the child was breastfeeding more than twice per day and therefore the quantity of energy and food intake could not be measured accurately. Sixteen mothers were not able to keep records on a weekend day, and 27 mothers had to complete the record mostly on the weekend, so included both weekend days and one weekday. These records were retained in the data set. Overall, the final number of three-day records analysed was 429, comprising an 80% response rate of those asked to keep records.

Raw data from the food records were checked, coded and entered into a nutrient analysis program (M & H Williams, Sydney, SERVE version 3.95 1998) based on Australian Composition of Foods (National Food Authority, Canberra, NUTTAB95 version 3.0 1995) to derive nutrient data for all foods eaten on each eating occasion. Food and nutrient data, as well as socio-demographic and anthropometric information for each subject were then exported into a Microsoft Access[TM] database (SPSS Inc, Chicago, SPSS for Windows release 11 2001) and Minitab release 12 for Windows (MINITAB Inc, State College PA, Minitab release 12 for Windows 1997) were used for statistical analyses. To calculate total intakes of meat and various sub-types of these, individual foods were classified into major food groups and subgroups following the 2-, 3- and 4-digit classification used for the 1995 National Nutrition Survey (1). Where names are given for individual foods, these are derived from NUTTAB 1995.

For the purpose of comparing the frequency with which different types of meat were consumed, the average frequency/month, that is per child-days of observation (total frequencyX30/429X3), was calculated for each type (Table 1).

Other variables

At the time of the 18-month assessment, study nurses measured weight in kilograms and recumbent length in centimetres, as described in the CAPS protocol (21).

Statistical analyses

Independent sample t-tests were used for comparing gender differences in anthropometric and dietary variables. Relationships between meat and nutrient intake were determined by Pearson correlation for meat intake expressed as a continuous variable and by one-way analysis of variance for quintiles of meat intake. Effect size, for the latter analysis, was expressed in standard deviation units by dividing the difference between the average meat intake for the highest and lowest quintiles by the larger standard deviation.

In view of the inter-relationships between nutrients and the large number of statistical tests carried out, P < 0.01 was used throughout to report statistical significance.

Results

The average age of both boys and girls at the time of measurement was 18.9 months and ranged from 16 to 24 months. Mean weight and length for boys and girls were, respectively 12.0 kg and 84.4 cm, and 11.2 kg and 82.8 cm (data not shown).

A total of 1563 meals containing meat or meat products (including meat dishes, cereal-based meat products and infant dinners), were recorded for the 429 children studied during the three days of weighed food records, indicating that meat was consumed, on average, just over once per day (Table 1). Over 97% of the children ate some meat (including meat in cereal-based products and infant dinners) during the recording period. Poultry, beef and sausages were the types of meats most frequently eaten and together accounted for over 50% of all meat meals including meals containing meat products such as pies, hamburgers and pizza. Cereal-based meat products accounted for about four meat meals per month and infant meat dinners for about two meals per month. The food categories in Table 1 are those used in the 1995 National Nutrition Survey. Excluding mixed meat dishes, mean intakes per consumer per meal, ranged from less than 10 g for organ meats to 45 g for sausages.

The 30 most commonly eaten individual meat items (including cereal-based products and infant dinners) are shown in order of frequency of consumption in Table 2. Together these items accounted for 66% of all meat meals consumed by the study children. Meat and poultry items most frequently consumed were simmered regular beef mince, McDonald's Chicken McNuggets[TM] and grilled beef sausages. These three items accounted for nearly a quarter of all meat items recorded in the study. The median intake of the most frequently consumed items were 15-20 g for bacon and ham, 40-50 g for chicken nuggets and sausages and 45-55 g for infant dinners and cereal-based items containing meat such as hamburgers, pies and pizza.

