European Journal of Clinical Nutrition (2004) 58, 10161021

& 2004 Nature Publishing Group All rights reserved 0954-3007/04 $30.00
www.nature.com/ejcn

ORIGINAL COMMUNICATION
Food habits in 1011-year-old children with functional gastrointestinal disorders
nen2, S Tikkanen1 and J Kokkonen1 M Haapalahti1,2*, H Mykka
1 Department of Pediatrics, University of Oulu, Oulu, Finland; and 2Department of Clinical Nutrition, University of Kuopio, Kuopio, Finland

Objective: To evaluate the role of food habits and nutrient intake in children with functional gastrointestinal disorders (FGIDs) considering their association with psychosocial factors. Design: Cross-sectional. Setting: A rural town in Mid-Western Finland. Subjects: After excluding organic gastrointestinal (GI) disorders, a total of 49 children with FGID and 78 control children without GI symptoms from a cohort of 422 children were studied. Methods: Food frequency questionnaire, GI-symptoms questionnaire and the Child Behaviour Check List (CBCL) filled in by parents together with their child, and 24-h dietary recall and anthropometric and haematological measurements. Results: Compared to control children, fewer children with FGID had daily family dinner (77 vs 91%, P 0.030) and they used less often vegetables (P 0.023), fruits (P 0.027) and berries (P 0.011), but more often ice cream (P 0.018) and soft drinks (P 0.027) and had a higher sucrose intake (9 vs 7E%, P 0.032) and lower lactose intake (27 vs 36 g, P 0.001). Reported food-related symptoms were more common among FGID group (69 vs 11%), as well as restricted milk use (31 vs 0%). Children with FGID had higher total problem scores in CBCL (P 0.002), and the behavioural/emotional problems associated with occurrence of FGID (OR 1.04, 95% CI 1.071.076) as did the sugar intake (OR 1.1, 95% CI 1.0041.215). Conclusions: This population-based study suggests that school-aged children with FGIDs have less-organized food habits and higher milk avoidance, and they display internalizing psychological characteristics. Sponsorship: Juho Vainio Foundation, Finland; the Alma and K.A. Snellman Foundation, Oulu, and the Pediatric Research Foundation, Finland.

European Journal of Clinical Nutrition (2004) 58, 10161021. doi:10.1038/sj.ejcn.1601925

Keywords: children; gastrointestinal; symptoms; abdominal pain; food habits; nutrient intake; psychosocial

Introduction
Functional gastrointestinal disorders (FGIDs) are defined as a combination of chronic or recurrent gastrointestinal (GI) symptoms in the absence of diagnosed disease. A number of common paediatric disorders falls into this category, for example, infant regurgitation, chronic nonspecific diarrhoea, irritable bowel syndrome (IBS), nonulcer dyspepsia,

*Correspondence: M Haapalahti, Department of Pediatrics, University of Oulu, PO Box 5000, FIN-90014 Oulu, Finland. E-mail: mila.haapalahti@uku.fi Guarantor: Docent Jorma Kokkonen. Contributors: MH was responsible for the data analysis and wrote the paper together with Professor HM and Docent JK. SH and JK were responsible for the clinical examinations. Received 11 April 2003; revised 29 August 2003; accepted 3 October 2003

infant dyschezia, and functional constipation (Hyams, 1999; Rasquin-Weber et al, 1999; Thompson et al, 1999). It is well known that psychopathology plays an important role in FGIDs. Gut function is controlled by a complex neural network and the current understanding is that FGIDs result from dysregulation of the bidirectional communication between the brain and the gut (ie the braingut axis), modulated by various psychosocial factors. Recently, instead of searching unidirectional relationship between psychosocial events and GI function, research has shifted to evaluate the reciprocal interaction between physiologic and psychosocial processes in these disorders (Drossman et al, 1999; Ringel and Drossman, 1999). More psychological problems and stressors have been found in children with recurrent abdominal pain (RAP) and in those with organic GI disorder as compared to GI patients without symptoms or healthy

