UMEÅ UNIVERSITY

ISBN 987-91-7264-615-5

_______________________________________________________

Faculty of Social Sciences, Department of Food and Nutrition

Faculty of Medicine, Department of Clinical Sciences, Pediatrics

Health implications of dietary intake

in infancy and early childhood

Inger Öhlund

Umeå 2008

Detta verk skyddas under lagen om upphovsrätt (URL(1960:729)

ISBN 978-91-7264-615-5

Illustrations Staffan Lidman and Magnus Burstedt

Printed by Solfjädern Offset AB, Umeå Sweden 2008

The key to the child health lies........in all round improvement of the competence of the mother - her physical condition, her economic state, her health and nutrition and her education. Such attention to the mother must start not after she has be-come a mother, not even when she is about to become one, but when she is herself a child” Goplan 1985

ABSTRACT

Introduction: Swedish children are the healthiest in Europe. Through regular

visits to well-baby clinics, infants and young children are checked and parents

given information and advice on diet and other relevant matters for their child.

For a long time, adequate nutrition during infancy and childhood has been fo-

cused on encouraging proper nutrition, preventing malnutrition and deficiency

states, and obtaining optimal growth. Today, malnutrition and deficiency states

in infants and children are rare. But other public health problems have arisen.

Nutrition early in life is now thought to influence health and diseases even in

adulthood. Thus promotion of a healthy diet in early life is important for pre-

venting public health diseases such as iron deficiency, cardiovascular disease,

obesity, and dental caries.

Aims: This study investigates health implications of dietary intake in infancy

and early childhood. More specific focus was on the associations between die-

tary fat intake and serum lipid levels in infants, early dietary intake, iron status,

dental caries, and Body Mass Index (BMI) at 4 years of age. In addition, heredi-

tary factors and changes over time were evaluated.

Methods: Before 6 month of age, 300 healthy infants were recruited from well-

baby clinics in Umeå. This thesis is based on secondary analysis of a prospec-

tive study in these infants run from 6-18 months and a follow-up of 127 of the

children at 4 years. Between 6-18 months and at 4 years, dietary intakes were

assessed, anthropometric measures performed, and venous blood samples

taken. At 4 years, a dental examination was also performed and anthropometric

data and blood samples were collected from parents and included in the study.

Results: All but two infants were ever breastfed and at 6 months 73% were still

breastfed. The quality of dietary fat was not within national recommendations.

At 4 years, intake of vitamin D and selenium were below and intake of sugar and

sweet products above the recommendations. In girls, but not boys, higher poly-

unsaturated fatty acid intake was associated with lower levels of total choles-

terol, low-density lipoprotein cholesterol, and apolipoprotein B levels. Iron

status of the children was generally good and no child had iron deficiency

anaemia (IDA). Children’s haemoglobin (Hb) levels tracked from infancy to 4

years and correlated with their mother’s Hb. Fortified infant products and meat

were important sources of iron at both 12 months and 4 years. Children with

frequent intake of cheese had less caries in this population with low caries

prevalence. We found higher protein intake over time to be associated with

higher Body Mass Index (BMI) at 4 years and high BMI at 4 years was associ-

ated with high BMI at 6 mo. There was also an association between the BMI of

the child and that of its parents.

Conclusions: BMI of the child and parents (especially the father), and iron

status at 6 months were predictors of these variables at 4 years of age. The qual-

ity rather than the quantity of dietary fat in infancy affected serum lipid values.

Even in a healthy and well-nourished group of Swedish infants and young chil-

dren, quality of food and intake of nutrients are important for current and later

health of the child.

ORGINAL PAPERS

This thesis is based on the following papers referred to in the text by their Ro-man numbers (I-IV).

I. Öhlund I, Hörnell A, Lind T, Hernell O. Dietary fat in infancy should be more focused on quality than on quantity. European J Clinical Nutri-tion 2007; Jun 20.10.1038/sj.ejcn.1602824

II. Öhlund I, Lind T, Hörnell A, Hernell O. Predictors of iron status in well-

nourished 4-y-old children. Am J Clin Nutr 2008;87:839-45.

III. Öhlund I, Holgerson PL, Backman B, Lind T, Hernell O, Johansson I. Diet intake and caries prevalence in four-year-old children living in a low-prevalence country. Caries Res 2007; 41:26-33.

IV. Öhlund I, Hernell O, Hörnell A, Stenlund H, Lind T. BMI in 4-year-old children is associated with previous and current protein intake and the father’s BMI. In manuscript.

ABBREVIATIONS

apoA Apolipoprotein A-1

apoB Apolipoprotein B

BMI Body Mass Index

Dfs Mean decayed-filled surfaces - a definition of caries experience

E% Percent of total energy

Fe Iron

Hb Haemoglobin

HDLC High density lipoprotein cholesterol

ID Iron deficiency

IDA Ion deficiency anaemia

LDLC Low density lipoprotein cholesterol

MCD Milk-based cereal drinks

MCV Mean corpuscular volume

MFC Milk-based fortified cereals (drinks and porridge)

MUFA Monosaturated fatty acid

PUFA Polyunsaturated fatty acid

SAFA Saturated fatty acid

S-Ft Serum ferritin

TG Triglycerides

TC Total cholesterol

CONTENT 

ABSTRACT .................................................................................  

ORGINAL PAPERS .......................................................................  

ABBREVIATIONS .........................................................................  

BACKGROUND ......................................................................... 11 

Diet and public health problems in a historical perspective ............................ 11 

Dietary intakes and health in children ............................................................. 13 

Energy and energy requirements .................................................................. 13 

Dietary fat ...................................................................................................... 14 

Iron ................................................................................................................ 15 

Cow’s milk ...................................................................................................... 16 

Carbohydrates ............................................................................................... 17 

Health implications of the diet in the childhood population .......................... 18 

Cardiovascular diseases ................................................................................ 18 

Iron deficiency .............................................................................................. 18 

Dental health ................................................................................................. 19 

Overweight .................................................................................................... 20 

OBJECTIVES ............................................................................ 22 

DESIGN .................................................................................. 23 

Subjects and methods ................................................................................... 23 

RESULTS ................................................................................ 24 

Food and nutrient intake ................................................................................. 24 

Dietary fat intake and serum lipids in infants ................................................. 28 

Iron intake and iron deficiency ........................................................................ 29 

Dietary factors and dental health ...................................................................... 31 

Factors associated with Body Mass Index ....................................................... 33 

DISCUSSION ........................................................................... 36 

Milk ................................................................................................................... 36 

Serum lipids and diet ....................................................................................... 37 

Iron status and diet .......................................................................................... 40 

Dental health and diet ...................................................................................... 42 

BMI and diet ..................................................................................................... 43 

Strengths and weakness of the study ............................................................... 46 

Future aspects .................................................................................................. 47 

CONCLUSIONS ........................................................................ 48 

POPULÄRVETENSKAPLIG SAMMANFATTNING ............................... 49 

ACKNOWLEDGMENTS ............................................................... 51 

REFERENCES ........................................................................... 54 

Background

11

BACKGROUND

Diet and public health problems in a historical perspective Nutrition during infancy and childhood has long been focused on meeting nutri-tional needs to prevent malnutrition and deficiency states and to obtain op-timal growth. A transition from malnu-trition to sufficient or even over nutri-tion has occurred in the north of Swe-den during the last century (Figure 1).

Figure 1 The map shows the city of Umeå, lo-cated in the county of Västerbotten in the northern area of Sweden The first Swedish guidelines on infant and childhood nutrition were produced in the 18th century (1). At the time, in-

fant mortality was high, probably above 200/1000. Poor nutrition was identi-fied as one cause. In northern Sweden, local doctors as well as Carolus Linnæus (Carl von Linné) during his extensive travels reported low preva-lence of breastfeeding and that breast milk from early age was replaced with whole cow’s milk causing high infant mortality (2). In 1749, Swedish population statistics was born when “Tabellverket” started to collect demographic and mortality data. Observations made on high infant mor-tality prompted the Royal Swedish Academy of Sciences to ask the father of paediatrics, Nils Rosén von Rosenstein, to write some articles in calendars with information on children’s health and promotion of breastfeeding. These cal-endars were circulated to parents with the aim to improve the health care dur-ing infancy and early childhood. In 1764, Rosén published the first known textbook in paediatrics anywhere in the world (1). In it he stated that infant nu-trition was important and mother’s milk was the best nutrition. However, if the mother, for some reason, was un-able to breastfeed her baby he sug-gested using a wet nurse. A wet nurse needed to be healthy, well nourished, and have enough milk of good quality. At the beginning of the 20th century, local doctors in the north of Sweden once again reported on poor living con-ditions including poor nutrition. These conditions resulted in tuberculosis, rickets, iron deficiency anaemia (IDA), gastrointestinal problems, caries, and high infant mortality. In 1901, the first charitable institution (Mjölkdroppen, “a drop of milk”) for protection of chil-

Background

12

dren’s health with focus on nutrition was started in Stockholm (3). At the beginning of the 20th century, local doctors in the north of Sweden once again reported on poor living con-ditions including poor nutrition. These conditions resulted in tuberculosis, rickets, iron deficiency anaemia (IDA), gastrointestinal problems, caries, and high infant mortality. In 1901, the first charitable institution (Mjölkdroppen, “a drop of milk”) for protection of chil-dren’s health with focus on nutrition was started in Stockholm (3). In 1929, the Odin study was started in Umeå, Sweden (4). The focus was on the prevalence of major diseases such as gastrointestinal diseases, oral health, and anaemia and how these diseases are influenced by diet. In the study group, 17 406 men, women, and chil-dren were included; 3 406 were school children. Major observations were that the diet was unbalanced, based on pota-toes, milk, and barley bread and con-tained low amounts of vitamins and iron. Malnutrition and poor dental health were common in the northern area of Sweden. This was a contributory cause of the 1937 governmental propo-sition to start well baby-clinics, where health and dietary information and ad-vices should be offered to parents. Pro-spective length and weight measure-ments were started to assess growth rate and thus identify poor growth and inadequate intake of a balanced diet. Later, growth charts were introduced to make it easier to assess growth and to identify more rapidly and accurately children with poor growth.

In 1938, another important reform – public dental clinics – was instituted by the Swedish Parliament. Sweden was the first country to start social dental care and the goal was “dental care for the people” (5). In the 1960s, caries prevention programmes were started with advice on oral hygiene habits that included reducing sugar consumption and using fluoride prophylaxis in chil-dren. In schools, dental nurses regularly administered fluorides to children.

In 1967, an epidemiological study on children was started in the Umeå area to study the relationship between diet and general and oral health (6). At the time, there were no Nordic or Swedish nutritional recommendations and the authors compared the dietary intakes with recommendations from the U.S. Food and Nutrition Board. The dietary intakes at 4 years were higher than or within recommendations for energy and nutrients except fo low intake of vitamin D. The intake of fat was high, comprising 40-42% of the energy (E%). Consumption of sweets was negatively associated with oral health. Iron status had improved since the Odin study in 1929 and the prevalence of IDA was low (7). Height and weight did not differ be-tween 4 year old boys and girls, but skin

Background

13

fold thickness was higher for girls than for boys (8). Further investigations of the children’s dietary intake and health were per-formed in the same area in the 1980s and no signs of nutritional deficiencies (IDA included) were found (9). In weaned infants at 6 and 12 months, the dietary fat intake was 29-33 E% and to-tal energy intake at 12 months was 80% of the recommended daily allowance (RDA), although the growth was nor-mal. For the first time, intake of vitamin D was reported to exceed the RDA to the degree that vitamin D supplementa-tion was questioned. Milk-based forti-fied cereals was an important food source, responding for 56 E% at 6 months and 26 E% at 12 months (10). Longitudinal data on children followed from 12 months to 3 years showed asso-ciations between caries and consump-tion of sucrose-rich foods such as cakes, bread, and sweet soups. The dietary patterns at 3 years were already estab-lished at 12 months, patterns that sug-gested that future health would be in-fluenced by early eating habits (11). In 1980, the first Nordic nutrition rec-ommendation (NNR) was published and then revised (1989, 1996, and 2004) due to new scientific data. At present a new revision is under discus-sion contributing to 2004 revision were new studies on energy requirements in infants (12, 13), associations between dietary fat, particularly saturated fat, and serum lipid levels in infants and young children (14), and the increasing prevalence of obesity in childhood (15) (16). The Swedish Nutrition Recom-mendations (SNR) are based on NNR

and the first SNR was published in 1981 and the most recent in 2005. In addi-tion to recommendations on nutrient intake, the SNR also includes recom-mendations on meal planning and physical activity.

Dietary intakes and health in children

Energy and energy require-

ments

Due to the rapid growth rate, the energy needs per kilo bodyweight are higher during infancy and early childhood than in adulthood. The diet should provide energy for total energy expenditure in-cluding basal metabolic rate (BMR), thermoregulation, and requirements for growth and physical activity. During the first 3 months of life, the cost of growth is highest, starting with 30-38% of the total energy need and decreasing rap-idly to about 2% at 2 years (12). Later in childhood and adolescence, the cost of growth varies between 1-4% of total en-ergy. If the energy intake for some rea-son becomes too low, the body has sev-eral defence mechanisms. One is to de-crease physical activity. If that is not enough, decreased body temperature, weight loss, and reduced linear growth (17) may be used to adapt to a lower en-ergy intake. In the revised NNR-04, the recom-mendation on total energy intake was reduced both for infants and children younger than 3 years. The reasons be-hind the changes were new studies us-ing the double-labelled water method, showing that previous estimation on

Background

14

energy requirements (ER) for infants and children had been too high (13, 18, 19) .

Fat, carbohydrate, and protein provide energy. Increasing the proportion of fat relative to carbohydrates and protein increases the energy density of the diet (20), which may affect the total energy intake (21). Although energy intake is within or even above the recommenda-tion, it is not clear if Swedish children have an optimal nutrient intake in other respects or even an optimal diet. In 4 year old Swedish children, sweet prod-ucts account for 25% of the total energy intake, often with low nutrient density (22).

Dietary fat

As mentioned, a diet with higher pro-portion of fat provides higher energy density and smaller volumes of food need to be consumed. Therefore, fat in-take and growth are expected to corre-late (23). Breast milk contains about 50 E% as fat and it is also high in choles-terol. Fat is necessary not only as a source of energy but also because it is a source of essential fatty acids and serves as carrier for absorption of fat-soluble vitamins. Historically, a diet rich in fat has been recommended for infants and young children to ensure sufficient energy in-take and adequate growth. In contrast, a diet rich in fat, particularly in satu-rated fat, is considered to have negative health effects in adults. The NNR (1980) encouraged adults to reduce fat intake and to increase dietary fibre in-take (24). Unintentionally, fat intake

decreased and fibre intake increased also among older infants and young children eating foods prepared by health-conscious parents. Some chil-dren presented at well-baby clinics and health care centres with diarrhoea. This phenomenon, called toddler’s diar-rhoea, often resulting from a diet low in fat and rich in complex carbohydrates, became more common. As a conse-quence, more comprehensive informa-tion was given to parents on the impor-tance of a diet with higher energy den-sity than for adults for infants from weaning through the first 2-3 years. Compared to NNR 1996, the revised recommendations, NNR 2004 recom-mend a lower maximum intake of die-tary fat for infants and children be-tween 2-3 years of age but with no change for the age group 12-23 months (Table 1). Globally, there is no consen-sus on the need to reduce dietary fat in children. On the contrary, it has been questioned if a low fat diet is safe for infants and young children (12, 25-27). Some studies on dietary fat intake in infants and young children have shown intakes of about 45-50 E% at 4 months, 38-40 E% at 6 months (28), and 37 E % at 12 months (29, 30). In Finnish in-fants, dietary fat intakes <30 E% (26-27 E%) (23) has been used in an interven-tion study with no adverse effect on growth or nutrient intakes. The Boga-lusa Heart study and a Norwegian study found that fat intake was replaced by sugar in low fat diets (31, 32). In Swed-ish children, the intake of dietary fat intake is 33 E% at 4 years (22, 33-36).