When meat and meat dishes only were grouped as red (1811, 1812, 1816, 184 and 187), white (1813, 183 and 189), or processed (1814, 1815, 185 and 186); red meats were found to be consumed less often (392 meals), than either white or processed (453 and 445 meals respectively). Of the 429 children, 58% had consumed some red meat during the three days of records compared with 62% and 54% who had consumed some white or processed meat from this group. Table 3, which shows the average nutrient contribution of the amount of red, white and processed meat eaten at a meal, clearly indicates that the average portion of red meat consumed at a meal by children aged around 18 months, while contributing significantly less total and saturated fat, cholesterol and sodium, contributes significantly more iron and zinc than a portion of white or processed meat.

Table 4 shows the percent contributed by each of the major food groups to the intake of selected nutrients. The meats group contributed only 7% of energy but 10% or more of the total nutrient intake for protein, total and monounsaturated fat, sodium, zinc, and niacin. By contrast

dairy foods contributed 39% of energy, and 40% or more of protein, total and monounsaturated fat, phosphorous, potassium, magnesium, and zinc. The greatest contributions to iron intake came from cereals (33%), infant foods (15%) and dairy foods (10%), while the main sources of zinc were dairy foods (45%), meat (17%), infant foods (10%) and cereals (10%).

The diets of the children in the highest quintile of meat intake were significantly higher in energy and had a significantly higher proportion of energy from protein, total and monounsaturated fat and a higher concentration per MJ for cholesterol, retinol and sodium than children in the lowest quintile (Table 5). They also had diets with a significantly lower proportion of energy from carbohydrate and with a lower concentration of calcium, thiamin, riboflavin and magnesium per MJ. The effect size for these differences was greatest for calcium and protein followed by sodium, energy and cholesterol. There was no significant difference between quintiles of meat intake in the proportion of energy provided by saturated fat, iron or zinc (data not shown).

Discussion

The CAPS study provided a unique opportunity to examine food and nutrient intakes in a sample of 18-month-old Australian children. Until such time as population-based Australian surveys of health and nutrition include infants and children less than two years of age, and utilise detailed methods of dietary assessment, we are reliant on data sets obtained from special studies such as CAPS to provide an indication of dietary intakes in this age group in Australia. While CAPS participants are not a random sample of the population, they are similar in many respects to the population of western Sydney, from where the sample was drawn (22). CAPS families do differ from the general population in that they are volunteers in a five-year health trial, have a family history of asthma, and a higher proportion have a tertiary education and are Australian-born than typically found in western Sydney. This may have biased our results towards better diets, and weight status, than we would find in a true random population sample of 18-month-old children. With this caveat, the data provide some interesting glimpses into the diets of young children in Australia.

Types of meat

Chicken nuggets, beef mince, beef sausages, and ham were the most commonly consumed meats. These types of meats were also the most popular in recent UK studies, the Children in Focus (CIF) study of 18-month-old children, and the UK National Diet and Nutrition Survey (NDNS) of 18-30-month-old children (23,5). The percent consuming these meats were almost identical for CAPS and UK samples. Small differences in selected meats were seen, with fewer CAPS children consuming meat pies 9% vs 11%, and organ meats (< 1% vs 7.5%). The reasons for the greater use of poultry, minced and processed meats in place of cuts of red meat in this age group may include: taste and texture preferences, and lack of developed dentition among some children of this age. Moreover, parents may lack skills in methods to prepare meats that appeal to young children, and/or be unaware of the relatively poorer nutrient profile of poultry and processed meats. The fact that some CAPS children did consume meat cuts in stews and other mixed dishes (coded as either the meat cut, or as a mixed dish) suggests that children of this age will eat such meat, if it is prepared to their liking.

Amounts/portion sizes of meat

The average quantities consumed of various meats and meat groups were relatively small, and are similar to the mean intakes per day reported by UK children of this age using similar measurement methods (5). Portion sizes per meal, of various foods, including meats, have not been previously reported in this age group. Our data provide a preliminary basis for use in tailoring portion sizes to this age group in survey questions about food intake, and in food selection recommendations. The median portion size can be used as a 'medium' or 'reference' portion, and the inter-quartile range (25th to 75th percentile) can be used to define small and large portions, as are sometimes used as the basis for designing food frequency questionnaires (9). For example, from our data, a 'medium' size portion of beef sausage, as indicated by the median, was 42 g, a small portion (25th percentile) was 23 g, and a large portion (75th percentile) 60 g.

