NEW EVIDENCE ON THE NUTRITION OF BRITISH SCHOOL CHILDREN AND CONCLUSIONS FOR SCHOOL MEALS

ABSTRACT

The National Diet and Nutrition Survey: Young people aged 4 to 18 years published in 2000 was based on the actual diets and intake of nutrients in 1997 of a representative sample of 2,000 British children.¹ One stated purpose of the Survey was to assist in the development of "dietary guidelines for food provided by schools." It was necessary to "monitor the extent of the deviation of the children's diet from that recommended as optimum for health." The Survey uses the Reference Nutrient Intakes (RNIs) set out in Dietary Reference Values for Food Energy and Nutrients for the United Kingdom published in 1991 by the Department of Health.² The Survey covers energy, protein, carbohydrates, fat and fatty acids, vitamins and minerals. Table 1 shows the percentage of children aged 4 to 18 years in four age brackets who failed to reach RNI levels for 8 vitamins and Table 2 shows those who failed to reach the level for 7 minerals. More than 10 years have passed since the RNI values were published and there are now many reasons for believing that they are out-of-date and in need of revision.

VITAMIN A IN THE DIET OF THE BOYS AND GIRLS

It is seen in Table I that vitamin A is the first line for boys and girls and that more than half of the boys and girls fail to reach the RNI intake level. This prompts the question: How certain is it that the RNI values for Vitamin A are correct? Table 3 compares the British RNI with the American RDA (Recom mended Dietary Allowance)³ and the French ANC (Apports Nutritionnels Conseilles).⁴

TABLE 1

Percentage of British children aged 4 to 18 failing to reach RNI levels for average daily intake of vitamins from food sources in 1997 boys

	boys
	4-6	7-10	11-14	15-18
vitamin A	42	57	65	69
folate B9	4	13	33	18
riboflavin B2	10	12	25	28
vitamin C	8	8	14	20
pyridoxine B6	3	1	11	5
thiamin B1	6	1	6	7
niacin B3	2	0	4	0
cobalamin B12	0	0	1	1
	girls
vitamin A	42	59	74	70
folate B9	5	22	50	52
riboflavin B2	7	21	40	40
vitamin C	9	9	20	26
thiamin B1	8	2	3	13
cobalamin B12	0	2	3	10
niacin B3	2	2	2	5
pyridoxine B6	5	1	3	18

The lower value of the British recommendations and their date of publication suggests the need for revision. This view is reinforced by the many publications on vitamin A since 1991. Throughout the last ten years more than one paper describing or discussing research results involving vitamin A has been published in some country every week. No one individual working full-time could keep up-to-date with the still unfolding story of the biological role of vitamin A. The research literature would, of course, not be so extensive if it were not for the multiplicity of roles of vitamin A. The growth and development of the cardiovascular system, for example of the smooth muscle cells of the arteries, require vitamin A. Proliferation and migration of the cells of the nervous system and the brain require vitamin A. The argument made in the US for their RDA for vitamin A places particular emphasis on the prevention of blindness or night blindness among children in the underdeveloped world which is caused by different degrees of vitamin A deficiency. The growth and development of the pancreas and the particular cells that produce insulin depend on adequate vitamin A and a deficiency is one of the many causes of diabetes. Vitamin A deficiency impairs fertility of men and women. Vitamin A deficiency increases the risk of cancer and is the subject of extensive cancer research programmes. The increasing knowledge and understanding of the role of vitamin A provide the argument for continuous review of recommendations.

TABLE 2

Percentage of British children aged 4 to 18 failing to reach RNI levels for average daily intake of minerals from food sources in 1997

	boys
	4-6	7-10	11-14	15-18
potassium	5	43	88	85
magnesium	11	56	86	75
zinc	79	73	84	68
iron	14	39	60	43
calcium	10	19	79	68
copper	38	33	35	50
iodine	19	24	37	33
	girls
zinc	89	83	96	72
magnesium	13	75	97	97
iron	28	59	96	93
potassium	5	46	97	99
calcium	15	29	79	76
copper	52	49	60	78
iodine	27	39	61	61

TABLE 3

	British RNI 1991 aged 15 to 18	US RDA 1989 aged 15 to 18	French ANC 1992 aged 13 to 19
girls	600	800	800
boys	700	1,000	1,000

There is another reason for a review of the RNI for vitamin A. The Survey of the diet and nutrition of children aged 4 to 18 divides vitamin A into "preformed retinol" and "carotene (beta-carotene equivalent)". The RNI for vitamin A assumes that 936 mcg of carotene can be regarded as equivalent to 1 mcg of retinol." Other research suggests that this conversion factor is wrong. A Dutch study, for example, suggests that 21 is more appropriate.⁵ Moreover, doubt is thrown on the validity of any conversion factor as the conversion of carotene to retinol depends on other dietary components and carotene has important nutritional roles without conversion to retinol.