Background

15

Table 1. Recommended intake of dietary fat in infants and young chil-dren, NNR 19961 to 20042

1 Nordic nutrition recommendations -96 (37) , 2 Nordic nutrition recommendation -04 (24). 3Saturated fatty acids, 4Monounsaturated fatty acids, 5Polyunsaturated fatty acids, 6not speci-fied

Iron The diet is considered to have great in-fluence on an individual’s iron status (38, 39). There are two types of iron in the diet: haem and non-haem iron. Haem iron is better absorbed than non-haem and in contrast to non-haem, not influenced much by other foods. Iron content and bioavailability vary confi-dentially between different food sources. The bioavailability of iron is influenced by inhibitors and promoters of its absorption. In the diet calcium, phytate, and tea decrease absorption, whereas meat, fish, and vitamin C en-hance absorption (28, 40, 42). Meat with its high content of iron of high bioavailability (haem iron) also pro-motes absorption of non-haem iron, the meat factor. It was recently sug-gested that meat should be included in

early complementary food in small amounts to improve iron status in in-fants (41). Whether calcium works as an inhibitor of iron absorption or not is debated. There is a potential risk of in-teractions between calcium and iron affecting absorption. An inhibiting ef-fect has only been shown in single meal and short-term studies (43). Long-term studies with calcium supplements, on both adults and children, have not con-firmed an association between high calcium intake and poor iron status (44-46) . In a cross sectional study of healthy 12 month old infants in Umeå, it was un-expectedly found that 26% were iron deficient as judged by serum ferritin (s-Ft values<12ųg/l). This was in spite of sufficient iron intake. The high intake of phytate-rich cereals, i.e milk-based ce-

NNR 96 NNR 04

E% 6-12 m 1-3 y >3y 6-11 m 12-23 m >2 y

Total fat 35-45 30-35 ≈30 30-45 30-35 25-35

SAFA3 na6 na6 ≤10 na6 ≤10 ≤10

MUFA4 na6 na6 10-15 na6 na6 10-15

PUFA5 na6 na6 5-10 5-10 5-10 5-10

n-6 fatty acids ≥4.5 ≥3 ≥2.5 >4 >3 ≈ 2-2.5

n-3 fatty acids >0.5 >0.5 >0.5 >1 >0.5 ≈ 0.5-1

Background

16

reals drinks (MCDs) and porridge of Swedish infants was suggested to be the culprit. In Sweden, MCDs, gruels, and porridges have long been part of the traditional diet in both adults and chil-dren. It is a very important part of the diet of infants and young children from 6 months (9,10,22). The manufactured commercial products are fortified with vitamins and mineral and the iron con-tent is 1.2-1.3 mg per 100 ml. To inves-tigate this, unexpectedly high preva-lence of ID a long-term intervention study in infants (SINUS, Study on In-fant Nutrition in Umeå, Sweden) was started in December 1995 (47). Healthy infants (n=300) were assigned to one of three study groups with weaning food varying in phytate content: regular in-fant cereals; low-phytate infant cereals or cow’s milk based infant formula; and regular porridge from 6 to 12 months. They were followed using dietary re-cords, anthropometric data, and blood samples with focus on iron and zinc status from 6 to 18 months. No associa-tion between the phytate content and iron status was found and it was con-cluded that the infants were not iron deficient, rather the definition of ID and IDA in infants need to be revised. Fur-thermore, iron intake was found to af-fect Hb during infancy but not during the second year of life. In contrast, S-Ft levels were not affected by dietary iron intake before 12 months but was posi-tively associated at 18 months (48). Hence iron metabolism develops during first year of life (48, 49). This study later formed the basis of the present study on infant and young child health. In British toddler and young children iron status was positively associated

with meat and fruit consumption and inversely with that of milk and milk products (50, 51). High milk intake was judged to displace iron-rich or iron-enhancing foods. In Finnish 3-4 years old children, iron intake was 8.6-8.8 mg per day. Out of them 10 % had S-Ft below 10ųg per li-tre (42). Swedish children who start out with good iron status at 12 mo seems to maintain sufficient iron status at 18 mo even with low iron intake (52). A Swedish national food survey in chil-dren was carried out 2003 (22). The main iron sources were meat fish and egg contributing 28.5 % of the iron in-take and MCDs 14 %. This was con-firmed by a recent study found meat products to contribute with 23 % and gruels with 17 % of total iron (34). Two other studies found iron intake in 4-y-old Swedish children close or within the recommendation (53, 54) which is 8 mg per day in infants and preschool children and based on the need to cover for basic losses and growth. Neither of the Swedish studies did analyse the as-sociations between diet, except for iron content, and iron status in the children.

Cow’s milk

In many countries, including Sweden, cow’s milk is an important food in healthy eating providing many nutri-ents and is an important source of cal-cium and vitamins A and D. Histori-cally, cow’s milk has been considered an important food source also for children although, as mentioned earlier, in the

Background

17

past also associated with poor nutrition in infants in northern Sweden. Cow’s milk is low in iron and the bioavailability is also low. In young in-fants, unmodified milk can cause blood loss from the intestinal tract and many studies have shown a negative effect of cow milk intake on iron status during the first year of life. In Icelandic infants, milk intake greater than 500 ml was as-sociated with poor iron status which seemed to remain at 6 years (55). Therefore, many countries, European Society for Pediatric Gastroneterolgy, Hepatology and Nutrition (ESPGHAN) and the WHO recommend that intake of undiluted cow’s milk should not be introduced before the end of the first year and then only in small amounts (28, 56). After one year of age, it is of-ten recommended that milk intake should be limited to about 500 ml per day. The new recommendation to delay the introduction of cow’s milk in Swed-ish infants resulted in debate as many judged milk as an important food also for infants (57, 58). The Swedish guide-lines from the National Food Admini-stration (SLV) recommend that the in-troduction of milk as a drink should not start before 10-12 months of age. From 12 months, 500 ml of low fat milk is recommended, which includes milk added in cooking.

Milk and cheese are important foods for mineralization of the teeth and reduce the risk of caries development. This is exemplified by specific bioactive pep-

tides in cheese and milk. Caseinophos-phopeptide, derived from β-casein, in-creases calcium levels at the tooth inter-face and reduces caries (59). The ca-seinoglycomacropeptide and kappacin, peptides derived from κ-casein, inhibit adhesion and metabolism of mutans streptococci, respectively.

Carbohydrates

A diet containing adequate amounts of dietary fibre from cereals, potatoes, vegetables, fruit, and berries are con-sidered important for children. Such a diet also provides vitamins and miner-als. Although cereals contain phytate, an inhibitor of iron absorption, the in-hibiting effect can be counter balanced by adequate amounts of iron and ascor-bic acid which enhances the absorption (47). In infants, too high intake of com-plex carbohydrates might cause gastro-intestinal problems and the bulk effect may result in low energy intake. NNR 2004 suggested that the intake of die-tary fibre should be at least 10 g per day by school age (24). In the 2005, the US Food and Nutrition Board guidelines for children recommended the intake of total fibre in children aged 1-3 years be 19 g per day (60). The SLV 2008 sug-gests an intake of 2-3 g per mega joule (MJ) from one year of age which will be about 15-19 g fibre per day in pre- school children (personal communica-tion). In a healthy diet, sucrose should be lim-ited and not exceed 10 E% (24). High sugar intake is associated with health problems such as dental caries, risk of obesity, and dyslipidemias as well as a poor diet (61). Some studies have re-ported that diets with low fat content

Background

18

may result in higher sugar intakes (31, 32). Recent studies report the intake of sucrose to be about 13 E% in 4 year old Swedish children (22). Sweet products such as soft drinks, sweets, ice cream, desserts, cakes, and biscuits account for 25 E%. Hence frequent consumption of sugar and sucrose-containing products has increased and seems to be intro-duced by the second half of infancy (11, 62).

Health implications of the diet in the childhood popu-lation Today, the World Health Organisation (WHO) classify Swedish children as the healthiest in Europe (63). During free regular visits to well-baby clinics and public dental clinics, infants and chil-dren are checked and the parents are given information and advice on how to keep their child healthy (64, 65). Pro-motion of a healthy lifestyle in early childhood is important (66) and infor-mation on breastfeeding and weaning recommendations is part of the infor-mation given to parents (67). Infants born in 1997, 74% and 78 % were breastfed at six months of age in Swe-den and in the county of Västerbotten, respectively (68). Today nutrition during foetal and post-natal life are considered to have not only short term but also long term health consequences (69-71). Thus early nutrition affects iron status during in-fancy and childhood but also the risk to develop obesity and cardiovascular dis-ease decades later through program-ming effects (72-74). In the present the-sis association between early diet and

risk indicators to cardiovascular dis-ease, caries, obesity and IDA has been studied.

Cardiovascular diseases

Sweden has a high prevalence of car-diovascular disease (CVD) in the adult population (75). To prevent CVD in adults, much emphasis is put on known risk factors such as smoking, obesity, raised serum cholesterol, low physical activity, and poor eating habits (76). Studies have shown that the atheroscle-rotic process starts already in child-hood. Fatty streaks and atherosclerotic lesions have been found in children, and childhood risk factors predict adult risk factors of cardiovascular disease (66, 77-79). Such risk factors are obe-sity, high blood pressure, high serum total cholesterol (TC), and high low-density lipoprotein cholesterol (LDL-C). These individual risk factors tend to cluster (66, 80). Promotion of a healthy lifestyle in early childhood is important (81) and includes regular physical activ-ity, healthy eating, and non-smoking. Intervention studies have shown that health promotion programs reduce risk factors among school children (81, 82), and other intervention studies in in-fants have shown the possibility to re-duce TC already during infancy with lasting effects into adolescence (83). However, in Swedish infants and young children, few studies have been per-formed (84, 85) on serum lipid levels and their relation to diet (35, 86).

Iron deficiency

ID, low s-Ft concentration, is globally the most prevalent nutritional disorder in young children and women. ID is the

Background

19

only nutrient deficiency that is preva-lent in both non-industrial and indus-trial countries. The WHO estimates that ID affects most preschool children and pregnant women in non-industrial countries and about 30-40% of pre-school children in industrial countries (87). IDA, anaemia resulting from ID, is also a common nutrition disorder in in-fants and children. ID results from various causes such as poor or inade-quate diet, infections, and other dis-eases (88). Growth and growth rate are linked to iron metabolism and iron status; in children, the highest preva-lence of IDA is found between 6 months and 2 years of age, a time of rapid growth. Thus rapid growth during the first year is associated with poor iron status. For infants and adolescents, gender-related differences have been proposed to be biological and not only caused by differences in iron intake or losses through menstrual bleedings (89, 90). Association between a mother’s and her child’s iron status has been described in areas with poor nutrition (91, 92). In a healthy, well-nourished, and iron-replete population, the child’s Hb-concentrations were associated with that of both parents (93). However, the prevalence of ID in children might be less than 2% in high-income countries like those in Scandinavia (48, 94, 95). In southern Sweden, the prevalence of IDA in healthy 2.5 year old children was 10%, with higher prevalence in immi-grants than Swedish children (96). For preschool and school children, three studies were undertaken in the area of Umeå (county of Västerbotten) between 1930 and 1980 (4,7,9). In 1980, not a

single case of IDA was found (9). How-ever, in the same area during 1991-1992, about 26% of 12-month-old in-fants was reported to be iron-depleted using s-Ft <12 ųg/l as definition, al-though IDA was rare (97).

Dental health

Dental caries is primarily the symptom of an infectious disease (98). Oral bac-teria – i.e., mutans streptococci – pro-duce acids through their metabolism of dietary carbohydrates, an action that demineralizes dental hard tissues. The disease became endemic in the western countries when sucrose consumption began to increase in these populations. The intake of sugars is the main known aetiological factor of dental caries but still many epidemiological studies fail to show that caries prevalence is ex-plained by frequency or amount of sugar intake (99). This failure may be because caries resistance is enhanced by protective components. Fluoride is one well-known protective agent found in different concentrations in drinking water and in specific products such as toothpaste and tablets. Foods that in-crease calcium or phosphate concentra-tion in the biofilm on teeth and inter-crystal fluids counteract colonisation of bacteria and may reduce caries (59, 100) (101, 102). Although dental health in Swedish children has improved, it is still a public health problem and differs between social groups (103). Further-more, caries prevalence is higher in overweight children (104) and in chil-dren with other diseases (105-108). Overall, the proportion of 3 year old and 6 year old Swedish children with-out caries has increased; however, be-cause the magnitude has increased in

Background

20

the population of children with caries, the frequency of caries is skewed. Be-tween 1985 and 2005, the prevalence of caries in Swedish pre-school children declined, but the decrease between 2000 and 2005 was low (109, 110). A goal for the WHO is that 80% of 6 year old children should be free from caries. In Swedish children, 73% of the 6 year olds are caries free (109). In the Umeå area, 54% of 4 year old children was caries free in 2002 (110) and Västerbot-ten county statistics show 79% of 5 year old children to be caries free in 2004.

Overweight

Body Mass Index [BMI, weight in kg/(height in m)2] is used as an esti-mate of overweight and obesity in adults and children. In adults, the cut-offs for underweight, overweight, and obesity are BMI < 18.5, ≥25, and ≥30 kg/m2, respectively. Those cut-offs are not possible to use for children. As chil-dren’s body composition changes as they grow, the corresponding age and sex BMI cut-offs for overweight and obesity are adjusted and termed iso-BMI (111). Iso-BMI is not yet used to estimate underweight, but in a recent study a cut-off of BMI 17 at age 18 years is suggested to be used as an interna-tional definition of thinness in children and adolescents (112), and new cut-offs for thinness/underweight in small chil-dren will be introduced. Children who are overweight compared with normal weight, spend more hours watching TV and playing computer games and less time with organised sports and other physical activities (113). Short-time health consequences

of obesity are mainly psychological and social. Some studies on quality of life have shown increased psychological and psychiatric problems in obese com-pared to non-obese children (114) while others have not (115). However, it seems that girls, compared with boys, are at greater risk of psychological mor-bidity and that the risk increases with age (80). Being overweight or obese is part of the metabolic syndrome, posing an increased risk of developing cardio-vascular diseases. Raised blood pres-sure, LDL-C and low s-triglycerides (TG), fasting insulin concentrations, and low high-density lipoprotein cho-lesterol (HDL-C) concentrations are also associated with overweight chil-dren (116). The Bogalusa Heart Study reported odds ratios in overweight schoolchildren of 2.4 for having ele-vated levels of TC, 4.5 for increased sys-tolic blood pressure, and 12.6 for raised fasting insulin level (80). Overweight and obesity have been in-creasing in children during the last dec-ades, particularly among girls and in areas with low socioeconomic standard (16, 117). Sweden in no exception, overweight and obesity has increased among children (118) and is today one of the main health problems. About 15-20% of children are overweight and 3-5% are obese (16, 118-120). In the Umeå area, the prevalence of overweight dou-bled to 18% in schoolchildren 6-13 years old and the frequency of obesity in-creased fivefold to 5% between 1986-2001 (15). Among 4 year old children born 1998-199 in the county of Väster-botten, 16.7% of the boys were classified as overweight and 3.1% obese. Corre-sponding figures for girls were 22.1%

Background

21

and 4.5 % (16). However, a recent study from Gothenburg, in the south-west of Sweden, indicates a decreasing trend in 10 year old children between 2001 and 2005, particularly among girls (121). In summary, well-baby clinics and pub-lic dental clinics free of charge for chil-dren have been important for health promotion among Swedish children. Food habits are founded early in life and influence health and diseases even in adulthood, but there are many im-

portant things we do not know about Swedish infants and preschool chil-dren’s diet and health. There are ques-tions about the associations between dietary intake in early life and public health problems such as serum lipid levels, iron status, BMI, and dental car-ies. Besides the current diet, tracking over time and associations between the parent’s and the child’s health needs to be evaluated.