Food habits in schoolchildren with FGIDs M Haapalahti et al

1017
control children (Sawyer et al, 1987; Walker and Greene, 1989; Robinson et al, 1990; Walker et al, 2001). There is also some evidence that FGIDs may be more common in children coming from low social class (Alfven, 1993b). Role of food habits in FGID is not clear. Many adults with IBS experience that diet, especially foods rich in carbohydrates and fat, triggers their GI symptoms (Ragnarsson and Bodemar, 1998; Jarrett et al, 2001; Simren et al, 2001). Similarly, many children with RAP have complained GI symptoms with meals (Alfven, 1993a). Irregular meal patterns may predispose to GI symptoms, producing irregular secretion of GI hormones, which may alter GI motility. One of the reasons behind unexplained GI symptoms may also be undiagnosed milk hypersensitivity (Kokkonen et al, 2001) or carbohydrate malabsorption (Hyams et al, 1988; Lifshitz et al, 1992). Undigested food may lead to food avoidance through nausea and other uncomfortable symptoms, and dyspeptic children with impaired gastric emptying have lower body weight for age (Papadopoulou et al, 2002), those with RAP a loss of appetite (Alfven, 1993a, b), and children with chronic constipation lower fibre and nutrient intakes than the control children (Morais et al, 1999; Roma et al, 1999). We postulate that children with FGIDs modify their food habits in manner, which may produce adverse nutritional effects and therefore food habits and health state of the children with FGIDs need to be evaluated in order to prevent nutritional disturbances and to establish a proper treatment of this entity. To our knowledge, it is not known whether these children have regular meal patterns and/or sufficient nutrient intake. The aim of the present population-based study was to determine the meal patterns, food use and nutrient intake of 1011-y-old children with FGIDs, considering the association of FGID with childrens emotional/ behavioural pattern and familys socioeconomic status.

Figure 1 Study population.

Subjects and methods

Participants and procedures A cohort of 422 children aged 1011 y living in the rural town of Ylivieska, located in mid-western Finland, were invited to participate in the study between September and October 1999. The protocol included three questionnaires (meal pattern and food use, GI symptoms and allergy, Child Behaviour Check List (CBCL), which were distributed by the school nurses to the children at the fourth and fifth grades in all 12 primary schools of the town. The children were asked to fill in the questionnaires at home together with their parents, and to return the questionnaires to the teachers, 404 (96%) returning (Figure 1). Of the 135 children having suffered from abdominal pain, persistent diarrhoea and/or constipation during the previous 2 y, 110 (81%) agreed to participate in the clinical examination, but complete information was available from only 100 subjects. The clinical examination included checking the GI symptoms during past 6 months by the medical officer (ST) and if so a

lactose load test, measurement of endomysium/transglutaminase and Helicobacter pylori antibodies. If symptoms were severe, a blind milk-protein challenge and/or gastro- and/or colonoscopy were undertaken. Based on this examination, a total of 26 children (24 with GI symptoms and two from the control group) had an organic GI disorder and were excluded. Of these, nine had milk allergy, eight had lactose intolerance, five had celiac disease, three had H. pylori infection, six had chronic constipation, and three had milk allergy with constipation and/or lactose intolerance. Of the remaining 76 in the symptomatic group, 27 children had symptoms less frequently than monthly and were excluded from further study. The 49 children (20 boys, 29 girls) having persistent abdominal pain at least once a month and/or persistent diarrhoea and/or constipation during the preceding 6 months were defined as the FGID group. Of these, three children had milk allergy in infancy, one child had diabetes mellitus and one epilepsy. Every third child with no GI symptoms (N 269) was selected to the control group attending same blood tests as those with GI symptoms. After exclusion of one child with celiac disease and one child with H. pylori infection, there were 78 children (44 boys, 34 girls) remaining in the control group. And of these, one child had epilepsy. The parents and the children were informed about the nature of the study and the parents signed an approval form. The Ethical Committee of the Oulu University Hospital accepted the study.