Several food frequency questionnaires in Australia use reference portion sizes for meats, which are as much as two to four times the median portion sizes of meats consumed by CAPS children. For example, a version of the CSIRO questionnaire gives six chicken nuggets (108 g) as the 'standard' serve, whereas in CAPS, the median portion of chicken nuggets was 48 g (2.5 nuggets), or about half of a 'standard serve' A standard serve of beef sausages on both the CSIRO and the Blue Mountains Eye Study questionnaires (24,25) is 'two thick or three thin' sausages, weighing 130-145 g, which is three times the median serve of sausage (42 g, or about half a thick sausage) for CAPS children. The medium portion size for beef stew listed on the FFQ developed by the Cancer Council of Victoria and used widely in Australia is 180 g, four times CAPS childrens' mean intake per meal of mixed beef dishes (26).

The practical importance of these differences between what is actually consumed by 18-month-old children, and the use of current 'standard' portions to assess diets, is that meat intakes, and nutrients such as zinc, iron and protein are likely to be substantially overestimated if portion sizes are unmodified when such questionnaires are used with children.

At present, the Australian Guide to Healthy Eating recommends that children aged four to seven years consume one-half to two sample serves of meat per day, with a sample serve defined as 65-100 g of meat (13). Thus, the lower limit recommended for a four-year-old is 32.5-50 g per meat per day, close to the median amount of many meats consumed by these 18-month-old children, and so would be an appropriate portion size range for recommendations developed for this age group.

Contribution of meat to nutrient intakes

Although meat is a good source of dietary iron and zinc it was not a major source of these nutrients in the diets of 18-month-old children in our study. Cereals, infant foods and cereal-based foods made a substantially larger contribution to iron than did meat, and dairy products were a considerably larger contributor to zinc intake than was meat. In our sample, the mean intake for iron was at the lower end of the range recommended for one- to three-year-olds, while the mean intake for zinc was below the RDI for one- to three-year-olds. Moreover, the bio-availability of the iron intakes might be expected to be lower from cereals than from meats, although precise information about bio-availability of iron in a mixed diet is a subject of current debate and research. It is well known that haem iron, the more bio-available form, is substantially higher in red muscle meats, than in poultry and processed meats (27). Low intakes of iron and zinc among this age group have been documented in US and UK surveys, and New Zealand studies, attributed to continuing a largely milk-based diet for too long during the second year of life (17-19).

Consumers have reduced their consumption of many types of meat in recent years, in part because of concern about the fat content (28). Although in this sample of young children, some of the most commonly eaten meats were higher fat varieties, only 10.6% of the total fat and 8% of the saturated fat came from meats mainly because all meats were consumed in relatively small amounts and some were prepared using lower fat cooking methods, for example, sausages were grilled far more often than fried (98 meals vs 13 meals). By contrast, dairy products contributed 50% of the total fat and 60% of the saturated fat intake of this group.

Selected diet indices and higher meat consumption

Children with higher intakes of meat (excluding cereal-based products and infant dinners) were taller and weighed more, and their diets were characterised by higher absolute intakes of energy, protein, and nutrients typically found in processed meats (fat, cholesterol, sodium). The expectation that higher meat eaters would have diets higher in iron and zinc per MJ was not observed in this study. Instead, higher meat eaters had diets with a significantly lower density for calcium, thiamin, riboflavin and magnesium, which may be partly attributable to the replacement of some of the milk and fortified infant foods with white and processed, rather than red meats, and without increasing the variety of other nutrient dense foods (8,17).

Conclusions

Australian children in the CAPS study, at 18 months of age, consumed relatively little meat, particularly red meats, in small portions. The most commonly consumed white and processed meats in this age group tended to be higher in energy, fat and sodium and lower in iron and zinc than red meats. The marginally low iron and zinc intakes in this group could be increased by the use of more cuts of red meat in place of some of the white and processed meats, in view of their higher contribution of bioavailable iron and zinc per portion. Portion sizes were typically about half to one-third the 'standard' adult serves on widely used FFQs. Dietary information on very young children such as these provide preliminary data for consideration of refinements of dietary assessment tools and more specific dietary guidance for this age group.