Objective

22

OBJECTIVES

The general objective for the present study was to investigate some the health impli-cations of dietary intakes in infancy and early childhood. Specific objectives were to evaluate the following:

• associations between dietary fat intake and serum lipid levels in infants (paper I);

• haemoglobin values and iron status at 4 years of age and their associations with dietary intake and body growth, parental values and possible tracking of iron status variables from infancy to early childhood (paper II);

• associations between dietary intakes and dental caries, levels of mutans strep-tococci and lactobacilli in saliva (paper III); and

• body mass index (BMI) at 4 years of age and its association with dietary in-takes from 6 months to 4 years, parental BMI, and changes over time. (paper IV)

Design

23

DESIGN This thesis report results from a longi-tudinal study originally focused on the effect of varying the phytate intake dur-ing infancy on iron and zinc status (47). Three hundred healthy term infants were recruited from well-baby clinics before 6 months of age and 234 were followed until 18 months. Recruited in-fants were all term with gestational age at birth between 38 and 42 weeks, birth

weights >2500 g, and with no chronic illness or feeding problems. From those 234 children who continued to 18 months, we recruited 127 to a follow-up at 4 years which also included their parents. The study was conducted in Umeå which is a university town of 110 000 inhabitants located in northern Sweden (Figure 2).

Subjects and methods

Figure 2 Flow chart for the different parts in the baseline study from 6 months to 18 months and the follow up study at 4 years. The methods used are described in the respective papers (I-IV)

BMI in 4-year old children is associated with previous and current protein intake and the father’s BMI (Paper IV)

Monthly five day food records from 6 mo to 18 mo Anthropometrics monthly 6-12 mo and bi-monthly 14-18 mo Blood samples collected at 6, 9, 12, 18 mo At 12 mo n=255 and at 18 mo n =234

Children who remained in the study until 18 mo were invited to a follow-up study at 4 y, together with their parents. This study included repeated five day food records, anthropometrics measurements, blood samples, and examination of dental health n= 127 children, 119 mothers, and 114 fathers The follow-up started in 1999

"Diet intake and caries prevalence in 4-year old children” (Paper III)

“Dietary fat in infancy should be more fo-cused on quality than on quanti-ty”(Paper I)

Predictors of iron sta-tus in well-nourished 4-year old children (Paper II)

Longitudinal study started in December 1995, including 300 infants

Results

24

RESULTS

Food and nutrient intake All but two infants were ever breast-fed and at 6 mo 73%, at 12 mo 15%, and at 18 mo 2% were still breastfed. They were all on diversified diets at 12 months. At 12 mo, 99% of the 255 infants were eating some commercial baby foods, such as formula, MCDs, porridge, pureed fruit, and jars with main courses (Figure 3). These baby foods accounted for 46% of the mean energy intake at 12 months. All but two infants had been eating fresh fruit, berries, or commercial pureed fruit for babies. Formula, MCD, and baby-porridge were important food sources: at 12 month consumed by 96% and at 4 years by 73% of the in-fants and children. These products were important sources of iron, en-ergy, and other nutrients. Of the home cooked foods prepared by the parents, potatoes were more com-mon than rice and pasta, and meat was more common than fish. The majority (96%) of the infants ate some kind of sweet products such as candies, chocolate, jams, soft drinks, biscuits, fruit syrups, or sugar at 12 months ice cream excluded. The su-crose intake was 5 E%.

Of the infants, 97% were drinking milk and both full fat (n=224) and low fat products (n=200) were used. The consumed volume was between 1.5 ml to 1240 ml per day with a mean daily intake of 237 ml. Milk in-take in infancy was negatively associ-ated with formula and MCD intake but positively associated with con-sumption of porridges. The mean in-take of full fat milk was lower, com-pared with low fat milk, 98 ml and 283 ml respectively.

Results

25

Figure 3 Food sources consumed by the infants at 12 months.* Includes can-dies, chocolate, sugar, cakes, jam, soft drinks, fruit syrup . ■ = commercial baby products □= home cooked foods. At 4 years age, milk intake was higher in children not consuming MCDs and varied from 40 ml to 940 ml per day with a mean intake of 360 ml per day. Among these healthy in-fants and children, mean intake of energy and most nutrients was in agreement with or higher than cur-rent recommendations, but intake of vitamin D and selenium was low at 4 years (5.1ųg and 20 ųg respectively) as compared to NNR-04. Only data of energy and macronutrients are shown (Table 2). The quality of dietary fat intake was not within the recommendations of NNR at 12 mo. Intake of saturated fatty acids (SAFA) was high and only

2 infants had a relative intake below the recommended upper level of 10 E%. Regarding intake of polyunsatu-rated fatty acids (PUFA), almost one-third of the infants had an intake be-low the recommended proportion of 5-10 E%. The proportion of SAFA and MUFA increased relatively more with increasing total fat intake than the proportion of PUFA (Figure 4). This situation remained at 4 years, but the proportion of SAFA had in-creased and that of PUFA slightly de-creased. At 4 years, nobody was within the recommended quality of the dietary fat intake. All children had a SAFA intake >10 E% and only 10% reached a total PUFA intake within the recommended 5-10 E%.

Results

26

Table 2. Mean energy and macronutrient intake per day of the chil-dren who fulfilled the follow-up at 4 y. Intakes are described for 12 months, 17/18 months and 4 years. 12 months 17/18 months 4 years

Girls (n=62)

Boys (n=62)

Girls (n=54)

Boys (n=53)

Girls (n=52)

Boys (n=47)

Energy (MJ) 3.7 (0.7) 4.1(0.9) 4.1(0.6) 4.3(0.8) 5.7 (1.0) 6.1 (1.0)

Carbohydrate (g) 113 (20) 124 (25) 123 (18) 132 (24) 185 (32) 188 (35) Carbohydrate(E%) 50.6 (4.1) 50.8 (3.7) 49.8 (4) 52.6(4.2) 53.5 (3.6) 51.4 (3.7)

Protein (g) 28.5 (6) 31.5 (7.8) 33.7 (6.5) 35 (7.6) 45.5 (10) 48.9 (11)

Protein (E %) 12.7 (1.9) 12.9 (1.9) 13.6 (1.6) 13.6 (1.6) 13.1 (1.6) 13.4 (1.8)

Fat (g) 35.9 (7) 39.2 (11) 39 (7.8) 40 (10) 50.4 (10) 54.7 (11)

Fat (E %) 36.3 (5.1) 35.9 (4.3) 35 (3.8) 34.6 (3.9) 32.8 (3.6) 33.8 (4.2)

Numbers are mean (SD)

55504540353025

Dietary fat intake (E%)

25

20

15

10

5

Fatty

aci

ds (E

%)

SAFA

MUFA

PUFA

Figure 4. Relative intake of polyunsaturated fatty acids (PUFA), monounsaturated fatty acids (MUFA), and saturated fatty acids (SAFA) at different levels of total fat intake in Swedish infants at 12 months. All as percent of energy (E%).

Results

27

At 4 y dietary fat intake and sucrose intake were positively associated with each other (r=0.23, p= 0.022). Furthermore, dietary fat intake was positively associated with energy in-take (r=0.812, p<0.001). The mean intake of dietary fibre was 9.5 g per day with a range between 4.5 -17 g per day. During the five recorded days, all children were eating meat, bread, and spreads as well as sweet products while 77% of the children were eating fruit and berries and 72%

fish. Potatoes were still more com-mon than rice or pasta. The main en-ergy source was dairy products and these were responsible for 16.3% of total energy intake and sweet prod-ucts for 15.7% (Figure 5). Dairy products were also the main source of protein intake and responsible for 30% of the mean daily protein in-take. In a quarter of the children, MCDs were responsible for more than 17% of total energy and in a few cases even 40%.

Figure 5. Proportions of total energy intake from different foods sources in the 4 years old children. Median, 25th - 75th percentiles, min-values, and max-values.

Results

28

To evaluate the accuracy of the re-ported energy intake in the present study, the intake data were compared with estimated average daily re-quirements (EAR) in the NNR 2004 at 12 months, 17-18 months, and 4 years. At both 12 months and 17-18 months, our intake data were higher compared with the EAR and this re-mained for boys at 4 years (Table 3).

The ratio of energy intake (EI) to basal metabolic rate (BMR) at 12 mo was 1.7 for girls and 1.75 for boys, at 17-18 mo it was 1.59 for girls and 1.55 for boys, and at 4 y it was 1.61 for both girls and boys.

Table 3. Reported energy intakes and estimated energy require-ments (EAR) at 12 months, 17-18 months, and 4 years.

1 Energy intake, MJ per day

2EAR calculated from actual kilo body weight with the use of NNR-04, 3 EAR for the age group and sex, with predicted weight NNR-04, 4 no specific EAR defined in NNR-04

Dietary fat intake and serum lipids in infants Paper I reported dietary intake and serum lipids at 12 months. An impor-tant finding from this study was that dietary fat intake affected serum lipid levels differently in girls and boys, although neither total dietary fat in-take, serum lipid levels, nor breast-feeding differed between genders. In girls, but not boys, higher intake of PUFA was significantly associated with lower levels of TC, LDL-C, and apo B levels. The mean total intake of saturated fatty acids (SAFA) was high (15.1 E%) and only two infants

had a relative SAFA intake below the recommended upper level of 10 E%. Although the mean PUFA intake was 5.6 E%, almost one-third (29 %) of the children were consuming less than the recommended proportion of 5-10 E% PUFA intake. PUFA intake was positively correlated with intake of commercial baby foods such as jars with main courses (r=0.386, p< 0.001) formula and MCDs, (r=0.291, p<0.001). As a consequence, SAFA was negatively correlated with com-mercial baby foods and MCDs, (r=-0.395, p< 0.001 and r=0.258 p=<0.001, respectively). Further-more, serum lipid levels TC, LDL-C,

12 mo 17 - 18 mo 4 y

Girls n=132

Boys n=121

Girls n=54

Boys n=56

Girls n=52

Boys n=47

EI1 3.7 (0.7) 4.1(0.9) 4.1 (0.6) 4.3(0.8) 5.7 (1.0) 6.1 (1.0)

EAR2 3.5 (0.4) 3.8 (0.8) 4.0 (0.4) 4.5 (0.8) 5.7 (0.7) 5.8(0.8)

EAR3 3.4 3.6 na4 na4 5.3 5.7

Results

29

and apo B were negatively correlated with intake of commercial baby products, whereas there was no cor-relation with intake of butter, oil, margarines, or fat from dairy prod-ucts.

Between 6 mo and 12 mo the serum lipid levels showed only small changes except for a decrease in apo A. Girls had higher apo B levels than boys at both 6 mo and 12 mo, but no other differences between boy’s and girl’s lipid levels were noted (Table 4).

TABLE 4. Serum lipid concentrations at 6 and 12 months of age in Swedish boys and girls. Boys (n=111)

6 mo 12 m Girls (n=125)

6 mo 12 mo TC mmol/l 3.39 (0.66) 3.34 (0.62) 3.45 (0.8) 3.47 (0.70)

LDL-C mmol/l 1.70 (0.51) 1.83 (0.64)

HDL-C mmol/l 0.94 (0.24) 0.91 (0.27) 0.90 (0.23) 0.90 (0.18)

ApoA g/l 1.02 (0.17) * 0.94 (0.18) * 0.99 (0.16)* * 0.95 (0.15)* *

ApoB g/l 0.68 (0.17) *** 0.69 (0.14) *** 0.73 (0.22) *** 0.73 (0.17) ***

Abbreviations: apoA, apolipoprotein A-1; apoB, apolipoprotein B; HDL-C; High-density lipoprotein cholesterol; LDLC, low-density-lipoprotein cholesterol (calculated from Fridewalds formula); TC, total cholesterol

Values are mean and (SD) *P < 0.001 in boys between 6 and 12 mo, * *P < 0.001 in girls between 6 and 12 mo, *** P < 0.05 between genders at 6 and 12 mo

Iron intake and iron de-ficiency Associations between iron status and dietary intake at 4 years were weak (Paper II). In this study group, which had an adequate intake of iron, calcium, and ascorbic acid, there were no associations between Hb concentrations and dietary intake such as mean daily intake of iron, meat, MCDs, ascorbic acid, calcium, or dairy products at 4 years. In boys, but not girls intake of meat products was a significant contributor to S-Ft.

Similarly, MCV was associated with intake of meat products but only in boys. Interestingly, we found tracking of iron status and a correlation between the child’s and the mother’s Hb lev-els. Hb as well as MCV levels tracked during infancy to childhood from 6 months to 12 and 18 months and fur-ther on to 4 years. For S-Ft levels, there was tracking during infancy, but this was not sustained to 4 years. The children’s Hb at 12 months,

Results

30

18 months, and 4 years were signifi-cantly associated with their mother’s but not father’s Hb (Figure 6). In the 4-year old children, no one had IDA, but two children (1.8 %) had Hb < 110 g/l and three (2.8 %) had S-Ft < 12 ųg/l. There were no differences in Hb or S-Ft between boys and girls, but MVC was higher in girls. Mean iron intake at 4 years was within recommendations although it ranged between 2.8-22.5 mg per day and in 57% the intake was below 8 mg per day. The mean iron intake at 12 months was 9.2 mg, at 18 months 9.6 mg, and decreased to 8.2 mg at 4

years. The main source of iron was formula or MCD at both 12 months and 4 years (Figure 7). Because 27% of the children did not eat these products at 4 years, we also com-pared children eating MCD with those who were not. Children with higher intake of MCD had a higher intake of both iron and calcium than others. In spite of the different intake of iron, iron status did not differ be-tween those consuming MCD and those that did not. Neither milk in-take nor calcium intake were associ-ated with iron status in this study group.

R Sq Linear = 0,069 R Sq Linear = 0,092

R Sq Linear = 0,069 R Sq Linear = 0,092

R Sq Linear = 0,069 R Sq Linear = 0,092

The child´s Hb at 12 mo and 4 y

4 y

12 mo

Mother's Hb

16014012010080

160

150

140

130

120

110

100

12 mo

4 y

Figure 6. Associations between the child’s Hb at 12 mo and 4 y and the moth-ers’ Hb. (R2 0.069 and 0.094, P= 0.001 and 0.004, respectively).

Results

31

Mean intake of iron (mg) at 4 y2,52,01,51,00,50,0

Sweets

Pasta

Potatoes

Bread and cereals

Blood and viscera

Meat products

Milk based fortifed cereals

Figure 7 Mean intake of iron from different food sources at 4 years of age

There was a positive association be-tween relative linear growth and Hb at 4 years in girls (r = 0.397 p = 0.002) but not boys and for S-Ft in boys (r=0.33, p= 0.02) but not girls. Weight gain and BMI were not asso-ciated to any measure of iron status (data not shown).