Dietary data Food habits were determined by the food frequency questionnaire, pre-tested at the Oulu University Hospital. It European Journal of Clinical Nutrition

Food habits in schoolchildren with FGIDs M Haapalahti et al

1018
consisted of 55 questions concerning on usual food use and meal patterns at home and school during weekdays. The meal patterns of the family dinner were canvassed using the statements: The whole family tends to eat together, We prepare a hot meal, and We tend to eat the same time of day. Food avoidance was asked by the question Does the child avoid any foodstuff now? Milk use was estimated as glasses, one glass being 1.7 dl of milk. The nutritional status of the children was determined by the 24-h dietary recall and based on relative body weight in the growth charts and biochemical measurements (haemoglobin, mean corpuscular volume, mean cell haemoglobin concentration, serum ferritin and serum pre-albumin). Data on food intake were calculated by the AIVO-program, version 2.10 (Aivo Finland Oy). The interviews were carried out on weekdays (Tuesday Friday), and the children were interviewed in the presence of one of their parents. Food picture booklets (Haapa et al, 1985) and household measures were used in estimating the portions, and information on the school lunch was collected directly from the school kitchen. Adequacy of the energy nutrients intake was assessed using the Nordic recommendations (fat r30E%, protein 1015E%, carbohydrates 55 60E%), and dietary fibre intake was considered sufficient if it was Z15 g/day (Williams et al, 1995). data (nutrient intakes, milk use, relative body weight, biochemical measurements, and emotional/behavioural parameters) independent samples t-test or MannWhitney U-test were used whenever appropriate. Factors associating with FGID were examined by calculating the adjusted OR using the logistic regression. Females, father/mother other than upper white-collar, lack of daily hot dinner, milk use, sugar intake, and total problem score were selected as covariates to the model. P-values o0.05 were regarded as significant.

Results
Background characteristics There was a tendency for female preponderance in FGID group (1.5 vs 0.7), approaching statistical significance between the groups (w2 2.9, P 0.087). When assessing the fathers and mothers occupation separately, no significant difference was observed between the FGID and control groups (w2 4.4, P 0.361 and w2 4.5, P 0.337), respectively, but after combining the occupation groups the difference was significant, 19% of FGID children vs 38% of the control children had upper white-collar father or mother (w2 4.7, P 0.029).

Socioeconomic status The socioeconomic group was determined using the standard occupational classification (Ammattiluokitus, 2001) as modified: (1) upper white-collar, (2) lower white-collar, (3) workers, (4) entrepreneurs, including farmers, (5) students, pensioners and other, including housewives. Fathers and mothers occupation associated strongly (w2 97.4, Po0.001) and were analysed separately and together.

Child Behaviour Checklist (CBCL) Since psychological factors are contributing to FGIDs, we used the CBCL, which is a standardized questionnaire, for screening the behavioural and emotional problems in the children during the preceding 6 months (Achenbach, 1991, 2001). We used the CBCL for children aged 418 y. It include an overall rating of emotional/behavioural problems, and ratings in two areas labelled as internalizing, externalizing characteristics and another category of characteristics including thought, social, and attention problems. The internalizing characteristics consist of withdrawal, somatic complaints and anxiousness/depression syndrome, and the externalizing characteristics of delinquent and aggression syndromes. Somatic complaints consist of following symptoms such as tiredness, dizziness, abdominal pain, headaches, and nausea.

GI symptoms Abdominal pain was the most common GI symptom in the FGID group: 59% had abdominal pain every month, but not weekly, 31% weekly and 2% had almost continuously. Before school age, 62% of the children with FGID and 4% of the control children had suffered from abdominal pain according the parents. Loose stools/diarrhoea was present in 25% FGID children, but no one reported constipation. Heartburn was reported by 33% children in the FGID group and by 12% of control children. Swelling of upper abdomen was reported by 7% of the FGID children. Skin rash was reported by 27% in the FGID and 12% in the control group. According to the parents, 69% of the FGID children and 11% of the control children had food-related symptoms, in FGID group mainly GI and in control group skin rash or oral symptoms.