FOLATE IN THE DIET OF BOYS AND GIRLS

The percentage of boys and girls failing to reach RNI levels for folate occupies the second line in Table I and girls are seen to be more deficient than boys. Congenital malformations and in particular neural tube defects can be caused by folate deficiency of women around the time of conception. Folate deficiency is also one cause of low birthweight and associated maldevelopment of body systems.

Folate is essential for DNA synthesis and for the replication of all cells. Research at the US Center for Environmental Health in the 1980s showed that folate deficiency increased chromosomal breakage and genetic mutations in women. It is not, therefore, surprising to discover that a reduced concentration of folate in the body fluids slows down growth and increases the mutation rate of somatic as well as germ cells. Research and discussion in the USA on the consequences of folate deficiency to the health of people of all ages, including the risk of heart disease, stroke, cancer and nervous system disorders and Alzheimer's disease, led to the conclusion that the average American could reduce his risk of all these diseases by increasing his intake of folate. The US Congress introduced mandatory fortification of all grain products with 140 mcg of folate per 100 g of grain from 1st January 1999.⁶

The Survey of schoolchildren shows that the lower 2.5 percentile for intake of folate by girls aged 15 to 18 was 88 mcg/d and 30 per cent of these girls had folate intakes between 100 and 150 mcg/d. The British RNI is 200 mcg/d and recent studies in the USA suggest that health improves up to 400 mcg/d, the level now recommended in the UK for women planning pregnancy or pregnant.⁷

REASONS FOR REVISING THE RNIs FOR PROTEIN

The Survey shows that the recorded daily intake of protein by children is higher than the levels recommended by the RNIs as seen in Table 4.

Is there any evidence that boys and girls would be healthier if they ate less protein? Should the careful parent reduce protein-rich foods in family meals? There is no such evidence to be found. Moreover, protein is more significantly associated with growth in children than any other dietary constituent.⁸ Protein intake influences growth over the whole time from preconception to the completion of growth in the teens. The Harvard Longitudinal Studies of Child Health and Development followed children from the first trimester of their mothers' pregnancy to the age of 18. One paper from this study published in 2000 showed the relationship of childhood diet to adolescent growth in girls and the greater growth promoting effect of animal protein compared to vegetable protein.⁹

TABLE 4

Actual daily intake of protein by British boys and girls in 1997 compared to recommended intakes (RNIS)

	boys
ages	4-6	7-10	11-14	15-18
RNI g/d	19.7	28.3	42.1	55.2
actual median intake g/d	47.9	54.2	64.8	75.5
	girls
RNI g/d	19.7	28.3	41.2	45.4
actual median intake g/d	42.7	50.7	51.7	53.9

Conformity with the RNIs for protein would require substantial differences in the protein content of food at different ages. Is it realistic to segregate Actual daily intake of protein recommended intakes (RNIs) by British boys and girls in 1997 compared to children in different age groups either at home, in family meals, or in institutional catering? Children eat to appetite or in other words they eat the energy they need. Table 5 shows the protein per megajoule proposed by the RNI using the "Estimated Average Requirement" of energy from the 1991 document on Dietary Reference Values and also the record in the Survey of actual protein intake of boys and girls.

It is seen in Table 5 that the actual intake of protein per unit of energy does not differ much through the 4 age groups. In contrast the recommendations, the RNIs, for the age range 4 to 6 are 36 per cent lower for boys and 41 per cent lower for girls than the recommendations for children aged 15 to 18 years. This is not compatible with real life wherever children of different ages sit round the family table sharing the same dishes with the only difference in the size or number of servings they eat. Can the lower protein content per unit of energy in the food of children aged 4 to 6 compared with higher protein food for the 15 to 18 year olds be in the best interest of children's health? This is another reason for revision of the RNIs.