Diet and dental health As reported earlier, most of the in-fants were consumers of sweet prod-ucts at 12 months although the amounts were low with an intake of sucrose about 5 E%. The intake of sucrose increased, and at 4 years was responsible for 13 E% of total energy. In total, sweet products (such as candies, chocolate, sugar, soft drinks, ice-cream, biscuits, and fruit syrup) were responsible for 17% of total en-ergy. In Paper III, the associations

between diet and dental health were analyzed using data on dietary in-takes, dental caries, oral hygiene in-cluding tooth brushing habits, fluo-ride habits, presence of plaque, and gingival inflammations, and levels of mutans strepotccoci and lactobacilli in salvia. Eighty-nine of the children took part in the oral examination and 69 children fulfilled both oral and dietary examinations. Of the 4-year old children, 70% were caries-free while 30% were affected by caries. A larger proportion of the boys than the girls had caries; however, in girls with caries, the caries experience [mean decayed-filled surfaces (dfs)] was greater.

Results

32

Cheese intake seemed to have a car-ies-protective effect. The children, who ate cheese daily in the 5 day pe-riod, were caries free (Figure 8) and caries correlated negatively with in-take frequency of cheese (rs=-0.25, P=0.004). One additional serving of cheese in a 5-day period reduced the risk to have one more caries-affected surface by 33 %. Black pudding seems to affect caries experience positively (rs=0.27,

p=0.03). Caries did not correlate with intake frequency or total intake of any other food group such as sweet products like biscuits, cakes, jam, marmalade, candies, and sugar. Nor did caries correlate with intake of particular nutrients as total carbo-hydrate or sucrose. Almost all chil-dren had a high sugar intake but were regularly exposed to fluoride in toothpaste.

Figure 8 The frequency of cheese intake and caries prevalence in 4-year old children. Data are means with upper 95% CI. The figures in the bars refer to number of children in each group.

Results

33

Factors associated with Body Mass Index In a univariate regression analyse we found significant positive associa-tions between BMI at 4 years and nu-trient intakes over time (Table 5). Correlations between the child’s BMI and father’s BMI was found at 12 and 18 months and at 4 years and with mother’s BMI at 4 years. Variables with a p value<0.05 were included in a multivariate analyse. In that linear multivariate regression, current and previous protein intake and father’s BMI remained positively associated with BMI in 4 years old Swedish children (Table 6). The univariate regression analyse showed positive associations be-tween BMI and nutrient intakes over time. Intake of total energy at 12 months, 17/18 months, and 4 years, protein at 17/18 months and 4 years, carbohydrate at 17/18 months and 4 years were all positively associated with BMI at 4 y. With respect to classes of fatty acids, there was a positive association between BMI at 4 years and the intake of n-6 fatty acids at 4 years, but not with intake of n-3 fatty acids. In the final multi-variate liner regression analysis we found that protein intake at 17-18 months and 4 years remained posi-tively associated with higher BMI in children at 4 years. During the analy-sis, we found macronutrients as well as energy to be associated with each other. Protein was highly associated with energy intake (r= 0.793, p< 0.001). There were also positive as-sociations between n-6 fatty acids

and energy intake at 4 years (r=0.556, p<0.001) as well as with total dietary fat, n-3 fatty acids, and protein intakes (r= 0.613, r= 0.508 and r= 0.369 respectively, p< 0.001). The n-3 fatty acid intake was associ-ated with total fat and energy intake (r= 0.271 and r=0.216 respectively, p< 0.05) but not with protein intake. The more the children were eating, the higher their protein intake. There were no correlations between BMI and any specific food source. Body Mass Index (BMI) is used to measure nutritional status and high BMI is an indicator of overweight and obesity. BMI was calculated in children over time from 6 months to 4 years and in parents when the child was 4 years (Paper IV). At 4 years, 16.4% of the children were classified as being overweight or obese with no gender differences (Figure 9)

103

17 OverweightNon-overweight

3 Obese

Figure 9 Numbers of the 4-year old child-ren classified as obese, over-weight, or non-overweight with the use of Coles cut offs for iso-BMI (n=122).

Results

34

Table 5. Associations between nutrient intake and BMI at 4 y in the participating children. Univariate regression analysis (n=99). BMI at 4 y Independed variable R2 B (s.e) P

Energy (kcal/day) 12 m 0.028 0.001(0.001) 0.07

Energy 17/18 m (kcal/day) 0.068 0.002(0.001) 0.006

Energy 4 y (kcal/day) 0.056 0.002(0.001) 0.02

Protein (g/day)12 m 0.023 0.034(0.02) 0.097

Protein (g/day) 17 -18 m 0.117 0.074(0.02) <0.001

Protein 4y (g/day) 0.087 0.045(0.015) 0.004

Fat intake 4 y (g/day) 0.023 0.023 (0.015) 0.14

n-6 fatty acids 4 y(g/day) 0.088 0.251( 0.083) 0.003

n-3 fatty acids 4 y (g/day) 0.21 0.175 (0.123) 0.16

Carbohydrate17-18m(g/day) 0.088 0.021(0.007) 0.002

Carbohydrate 4 y (g/day) 0.062 0.012 (0.005) 0.015 Macronutrients with p< 0.2 are shown. Table 6. Final multiple stepwise linear regression model showing as-sociations between BMI at 4 y, intake of protein at 17/18 mo and 4 y, and father’s BMI. BMI at 4 y Independed variable R2 B (s.e) p-value 0.28 Protein intake (g) 17/18 m 0.072 (0.025) 0.005 Protein intake (g) 4y 0.053 (0.017) 0.002 Father’s BMI 0.108 (0.054) 0.049 Independed variables entered into the analyses were variables with p< 0.05 in the uni-variate regression analyse, i.e. intake of protein at, 17/18 mo and 4 y, energy intake at, 17-18 mo and 4 y, carbohydrates at 17/18 mo and 4 y and n-6 fatty acids at 4 y and BMI of mothers and fathers. Adjusted for gender.

Results

35

BMI was positively correlated with fat mass percent, confirming that high BMI was associated with over-weight/obesity. Although we found no significant gender differences in BMI or overweight using Cole’s definition, girls’ fat mass percent was higher than boys’ fat mass per-cent. In parents, mean BMI was higher in fathers than mothers [25.3 (3.1) vs. 23.8 (3.5), p<0.001]. Of the fathers, 34 % and 8.6% were overweight and obese, respectively and among the mothers, 19.5 % were overweight and 7% were obese. Furthermore, BMI at 4 y tracked significantly over time from 6 mo of age.

Discussion

36

DISCUSSION The most interesting observations in this study were the association be-tween the child’s and the mother’s Hb concentrations and the child’s and the father’s BMI and that there was strong tracking of both Hb con-centrations and of BMI. Furthermore we found a strong association be-tween early protein intake and later BMI, but not with total fat, carbohy-drates, or energy intake. Another finding was that the quality of dietary fat affected serum lipid levels differ-ently in girls and boys at 12 months. Finally, regular intake of cheese seemed to be protective against car-ies. The longitudinally-followed children in our study group were healthy and well-nourished. Although parents and other carers often are concerned that children do not eat enough, our data clearly indicate that the quality of the diet rather than the quantity of food should be of concern. The quality of dietary fat did not meet the recommendation of the NNR. For almost all infants, the in-take of SAFA was too high and in one-third of the infants the PUFA intake was below the recommended 5-10 E%. This situation worsened at 4 years as the proportion of SAFA had increased and PUFA decreased. At 4 years, no child met the national or Nordic recommended with respect to the fat composition of the diet. We confirm previous studies that noted that sweet products are one of

the main sources of energy in 4-year old Swedish children (22, 34, 54). Already at 12 months most of the children had been introduced to sweet products that were responsible for 5-10 % of daily energy intake, partly depending on which products were classified as “sweet products”. Besides intake of dietary fat and su-crose, most nutrients were within the recommendations, but intake of vi-tamin D and selenium were below. Unfortunately we did not measure vitamin D status. Low vitamin D status does negatively affect health and studies in Swedish children are required (122-124).

Milk Cultural factors and available food sources influence current feeding practise in infants and children (56). In Sweden, cow’s milk and products based on cow’s milk have long been used as an important food source during childhood (125). The present study was started before the current recommendation to introduce infants to cow's milk not until the end of the first year [45, 46]; hence milk was introduced without restriction before the age of 12 months (57, 58). Whether milk as such, milk fat, or milk protein (126) as well as calcium from milk has only negative or only positive influence on health is a mat-ter of continuing debate (45, 51, 58, 127-129). Moreover, the exact role of specific fatty acids present in milk fat is not clear (130, 131). Of the 255 in-fants at 12 months, 248 were drink-ing milk, and both full fat and low fat dairy products were consumed. In-terestingly, those who were drinking

Discussion

37

low fat milk had a higher mean and total intake of milk. Dairy products were the main source of both energy and protein at 4 years even if MCD were as not included. As MCD con-tain both milk and cereal, they were analysed separately and not included in dairy products in the analysis.

Serum lipids and diet An important finding from this study was that dietary fat intake affected serum lipid levels differently in girls and boys, although neither total die-tary fat intake, nor serum lipid levels or duration of being breastfed dif-fered between genders. As breastfed infants compared to formula-fed infants have higher TC levels until breastfeeding is discon-tinued (84,132) we were surprised by the low TC levels in our study group compared to Finish and Icelandic in-fants (see below), and the small change in TC from 6 months to 12 months (Tables 4 and 7). The num-ber of infants’ breastfed at 6 months was close to official national statistics of 76% (The National Board of Health and Welfare, 1999). A likely explanation is that the expected change in TC from 6 to 12 months of age was counterbalanced by a grad-ual decrease in breastfeeding rate during infancy. Besides the apoB level that was higher in girls compared to boys, there were no significant differences in serum lipid levels between gender. ApoA has been reported to be higher in boys during the first decade of life (133), but this was not confirmed in our study.

In Paper I, we compared our data with similar data from Finland and Iceland and an interesting picture emerged (Figure 10, Table 7). Ice-landic infants had a relative (E%) to-tal fat intake comparable to the Swedish infants but the proportion of SAFA was higher and that of both MUFA and PUFA lower. At the time of the Icelandic study, cow milk was used rather than infant or follow-on formula, which explains the com-paratively high intake of saturated fat. In the Finnish study, infants had lower intake of total fat, SAFA and MUFA but equal intake of PUFA (all expressed as E%) compared to the Swedish infants. The Finnish infants had significantly higher levels of TC (P<0.001) and non-HDL-C, but the level of HDL-C was equal to the level in the Swedish study population. This is also supported by the com-parison of the ratio apoA/apoB that was significantly higher in Swedish girls compared to Finnish girls but not boys. In total, Swedish infants showed lower TC, non-HDL-C and LDL-C levels than both the Finnish and Icelandic infants while HDL-C were higher in Icelandic infants compared with those in our study group. Thus, we could conclude that the quality of dietary fat rather than quantity is associated with what is considered favourable serum lipid levels. Despite a higher total dietary fat intake in Swedish, as compared to Finnish infants, the Swedish infants had more favourable serum lipid lev-els.

Discussion

38

Figure 10 Intake of total fat, saturated fatty acids (SAFA), monoun-saturated fatty acids (MUFA) and polyunsaturated fatty acids (PUFA) in Swedish, Finnish(134), and Icelandic (135) infants 12-13 month of age.

39

Discussion

TABLE 7. Serum lipid concentrations in Swedish, Finnish, and Icelandic infants1.

Sweden 1995-1997

12 mo

Finland*

1990-1992 13 mo

Iceland†

12 mo

Boys (n=120)

Girls (n=128)

Boys (n=201)

Girls (n=173)

Boys (n=51)

Girls (n=52)

TC¶ (mmol/l) 3.33 (0.60)‡§ 3.48 0.(70)‡§ 3.74 (0.60) ‡ 4.29 (0.91) ‡ 4.62 (0.8║)§ 4.48 (0.77)§

LDL-C** (mmol/l) 1.70 (0.51)§ 1.83(0.64)§ - - 2.84 (0.83)§ 2.74 (0.73)§

Non-HDL-C†† (mmol/l) 2.41(0.57) ‡ 2.58(0.65) ‡ 2.87 (0.57) ‡ 3.38 (0.89) ‡ 3.61 3.49

HDL-C‡‡ (mmol/l) 0.92 (0.27) 0.90(0.18) *** 0.87 (0.2) 0.86 (0.2) 1.01 (0.33) 0.99 (0.29) ***

apoA§§ (g/l) 0.94 (0.18) ‡ 0.95 (0.12) ‡ 1.05 (0.16) ‡ 1.09 (0.15) ‡ - -

apoB║║(g/l) 0.69 (0.14) ‡ 0.73(0.17) ‡ 0.75 (0.16) ‡ 0.87 (0.21) ‡ - -

Ratio apoA/apoB 1.44 (0.39) 1.36 (0.32) ††† 1.47 (0.47) 1.26 (0.34) †††

1Data collected from the SINUS study and published comparable data from studies in *Finland (Lagström H. 1997) and †Iceland (Thorsdottir I. 2003). Values are presented as mean and SD. ║Corrected SD after communication with the authors. ¶Total cholesterol, **Low- density-lipoprotein cholesterol calculated from Friedewalds formula in Swedish and Icelandic infants, ††Non- high-density lipoprotein cholesterol, re-calculated in Swedish and Icelandic infants from numbers in the table, ‡‡High- density lipoprotein cholesterol, §§Apolipoprotein A, ║║Apolipoprotein B, ,§p <0.001 between Sweden and Iceland, ***p <0.05 between Swedish and Icelandic girls, ††† <0.05 between Swedish and Finnish girls.

Discussion

40

We found no correlations between intakes of total milk or of milk fat and serum lipid levels. Surprisingly, we could not see any impact of SAFA intake on serum lipid values, al-though there is a general agreement that high SAFA intakes are positively linked to high TC and LDL-C levels in children (24,25). A possible expla-nation for the lack of associations could be the relatively high SAFA in-take in all studied infants. Another possibility is that serum cholesterol levels are not affected by total SAFA intake but specifically by palmitic and myristic acids (129). To enable a decrease of SAFA intake below the recommended upper level of 10 E%, it seems necessary to re-duce the contribution of fat from dairy products in infants and chil-dren. In older children, other main sources of total dietary fat and SAFA were snacks and sweet products (22, 34). In line with recent studies from Finland (62), sucrose intake was positively correlated with SAFA. In addition, to increase the intake of PUFA in accord with the recommen-dations for infants and children, the intake of fish, soft margarines, and oils needs to be increased. Even if the intake of SAFA from milk products needs to decrease in children, milk and milk products are valuable con-stituents of the diet with high nutri-tional density. The nutrient intake of most children would improve if full fat milk is exchanged with low-fat milk and the fat replaced by soft spreads and oils. Another possibility could be to change the fatty acid

composition in cow’s milk (130, 136, 137). Studies have shown that the fat composition of cows’ milk in the Nordic countries differs due to dif-ferences in cows’ fodder. As a result of these comparisons, we concluded that it seems appropriate to focus on quality of fat rather than quantity, although overfeeding should be avoided. The Finnish study group is still followed, but the goal for the ratio of unsaturated to satu-rated fatty acids has not been reached as current intake of SAFA accounts for more than 10 E% in spite of the individualized dietary counselling (138, 139). This impli-cates a difficulty of changing the fatty acid composition of the diet mainly by lowering total fat intake in ordi-nary diets. In addition, we need more data to clarify the optimal intake of dietary fat during infancy and early child-hood.