Statistics Statistical analyses were performed using the SPSS 11.0. for Windows. Categorical variables (meal patterns, food use) were examined using w2-test or Fishers exact test. For continuous European Journal of Clinical Nutrition

Meal patterns and food use In Table 1 is seen that the children with FGID came more often from families having less-organized meal patterns during weekdays, preparing less often daily dinner (77 vs 91%, w2 4.7, P 0.030) and tended to eat less often together (46 vs 65%, w2 4.4, P 0.036). In addition, FGID children seemed to skip school lunch more often than the control children (14 vs 5%, w2 3.2, P 0.075). The FGID group more frequently ate ice-cream (23 vs 8%, w2 5.6, P 0.018), but vegetables (48 vs 68%, w2 5.2, P 0.023), fruits (55 vs 75%, w2 4.9, P 0.027), berries (17 vs 37%, w2 6.5, P 0.011) less frequently (Table 2). Daily use of sugar containing soft drinks was also more common among FGID group (15 vs 3%, Fishers exact test, P 0.027).

Food habits in schoolchildren with FGIDs M Haapalahti et al

1019
Table 1 Daily meal patterns during weekdays among the children with FGIDs and the control childrena FGID n 4748 (%) School lunch Main meal Salad Water or milk Bread Family dinner Hot dinner is prepared Family is eating together
a

Table 3 Intakes of energy nutrients and dietary fiber by the FGID children and control childrena FGID n 48 Control n 72 7.5 16 31 48 7 18 36 (1.9) (3.3) (6.7) (7.0) (4.2) (7.1) (16.5) P-valueb 0.261 0.107 0.937 0.529 0.037 0.263 0.001

Control n 7277 (%) 95 96 53 93 73 91 65

w2 3.2 3.0 1.9 1.9 2.7 4.7 4.4

P-value 0.075 0.085 0.164 0.172 0.098 0.030 0.036

86 88 40 85 86 77 46

Energy (MJ) Protein (E%) Fat (E%) Carbohydrate (E%) Sucrose (E%) Dietary fibre (g) Lactose (g)
a b

7.1 15 30 50 9 17 27

(2.0) (3.0) (7.7) (7.7) (4.8) (8.1) (15.0)

The values are given as mean (s.d.), E% energy percent. Independent samples T-test P-values, nonsignificant values are also tested by MannWhitney U-test.

Data collected by the food frequency questionnaire.

Table 2 Food use at home during weekdays by the children with FGIDs and the control childrena,b FGID n 4448 (%) Yogurt/sour milk Ice cream Beef/pork Cereals Vegetables Fruits Berries Sweet pastries/buns/ biscuits Candies/chocolate Fast food 64 23 40 63 48 55 17 52 28 4 Control n 7376 (%) 67 8 53 53 68 75 37 43 23 4

w2 0.1 5.6 2.0 1.0 5.2 4.9 6.5 0.8 0.4 0.011

P-value 0.710 0.018 0.156 0.317 0.023 0.027 0.011 0.346 0.544 0.918

a Values are given as percent of children who use the listed foods daily or almost daily. b Data collected by the food frequency questionnaire.

t 3.4 (df 107), P 0.001). Dietary fibre intake was similar between the groups, but 48% of the FGID and 35% of the control children did not reach the recommended minimum level of their fibre intake. Similarly, 14% of both the FGID and the control children used regularly vitamin or mineral, mainly calcium supplements. The nutritional status was similar in the groups and no significant differences were found in the relative body weights or haematological measurements. According the growth chart, 17% of the FGID and 11% of the control children were overweight by 20% or more, and 2% of the FGID and 4% of the control children were underweight by 15% or more. The haematological values (haemoglobin, mean corpuscular volume, and mean cell haemoglobin concentration) were within the normal range in both groups, except one child in FGID group who had a serum ferritin value o5 mg/l. One child with FGID and one control child had a prealbumin value below 0.10 g/l at the first measurement, but within the reference value at the second measurement.

Food avoidance was more common among the children with FGID (48 vs 11%, w2 21.9, Po0.001), cows milk being the most commonly avoided foodstuff. Compared to controls, a third (31% vs 0%) of the FGID children had restricted their use of dairy products, while 13% (vs 1%) of the FGID children reported that that they drink no milk and 17% (vs 0%) used low lactose milk. Milk consumption was lower among the FGID children (mean 5.7 vs 7.1 dl/day, t 2.4, df 122, P 0.020). Other foodstuffs than milk were also avoided, 6% (vs 5%) of FGID children avoided some fruits, 8% (vs 1%) some vegetables, and 10% (vs 4%) some other food.