Table 1. Consumption of meats and foods containing meat by type: percent
consuming, frequency of intake per 30 child-days (i.e. per month) and
mean intake (g per meal) (95% CI), among CAPS children aged around 18
months (n = 429)

                                      Percent    Frequency   Mean g/meal
Type of meat                          consuming  per month   (95% CI)

Meat and meat products (18)
Beef (1811) (a)                        45          6.7        30 (28-32)
Lamb (1812)                            24          1.7        24 (20-29)
Pork (1813)                            12          1.0        24 (16-23)
Bacon (1814)                           21          1.5        16 (11-20)
Ham (1815)                             31          2.5        19 (16-21)
Veal (1816)                           < 0.1        0.2        27 (8-45)
Poultry (183)                          59          9.4        39 (36-42)
Organ meats (184)                     < 0.1        0.1         8 (6-9)
Sausages, frankfurts, saveloys         31          4.0        43 (38-48)
  (185)
Processed meats (186)                  15          2.3        21 (17-25)
Mixed beef/veal dishes (187)            4          0.5        44 (30-58)
Mixed lamb/pork dishes (188)          < 0.1      < 0.1       171 (-)
Mixed poultry/game dishes (189)         1          0.1        48 (13-83)
Cereal-based meat products (13)
Meat pies & sausage rolls (1345)        9          1.1        61 (46-76)
Pizza (1352)                           11          1.3        56 (47-64)
Hamburgers (1354)                     114          1.1        64 (56-72)
Chiko/spring rolls & dim sims                      0.5        48 (29-68)
  (1356)
Infant foods (31)                      19
Infant poultry & meat based dinners                2.4        59 (40-73)
  (3133)
All types of meats & products          97         36

(a) Numbers in parentheses are 1995 NNS 3- and 4-digit food codes.

Table 2. The 30 most commonly eaten meats and foods containing meat
(includes cereal-based products and infant foods) and portion sizes
(median and inter-quartile range) consumed per meal by CAPS children
aged around 18 months

Food name                                        Amount in g/meal
                                    No. of
                                    meals   Median  Inter-quartile range

Beef mince, regular simmered,       149     27      15             45
  drained
Chicken McNuggets (McDonald's)      118     48      29             66
Sausage beef, grilled                98     42      23             60
Ham leg, lean only                   49     15       6             24
Devon/fritz                          44     24      13             38
Chicken breast, baked lean & skin    42     21      10             36
Frankfurt, simmered                  39     39      21             54
Hamburger, cheeseburger              36     57      39             87
  (McDonald's)
Chicken breast, baked lean only      35     18       9             51
Bacon breakfast rasher, fried        34     12       6             23
Chicken breast quarter, rotisserie   33     33      12             60
  lean & skin
Chicken, unspecified baked lean &    29     24       9             41
  skin
Ham leg, lean & fat                  27     15       9             21
Chicken drumstick, baked lean only   27     39       9             54
Chicken noodle junior canned         25     52      40             67
  dinner
Chicken, unspecified baked lean      22     26      11             60
  only
Chicken drumstick, baked lean &      22     33      14             54
  skin
Sausage roll, individual size        20     57      26             69
Ham shoulder, lean & fat             19     15       9             24
Chicken roll                         19     18       9             24
Pasta, beef and tomato junior        19     53      35             70
  canned dinner
Beef boneless, unspecified cut       17     21      15             45
  cooked lean
Pizza supreme                        16     47      39             93
Meat pie, individual size            15     45      24             90
Chicken, unspecified rotisserie      15     21       9             51
  commercial lean & skin
Chicken and vegetable junior         15     73      51             73
  canned dinner
Pizza thick crust commercial         14     51      30             66
Steak and vegetable junior canned    14     71      51             75
  dinner
Ham, unspecified cut lean & fat      13     18      14             36
Lamb, boneless unspecified cut       13     15       9             39
  cooked lean & fat