Iron status and diet The main findings from the study on iron status (Paper II) were the tracking of Hb from infancy to 4 years of age, and the correlation be-tween the mother’s and her child’s Hb concentrations. As mentioned, the children studied at 4 years were followed-up from a baseline double-blinded intervention study from 6-12 months of age on effect of the phytate content in com-plementary food on iron status dur-

Discussion

41

ing infancy. Irrespective to phytate content in the diet, the prevalence of IDA was low, 1.2% and 0.4 % at 12 and 18 months respectively (47). Al-though the infants were not iron de-ficient, total iron intake correlated positively with the Hb concentration between 6-12 months but with s-Ft between 12 months and 18 months. This was interpreted as an ongoing development of iron metabolism during that age period (48). At 4 years no child was classified as having IDA. The Hb concentrations tracked from 6 mo to 4 y. Tracking of Hb from infancy to childhood has been reported by others (91, 95, 140) as well as an association between the child´s and the mother’s Hb(91, 93). For 4-year old children there were no associations between Hb and dietary intake such as the mean daily intake of iron, meat, MCD, vitamin C, cal-cium, or dairy products. As reported earlier, mean iron intake at 4 years met the recommendations although the intake varied considerably be-tween 2.8 and 22.5 mg per day. Children consuming MCD had a higher intake of both iron and cal-cium than those who did not, but iron status did not differ. This might be due to the high bioavailability in meat, which also is as an enhancer of iron absorption and was consumed by all children and an average high intake of vitamin C which is another enhancer. In boys, but not girls, in-take of meat products was associated

with higher S-Ft and MCV concen-trations. Although other studies (96, 141-144) have reported negative associations between intake of cow’s milk and iron status, we found no such rela-tion, including associations between s-Ft or Hb concentrations and intake of all dairy products or calcium. Globally, in areas with low total iron intake and high prevalence of ID, milk intake is inversely associated with iron status (145). Iron-fortified infant formula during the complementary feeding period and later MCDs as well as meat products and high intake of vitamin C are likely reasons for lack of ID in this study group. The prevalence of IDA in infants and pre-school chil-dren varies even in well-nourished populations (51, 91, 94, 96, 143, 144, 146). In a British group of 18 month old children, mean iron intake was 4.9 mg in girls and 5.2 mg in boys and 1.7% had IDA (144). Concurring with Lind et al. (48), they reported a positive association between total iron intake and S-Ft levels but not with Hb. In contrast, the positive associations between vitamin C and Hb concen-trations found by Cowin et al. (144) were not confirmed in the present study, probably due to the high over-all intake and low prevalence of ID and anaemia. In Cowin’s study group the prevalence of low Hb levels was higher in children who consumed no meat (144).Thus the prevalence of ID

Discussion

42

is possible to reduce (147) and even prevent by iron fortification of infant products and by introducing meat early during the weaning period in healthy infants (41, 145). Oral iron drops given to iron replete infants are suggested to negatively affect growth and highlight a possible risk with too high iron consumption (148). Such negative effects have not been seen when iron is given in forti-fied infant products. The reason behind the association between the child’s and the mother’s Hb is unclear. Only one previous study suggests that hereditary factors affect Hb concentrations in popula-tions with low prevalence of ID (93). However, the lack of association be-tween the child’s dietary intake and Hb concentration suggests that fam-ily factors besides shared diet are important for the children’s Hb. One such factor is heredity which would be compatible with the tracking of Hb from infancy. It was also surprising to find, even in these healthy well-nourished chil-dren, a positive relation between iron status indices and growth at 4 years. The relative linear growth was posi-tively associated with the Hb concen-trations in girls but not boys, and for S-Ft, the association was seen in boys but not girls. Weight gain and BMI were not associated to any measure of iron status (data not shown).

Dental health and diet Dental caries experience in the pre-sent study group (Paper III) was similar to County Council data, but not quite meeting the WHO goal of 80% caries free-children at 6 years. The study indicated that cheese con-sumption is protective against caries. Daily intake of cheese during a five-day period reduced the risk to have one more caries-affected surface by 33% in pre-school children compared to no cheese. This effect agrees with previous studies in children, adult dry-mouth patients, and elderly sub-jects (149-151). The frequency of black pudding consumption was as-sociated with increased caries ex-perience. However, only 19% of the children were consumers and of them, two had black pudding twice during the five days record. Swedish black pudding is cariogenic since it contains about 8% sucrose and al-most always eaten with jam. Because black pudding clings to the teeth, it is important to brush the teeth after the meal. Sugar is well known as a risk factor of dental caries (152). How-ever, although sugar consumption was high among the children, no as-sociation with caries could be found. This might be due to the generous use of fluoridated toothpaste as well as from the small differences in sugar consumption between the children. In non-fluoridated areas, with de-clining caries prevalence, higher daily consumption of sugars and higher between meals consumption is still a risk factor for caries (153, 154). Reducing the sugar intake should promote dental health as well

Discussion

43

as other health aspects. In many in-dustrial countries, dental health has improved but without a concomitant decrease in sugar consumption. In a recent longitudinal study from Finland, persistently high sucrose intake increased the risk of dental caries in children at 10 years (62). A conclusion from that study was that the habit of consuming excessive amount of sucrose daily starts early in childhood. Thus early childhood oral health needs to be protected. In Sweden, dental care for children is free of charge and preventive strate-gies have been used for more than 30 years. In spite of that, dental caries is still a public health problem, affect-ing more than 20% of young chil-dren. In addition to caries, there are other reasons to reduce the intake of su-crose and sweet products. One such reason is the increasing prevalence of overweight and obesity.

BMI and diet In the present study, we found higher protein intake over time to be associ-ated with higher BMI at 4 years. BMI also tracked over time from 6 months and BMI in infancy was in fact a predictor of BMI at 4 years. There was also an association be-tween the child and the parental BMI; in particular, that of the father. We were surprised to find a stronger association between the child’s BMI and the father’s BMI than the mother’s BMI. Associations between maternal BMI, or obesity, and her

child’s BMI have been described al-though most studies were conducted before the “obesity epidemic” started (155). Regarding the associations with the father’s BMI, few studies have compared the maternal-offspring and paternal-offspring as-sociations of BMI separately. Obesity in both parents increases the risk of obesity 10-fold for obesity in the child (142), compared to two-fold if only the mother is obese (156). Our data is supported by a previous Swedish study where the impact of obesity in fathers was higher than that in mothers on the prevalence of childhood overweight (157). BMI of normal weight girls was found to be associated to that BMI of both par-ents (158). And paternal BMI, inde-pendently of maternal, was reported to be associated with the child’s birth weight (159). Our findings seem to agree with at least some previous ob-servations (160). One possible expla-nation of the positive association be-tween the child’s BMI and the fa-ther’s BMI is suggested to be mother’s dieting behaviour (161). Di-eting is more common in young women than men as is fruit and vege-table consumption (162, 163). On the other hand, young men are eating more meat and drinking more milk, products linked to protein intake (164, 165). In our group, we have no data on parents’ diet and can only speculate. Swedish fathers are per-haps more involved in the care of their children than fathers in many other countries due to the social se-curity system ensuring paid parental leave to both parents. Of the 480

Discussion

44

days parental benefit, 60 days can only be used by the father and 60 on-ly by the mother. The remaining 360 days can be used as the parents see fit. Ii is also of note that in our study group, mean BMI and both over-weight and obesity rates were higher in fathers as compared to mothers, which may have increased the power in the analysis. Although we found higher intake of protein to be associated with higher BMI at 4 years, we could not confirm the suggested association between higher milk intake and higher BMI (126). Higher milk intake at 9-12 months and 12-18 months was posi-tively associated with height and weight at 4 years. Similar results were reported from Denmark where protein intake at 9 months correlated with body size but not body fat at 10 years (166). High early protein intake enhances weight gain in infancy and is suggested to be a risk for obesity later in life (167). BMI increases more in tall children than in short and the authors suggest that this probably is mediated via appetite (168). Protein stimulates secretion of insulin like growth-factor (IGF-1), which might explain the associations between growth and protein intake (169). Often milk protein is sug-gested to be the stimulating factor, but this remains to be proven. More-over it seems that the quantity and quality of protein intake matters (170-172). In our study group, dairy products were the main source of protein and because of that probably

involved in the positive associations between protein intake and BMI. Intake of carbohydrate and protein, as well as of total energy, but not to-tal fat was associated with BMI in the univariate analysis. A major source of energy is dietary fat, but in the present study we found no relation-ship between BMI and dietary fat in-take expressed as total fat, saturated fatty acids, monounsaturated fatty acids, polyunsaturated fatty acids, or as n-3 fatty acid intake. The only sig-nificant univariate association be-tween BMI and fat intake in our study was with n-6 fatty acids, but this association did not remain sig-nificant in the final multivariate re-gression analysis, where only the protein intake and the father’s BMI remained associated with the child’s BMI. In a previously well-described study, energy-dense, low-fibre, and high fat diet was associated with increased fatness in childhood (173). Energy density is increased by fat and it has been demonstrated that high-fat di-ets may lead to an accidental over-consumption of energy and thereby overweight (20, 21). BMI seems to increase more in the tallest children than in the shortest (168), and in in-fants with high growth rate a consis-tently high fat intake during the sec-ond year is positively associated with higher body fat percent between 2 and 5 years of age (174). Disregarding that reduced intake of dietary fat, particularly of saturated

Discussion

45

fat, is generally recommended in healthy diets to reduce the risk for overweight and cardiovascular dis-ease. These recommendations were disputed in a recent study on 4-year olds where a low fat intake seemed to be associated with an increased risk of overweight (35). The reason for these different results is confusing, but may be due to differences in the statistical analyses. We used BMI as a continuous variable that may in-crease the statistical power compared to group analyses when evaluating data sets with small sample sizes (175). Other possible reasons could be differences in how the intake of macronutrients is described and which quantifying measure–gram per kilo body weight (g/kg b.w.), as percent of total energy (E%), or as total intake (g)–is used in the analy-sis. When calculations on intakes are expressed as g/kg b.w., the intake becomes relatively lower with in-creasing body weight, making it diffi-cult to use in comparisons of groups with different body composition. Us-ing E%, the intake of fat E% depends on the intakes of protein and carbo-hydrate; e.g., increased carbohydrate intake reduces the fat E% even if the total intake of fat in gram is constant. Thus collinearity is an important fac-tor to consider in nutritional studies (176, 177). For that reason, we used total intake (g) of macronutrients in the final analyses in the present study, as compared to Garemo who used E% of fat. Surprisingly, we could not confirm any associations between intake of

sucrose and high BMI. This is proba-bly explained by the relatively high average intake of sucrose and small differences within the study group. As mentioned, there was an interac-tion between the intakes of various nutrients. However, contrary to oth-ers (31, 32), we found that the chil-dren with higher dietary fat intake also had a higher sucrose intake. In these analyses, we used nutrients and some foods as single variables. In further analysis, it will be interest-ing to build clusters or models of die-tary patterns and to consider correla-tions between various nutrients and foods. In the present study, a large propor-tion of foods and nutrients came from the commercial baby foods, not only in infancy but also at 4 years. Moreover many children are eating several daily meals at day care cen-tres during weekdays and these are therefore important for the pre-school children’s dietary intake. For children spending full time at a day care centre, about 75% of recom-mended energy and nutrient intake are provided. Both in the day care centre and at home, meals are often based on convenience food. There is an increasing trend to use more of convenience food products. In Ger-man children and adolescents, such foods were high in fat content and high in number of food additives (178). This highlights the importance of food industry to produce high quality foods. Furthermore, many people (parents, chefs, teachers, and other staff) contribute to a child’s

Discussion

46

food habits. Information for parents at well-baby clinics is mostly pro-vided by nurses with paediatric edu-cation (179-181). A possibility to im-prove nutrition knowledge at well-baby clinics is to involve more paedi-atric dieticians. This now happens in Sweden. Another possibility is of course to enhance nutrition educa-tion of the nurses, cooks, chefs, teachers, and other staff of important to maintain high standards in chil-dren’s nutrition care.

Strengths and weakness of the study The strength in this study is the lon-gitudinally design with children fol-lowed from 6 months to 4 years and the careful collection of dietary data, anthropometrics, and blood samples. During the study, there was a close and regular contact between the re-search team and parents, records were continuously checked, and questions asked. During the study, the parents became experienced re-corders. Another important factor is that data of parents were included in the follow-up at 4 years. To validate reported dietary intakes, different methods can be used (13, 131).We calculated the ratio of energy intake (EI) to basal metabolic rate (BMR); EI:BMR at 12 months, 17-18 months, and 4 years and at all ages it was >1.55, which supports the valid-ity of our data (182). In studies on small children, a cut-off ≤ 1.55 might be considered as miss-reporting. However, in small children underre-

porting is considered to be less common than in adults (183). We also believe we have high validity of our data because the families who volunteered in this study had a posi-tive attitude to diet and health re-search in infants and children. In in-fants and small children, fewer re-cording days are needed compared with adults to assess acceptable de-gree of energy and nutrients (184). Reported energy intake in our study is comparable with other studies at 12 months (14, 95, 185) as well as at 18 months (185). In addition, our food intake data at 4 years are simi-lar to previous Swedish studies (22, 34). The weakness is that the study is not population based and the high num-ber of children and parents who did not participate in the follow up at 4 years. The baseline study was planned to continue from 6 to 18 months and the follow-up was planned at a later stage. The parents were asked to participate in the fol-low-up when the child was >3 years. Reported reasons for not participat-ing in the follow-up were lack of time and willingness to do the dietary re-cords. This could be a selection bias. However, children with complete data at 4 years were not different from the not participating children with respect to data at 12 months on dietary intake body weight and length and serum lipid levels. Educa-tion level i.e. parents university stud-ies were more frequent among chil-dren in the follow-up study com-

Discussion

47

pared with the non-participants (32.6% vs.25.4% p < 0.001).

Future aspects This thesis shows important associa-tions between dietary intake and health in Swedish infants and young children. The findings raise new questions such as the impact of die-tary fat quality in infancy on health later in life. Furthermore, another unsolved question is if and how diet, from a practical perspective, will fit recommendations. A better understanding of the rea-sons behind the associations found between the child’s and the parent’s BMI at 4 years is important because it might be used to improve preven-tive strategies. Iron intake in young Swedish chil-dren is adequate and increased in-take is not supposed to affect iron status. As we found the child’s Hb to be associated with that of its mother’s, it seems possible to use in-formation of mother’s Hb as a pre-dictor of the child’s Hb. This could be useful for detection of children at risk and early intervention Another question based on our find-ings is if there is any health implica-tion of the low intake of vitamin D in young children and if there is any need of further interventions.

The Nordic (NNR) and national (SNR), nutrition recommendation are valid for healthy groups. Improv-ing nutrition and health for individu-als, we need more knowledge of in-teractions between diet and genetics, gender and age groups.

Conclusions

48

CONCLUSIONS

In this healthy and well-nourished group of infants and children, we found the quality of food and nutrient composition to be important indicators for the child’s health.

• Higher PUFA and lower SAFA intakes may reduce total cholesterol and low-density lipoprotein cholesterol early in life, particularly in girls. In respect of lowering serum lipid levels in early childhood, it seems appro-priate to set focus on the quality rather than quantity of dietary fat intake although overfeeding should be avoided and energy balance kept. The long term consequences are largely unknown.

• The strongest predictors of Hb concentrations at 4 years were Hb con-

centration in infancy and the mother’s Hb. Milk-based fortified cereals and meat products are important sources of iron in infancy and early childhood. Meat was positively associated with S-Ft and MCV in boys but not in girls.

• Intake of sugar and sweet products were high and many of the sweet

products consumed by children also contain dietary fat with high levels of saturated fatty acids.