TABLE 5

Protein daily intake of British boys and girls aged 4 to 18 per megajoule energy intake as recommended (RNIS) and as recorded in 1997

	boys
ages	4-6	7-10	11-14	15-18
RNI gIMJ	3.05	3.43	4.54	4.80
actual median intake gIMJ	7.46	7.41	7.79	7.82
	girls
RNII g/MJ	3.05	3.89	5.20	5.14
actual median intake g/MJ	7.45	7.54	7.41	8.08

MINERALS IN THE DIET OF BOYS AND GIRLS

The Survey shows that boys and girls in the age range 4 to 6 eat food on average similar to the 15 to 18 year olds but that they eat less. The older boys eat about 50 per cent more and the older girls eat about 20 per cent more than when they were 4 to 6 years of age. The RNIs, however, recommend very much larger differences in food composition, particularly in the protein and mineral content, for children in the younger as compared to the older age groups. Table 6 shows the potassium intake per megajoule at different ages as recommended and as actually recorded:

TABLE 6

Potassium daily intake of British boys and girls aged 4 to 18 per megajoule energy intake as recommended (RNIS) and as recorded in 1997

	boys
ages	4-6	7-10	11-14	15-18
RNI g/MJ	154	243	334	304
actual median intake gIMJ	297	85	282	288
	girls
RNIg/MJ	170	275	391	396
actual median intake g/MJ	290	302	308	301

Growth at all ages is partly controlled by hormones including recombinant growth hormone (GH) and the insulin-like growth factor (IGF) system.10 The blood levels of these hormones depend upon energy intake and also on consumption of protein and the minerals potassium, magnesium and zinc. Is the diet with substantially lower potassium content recommended by the RNIs for boys and girls aged 4 to 6 as seen in Table 6, compatible with the growth desirable at this age? The different recommendations of other countries, as shown for young school boys in Table 7, strengthens the case for revision of British RNIs for several essential minerals.

Table 2 showed that a much lower percentage of children aged 4 to 6 fail to reach RNI levels compared to children aged 15 to 18. This is largely explained by the low levels of the RNIs for children aged 4 to 6 and not by any evidence of a much reduced requirement for minerals in this age group. Nutrient intakes in relation to energy intake at different ages of children, and indeed of adults, merits study in the future when the dietary reference values come to be revised.

TABLE 7

	United Kingdom	USA	France
calcium mg/d	450	800	700
phosphorus mg/d	350	800	800
magnesium mg/d	120	120	180
zinc mg/d	6.5	10	10
selenium mcgld	20	20	30

CHRONIC SICKNESS OF CHILDREN IN BRITAIN

There is a serious background to the study of 1997 showing low intakes of essential nutrients by school children. There was an increase in long-standing illness among children in Britain during the last 25 years of the 20th century. "Long-standing illness" includes disabilities and is also called "chronic sickness" or "long-term illness or disability" in the reports of the National Household Survey and elsewhere." The chronic sickness reported by school children or their parents in the National Household Survey is shown in Table 8.

Studies of the health of the children in individual cities show that the decline in child health was concentrated among the children living in the poorer wards of the city and increased with the increasing numbers of poor children. The health of children in better off wards improved but the decline in health of children in some city wards was more than enough to dominate national statistics. Three quarters of the increase in the chronic sickness of school children shown in Table 8 had its origin before the age of 5 as shown in Table 9.

TABLE 8

Chronic sickness of British school children aged 5 to 15 in 1972 and 1998

	1972	percentage 1998	estimated number 1998
boys	9	21	790,000
girls	6	19	680,000
all schoolchildren	8	20	1,470,000

TABLE 9

	1972	percentage 1998	estimated number 1998
boys	5	15	280,000
girls	3	15	270,000
all children under 5	4	15	550,000

Some of this chronic sickness or disability may be curable but the statistics show chronic sickness to be cumulative and passed on from age group to age group. The chronic sickness of British children becomes chronic sickness of adults which can be either ameliorated or aggravated by life style and treatment as the years pass.

CHRONIC SICKNESS AND THE ECONOMY

Chronic sickness seriously affects economic activity. In the 1991 Census 35 per cent of men recorded as economically inactive, or over I million men under 65, were chronically sick. There were also I million women under age 65 who were economically inactive and were also sick. The chronically sick were also over-represented among the unemployed seeking work. The Census showed that 8.3 per cent of unemployed men and women were chronically sick whereas only 4.5 per cent of men and 3.5 per cent of women who were working were chronically sick or disabled. The unemployed chronically sick numbered 202,000.