Table 3. Comparison of the nutrient contribution (mean & 95% CI for
mean) of the average amount of red, white and processed meats consumed
per meal by CAPS children aged around 18 months

                                             Type of meat (a)
                     Red                     White

No. of meals         392                     453
Amount/meal (g)       29 (27-31)              38 (35-41)
Nutrient:
Energy (kJ)          243 (224-262)           394 (361-427)
Protein (g)            7.2 (6.6-7.8)           8.9 (8.2-9.6)
Total fat (g)          3.2 (2.9-3.5)           5.8 (5.5-6.1)
  -- Sat fat (g)       1.4 (1.3-1.5)           2.0 (1.8-2.2)
  -- Mono fat (g)      1.4 (1.3-15)            2.5 (2.3-2.7)
  -- Poly fat (g)      0.14 (0.10-0.18)        0.60 (0.54-0.66)
Cholesterol (g)       22 (20-24)              40 (37-43)
Sodium (mg)           37 (29-45)              85 (74-96)
Iron (mg)              0.72 (0.69-0.75)        0.37 (0.33-0.41)
Zinc (mg)              1.37 (1.26-1.48)        0.51 (0.46-0.56)

                        Processed

No. of meals            445
Amount/meal (g)          28 (26-31)
Nutrient:
Energy (kJ)             268 (239-297)
Protein (g)               4.8 (4.3-5.3)
Total fat (g)             4.6 (4.1-5.1)
  -- Sat fat (g)          1.9 (1.7-2.1)
  -- Mono fat (g)         2.1 (1.8-2.4)
  -- Poly fat (g)         0.24 (0.18-0.30)
Cholesterol (g)          17 (15-19)
Sodium (mg)             301 (275-327)
Iron (mg)                 0.57 (0.51-0.63)
Zinc (mg)                 0.81 (0.72-0.90)

(a) Red (beef 1811/1816, lamb 1812 and organ meats 184 & dishes 187);
white (pork 1813, poultry 183 & dishes 189) and processed (bacon 1814,
ham 1815, frankfurts/sausages/saveloys 185 and processed deli
meats and meat pastes 186).

Table 4. Percent consuming and contribution of core food groups to
energy, macronutrients and micronutrients for which meat contributed 5%
or more of total intake among CAPS children aged around 18 months

                       Cereal-
Food group    Cereals  based    Dairy   Meat   Fish   Eggs   Fruit
Code no (a)   12       13       19      18     15     17     16

Consumers     100      93       100     92     29     33     88
Energy         15.0     9.1      39.3    7.0  < 1.0  < 1.0    4.9
Protein        12.4     5.8      47.8   17.4    1.7    1.4    2.0
Carbohydrate   24.0     9.4      20.5  < 1.0  < 1.0  < 0.1    9.3
Total Fat       3.1     9.9      49.7   10.6  < 1.0    1.3  < 1.0
  Sat           1.4     9.7      61.1    8.0  < 1.0  < 1.0  < 0.1
  Mono          1.8    10.3      40.4   14.2    1.0    1.3    0
  Poly          9.5     7.1      21.4    7.0    2.3    4.8  < 0.1
Phosphorus     10.1     4.9      60.6    7.9    1.0  < 1.0    1.9
Sodium         20.8    13.3      27.2   12.5    1.2  < 1.0  < 1.0
Potassium       5.8     2.4      46.9    5.0    1.0  < 1.0   10.6
Magnesium      18.0     4.0      41.2    5.2  < 1.0  < 1.0    7.2
Iron           32.9     6.8      10.0    9.4  < 1.0    1.2    5.4
Zinc           10.0     5.1      44.6   16.9  < 1.0  < 0.1    2.6
Niacin         25.3     5.3      12.8   15.8    1.5  < 0.1    2.8

                                                 Non-milk
Food group       Vegetables     Infant foods     beverages
Code no (a)      23             31                11