Psychological factors (CBCL) The children with FGID had higher total problem (mean 23.6. vs 14.6, t 3.2 (df 86), P 0.002), internalizing (mean 7.9 vs 3.4, t 4.1 (df 79), Po0.001) and externalizing scores (mean 7.9 vs 3.4, t 2.1 (df 102), P 0.036) in CBCL than the control children. However, only 4% of the FGID children (vs 0% of the control children) had the total problem score exceeding the clinical borderline (score467), which indicates a need of more detailed assessment of the emotional/ behavioural problems. In specific problem items, the scores of somatic complaints exceeded the clinical borderline in the FGID group more frequently than in the control group (29 vs 4%, w2 16, Po0.001).

Nutrient intake and nutritional status Children with FGID had higher sucrose intake than the control children (9 vs 7E%, t 2.1 (df 91), P=0.037), even though it did not exceed the recommended maximum level 10% of total energy in either group (Table 3). No other significant differences in energy intake or energy nutrient intakes were observed, except that of lactose (27 vs 36 g,

Factors associated with FGID In the regression model, high total problem score in CBCL associated with the occurrence of the FGIDs (OR 1.04, 95% CI 1.0071.076), but also a high sugar intake increase the risk to belong to the FGID group (OR 1.1, 95% CI European Journal of Clinical Nutrition

Food habits in schoolchildren with FGIDs M Haapalahti et al

1020
Table 4 Factors associated to FGIDs in 1011-y-old children OR adjusted Female vs male Milk use (dl/day) Sugar intake (E%) No daily hot dinner Father or mother other than upper-white collar Total problem score in CBCL 1.6 0.9 1.1 2.1 2.5 1.04 95% CI 0.64.0 0.81.02 1.0041.215 0.77.3 0.96.7 1.0071.076 P-value 0.297 0.088 0.042 0.238 0.070 0.017

Logistic regression analysis was conducted.

1.0041.215) (Table 4). There was also a weak association between the low socioeconomic status of the family and the FGID.

Discussion
Our study population was a cohort of 1011-y-old children from a rural town in mid-western Finland and we suggest that the results could be generalized to schoolchildren living in Finnish rural towns. The merits to this study are the high rate of the questionnaires compliance, careful clinical examination of all study subjects and exclusion of definite GI diseases from the FGID group and control group. In addition, this population-based study gives a picture of FGID in relation to nutrition and psychosocial factors in general community, whereas most of the previous studies have had highly selected case material. We found lessorganized meal patterns with infrequent use of vegetables, fruits and berries, but frequent use of ice cream and soft drinks in the 1011 y-old schoolchildren with FGIDs. Restricted milk use was also common among this group. The FGID children had higher sucrose intake than the control children, and only a half of the children had the recommended intake of dietary fibre. Although these children appeared to be healthy in terms of having similar nutritional status as compared to the control children, they had higher psychological problem scores. In the regression model, behavioural problems associated significantly with functional GI complaints. Nearly two-thirds (69%) of the children with FGID had food-related, mainly GI symptoms. Cows milk was restricted by a third of children in the FGID group, but only one in 10 was avoiding milk as drink. In a recent study, GI and dermal symptoms were shown to be common reasons for food avoidance among children referred to outpatient clinic, cows milk being also the most commonly avoided (de Vries et al, 2001). Our recent studies have shown a delayed type of immunological reactivity occurring against milk proteins in the children with abdominal pain (Kokkonen et al, 2001). This clinical entity may be one of the main reasons behind RAP and other GI discomfort in children and should be considered when examining children with functional GI disorders. Although milk hypersensitivity was tested for using a blinded milk challenge, we probably could not European Journal of Clinical Nutrition