The costs to the economy of the economically inactive and the economically active who are at any time unemployed due to sickness include opportunity costs or in other words the national income that might have been but for chronic sickness. In addition the cost of chronic sickness includes the cost to the social security budget, to the National Health Service and to Local Authorities for caring services.

Tables 8 and 9 summarize time trends in childhood chronic sickness and the tables prompt the question: What will happen if there is no change of direction? The answer may be summarised: There will be a gradual increase in chronic sickness and disability at all ages as the chronically sick children now living and their successors grow up. It must be accepted that there has been something profoundly wrong with the public health policies of Governments during the last 25 years, and perhaps longer, which have led to such time trends. There must be a change in direction. What then must be done?

THE INCREASE IN THE NUMBERS OF POOR CHILDREN AND THEIR POOR MOTHERS

The Department of Social Security published a report in 1997 which said that the proportion of children in households with incomes less than half the national average had g(own from 10 per cent in 1979 to 32 per cent in 1994/5.¹² The journal Poverty reported a further increase to 35 per cent in 1998/99. The number of poor children defined in this way increased from 1.4 to 4.5 million during the 20 years. The number of children in low income households was one reason why there was an increase in the need for school meals, but this was not acknowledged during the same years.

About 70 per cent of these low income households in 1979 and 1994/5 had no member with an earned income. There was also a great increase in the number of lone parents during these years. Social security statistics show an increase in the number of lone parents on income support from 250,000 in 1971 to 1.1 million in 1996.¹² The percentage of children living in fatherless families increased from 7 per cent in 1972 to 21 per cent in 1998. One-parent families can often be particularly dependent on adequate school meals.

Studies of diets in poor households have repeatedly shown that mothers give priority to their children and eat a less nutritious meal themselves. In 1998 the Department of Social Security published a study entitled What happens to lone parents?¹⁴ What happened was that the poverty of lone mothers resulted in a 100 per cent increase in long-standing or chronic illness in an average time of about 4 years. The World Health Organisation in 1998 showed that the United Kingdom was placed 23rd among member countries ranked for untimely death of women aged 15 to 59. In 1950 the UK was placed 5th. The United Kingdom has been overtaken in the health of its younger women by nearly all developed countries.

One consequence of poor health of women is low birthweight of their babies which increases the risk of many kinds of sickness and disability. The British Medical Association said in its book Growing Up in Britain in 1999:

"Figures from the World Health Organisation show that in 1992 the UK had the highest rate of low birthweight babies in the European Union; the UK's rate was on a par with Albania."

Recent WHO reports confirm this lowly position of the UK. In the UK the percentage of low birthweight babies, born under 2,500g. varies from 3.0 per cent in the more prosperous wards of cities to 12 per cent or more in the poorer wards and figures as high as 20 per cent have recently been recorded.¹⁵ Risk continues as baby weight increases over the traditional definition of low birthweight as 2,500g and up to birthweight of 3,000g which has been called the safety threshold below which the risks of illness and disability increase steadily with failing weight. About one quarter of British babies, 24 per cent in 1997, are born under 3,000g. The majority of the men and women who have babies have never been taught about the importance of health before conception and how to reduce the chances of having a child with long-standing illness or disability. There are few schools that convey this knowledge.

A report to the US Congress found that 12 per cent of American children were suffering from one or more neurological disorders and in 75 per cent of cases the disorder had its origin before birth in the defective health of parents including defective nutrition.¹⁶ Too small a weight at birth increases the risk of neurological disorders including cerebral palsy, dyspraxia, epilepsy and visual and auditory handicap and learning disability which is defined for school children in the UK as Special Educational Needs (SEN). In 1997 18 per cent of children or 1.5 million children had SEN.