• Higher protein intake over time was associated with higher BMI at 4 years. BMI tracked from 6 months to 4 years and BMI at 6 months was in fact a predictor of BMI at 4 years. There was also an association between child and parental BMI, in particular that of the father.

Populärvetenskaplig sammanfattning

49

POPULÄRVETENSKAPLIG SAMMANFATTNING

Historiskt och globalt har målet inom barnnutrition varit att ge barn tillräckligt med mat av god kvalitet för att säkerställa adekvat tillväxt och undvika närings-brister. Rätt mat är en förutsättning för god hälsa både i barndomen och senare i livet. Från att ha varit ett land där näringsbrister var vanliga så har vi idag andra folkhälsoproblem. Det finns flera folkhälsosjukdomar som kan påverkas av kostintaget redan under barndomen. Dit hör övervikt, för höga blodfetter, högt blodtryck, diabetes typ 2 och karies. Vi har undersökt hur matintaget hos barn från 6 månaders ålder och upp till 4 år kan påverka blodfetter, järnstatus, karies och Body Mass Index (BMI) (kroppssammansättning). 300 friska barn ingick i studien från 6 månader till 18-månader. Varje månad gjordes kostregistreringar, tillväxt mättes varje må-nad upp till 12 månaders ålder och därefter varannan månad. Barnen lämnade blodprov vid fyra tillfällen. Vid 4 år inbjöds de att delta i en uppföljning till-sammans med sina föräldrar och 127 familjer tackade ja. Förnyade blodprover togs liksom kostregistreringar och mätningar av längd, vikt och underhudsfett men också en tandhälsoundersökning. I denna uppföljning kunde vi mäta längd och vikt och ta blodprover även på föräldrarna. Vid 12 månaders ålder åt alla utom två barn mycket mättat fett och bara 30% fick tillräcklig med fleromättat fett jämfört med aktuella rekommendationer. Vid fyra år var intaget av socker för högt medan intaget av D-vitamin och selen var lägre än de rekommenderade. Den genomsnittligt största järnkällan var väl-ling, men de barn som hade det enskilt högsta järnintaget hade ätit blodpud-ding. Fettkvalitén hade betydelse för andel blodfetter redan vid 12 månaders ålder. Högre andel fleromättade- och lägre mättade fettsyror gav lägre nivåer totalko-lesterol, low-density-lipoprotein cholesterol (LDL-C) och apolipoprotein B, sär-skilt bland flickor. Det fanns en tendens som visade att barn med högre fettintag hade högre BMI vid 4 år men det var signifikant enbart för n-6 fettsyror. Barnens järnstatus var bra och det fanns ett samband med Hb-värdet över tid; från 6 månader upp till 4 år. Dessutom fanns ett starkt samband mellan mam-mornas Hb och deras barns Hb. Vid 4 års ålder var 70 % av barnen kariesfria. Att äta ost minskade risken att få karies.

Populärvetenskaplig sammanfattning

50

Vi såg att BMI vid 6 månaders ålder hade ett samband med BMI vid 4 år. Ett intressant fynd var sambanden mellan pappornas BMI och barnens BMI. Den starkaste kopplingen till kostintaget var att barn med högre proteinintag hade ett högre BMI vid 4 år. Däremot kunde vi inte styrka påståendet att ett lågt fett-intag eller lågt intag av omgea-3 fettsyror skulle öka risken för övervikt hos barn.

Acknowledgements

51

ACKNOWLEDGMENTS

This project is a team work with many people involved in which I had the bene-

fit of being a part and learn a lot. During this wonderful journey I have met

many inspiring people and new windows are opened in my mind. First of all I

would like to thank all the parents and children for your contribution to this

project. Without your participation by leaving blood samples, writing food re-

cords and measurements of weight, height and skin folds; all this would not

have happened to me.

In particular I will extend my thanks to all my supervisors who have spent so

much time in this project guiding me into and through this scientific area of

paediatric research.

Agneta Hörnell, my main supervisor for your calm patience and everlasting

support in scientific questions and writing and for sharing your excellent ex-

perience in infant feeding with me. Besides that, your memory is clear and

bright, like a sunny winter day.

Olle Hernell, my co-supervisor, for bringing me into this area of nutrition re-

search. Your great experience and knowledge seems to be a without end and

your ability to see beyond details, focus on the entirety and draw clear conclu-

sions are outstanding. Besides all that, your gentle and careful methods have

kept me in direction.

Torbjörn Lind, my co-supervisor, for always sharing your vast knowledge.

Your characteristic way of pointing in the right direction has helped me keep fo-

cus. You know like nobody else how to turn a short question into result. When-

Acknowledgements

52

ever there was a problem you gave an answer, mostly followed by some philo-

sophical words of wisdom.

Ingegerd Johansson, for cooperating, introducing and supporting me in the

new world of dental caries aspects. You made me feel like a kid in a candy store.

Hans Stenlund, for sharing your statistical knowledge and experience and for

your invaluable statistical advices.

Catharina Tennefors who invited me into the project by offering me to work

in the baseline study and contributing in the planning of the study.

Margareta Henriksson and Margareta Beckman, the research nurses

who collected data and worked close together with me. Your professional and

organization skills are priceless.

Phil Lyon, professor at the department of food and nutrition, for your gently

manners and support for my English writing skills.

My colleges and friends, Lisbeth Nordström, Lena Hansson and Ann-Kristin

Sandström, who were working hard at the clinic during my absences. You have

such a supporting manner and are always quick to lend a helping hand.

Ingalill Sandström, a former college, for being such an exemplary pediatric

dietician and a dear friend.

The dieticians, Maria Sehlsted, Maj-Britt Nyberg, Agneta Frängsmyr, Anna

Karlsson for your work with preparing the food records and database.

Ulla Cederholm, my former college, for all your happy, supporting and stimu-

lating questions and messages and for being such a friend.

All friends and colleges at the pediatric clinic and the department of pedia-

trics.

Acknowledgements

53

All my colleges at the department of food and nutrition, for sharing my

interest in pediatric nutrition and supporting the field of science.

Anders, my husband for all support words and your positive attitude, full of

never-ending optimism.

Anton, Karin and Sara, our children, you are making life fun and bringing

me happiness. Besides all that, you also have added me some experiences in

English.

All supporting friends for being friends during my travel in the scientific

area.

References

54

REFERENCES 1. Jägervall M. Nils Rosén von Rosenstein och hans lärobok i pediatrik. Lund: Studentlitteratur; 1990. 2. Brändström A. De kärlekslösa mödrarna (The loveless mothers). Umeå; 1984. 3. Hagelin E, Magnusson M, Sundelin C. Barn, hälsa, vård (Child, health, care). Stockholm: Liber AB; 2000. 4. Odin M. An investigation into questions of social hygiene in the counties of Västerbotten and Norrbotten, Sweden. P. 2, Diseases and the frequency of diseases in upper Norrland, with special reference to the composition of the diet Medicinalstyrelsen, editor. Lund; 1937/1934. 5. Bäckman N, Grahnen H, Ollinen P. Tandvård för folket (dental health services for the people). Umeå; 1988. 6. Samuelson G, Grahnen H, Arvidsson E. An epidemiological study of child health and nutrition in a northern Swedish county. VI. Relationship between general and oral health, food habits, and socioeconomic conditions. Am J Clin Nutr. 1971 Nov;24(11):1361-73. 7. Samuelson G, Sjölin S. An epidemiological study of child health and nu-trition in a northern Swedish county. IV. Haematological investigations, espe-cially in regard to iron deficiency anaemia. Acta Paediatr Scand. 1972 Jan;61(1):63-72. 8. Samuelson G. An epidemiological study of child health and nutrition in a northern Swedish county. 3. Medical and anthropometrical examinations. Acta Paediatr Scand. 1971 Nov;60(6):653-65. 9. Hagman U, Bruce A, Persson LÅ, Samuelson G, Sjölin S. Food habits and nutrient intake in childhood in relation to health and socio-economic condi-tions. A Swedish Multicentre Study 1980-81. Acta Paediatr Scand Suppl. 1986;328:1-56. 10. Persson LÅ, Johansson E, Samuelson G. Dietary intake of weaned infants in a Swedish community. Hum Nutr Appl Nutr. 1984 Aug;38:247-54. 11. Persson LÅ, Holm AK, Arvidsson S, Samuelson G. Infant feeding and dental caries--a longitudinal study of Swedish children. Swed Dent J. 1985;9(5):201-6. 12. Butte NF. Fat intake of children in relation to energy requirements. Am J Clin Nutr. 2000 Nov;72(5 Suppl):1246S-52S. 13. Tennefors C, Coward WA, Hernell O, Wright A, Forsum E. Total energy expenditure and physical activity level in healthy young Swedish children 9 or 14 months of age. Eur J Clin Nutr. 2003 May;57(5):647-53. 14. Simell O, Niinikoski H, Rönnemaa T, Lapinleimu H, Touti T, Lagström H et al. Special Turku Coronary Risk Factor Intervention Project for Babies (STRIP). Am J Clin Nutr. 2000;72 (5 Suppl):1316S-31S.

References

55

15. Petersen S, Brulin C, Bergström E. Increasing prevalence of overweight in young schoolchildren in Umeå, Sweden, from 1986 to 2001. Acta Paediatr. 2003 Jul;92(7):848-53. 16. Blomquist HK, Bergström E. Obesity in 4-year-old children more preva-lent in girls and in municipalities with a low socioeconomic level. Acta Paediatr. 2007 Jan;96(1):113-6. 17. Shetty PS. Adaptation to low energy intakes: the responses and limits to low intakes in infants, children and adults. Eur J Clin Nutr. 1999 Apr;53 Suppl 1:S14-33. 18. Torun B. Energy requirements of children and adolescents. Public Health Nutr. 2005 Oct;8(7A):968-93. 19. Butte NF, Wong WW, Hopkinson JM, Heinz CJ, Mehta NR, Smith EO. Energy requirements derived from total energy expenditure and energy deposi-tion during the first 2 y of life. Am J Clin Nutr. 2000 Dec;72(6):1558-69. 20. Prentice AM. Macronutrients as sources of food energy. Public Health Nutr. 2005 Oct;8(7A):932-9. 21. Prentice AM. Manipulation of dietary fat and energy density and subse-quent effects on substrate flux and food intake. Am J Clin Nutr. 1998 Mar;67(3 Suppl):535S-41S. 22. Enghardt Barbeieri H. Pearson M, Becker W. Dietary habits and nutrient intake in Swedish children 4 year old and school children in grade 2 and 5 (Riksmaten - barn 2003) Uppsala: National Food Administration (Livsmedels-verket), Uppsala, Sweden.; 2006 Contract No.: Document Number|. 23. Niinikoski H, Viikari J, Rönnemaa T, Helenius H, Jokinen E, Lapinleimu H, et al. Regulation of growth of 7- to 36-month-old children by energy and fat intake in the prospective, randomized STRIP baby trial. Pediatrics. 1997 Nov;100(5):810-6. 24. NNR. Nordic Nutrition recommendations 2004. 4 ed. Copenhagen: Nor-dic council of ministers; 2004. 25. Aggett PJ, Haschke F, Heine W, Hernell O, Koletzko B, Lafeber H, et al. Committee report: childhood diet and prevention of coronary heart disease. ESPGAN Committee on Nutrition. European Society of Pediatric Gastroenterol-ogy and Nutrition. J Pediatr Gastroenterol Nutr. 1994 Oct;19(3):261-9. 26. Bier DM, Brosnan JT, Flatt JP, Hanson RW, Heird W, Hellerstein MK, et al. Report of the IDECG Working Group on lower and upper limits of carbohy-drate and fat intake. International Dietary Energy Consultative Group. Eur J Clin Nutr. 1999 Apr;53 Suppl 1:S177-8. 27. Koletzko B. Response to and range of acceptable fat intakes in infants and children. Eur J Clin Nutr. 1999 Apr;53 Suppl 1:S78-83. 28. Michaelsen K.F. Feeding and nutrition of infants and young children. In: (WHO) WHO, editor: WHO Library Cataloguing in Publication Data; 2000.

References

56

29. Thorsdottir. I, Gunnarsdottir I, Paulsson GI. Birth Weight, growth and feeding in infancy: relation to serum lipid concentration in 12-month-old in-fants. European Clin Nutr. 2003 ;57:1479-85. 30. Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech Rep Ser. 2000;894:i-xii, 1-253. 31. Nicklas TA, Webber LS, Koschak M, Berenson GS. Nutrient adequacy of low fat intakes for children: the Bogalusa Heart Study. Pediatrics. 1992 Feb;89(2):221-8. 32. Tonstad S, Sivertsen M. Relation between dietary fat and energy and mi-cronutrient intakes. Arch Dis Child. 1997 May;76(5):416-20. 33. Forsander G, Malmodin B, Eklund C, Persson B. Relationship between dietary intake in children with diabetes mellitus type I, their management at di-agnosis, social factors, anthropometry and glycaemic controls. Scandinavian Journal of Nutrition. 2003;47(2):75 - 83. 34. Garemo M, Lenner RA, Strandvik B. Swedish pre-school children eat too much junk food and sucrose. Acta Paediatr. 2007 Feb;96(2):266-72. 35. Garemo M, Palsdottir V, Strandvik B. Metabolic markers in relation to nutrition and growth in healthy 4-y-old children in Sweden. Am J Clin Nutr. 2006;84(5):1021-6. 36. Öhlund I, Hörnell A, Lind T, Hernell O. Dietary fat in infancy should be more focused on quality than on quantity. 2007 (June)10.1038/sj.ejcn.1602824 37. Nordiska Ministerrådet Köpenhamn (Nordic Council of Ministers) 1996. Nordiska Näringsrekommendationer 1996. 38. Zimmermann MB, Hurrell RF. Nutritional iron deficiency. Lancet. 2007 Aug 11;370(9586):511-20. 39. Rodriguez SC, Hotz C, Rivera JA. Bioavailable dietary iron is associated with hemoglobin concentration in Mexican preschool children. J Nutr. 2007 Oct;137(10):2304-10. 40. Navas-Carretero S, Perez-Granados AM, Sarria B, Carbajal A, Pedrosa MM, Roe MA, et al. Oily fish increases iron bioavailability of a phytate rich meal in young iron deficient women. J Am Coll Nutr. 2008 Feb;27(1):96-101. 41. Engelmann MD, Davidsson L, Sandström B, Walczyk T, Hurrell RF, Mi-chaelsen KF. The influence of meat on nonheme iron absorption in infants. Pe-diatr Res. 1998 Jun;43(6):768-73. 42. Niinikoski H, Koskinen P, Punnonen K, Seppanen R, Viikari J, Rönne-maa T, et al. Intake and indicators of iron and zinc status in children consuming diets low in saturated fat and cholesterol: the STRIP baby study. Special Turku Coronary Risk Factor Intervention Project for Babies. Am J Clin Nutr. 1997 Sep;66(3):569-74. 43. Gleerup A, Rossander-Hulthen L, Gramatkovski E, Hallberg L. Iron ab-sorption from the whole diet: comparison of the effect of two different distribu-tions of daily calcium intake. Am J Clin Nutr. 1995 Jan;61(1):97-104.