identify all those reacting to milk protein within a challenge lasting only one week. It is possible that also other food hypersensitivities (fruits, vegetables, cereals) or even sorbitol/fructose malabsorption may occur among these children (Hyams et al, 1988). Our study suggests that the degree of food avoidance and GI symptoms in relation to food use among children with unexplained GI symptoms needs to be evaluated in more detail. We found that lack of organized meal patterns in the family associates with the occurrence of functional GI disorders in children. The families having a child with FGID tended not to eat together or prepare hot dinner regularly, and at school, the FGID children seemed also to skip free lunch more frequently, which is offered every day to Finnish schoolchildren and adolescents. Less-organized meal patterns may force the child to replace proper meals by snacks and this could explain frequent use of ice cream and soft drinks among the FGID children. A recent study showed that healthy food choices associate with regular family dinner in children and adolescents (Gillman et al, 2000) and lack of proper meals at home may increase the risk for imbalanced diet and even inadequate nutrient intake. In our study, intake of energy nutrients was within recommendations in both groups, even if sucrose intake was higher in the FGID group. Based on our findings on food use, it is, however, possible that the intake of micronutrients by the FGID children is lower than by the control children. Irregular meal patterns in family may reflect irregular lifestyle in general or other family problems (marital, unemployment, or illness). Similarly, behavioural problems of the children with FGID may reflect problems in familys life. In our previous study on the same cohort, we found that children without regular family dinner had higher level of psychological problems according CBCL than the children a more regular pattern (Haapalahti et al, 2003). In the present study, children with FGID had higher total problem scores and tendency to internalize shown as higher somatic complaints and anxiousness/depression scores than the children without GI symptoms. However, nearly all scores fell within the normal range, as did also the CBCL scores in another study on children with RAP (Stordal et al, 2001). Stressful life events have been shown to associate with functional GI disorders in children and in adolescents (Aro, 1987; Robinson et al, 1990; Walker et al, 1994, 2001). Higher rates of anxiety, depression, and somatic complaints are also found in patients with organic GI problems than in control patients (Walker and Greene, 1989). Somatic nature of the reactions to stressors has been suggested to be a factor that differentiates the children with RAP from healthy controls (Walker et al, 2001). In conclusion, this population-based cross-sectional study suggests that together with psychosocial factors, eating habits within the family associate with FGID. However, more long-term studies on the relationship of eating habits with fluctuation of the functional GI disorders are needed to develop better understanding and treatment of FGID problems.