DIET OF CHILDREN AND THE INCOME OF THEIR FAMILIES

The Survey shows the intake of dietary components and all nutrients for children in households at 5 different levels of income not allowing for numbers of children. The lowest weekly household income bracket was "less than £160" and the highest "£600 and over." The daily vitamin C intake of girls declined from 83.3 mg/day in the highest income bracket to 46.9 mg/day in the lowest, a decline by 44 per cent. This is the biggest decline with lowered income of any nutrient but there is some decline of nearly every dietary component. Vitamin A intake, for example, declined by 19 per cent from the highest to lowest income bracket. In some cases, for example thiamin BI, the children's nutrient intake was still above recommended level, the RNI, even at the lowest income level. In contrast vitamin A intake of the majority of children was less than the RNI level even in the highest income bracket but the deficiency was greater and more children had diets deficient in vitamin A in the lowest income brackets.

The Survey shows an average household income of about £400 a week. Below £400 a week the median intake of most micro-nutrients, of protein and energy began to decline. A household income of half average income is one commonly used definition of poverty. In the context of the Survey this is only £200 a week. An increase of the income of all households to £400 a week is an aim for the future. Can the improvement in the nutrition and diet of children be given priority in face of these figures? School lunches are the most cost-effective way of giving priority to the nutrition of children. Today school breakfasts and meals as part of after-school clubs are beginning to extend this priority.

The National Food Survey shows the expenditure per head on food according to the number of children in a household. In prosperous households with a weekly income of £640 and over the expenditure on food per head is little affected by the number of children. However, in poorer households with a weekly income between £160 and £330, and in households with less than £160 a week, the expenditure on food per head declines steeply as the number of children per household increases from 1 to 2, to 3 and more. Thus in three child households the prosperous family is recorded as spending £16.14 per head on food while in the poorest families with 3 children only £7.26 per head was spent on food. School meals can feed children according to their needs and appetites without any bias against those who have brothers and sisters.

THE DECLINE OF SCHOOL MEALS

In the United Kingdom an Act of 1944 required all local educational authorities to provide school meals "suitable in all respects for those pupils who wanted them." In May 1975 in England 5,147,840 children or 67.9 per cent ate a lunch provided at school.¹⁷ In 1980, however, the Education Act removed all obligation to provide school meals except for pupils whose parents received social security benefits. The 1980 Act was followed by a decline in the number of meals provided each day from 4,855,000 in 1979-80 to 3,010,000 in 1989-9, or by 38 per cent.¹⁸ In June 1997 David Blunkett, Secretary of State for Education and Employment, said:¹⁹

"For nearly 18 years we have seen the quality of school dinners deteriorate and the number of children eating them drop significantly. Last year 43 per cent of children took meals compared with 64 per cent in 1-9."

The cost of school meals and milk fell by an even larger 55 per cent from £796 million in 19799780 to £359 million in 1989-90 at 1989-90 prices. Moreover, the school meals provided in 1989-90 were cheaper and simpler than provided in the earlier year costing only £119 per child-year compared to £179 in 1980-81. In 1999 a memorandum from the Department of Education and Employment to the House of Commons Education and Employment Committee inquiry on school meals said that the cost to parents of a school lunch was £1.28 and in the inquiry free school lunches were said to cost local authorities £1.07, both figures substantially below those for 1979-80 in real terms.

The cost of school meals was reduced in particular by reducing their protein content. The Nutritional Standard of the School Dinner, published by the Department of Education and Science in 1965, recommended that the meals should contain 9329 g (including 18.5 g animal protein)". The 1997 Survey found that the average school meal had about 17 or 18 g of protein. In 2000 the Department of Education and Employment published two brochures entitled Healthy School Lunches as a guide for school caterers, one for primary schools and one, for secondary schools. The protein recommendations in these two brochures are shown in Table 10.

TABLE 10

Protein in school lunches in the UK: grams per day recommended in 2000 and as percentage of energy

infant school 5-6	junior school 7-10	middle school 9-13	secondary school 11-15	sixth form 17-18	secondary plus sixth 11-18	upper school 14-18
recommended protein not less than g/meal
5.9	8.5	10.9	13.0	15.0	13.3	14.1
percentage of energy from protein
4.8	6.1	7.4	8.2	8.2	8.2	8.3

The percentage of energy from protein in Table 10 has been calculated from the protein and the energy levels recommended in the brochures on school lunches. It is seen that these recent recommendations follow the RNIs of 1991 in advocating especially low protein density in the diets of the youngest school children. Other countries express the protein requirements of children or in school meals by a single number, the percentage of energy to be provided by protein. In Sweden, for example, the number is 15 per cent of energy from protein. This contrasts with the last line in Table 10.