Consumers        97             43                98
Energy            4.5            3.2               8.3
Protein           3.4            2.9             < 1.0
Carbohydrate      5.0            4.0              17.3
Total Fat         4.1            2.3             < 0.1
  Sat             3.1            4.3               0
  Mono            4.5            4.9               0
  Poly            4.2            8.0               0
Phosphorus        3.2            3.1               1.3
Sodium            2.9            2.0               1.8
Potassium        10.6            3.4               7.5
Magnesium         7.2            3.4               4.6
Iron              7.0           15.1               1.5
Zinc              4.3           10.1             < 1.0
Niacin            5.9            4.7               8.1

(a) Two-digit codes as used in the 1995 NNS.

Table 5. Effect size for statistically significant relationships between
quintile of meat intake and anthropometric and dietary indices among
CAPS children aged around 18 months (n = 429)

                     Q1         Q2         Q3         Q4         Q5

Weight (kg)          11.2       11.8       11.5       11.6       11.9
                     (1.3)      (1.3)      (1.5)      (1.4)      (1.4)
Height (cm)          83.1       84.0       82.6       83.5       84.5
                     (2.9)      (2.9)      (4.9)      (3.0)      (2.8)
Energy (MJ)        4020.0     4211.0     4324.0     4402.0     4962.0
                   (928.0)    (859.0)    (864.0)    (960.0)    (108.0)
% Energy from:
Protein              14.3       15.2       15.5       15.9       16.9
                     (2.7)      (2.4)      (2.2)      (2.2)      (3.0)
CHO                  52.7       50.5       50.0       49.0       47.0
                     (8.0)      (6.8)      (6.4)      (7.0)      (6.6)
Total fat            34.7       36.0       36.2       36.8       37.6
                     (6.3)      (5.4)      (5.1)      (5.3)      (5.1)
-MUFA                11.1       11.3       11.6       12.5       12.8
                     (4.3)      (2.2)      (2.6)      (3.5)      (1.9)
Cholesterol mg/MJ    24.7       28.4       28.6       32.2       31.9
                    (11.9)      (9.9)      (9.2)     (13.0)     (11.3)
Calcium mg/MJ       194.0      195.0      184.0      163.0      151.0
                    (67.0)     (56.0)     (45.0)     (51.0)     (51.0)
Magnesium mg/MJ      33.7       33.3       33.1       31.0       31.0
                     (5.3)      (5.7)      (5.0)      (4.8)      (5.9)
Sodium mg/MJ        235.0       25.0       26.0       28.0      292.0
                    (56.0)     (53.0)     (51.0)     (80.0)     (75.0)
Riboflavin mg/MJ      0.436      0.439      0.388      0.373      0.324
                     (0.177)    (0.179)    (0.097)    (0.136)    (0.125)
Thiamine mg/MJ        0.222      0.221      0.199      0.194      0.177
                     (0.093)    (0.100)    (0.060)    (0.095)    (0.076)
Retinol mg/MJ        70.4       76.1       72.9       65.9       59.3
                    (29.9)     (35.1)     (29.8)     (24.2)     (19.2)

                   P-value  Effect size (a)

Weight (kg)         0.008       0.51
Height (cm)         0.003       0.47
Energy (MJ)         0.000       0.69
% Energy from:
Protein             0.000       0.87
CHO                 0.000      -0.71
Total fat           0.013       0.46
-MUFA               0.001       0.40
Cholesterol mg/MJ   0.000       0.61
Calcium mg/MJ       0.000      -1.04
Magnesium mg/MJ     0.000      -0.46
Sodium mg/MJ        0.000       0.75
Riboflavin mg/MJ    0.000      -0.64
Thiamine mg/MJ      0.003      -0.48
Retinol mg/MJ       0.002       0.37

(a) (Q5-Q1 mean)/larger SD.