Food habits in schoolchildren with FGIDs M Haapalahti et al

1021
Acknowledgements This study was supported by the the Juho Vainio Foundation, the Alma and K.A. Snellman Foundation, Oulu, and the Pediatric Research Foundation, Finland. References
Ammattiluokitus 2001: (The classification of the occupations, 2001, in Finnish). Ka sikirjoja 14. Helsinki: Tilastokeskus. Achenbach TM (1991): Manual for the Child Behaviour Checklist/4-18 &1991 Profile. Burlington, VT: University of Vermont. Achenbach TM (2001): Manual for the ASEBA School-Age Forms & Profiles. Burlington, VT: University of Vermont, Research Center for Children, Youth, & Families. Alfven G (1993a): Preliminary findings on increased muscle tension and tenderness, and recurrent abdominal pain in children. A clinical study. Acta Paediatr. 82, 400403. Alfven G (1993b): The covariation of common psychosomatic symptoms among children from socio-economically differing residential areas. An epidemiological study. Acta Paediatr. 82, 484487. Aro H (1987): Life stress and psychosomatic symptoms among 14 to 16-year old Finnish adolescents. Psychol. Med. 17, 191201. de Vries TW, Wierdsma N, van Ede J & Heijmans HS (2001): Dieting in children referred to the paediatric outpatient clinic. Eur. J. Pediatr. 160, 595598. Drossman DA, Creed FH, Olden KW, Svedlund J, Toner BB & Whitehead WE (1999): Psychosocial aspects of the functional gastrointestinal disorders. Gut 45(Suppl 3), S25S30. Gillman MW, Rifas-Shiman SL, Frazier AL, Rockett HR, Camargo Jr CA, Field AE, Berkey CS & Colditz GA (2000): Family dinner and diet quality among older children and adolescents. Arch. Fam. Med. 9, 235240. sa nen L (1985): Picture Booklet. Haapa E, Toponen T, Pietinen P & Ra Helsinki: National Public Health Institute. Haapalahti M, Mykkanen H, Tikkanen S & Kokkonen J (2003): Meal patterns and food use in 10- to 11-year-old Finnish children. Public Health Nutr. 6, 365370. Hyams JS (1999): Functional gastrointestinal disorders. Curr. Opin. Pediatr. 11, 375378. Hyams JS, Etienne NL, Leichtner AM & Theuer RC (1988): Carbohydrate malabsorption following fruit juice ingestion in young children. Pediatrics 82, 6468. Jarrett M, Visser R & Heitkemper M (2001): Diet triggers symptoms in women with irritable bowel syndrome. The patients perspective. Gastroenterol. Nurs. 24, 246252. Kokkonen J, Ruuska T, Karttunen TJ & Niinimaki A (2001): Mucosal pathology of the foregut associated with food allergy and recurrent abdominal pains in children. Acta Paediatr. 90, 1621. Lifshitz F, Ament ME, Kleinman RE, Klish W, Lebenthal E, Perman J & Udall Jr JN (1992): Role of juice carbohydrate malabsorption in chronic nonspecific diarrhea in children. J. Pediatr. 120, 825829. Morais MB, Vitolo MR, Aguirre AN & Fagundes-Neto U (1999): Measurement of low dietary fiber intake as a risk factor for chronic constipation in children. J. Pediatr. Gastroenterol. Nutr. 29, 132135. Papadopoulou A, Vazeou A, Siafakas C, Papadimitriou A & Bartsocas C (2002): Nutritional implications of chronic dyspepsia in childhood. Dig. Dis. Sci. 47, 3237. Ragnarsson G & Bodemar G (1998): Pain is temporally related to eating but not to defaecation in the irritable bowel syndrome (IBS). Patients description of diarrhea, constipation and symptom variation during a prospective 6-week study. Eur. J. Gastroenterol. Hepatol. 10, 415421. Rasquin-Weber A, Hyman PE, Cucchiara S, Fleisher DR, Hyams JS, Milla PJ & Staiano A (1999): Childhood functional gastrointestinal disorders. Gut 45(Suppl 2), S60S68. Ringel Y & Drossman DA (1999): From gut to brain and backa new perspective into functional gastrointestinal disorders. J. Psychosom. Res. 47, 205210. Robinson JO, Alverez JH & Dodge JA (1990): Life events and family history in children with recurrent abdominal pain. J. Psychosom. Res. 34, 171181. Roma E, Adamidis D, Nikolara R, Constantopoulos A & Messaritakis J (1999): Diet and chronic constipation in children: the role of fiber. J. Pediatr. Gastroenterol. Nutr. 28, 169174. Sawyer MG, Davidson GP, Goodwin D & Crettenden AD (1987): Recurrent abdominal pain in childhood. Relationship to psychological adjustment of children and families: a preliminary study. Aust. Paediatr. J. 23, 121124. Simren M, Mansson A, Langkilde AM, Svedlund J, Abrahamsson H, Bengtsson U & Bjornsson ES (2001): Food-related gastrointestinal symptoms in the irritable bowel syndrome. Digestion 63, 108115. Stordal K, Nygaard EA & Bentsen B (2001): Organic abnormalities in recurrent abdominal pain in children. Acta. Paediatr. 90, 638642. Thompson WG, Longstreth GF, Drossman DA, Heaton KW, Irvine EJ & Muller-Lissner SA (1999): Functional bowel disorders and functional abdominal pain. Gut 45(Suppl, 2), S43S47. Walker LS, Garber J & Greene JW (1994): Somatic complaints in pediatric patients: a prospective study of the role of negative life events, child social and academic competence, and parental somatic symptoms. J. Consult. Clin. Psychol. 62, 12131221. Walker LS, Garber J, Smith CA, Van Slyke DA & Claar RL (2001): The relation of daily stressors to somatic and emotional symptoms in children with and without recurrent abdominal pain. J. Consult. Clin. Psychol. 69, 8591. Walker LS & Greene JW (1989): Children with recurrent abdominal pain and their parents: more somatic complaints, anxiety, and depression than other patient families? J. Pediatr. Psychol. 14, 231243. Williams CL, Bollella M & Wynder EL (1995): A new recommendation for dietary fiber in childhood. Pediatrics 96, 985988.

European Journal of Clinical Nutrition