A new Regulation Education (Nutritional Standards for School Lunches) was laid before Parliament on 12 July 2000 and came into force on 1 April 2001. This new Regulation reverses the effect of the 1980 Education Act and makes it necessary for all schools once again to offer school lunches to all their pupils. However, the new Regulation is based on food groups and says nothing at all about either the quantity or the quality of the foods which must be served at lunchtime and the meals can be grossly inadequate in amount or in their balance of essential nutrients without contravening the Regulation. This only states some of the foods that lunches must include. For example meals containing red meat must be offered twice a week to primary school children and three times a week to secondary school children. Similarly primary school children must be offered fish once a week but secondary school children must find it on offer twice a week. The younger the children the less often they are to be offered animal protein. The new Regulation does not require any monitoring of school lunches.

In contrast in the USA recent legislation requires that all States provide meals, not only for lunch but or breakfast and for snacks and for after-school which are consistent with the Recommended Dietary Allowances (RDAs).²⁰ Moreover, standard recipes and servings with known nutrient content must be used. These standards are required on a weekly basis but with recommended procedures if the consumption of, for example, protein or vitamin A, is not up to requirement in any one week.

In Britain a computer menu planner called CHOMP for under-5s in day care and a menu planner for older people in residential homes called CORA have been developed by the Caroline Walker Trust.²¹ Both CHOMP and CORA have databases of over 800 recipes and foods and substantial help files which give information about nutrition and sources of nutrients and issues around eating well. They e~iable the users to plan menus over a week or more and compare them with established nutritional guidelines. The programme then provides suggestions for alternative foods and drinks where necessary to improve the nutritional balance. Menus can be stored for future use. CHOMP also has a costing system and shopping list built into the programme.

The Survey of school children shows that in Great Britain in 1997 13 per cent of boys and 14 per cent of girls had free school meals. A further 1 per cent had meals that were subsidised and 31 per cent of boys and girls had a school meal paid for by their parents. In summary, therefore, 44 per cent of boys and 45 per cent of girls had a lunch-time meal provided by their school. Forty four per cent of boys and 45 per cent of girls went to school with packed lunches, 4 per cent had lunch at home and 4 per cent bought lunches elsewhere.

The latest available figures for free school lunches in England in January 2000 show that in primary schools 18.3 per cent of children were eligible for free lunches but only 15 per cent took them. In secondary schools 16.5 per cent were eligible but only 11.6 per cent took free lunches.22 The total of English school children taking a free lunch was 1,039,789 which may be compared to an estimate of 3.76 million English children who are poor defined by the fact that household income is less than half national average. No figures are available for England or Wales for 2000 of the number of children taking paid-for lunches or taking packed lunches to school. In Wales 20 per cent of children, or 100,437 were entitled to free school lunches in 1999-2000 but only 18 per cent took the free lunch.²³ In Scotland, however, there is an annual survey of information about school meals collected from local authorities and in January 2000 16 per cent of children were reported as taking free school meals out of the 20.3 per cent who were eligible. A further 30.0 per cent of Scottish school children paid for the lunches provided by their school.²⁴ No national figures are available on school breakfasts or after school meals or snacks neither of which are subject to regulations on nutritional content of any kind. In the absence of information about the food children eat during their school day any action by the Department must be severely limited.

National standards covering all food provided by schools and a national monitoring service is needed and would require legislation. The majority of parents or children today think that school meals are inadequate, or unattractive or too expensive. Packed lunches are thought by nearly half of all parents to be better than school meals in spite of the expense and trouble day after day of preparing them. The Survey found, however, that the packed meals were nutritionally unsatisfactory. Perhaps the most surprising finding in 1997 was that 27 per cent of boys and 23 per cent of girls were entitled to free school meals but nearly one half did not take them.

A research report prepared by the Child Poverty Action Group for the Department of Education ancf Employment published in 2001 showed that the main reason why children did not take free meals was the stigma attached to doing so as the children saw it and proposals for removing this stigma are suggested. The value of free meals varied from £0.90 to £1 .65.²⁵ Most of the meals in the seven schools in the report were unsatisfactory from a variety of different points of view, they were not nutritionally sound, they were unattractive to the children or did not even satisfy appetite. In some of these seven schools it was not only the free meals which were unsatisfactory but also the meals paid for by parents. UNISON, which includes school meals staff among its members, said in evidence to the House of Commons Committee on school meals that:¹⁹

"Our members are aware that for many of the children who miss out on their entitlement, the school dinner is likely to be their one opportunity in the day to eat a hot nutritious meal ... One in nine children start the day without breakfast and one in six goes home from school to no cooked meal."