References

57

44. Minihane AM, Fairweather-Tait SJ. Effect of calcium supplementation on daily nonheme-iron absorption and long-term iron status. Am J Clin Nutr. 1998 Jul;68(1):96-102. 45. Molgaard C, Kaestel P, Michaelsen KF. Long-term calcium supplementa-tion does not affect the iron status of 12-14-y-old girls. Am J Clin Nutr. 2005 Jul;82(1):98-102. 46. Sandström B. Micronutrient interactions: effects on absorption and bio-availability. Br J Nutr. 2001 May;85 Suppl 2:S181-5. 47. Lind T, Lönnerdal B, Persson LÅ, Stenlund H, Tennefors C, Hernell O. Effects of weaning cereals with different phytate contents on hemoglobin, iron stores, and serum zinc: a randomized intervention in infants from 6 to 12 mo of age. Am J Clin Nutr. 2003 Jul;78(1):168-75. 48. Lind T, Hernell O, Lönnerdal B, Stenlund H, Domellöf M, Persson LÅ. Dietary iron intake is positively associated with hemoglobin concentration dur-ing infancy but not during the second year of life. J Nutr. 2004 May;134(5):1064-70. 49. Domellöf M. Iron requirements, absorption and metabolism in infancy and childhood. Curr Opin Clin Nutr Metab Care. 2007 May;10(3):329-35. 50. 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 Dec;3(4):433-40. 51. Thane CW, Bates CJ, Prentice A. Risk factors for low iron intake and poor iron status in a national sample of British young people aged 4-18 years. Public Health Nutr. 2003 Aug;6(5):485-96. 52. Virtanen MA, Svahn CJ, Viinikka LU, Raiha NC, Siimes MA, Axelsson IE. Iron-fortified and unfortified cow's milk: effects on iron intakes and iron status in young children. Acta Paediatr. 2001 Jul;90(7):724-31. 53. Garemo M, Arvidsson Lenner R, Nilsson EK, Borres MP, Strandvik B. Food choice, socio-economic characteristics and health in 4-year olds in a well-educated urban Swedish community. Clin Nutr. 2007 Feb;26(1):133-40. 54. Sepp H, Lennernäs M, Pettersson R, Abrahamsson L. Children's nutrient intake at preschool and at home. Acta Paediatr. 2001 May;90(5):483-91. 55. Thorsdottir I, Gunnarsson BS, Atladottir H, Michaelsen KF, Palsson G. Iron status at 12 months of age -- effects of body size, growth and diet in a popu-lation with high birth weight. Eur J Clin Nutr. 2003 Apr;57(4):505-13. 56. Agostoni C, Decsi T, Fewtrell M, Goulet O, Kolacek S, Koletzko B, et al. Complementary feeding: a commentary by the ESPGHAN Committee on Nutri-tion. J Pediatr Gastroenterol Nutr. 2008;46(1):99-110. 57. Axelsson I, Gebre-Medhin M, Hernell O, Jakobsonn I, Michaelsen KF, Samuelson G. [Recommendations for prevention of iron deficiency. Delay cow's milk intake as a beverage to infants until 10-12 months of age!]. Lakartidningen. 1999 May 5;96(18):2206-8.

References

58

58. Axelsson I, Gebre-Medhin M, Hernell O, Jakobsson I, Michaelsen KF, Samuelson G. [A reply about milk porridge. Infant food is also a question of nu-tritional physiology]. Lakartidningen. 1999;96(21):2624-5. 59. Cai F, Shen P, Morgan MV, Reynolds EC. Remineralization of enamel subsurface lesions in situ by sugar-free lozenges containing casein phosphopep-tide-amorphous calcium phosphate. Aust Dent J. 2003 Dec;48(4):240-3. 60. Nicklas TA, Hayes D. Position of the American Dietetic Association: nu-trition guidance for healthy children ages 2 to 11 years. J Am Diet Assoc. 2008 Jun;108(6):1038-44, 1046-7. 61. Johnson RK, Frary C. Choose beverages and foods to moderate your in-take of sugars: the 2000 dietary guidelines for Americans--what's all the fuss about? J Nutr. 2001 Oct;131(10):2766S-71S. 62. Ruottinen S, Karjalainen S, Pienihakkinen K, Lagström H, Niinikoski H, Salminen M, et al. Sucrose intake since infancy and dental health in 10-year-old children. Caries Res. 2004 Mar-Apr;38(2):142-8. 63. Hjern A. Chapter 7: children's and young people's health. Scand J Public Health Suppl. 2006 Jun;67:165-83. 64. Magnusson M, Garrett MJ, Sundelin C. Child health service's response to early feeding and behaviour problems: changes in parental perceptions 1970-93. Acta Paediatr. 1999 Oct;88(10):1094-100. 65. Sundelin C, Håkansson A. The importance of the Child Health Services to the health of children: summary of the state-of-the-art document from the Sig-tuna conference on Child Health Services with a view to the future. Acta Pae-diatr Suppl. 2000 Sep;89(434):76-9. 66. Berenson GS. Childhood risk factors predict adult risk associated with subclinical cardiovascular disease. The Bogalusa Heart Study. Am J Cardiol. 2002;90(10C):3L-7L. 67. Mikkelsen A, Rinne-Ljungqvist L, Borres MP, van Odijk J. Do parents follow breastfeeding and weaning recommendations given by pediatric nurses? A study with emphasis on introduction of cow's milk protein in allergy risk fami-lies. J Pediatr Health Care. 2007 Jul-Aug;21(4):238-44. 68. Amning av barn födda 1997 (Breastfeeding, children born 1997). Stock-holm: Socialstyrelsen (The National Board of Health and Welfare); 1999. 69. Singhal A, Lucas A. Early origins of cardiovascular disease: is there a un-ifying hypothesis? Lancet. 2004 May 15;363(9421):1642-5. 70. Barker DJ. Adult consequences of fetal growth restriction. Clin Obstet Gynecol. 2006 Jun;49(2):270-83. 71. Forsen T, Eriksson J, Tuomilehto J, Reunanen A, Osmond C, Barker D. The fetal and childhood growth of persons who develop type 2 diabetes. Ann In-tern Med. 2000 Aug 1;133(3):176-82. 72. Singhal A, Cole TJ, Lucas A. Early nutrition in preterm infants and later blood pressure: two cohorts after randomised trials. Lancet. 2001 Feb 10;357(9254):413-9.

References

59

73. Demmelmair H, von Rosen J, Koletzko B. Long-term consequences of early nutrition. Early Hum Dev. 2006 Aug;82(8):567-74. 74. Koletzko B, Broekaert I, Demmelmair H, Franke J, Hannibal I, Oberle D, et al. Protein intake in the first year of life: a risk factor for later obesity? The E.U. childhood obesity project. Adv Exp Med Biol. 2005;569:69-79. 75. Socialstyrelsen. Hjärt- och kärlsjukdomar fortfarande vanligaste dödsor-saken. http://www.socialstyrelsen.se. 2008 76. Socialstyrelsen. Nationella riktlinjer för hjärtsjukvård. http://www.socialstyrelsen.se . 2008 77. Simoes C, Bianciardi G, Toti P, Weber G, Tanganelli P. World Health Or-ganization (WHO) and the World Heart Federation (WHF) Pathobiological De-terminants of Atherosclerosis in Youth (PBDAY) Study. Lipid and raised lesion distribution in the right coronary artery of young people. Nutr Metab Cardio-vasc Dis. 1999 Dec;9(6):277-83. 78. Myers L, Coughlin SS, Webber LS, Srinivasan SR, Berensson GS . Predic-tion of Adult Cardiovascular Multifactorial Risk Status from Childhood Risk Factor Levels The Bogalusa Heart Study. American journal of epidemiolo-gy. 1995;142(9):918-24. 79. Morrison JA, Friedman LA, Gray-McGuire C. Metabolic syndrome in childhood predicts adult cardiovascular disease 25 years later: the Princeton Li-pid Research Clinics Follow-up Study. Pediatrics. 2007 Aug;120(2):340-5. 80. Freedman DS, Dietz WH, Srinivasan SR, Berenson GS. The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics. 1999 Jun;103(6 Pt 1):1175-82. 81. Berenson GS, Pickoff AS. Preventive cardiology and its potential influ-ence on the early natural history of adult heart diseases: the Bogalusa Heart Study and the Heart Smart Program. Am J Med Sci. 1995 Dec;310 Suppl 1:S133-8. 82. Arbeit ML, Johnson CC, Mott DS, Harsha DW, Nicklas TA, Webber LS, et al. The Heart Smart cardiovascular school health promotion: behavior corre-lates of risk factor change. Prev Med. 1992 Jan;21(1):18-32. 83. Niinikoski H, Lagstrom H, Jokinen E, Siltala M, Ronnemaa T, Viikari J, et al. Impact of repeated dietary counseling between infancy and 14 years of age on dietary intakes and serum lipids and lipoproteins: the STRIP study. Circula-tion. 2007 Aug 28;116(9):1032-40. 84. Åkeson PM, Axelsson IE, Raiha NC. Plasma lipids and apolipoproteins in breastfed and formula-fed Swedish infants. Acta Paediatr. 1999 Jan;88(1):1-6. 85. Svahn JC, Feldl F, Raiha NC, Koletzko B, Axelsson IE. Different quanti-ties and quality of fat in milk products given to young children: effects on long chain polyunsaturated fatty acids and trans fatty acids in plasma. Acta Paediatr. 2002;91(1):20-9. 86. Wennlöf AH, Yngve A, Nilsson TK, Sjöström M. Serum lipids, glucose and insulin levels in healthy schoolchildren aged 9 and 15 years from Central

References

60

Sweden: reference values in relation to biological, social and lifestyle factors. Scand J Clin Lab Invest. 2005;65(1):65-76. 87. WHO. Iron deficiency anemia: assement, prevention and control. A guide for programme managers. Geneva: World Health Organization (WHO),United nations Children´s Fund (Unicef) and United Nations University;2001. 88. Aggett PJ, Agostoni C, Axelsson I, Bresson JL, Goulet O, Hernell O, et al. Iron metabolism and requirements in early childhood: do we know enough?: a commentary by the ESPGHAN Committee on Nutrition. J Pediatr Gastroenterol Nutr. 2002;34(4):337-45. 89. Domellöf M, Lönnerdal B, Dewey KG, Cohen RJ, Rivera LL, Hernell O. Sex differences in iron status during infancy. Pediatrics. 2002 Sep;110(3):545-52. 90. Bergström E, Hernell O, Lönnerdal B, Persson LÅ. Sex differences in iron stores of adolescents: what is normal? J Pediatr Gastroenterol Nutr. 1995 Feb;20(2):215-24. 91. Freeman VE, Mulder J, van't Hof MA, Hoey HM, Gibney MJ. A longitu-dinal study of iron status in children at 12, 24 and 36 months. Public Health Nutr. 1998 Jun;1(2):93-100. 92. Faber M, Swanevelder S, Benade AJ. Is there an association between the nutritional status of the mother and that of her 2-year-old to 5-year-old child? Int J Food Sci Nutr. 2005 Jun;56(4):237-44. 93. Siimes MA, Kallio MJ, Salmenpera L, Perheentupa J. Effect of heredity on hemoglobin concentration. J Pediatr. 1994 Jan;124(1):100-2. 94. Hernell O, Lönnerdal B. Is iron deficiency in infants and young children common in Scandinavia and is there a need for enforced primary prevention? Acta Paediatr. 2004 Aug;93(8):1024-6. 95. Gunnarsson BS, Thorsdottir I, Pålsson G. Iron status in 6-y-old children: associations with growth and earlier iron status. Eur J Clin Nutr. 2005 Jun;59(6):761-7. 96. Bramhagen AC, Axelsson I. Iron status of children in southern Sweden: effects of cow's milk and follow-on formula. Acta Paediatr. 1999 Dec;88(12):1333-7. 97. Persson LÅ, Lundström M, Lönnerdal B, Hernell O. Are weaning foods causing impaired iron and zinc status in 1-year-old Swedish infants? A cohort study. Acta Paediatr. 1998 Jun;87(6):618-22. 98. Touger-Decker R, van Loveren C. Sugars and dental caries. Am J Clin Nutr. 2003 Oct;78(4):881S-92S. 99. Garcia-Closas R, Garcia-Closas M, Serra-Majem L. A cross-sectional study of dental caries, intake of confectionery and foods rich in starch and su-gars, and salivary counts of Streptococcus mutans in children in Spain. Am J Clin Nutr. 1997 Nov;66(5):1257-63.

References

61

100. Ahola AJ, Yli-Knuuttila H, Suomalainen T, Poussa T, Ahlstrom A, Meur-man JH, et al. Short-term consumption of probiotic-containing cheese and its effect on dental caries risk factors. Arch Oral Biol. 2002 Nov;47(11):799-804. 101. Alanen P, Isokangas P, Gutmann K. Xylitol candies in caries prevention: results of a field study in Estonian children. Community Dent Oral Epidemiol. 2000 Jun;28(3):218-24. 102. Nase L, Hatakka K, Savilahti E, Saxelin M, Ponka A, Poussa T, et al. Ef-fect of long-term consumption of a probiotic bacterium, Lactobacillus rhamno-sus GG, in milk on dental caries and caries risk in children. Caries Res. 2001 Nov-Dec;35(6):412-20. 103. Källestål C, Wall S. Socio-economic effect on caries. Incidence data among Swedish 12-14-year-olds. Community Dent Oral Epidemiol. 2002 Apr;30(2):108-14. 104. Gerdin EW, Angbratt M, Aronsson K, Eriksson E, Johansson I. Dental caries and body mass index by socio-economic status in Swedish children. Community Dent Oral Epidemiol. 2008 Feb 18. 105. Stecksen-Blicks C, Rydberg A, Nyman L, Asplund S, Svanberg C. Dental caries experience in children with congenital heart disease: a case-control study. Int J Paediatr Dent. 2004 Mar;14(2):94-100. 106. Rosen L, Stecksen-Blicks C. Experience of dental care for children with congenital heart disease among Swedish dentists. Swed Dent J. 2007;31(2):85-90. 107. Touger-Decker R, Mobley CC. Position of the American Dietetic Associa-tion: oral health and nutrition. J Am Diet Assoc. 2007 Aug;107(8):1418-28. 108. Twetman S, Petersson GH, Bratthall D. Caries risk assessment as a pre-dictor of metabolic control in young Type 1 diabetics. Diabet Med. 2005 Mar;22(3):312-5. 109. Socialstyrelsen. Tandhälsan hos barn och ungdomar 1985-2005. www.socialstyrelsen.se; 2006 110. Stecksen-Blicks C, Stenlund H, Twetman S. Caries distribution in the dentition and significant caries index in Swedish 4-year-old children 1980-2002. Oral Health Prev Dent. 2006;4(3):209-14. 111. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard defi-nition for child overweight and obesity worldwide: international survey. Bmj. 2000 May 6;320(7244):1240-3. 112. Cole TJ, Flegal KM, Nicholls D, Jackson AA. Body mass index cut offs to define thinness in children and adolescents: international survey. BMJ. 2007 Jul 28;335(7612):194. 113. Aeberli I, Kaspar M, Zimmermann MB. Dietary intake and physical activ-ity of normal weight and overweight 6 to 14 year old Swiss children. Swiss Med Wkly. 2007 Jul 28;137(29-30):424-30. 114. Hughes AR, Farewell K, Harris D, Reilly JJ. Quality of life in a clinical sample of obese children. Int J Obes (Lond). 2007 Jan;31(1):39-44.