Acknowledgments

We are grateful to the CAPS families who supplied dietary and other data for this study, and the CAPS research team, particularly Stephen Leeder, in providing scientific and organisational support. The weighed food record collection for CAPS was funded by the Commonwealth Department of Health and Aged Care, the Cooperative Research Centre for Asthma and the Woolcock Institute of Medical Research, University of Sydney. The analysis for this paper was partly funded by Meat and Livestock Australia and the NSW Centre for Public Health Nutrition (University of Sydney and NSW Department of Health). Dr Lahti-Koski was supported by the Academy of Finland, the Juho Vainio Foundation, the Finnish Food Research Foundation and the Finnish Cultural Foundation.

References

1. Australian Bureau of Statistics. National Nutrition Survey. Nutrient intakes and physical measurements 1995. Catalogue No 4805.0. Canberra: ABS; 1998.

2. USDA. Food Surveys Research Group. Food and Nutrient Intakes by Children 1994-96, 1998, Table set 17, USDA Continuing Survey of Food Intakes by Individuals. Beltsville: Maryland. http://www.barc.usda.gov/bhnrc/foodsurvey/home.htm.2002

3. Federation of American Societies for Experimental Biology (FASEB) Life Sciences Research Office. Third Report on Nutrition Monitoring in the United States, Volume 2. Washington: US Government Printing Office; 1996. p. VA62-4, VA96-150.

4. Ganji V, Betts N, Whitehead MS. Nutrient intakes of 1-3, 4-6 and 7-10 year age group children: analysis of diets reported in 1987-88 Nationwide Food Consumption Survey. Nutrition Research 1996;15:623-31.

5. Cowin I, Emmett and the ALSPAC study team. Diet in a group of 18-month-old children in South West England, and comparison with the results of a national survey. J Human Nutr Diet 2000;13:87-100.

6. Emmett-P, Rogers I, Symes C and the ALSPAC Study Team. Food and nutrient intakes of a population sample of 3-year-old children in the South West of England in 1996. Public Health Nutr 2002;5:55-64.

7. Marks G, Webb K, Rutishauser IHE, Riley M. Monitoring food habits in the Australian population using short questions. Canberra: Commonwealth Department of Health and Aged Care; 2001.

8. Picciano MF, Smiciklas Wright H, Birch LL, Mitchell DC, Murray-Kolk L, McConahy KL. Nutritional guidance is needed during dietary transition in early childhood. Pediatrics 2000;106:109-14.

9. Thompson FE, Byers T. Dietary Assessment Resource Manual. J Nutr 1994;124:2245S-317S.

10. Australian Bureau of Statistics. Measuring Dietary Habits in the 2001 National Health Survey. Australia. Occasional Paper No 4814.0.55.001. Canberra: ABS; 2003.

11. Wilson D, Taylor A, Chittleborough C. The second computer assisted telephone interview (CATI) forum: the state of play of CATI survey methods in Australia. Aust N Z J Public Health 2001;25:272-4.

12. New South Wales Department of Health. NSW Child Health Survey 2001. NSW Public Health Bulletin (Supplement) 2002;13(S-3).

13. Smith A, Kellett E, Schmerliab Y. The Australian Guide to Healthy Eating. Commonwealth Department of Health and Family Services. Canberra: CDHFS; 1998.

14. National Health and Medical Research Council (NHMRC). Dietary guidelines for children and adolescents. Canberra: Australian Government Publishing Service; 2003.

15. Skinner JD, Carruth BR, Houck K, Coletta F, Cotter R, Ott D, et al. Longitudinal study of nutrient and food intakes of infants aged 2 to 24 months. J Am Diet Assoc 1997;97:496-504.

16. Munoz KA, Krebs-Smith S, Ballard-Barbash R, Cleveland L. Food intakes of US children and adolescents compared with recommendations. Pediatrics 2000;100:323-9.

17. Cowin I, Emond A, Emmett P, and the ALSPAC study team. Association between composition of the diet and haemoglobin and ferritin levels in 18-month-old children. Eur J Clin Nutr 2001;55:278-86.

18. Thane CW, Walmsley CM, Bates CJ, Prentice A, Cole TJ. Risk factors for poor iron status in British toddlers: further analysis of data from the National Diet and Nutrition Survey of children aged 1.5-4.5 years. Public Health Nutr 2000;3:433-40.