Tables 1 and 2 show that a majority of school children might profit from a better school meal service that took account of the contemporary low intakes of particular nutrients. The Acheson report on Inequalities in Health recommended that school food policies should include free school milk and free school fruit.²⁶

PROMOTING THE BALANCE OF HOME AND WORK

Traditionally the school lunch should provide 30 per cent of a child's daily nutrients. The third edition of French nutritional allowances published in 2001 discusses the division between three daily meals.²⁷ Investigations are quoted showing that meals later in the day cannot fully compensate for missing or inadequate breakfast. The coverage of 25 per cent of daily requirements at breakfast has been advocated in France but has not and cannot always be achieved. Assuming breakfast covers a more modest 20 per cent only of requirements 80 per cent remains to be covered. If, say, 33 per cent of requirements is met by a school lunch then 47 per cent remains to be satisfied by an evening meal; generally provided by mothers. It would help working parents if they could be confident that their children have had a main meal at school which provided at least 47 per cent of requirements.

The Working Families Tax Credit (WFTC) has raised the income of poor families in work to a level at which many can achieve the "low cost but acceptable" budget of the Family Budget Unit.²⁸ WFTC is means tested and the credit, like other benefits, is tapered as income rises so that families are still in the so called "poverty trap." School meals entitlement is, however, lost when WFTC is first received so that support for child nutrition is withdrawn before the poverty trap redtrces other benefits.

More mothers, including single parents, would seek employment and more parents would take advantage of Working Families Tax Credit if they could rely on the main meal of the day for children being provided at school. Mothers with modest or low incomes when deciding whether or not to work outside the home may conclude that the loss of five free school lunches a week makes it less valuable for her to work than she thought it would be. The tasks that working parents face every evening should not be underestimated. Cooking the main meal of the day is one task which could be lightened if parents could rely on the school lunch. Parents often get a meal at work. A secondary meal, a supper, need not always be cooked and would be quicker to prepare. We conclude that the school lunch provision does not presently play the part it should play in promoting the balance of home and work. School breakfasts which have developed in many local authority areas to improve the nutrition of children are a help to working parents. After school-clubs which provide a meal or substantial snack are another essential help to some working parents.

CONCLUSION: SCHOOL MEALS SHOULD BE FREE LIKE EDUCATION

Conclusion 1. The conclusion from the Survey of children aged 4 to 18 is that many more children need school meals than the children now eligible for free lunches. According to the House of Commons Committee report on school meals free school lunches were taken by only 15 per cent of all children and not more than half of the children classified as living in poor households. Moreover, the Survey reviewed in this paper shows that many children in households not classified as poor but which have incomes below the median income also have deficient diets, with intakes of several important essential nutrients below the recommended levels. The system of provision of some free and some paid-for school meals has failed to meet the needs of children in households with low or modest incomes.

Conclusion 2.It is cost effective to feed children in schools. Evidence to the House of Commons Committee noted that good nutrition including a good breakfast improves children's ability to concentrate and their scholastic achievement and school meals help the teaching of good nutrition. It is very desirable that girls and boys too should leave school, when they are at the beginning of their reproductive years, not only well fed but with an interest in good food and the knowledge of how to prepare it.

Conclusion 3. Free school meals would not only be a present benefit but would be an important investment. The cost of paid-for school meals varies widely and the House of Commons Committee gave an estimate of £1.28 a meal. This suggests a rough estimate of £2.0 billion a year as the cost of providing all children with a free school lunch, or say, £3 billion a year with only a small increase in quality and service. This would be only a small item compared to the £100 billion plus budget of the Department of Social Security and would reduce the number of claimants and increase the number of mothers who work. Improvement in the nutrition of children would have a short term financial benefit by a decrease in the number of children sick enough to need to see the doctor. If school meals were to be carefully designed to correct what is now known about common deficiencies in children's diets substantial improvement in health may be anticipated. Financial benefits would accrue after children left school in a healthier population.