References

62

115. Flodmark CE. The happy obese child. Int J Obes (Lond). 2005 Sep;29 Suppl 2:S31-3. 116. Reilly JJ, Methven E, McDowell ZC, Hacking B, Alexander D, Stewart L, et al. Health consequences of obesity. Arch Dis Child. 2003 Sep;88(9):748-52. 117. Holmback U, Fridman J, Gustafsson J, Proos L, Sundelin C, Forslund A. Overweight more prevalent among children than among adolescents. Acta Pae-diatr. 2007 Apr;96(4):577-81. 118. Perlhagen J. Flodmark CE, Hernell O. Fetma hos barn-prevention enda realistiska lösningen på problemet. Läkartidningen. 2007(3):138-41. 119. Ekblom O, Oddsson K, Ekblom B. Prevalence and regional differences in overweight in 2001 and trends in BMI distribution in Swedish children from 1987 to 2001. Scand J Public Health. 2004;32(4):257-63. 120. Huus K, Ludvigsson JF, Enskar K, Ludvigsson J. Risk factors in child-hood obesity-findings from the All Babies In Southeast Sweden (ABIS) cohort. Acta Paediatr. 2007 Sep;96(9):1321-5. 121. Sjöberg A, Lissner L, Albertsson-Wikland K, Marild S. Recent anthropo-metric trends among Swedish school children: evidence for decreasing preva-lence of overweight in girls. Acta Paediatr. 2008 Jan;97(1):118-23. 122. Rovner AJ, O'Brien KO. Hypovitaminosis D among healthy children in the United States: a review of the current evidence. Arch Pediatr Adolesc Med. 2008 Jun;162(6):513-9. 123. Gordon CM, Feldman HA, Sinclair L, Williams AL, Kleinman PK, Perez-Rossello J, et al. Prevalence of vitamin D deficiency among healthy infants and toddlers. Arch Pediatr Adolesc Med. 2008 Jun;162(6):505-12. 124. Huh SY, Gordon CM. Vitamin D deficiency in children and adolescents: epidemiology, impact and treatment. Rev Endocr Metab Disord. 2008 Jun;9(2):161-70. 125. Jönsson H. Mjölk - en kulturanalys av mejeridiskens nya ekonomi. Stockholm: Brutus Östlings Bokförlag; 2005. 126. Gunther AL, Remer T, Kroke A, Buyken AE. Early protein intake and lat-er obesity risk: which protein sources at which time points throughout infancy and childhood are important for body mass index and body fat percentage at 7 y of age? Am J Clin Nutr. 2007 Dec;86(6):1765-72. 127. Gunnarsson BS, Thorsdottir I, Palsson G. Associations of iron status with dietary and other factors in 6-year-old children. Eur J Clin Nutr. 2007 Mar;61(3):398-403. 128. Stintzing G, Sundelin C. [Infant food is not only a question of nutritional physiology. Doubtful to advise against milk with porridge]. Läkartidningen. 1999 May 12;96(19):2374-6. 129. Samuelson G, Bratteby LE, Mohsen R, Vessby B. Dietary fat intake in healthy adolescents: inverse relationships between the estimated intake of satu-rated fatty acids and serum cholesterol. Br J Nutr. 2001 Mar;85(3):333-41.

References

63

130. Iggman D, BirgitsdottirB, Ramel A, Hill J, Thorsdottir. Differences in cow´s milk composition between Island and the other Nordic countries and possible connections to public health. Scand J Nutr. 2003;47(4):194-8. 131. Biong AS, Veierod MB, Ringstad J, Thelle DS, Pedersen JI. Intake of milk fat, reflected in adipose tissue fatty acids and risk of myocardial infarction: a case-control study. Eur J Clin Nutr. 2006 Feb;60(2):236-44. 132. Owen CG, Whincup PH, Odoki K, Gilg JA, Cook DG. Infant feeding and blood cholesterol: a study in adolescents and a systematic review. Pediatrics. 2002 Sep;110(3):597-608. 133. Ritchie RF, Palomaki GE, Neveux LM, Ledue TB, Marcovina S, Navolots-kaia O. Reference distributions for apolipoproteins AI and B and B/AI ratios: comparison of a large cohort to the world's literature. J Clin Lab Anal. 2006;20(5):218-26. 134. Lagström H, Jokinen E, Seppanen R, Ronnemaa T, Viikari J, Valimaki I, et al. Nutrient intakes by young children in a prospective randomized trial of a low-saturated fat, low-cholesterol diet. The STRIP Baby Project. Special Turku Coronary Risk Factor Intervention Project for Babies. Arch Pediatr Adolesc Med. 1997 Feb;151(2):181-8. 135. Thorsdottir I, Gunnarsdottir I, Palsson GI. Birth weight, growth and feeding in infancy: relation to serum lipid concentration in 12-month-old in-fants. Eur J Clin Nutr. 2003 Nov;57(11):1479-85. 136. Thorsdottir I, Hill J, Ramel A. Seasonal variation in cis-9, trans-11 conju-gated linoleic acid content in milk fat from Nordic countries. J Dairy Sci. 2004 Sep;87(9):2800-2. 137. Thorsdottir I, Hill J, Ramel A. Omega-3 fatty acid supply from milk asso-ciates with lower type 2 diabetes in men and coronary heart disease in women. Prev Med. 2004 Sep;39(3):630-4. 138. Talvia S, Lagström H, Rasanen M, Salminen M, Rasanen L, Salo P, et al. A randomized intervention since infancy to reduce intake of saturated fat: calo-rie (energy) and nutrient intakes up to the age of 10 years in the Special Turku Coronary Risk Factor Intervention Project. Arch Pediatr Adolesc Med. 2004 Jan;158(1):41-7. 139. Rask-Nissila L, Jokinen E, Viikari J, Tammi A, Rönnemaa T, Marniemi J, et al. Impact of dietary intervention, sex, and apolipoprotein E phenotype on tracking of serum lipids and apolipoproteins in 1- to 5-year-old children: the Special Turku Coronary Risk Factor Intervention Project (STRIP). Arterioscler Thromb Vasc Biol. 2002 Mar 1;22(3):492-8. 140. Lozoff B, Kaciroti N, Walter T. Iron deficiency in infancy: applying a phy-siologic framework for prediction. Am J Clin Nutr. 2006 Dec;84(6):1412-21. 141. Bramhagen AC, Virtanen M, Siimes MA, Axelsson I. Transferrin receptor in children and its correlation with iron status and types of milk consumption. Acta Paediatr. 2003 Jun;92(6):671-5.

References

64

142. Reilly JJ, Armstrong J, Dorosty AR, Emmett PM, Ness A, Rogers I, et al. Early life risk factors for obesity in childhood: cohort study. BMJ. 2005 Jun 11;330(7504):1357. 143. Hay G, Sandstad B, Whitelaw A, Borch-Iohnsen B. Iron status in a group of Norwegian children aged 6-24 months. Acta Paediatr. 2004 May;93(5):592-8. 144. Cowin I, Emond A, Emmett P. Association between composition of the diet and haemoglobin and ferritin levels in 18-month-old children. Eur J Clin Nutr. 2001 Apr;55(4):278-86. 145. Fitch CW, Cannon MS, Seidel GE, Krummel DA. Dietary factors affecting iron status of children residing in rural West Virginia. W V Med J. 2008 May-Jun;104(3):19-22. 146. Schneider JM, Fujii ML, Lamp CL, Lönnerdal B, Dewey KG, Zidenberg-Cherr S. Anemia, iron deficiency, and iron deficiency anemia in 12-36-mo-old children from low-income families. Am J Clin Nutr. 2005 Dec;82(6):1269-75. 147. Lutter CK, Rodriguez A, Fuenmayor G, Avila L, Sempertegui F, Escobar J. Growth and micronutrient status in children receiving a fortified complemen-tary food. J Nutr. 2008 Feb;138(2):379-88. 148. Domellöf M, Lind T, Lönnerdal B, Persson LÅ, Dewey KG, Hernell O. Ef-fects of mode of oral iron administration on serum ferritin and haemoglobin in infants. Acta Paediatr. 2008 Jun 19. 149. Sela M, Gedalia I, Shah L, Skobe Z, Kashket S, Lewinstein I. Enamel re-hardening with cheese in irradiated patients. Am J Dent. 1994 Jun;7(3):134-6. 150. Papas AS, Joshi A, Palmer CA, Giunta JL, Dwyer JT. Relationship of diet to root caries. Am J Clin Nutr. 1995 Feb;61(2):423S-9S. 151. Kashket S, DePaola DP. Cheese consumption and the development and progression of dental caries. Nutr Rev. 2002 Apr;60(4):97-103. 152. Sheiham A. Dietary effects on dental diseases. Public Health Nutr. 2001 Apr;4(2B):569-91. 153. Burt BA, Eklund SA, Morgan KJ, Larkin FE, Guire KE, Brown LO, et al. The effects of sugars intake and frequency of ingestion on dental caries incre-ment in a three-year longitudinal study. J Dent Res. 1988 Nov;67(11):1422-9. 154. Szpunar SM, Eklund SA, Burt BA. Sugar consumption and caries risk in schoolchildren with low caries experience. Community Dent Oral Epidemiol. 1995 Jun;23(3):142-6. 155. Lawlor DA, Smith GD, O'Callaghan M, Alati R, Mamun AA, Williams GM, et al. Epidemiologic evidence for the fetal overnutrition hypothesis: findings from the mater-university study of pregnancy and its outcomes. Am J Epide-miol. 2007 Feb 15;165(4):418-24. 156. Whitaker RC. Predicting preschooler obesity at birth: the role of maternal obesity in early pregnancy. Pediatrics. 2004 Jul;114(1):e29-36. 157. Mårild S, Bondestam M, Bergström R, Ehnberg S, Hollsing A, Al-bertsson-Wikland K. Prevalence trends of obesity and overweight among 10-

References

65

year-old children in western Sweden and relationship with parental body mass index. Acta Paediatr. 2004 Dec;93(12):1588-95. 158. Treuth MS, Butte NF, Ellis KJ, Martin LJ, Comuzzie AG. Familial resem-blance of body composition in prepubertal girls and their biological parents. Am J Clin Nutr. 2001 Oct;74(4):529-33. 159. Klebanoff MA, Mednick BR, Schulsinger C, Secher NJ, Shiono PH. Fa-ther's effect on infant birth weight. Am J Obstet Gynecol. 1998 May;178(5):1022-6. 160. Kivimaki M, Lawlor DA, Smith GD, Elovainio M, Jokela M, Keltikangas-Jarvinen L, et al. Substantial intergenerational increases in body mass index are not explained by the fetal overnutrition hypothesis: the Cardiovascular Risk in Young Finns Study. Am J Clin Nutr. 2007 Nov;86(5):1509-14. 161. Vogels N, Posthumus DL, Mariman EC, Bouwman F, Kester AD, Rump P, et al. Determinants of overweight in a cohort of Dutch children. Am J Clin Nutr. 2006 Oct;84(4):717-24. 162. Baker AH, Wardle J. Sex differences in fruit and vegetable intake in older adults. Appetite. 2003 Jun;40(3):269-75. 163. Nowak M, Speare R. Gender differences in food-related concerns, beliefs and behaviours of north Queensland adolescents. J Paediatr Child Health. 1996 Oct;32(5):424-7. 164. Von Post-Skagegard M, Samuelson G, Karlstrom B, Mohsen R, Berglund L, Bratteby LE. Changes in food habits in healthy Swedish adolescents during the transition from adolescence to adulthood. Eur J Clin Nutr. 2002 Jun;56(6):532-8. 165. Roos G, Prattala R, Koski K. Men, masculinity and food: interviews with Finnish carpenters and engineers. Appetite. 2001 Aug;37(1):47-56. 166. Hoppe C, Molgaard C, Thomsen BL, Juul A, Michaelsen KF. Protein in-take at 9 mo of age is associated with body size but not with body fat in 10-y-old Danish children. Am J Clin Nutr. 2004 Mar;79(3):494-501. 167. Koletzko B. Long-term consequences of early feeding on later obesity risk. Nestle Nutr Workshop Ser Pediatr Program. 2006;58:1-18. 168. Buchan IE, Bundred PE, Kitchiner DJ, Cole TJ. Body mass index has ris-en more steeply in tall than in short 3-year olds: serial cross-sectional surveys 1988-2003. Int J Obes (Lond). 2007 Jan;31(1):23-9. 169. Bertram HC, Hoppe C, Petersen BO, Duus JO, Molgaard C, Michaelsen KF. An NMR-based metabonomic investigation on effects of milk and meat pro-tein diets given to 8-year-old boys. Br J Nutr. 2007 Apr;97(4):758-63. 170. Rolland-Cachera MF, Deheeger M, Akrout M, Bellisle F. Influence of ma-cronutrients on adiposity development: a follow up study of nutrition and growth from 10 months to 8 years of age. Int J Obes Relat Metab Disord. 1995 Aug;19(8):573-8.

References

66

171. Rogers I, Emmett P, Gunnell D, Dunger D, Holly J. Milk as a food for growth? The insulin-like growth factors link. Public Health Nutr. 2006 May;9(3):359-68. 172. Hoppe C, Molgaard C, Juul A, Michaelsen KF. High intakes of skimmed milk, but not meat, increase serum IGF-I and IGFBP-3 in eight-year-old boys. Eur J Clin Nutr. 2004 Sep;58(9):1211-6. 173. Johnson L, Mander AP, Jones LR, Emmett PM, Jebb SA. Energy-dense, low-fiber, high-fat dietary pattern is associated with increased fatness in child-hood. Am J Clin Nutr. 2008 Apr;87(4):846-54. 174. Karaolis-Danckert N, Gunther AL, Kroke A, Hornberg C, Buyken AE. How early dietary factors modify the effect of rapid weight gain in infancy on subsequent body-composition development in term children whose birth weight was appropriate for gestational age. Am J Clin Nutr. 2007 Dec;86(6):1700-8. 175. Altman DG, Royston P. The cost of dichotomising continuous variables. BMJ. 2006;332(7549):1080. 176. Marshall JR, Zielezny MA. Fat and calories in the epidemiology of colon cancer in western New York. Prev Med. 1993 Sep;22(5):775-82. 177. Elmståhl S, Gullberg B. Bias in diet assessment methods--consequences of collinearity and measurement errors on power and observed relative risks. Int J Epidemiol. 1997 Oct;26(5):1071-9. 178. Alexy U, Sichert-Hellert W, Rode T, Kersting M. Convenience food in the diet of children and adolescents: consumption and composition. Br J Nutr. 2008 Feb;99(2):345-51. 179. Williams A, Pinnington LL. Nurses' knowledge of current guidelines for infant feeding and weaning. J Hum Nutr Diet. 2003 Apr;16(2):73-80. 180. Mowe M, Bosaeus I, Rasmussen HH, Kondrup J, Unosson M, Rothen-berg E, et al. Insufficient nutritional knowledge among health care workers? Clin Nutr. 2008 Apr;27(2):196-202. 181. Bramhagen AC, Axelsson I, Hallström I. Mothers' experiences of feeding situations - an interview study. J Clin Nurs. 2006 Jan;15(1):29-34. 182. Black AE. Critical evaluation of energy intake using the Goldberg cut-off for energy intake:basal metabolic rate. A practical guide to its calculation, use and limitations. Int J Obes Relat Metab Disord. 2000 Sep;24(9):1119-30. 183. Sichert-Hellert W, Kersting M, Schoch G. Underreporting of energy in-take in 1 to 18 year old German children and adolescents. Z Ernahrungswiss. 1998 Sep;37(3):242-51. 184. Lanigan JA, Wells JC, Lawson MS, Cole TJ, Lucas A. Number of days needed to assess energy and nutrient intake in infants and young children be-tween 6 months and 2 years of age. Eur J Clin Nutr. 2004 May;58(5):745-50. 185. Svahn JC, Axelsson IE, Raiha NC. Macronutrient and energy intakes in young children fed milk products containing different quantities and qualities of fat and protein. J Pediatr Gastroenterol Nutr. 1999 Sep;29(3):273-81.

References

67