19. Soh P, Ferguson EL, McKenzie JE, Skeaff S, Parnell W, Gibson RS. Dietary intakes of 6-24-month-old urban South Island New Zealand children in relation to biochemical iron status. Public Health Nutr 2002;5:339-46.

20. Oti-Boateng P, Seshadri R, Petrick S, Gibson RA, Simmer K. Iron status and dietary iron intake of 6-24-month-old children in Adelaide. J Paediatr Child Health. 1998;34:250-3.

21. Mirshahi S, Peat JK, Webb K, Tovey E, Marks G B, Mellis C, et al. The childhood asthma prevention study (CAPS): design and research protocol of a randomized trial for the primary prevention of asthma. Control Clin Trials 2001;22:333-54.

22. Mihrshahi S, Vukasin N, Forbes S, Wainwright C, Krause W, Ampon R, et al. Are you busy for the next 5 years? recruitment in the Childhood Asthma Prevention Study (CAPS). Respirology 2002;7:147-51.

23. Gregory JR, Collins DL, Davies PSW, Hughes JM, Clarke PC. National Diet and Nutrition Survey: Children aged 1.5 to 4.5 years. Volume 1: Report of the Diet and Nutrition Survey. London: HMSO;1995.

24. Baghurst KI, Baghurst PA. The measurement of usual dietary intake in individuals and groups. Trans Menzies Foundation 1981;3:139-60.

25. Smith W, Mitchell P, Reay E, Webb K. Validity and reproducibility of a self-administered food frequency questionnaire in older people [The Blue Mountains Eye Study]. Aust N Z J Public Health 1998;22:456-63.

26. Hodge A, Patterson AJ, Brown WJ, Ireland P, Giles G. The Anti Cancer Council of Victoria FFQ: relative validity of nutrient intakes compared with weighed food records in young to middle-aged women in a study of iron supplementation. Aust N Z J Public Health 2000;24:576-83.

27. Rangan AM, Ho RWL, Blight GD, Binns CW. Haem iron content of Australian meats and fish. Food Australia 1997;49:508-11.

28. Baghurst K, Record S, Leppard P. Red meat consumption in Australia: intakes, nutrient contribution and changes over time. Aust J Nutr Diet 2000;57:3S-35S.

University of Sydney

K. Webb, PhD, MPH, BSc, Co-Director, NSW Centre for Public Health Nutrition, Senior Lecturer, School of Public Health and Human Nutrition Unit, School of Molecular and Microbial Biosciences

T. Katz, MSc (Nutr & Diet), Research Officer and Nutritionist, NSW Centre for Public Health Nutrition

J. Peat, PhD, Associate Professor, School of Paediatrics and Child Health

Deakin University, Melbourne

I. Rutishauser, MSc, Honorary Fellow, School of Health Sciences

Department of Epidemiology and Health Promotion, National Public Health Institute, Helsinki, Finland

M. Lahti-Koski, PhD, MSc, Research Fellow

The Children's Hospital, Westmead

N. Knezevic, BAppSc (Hons), MND, Research Dietitian, Clinical Epidemiology Unit

S. Mihrshahi, BSc(Hons), MPH, Research Officer, Clinical Epidemiology Unit

Correspondence: K. Webb, Human Nutrition Unit, Building GO8, University of Sydney, NSW 2006. E-mail: [e-mail] K. Webb oversaw the dietary component of CAPS, and the initial processing of the dietary data, drafted the manuscript, and collaborated on the analysis. I. Rutishauser conducted the data analysis and contributed to preparation of the manuscript. T. Katz conducted the literature review and assisted in preparing the manuscript. N. Knezevic collected, processed and cleaned the dietary data and collaborated on planning the paper. M. Lahti-Koski conducted some further data cleaning and processing. J. Peat advised on the statistical analyses. S. Mihrshahi was CAPS study manager and oversaw the recruitment and data collection for the 18 months' assessments.