The school years are the years outstanding in the whole human life cycle when governments have a once in a lifetime never to be repeated opportunity to feed their population, to influence eating habits and to teach the importance of good food for good health. In particular the school years are the years when governments can safeguard the nutritional status of the young women who will one day bear the country's children.

REFERENCES

1. Gregory, A. and Lowe, S. (2000). National diet and nutrition survey: Young people aged 4 to 18 years. London: Stationery Office.

2. Department of Health (1991). Dietary reference values for food energy and nutrients for the United Kingdom. London: HMSO.

3. US National Research Council (1989). Recommended dietary allowances. 10th Edition. Washington DC: National Academy Press.

4. Dupin, H., Abraham, J. and Giachetti, I. (1992). Apports nutritionnels conseill~s pour Ia population fran~aise. Paris: Tec & Doc Lavoisier.

5. West, C.E. (2000). Meeting requirements for vitamin A. Nutrition Reviews, 58(11), 341-5.

6. Wynn, M. and Wynn, A. (1998). Fortification of grain products with folate: should Britain follow the American example? Nutr Health, 12, 147-6 1.

7. Bailey, L.B. ed. (1995). Folate in health and disease. New York: Marcel Decker, chapter 5 pp. 123-51.

8. Wynn, M. and Wynn, A. (1996). Protein in the diet of children: A discussion of current recommendations. J Nutr Environ Med, 6, 161-77.

9. Berkey, CS., Gardner, J.D., Frazier, A.L. and Colditz, G.A. (2000). Relation of childhood diet and body size to menarche and adolescent growth in girls. Am J Epidemiol, 152/5, 446-52.

10. Estivariz, C.E.D. and Ziegler, T.R. (1997). Nutrition and the insulin-like growth factor system. Endocrine, 7(1), 65 -71.

11. Bridgewood, A., Lilley, R., Thomas, M., Bacon, J., Sykes, W. and Morris, 5. (2000). Living in Britain. Office for National Statistics. London: Stationery Office.

12. Department of Social Security (1997 and 1998). Welfare Reform Focus File. Central Office of Information, November.

13. Poverty Watch (2000) Poverty. Child Poverty Action Group 107, p. 20.

14. Ford, R., Marsh, A. and Finlayson, L. (1998). What happens to lone parents? Department of Social Security. London: Stationery Office.

15. Wynn, M. and Wynn, A. (2000). Low birthweight in electoral wards; a useful health and social indicator at local level. Nutr Health, 14, 77-87.

16. US House of Representatives Subcommittee on Appropriations. Hearings of Departments of Labor, and Health. Education and Welfare. Appropriations for 1962, p. 756.

17. Hollingsworth, D. (1971). Development of the school meals and welfare food services. The British Nutrition Foundation.

18. Cole-Hamilton, I, Dibb, S. and O92Rourke, J. (1991). School meals fact sheet. Food Commission and Child Poverty Action Group.

19. House of Commons Education and Employment Committee. School Meals. First Report Session 1999-2000 HC 96.

20. US Department of Agriculture (1996). Nutrient analysis protocols for the school nutrition program. Healthy school meals resources system.

21. Caroline Walker Trust. Eating Well for Older People 1995. The CORA Menu Planner 1998. Eating Well for Under-5s in Childcare 1998: The CHOMP Menu Planner 2000. All from CWT, 22 Kindersley Way, Abbots Langley, Herts WD5 ODQ, Tel: 01923 269902.

22. Department of Education and Employment (2000). Statistics of Education. Schools in England. London: Stationery Office.

23. National Assembly for Wales (2000). Schools in Wales, General Statistics.

24. Scottish Executive Education Department. School meals in Education Authority schools 1999-00. Edinburgh 29 June 2000.

25. Story, P. and Chamberlin, R. (2001). Improving the take up of free school meals. Child Poverty Action Group and Department of Education and Employment. Report RR 270. London: DfEE.

26. Acheson, D. (1998). Independent Inquiry into Inequalities in Health. Department of Health. London: Stationery Office.

27. Mann, A., ed. (2001). Apports nutritionnels conseilles pour Ia population fran~aise. 3rd edition. Paris: Tec & Doc Lavoisier.

28. Parker, H., ed. (1998). Low Cost but Acceptable. A minimum income standard for the UK: Families with young children. Family Budget Unit. Policy Press & Zacchaeus Trust.

Published by: Journal of Nutrition and Health, 2002, Volume 16, pages 55-71.