234 Archives of Disease in Childhood 1994; 70: 234-236 Eating habits and attitudes of mothers of children

with non-organic failure to thrive Arch Dis Child: first published as 10.1136/adc.70.3.234 on 1 March 1994. Downloaded from

Jacinta B McCann, Alan Stein, Christopher G Fairburn, David B Dunger

Abstract sidered whether particular diets (for example, a The eating habits and attitudes concern- macrobiotic diet), adversely influence the ing body shape and weight among 26 growth of young children.9 10 In one case mothers of children with non-organic series consisting of seven children, parental failure to thrive (the index group) were health beliefs appeared to lead to restriction of studied using the eating disorder exami- children's food intake."I nation. They were compared with equiva- Two case series have reported that the lent data on 26 individually matched children of mothers with eating disorders women who participated in a large com- (anorexia nervosa or bulimia nervosa) have munity survey. The index mothers' views feeding difficulties or poor weight gain.12 13 In oftheir child's weight and shape were also one, some mothers with bulimia nervosa studied. admitted to slimming their babies. 14 In a The principal findings were, firstly, that Danish retrospective study of mothers with when compared with the comparison anorexia nervosa a significant number recalled group, mothers of children with non- that their children had growth and/or feeding organic failure to thrive had higher levels difficulties in the first year of life.15 These of dietary restraint. Secondly, despite findings raise the question as to whether the their child's low weight, 50% of the index presence of eating disorder psychopathology in mothers were restricting their child's mothers may be an aetiological factor in the intake of 'sweet' foods, and a further 30% development of non-organic failure to thrive in were restricting foods they considered their children. 'fattening' or 'unhealthy'. The aim of this study was to determine These results raise the question of whether disturbed eating habits and attitudes whether maternal eating habits and atti- to body shape, weight, and food are more tudes have a causal role in the genesis of common among mothers of children with non- non-organic failure to thrive. They organic failure to thrive than among mothers suggest that carefui inquiry about the from the general population. http://adc.bmj.com/ mothers' eating habits and attitudes is needed when assessing children with non-organic failure to thrive. Subjects and methods (Arch Dis Child 1994; 70: 234-236) SUBJECTS Twenty six children with non-organic failure to thrive were identified from the patient lists of

Failure to thrive in young children accounts for two Oxfordshire consultant paediatricians (the on September 30, 2021 by guest. Protected copyright. between 1% and 5% of paediatric admissions index group) having been referred by family to hospital.' Thirty to forty per cent of such doctors in the previous three years. The criteria children are given the diagnosis of 'non- for entry into the study were that the child's organic failure to thrive', with no sufficient weight at referral was at, or below, the third physiological cause being found.2 population centile.16 17 In addition, the study Section of Child and The aetiology of non-organic failure to was restricted to singleton children whose birth Adolescent Psychiatry, thrive has received much attention over recent weight was above 2500 g, with a gestation of Oxford University Department of years. It has sometimes been viewed as result- greater than 37 weeks. Psychiatry, Park ing from emotional deprivation,3 4 although Hospital for Children, this concept has increasingly been called into Old Road, Headington, ASSESSMENT Oxford OX3 7LQ question. The presence of adverse environ- Jacinta B McCann mental factors influencing non-organic failure Subjects were assessed by interview which Alan Stein to thrive has also been observed.5 Another comprised of two parts. The first consisted of from the that when an examination of the mothers' eating habits Oxford University viewpoint stems finding Department of children present with non-organic failure to and attitudes, using the eating disorder exami- Psychiatry, Warneford thrive, their nutritional intake appears to be nation. This is a validated investigator based Hospital, Oxford authors take the view that interview which assesses the clinical features of Fairbum deficient.6 Many Christopher G whatever the primary aetiology, inadequate eating disorders.18 Five subscales may be Department of energy intake is the final common pathway of derived from its ratings, the key ones being Paediatrics, John the growth failure in non-organic failure to eating restraint, shape concern, and weight Radcliffe Hospital, disorder examination also Oxford thrive.7 8 Some studies have focused on how concern. The eating David B Dunger the nutritional intake of children with non- generates operational Diagnostic and Statistical failure to thrive becomes inadequate, Manual of Mental Disorders, third edition, Correspondence to: organic Dr McCann. although no systematic research has been revised (DSM-III-R) eating disorder Accepted 3 November 1993 undertaken. In a Dutch study, researchers con- diagnoses. Eating habits and attitudes ofmothers ofchildren with non-organicfailure to thrive 235

For comparison purposes, equivalent data concern; 31% (eight mothers) were relieved were extracted on an individually matched that their child's difficulties with poor weight

subgroup of mothers who participated in a gain were being acknowledged. The time Arch Dis Child: first published as 10.1136/adc.70.3.234 on 1 March 1994. Downloaded from large community based study ofthe prevalence interval between the child's poor weight gain of eating disorder features among women being first noted by the primary health care (C G Fairburn et al, unpublished). Each index team and subsequent referral ranged from one mother was paired with the mother who month to three years. The mothers who had to matched her most closely on the following wait longer for referral became particularly criteria: (i) number of children, (ii) social class, concerned about their child's poor weight gain. (iii) maternal age. Sixty one per cent (16 mothers) reported The second part of the interview addressed blaming themselves for their child's poor the following topics: (A) The events leading up weight gain, feeling that they had not done suf- to the referral to the paediatrician. (B) ficient to ensure adequate weight gain. They Attitudes and feelings of the mother towards recalled feeling angry towards their child, as the referral and the follow up procedure. (C) they had devoted much effort in attempting to The mother's views on her child's weight and increase their child's weight without success. shape: for example, they were asked specifi- After referral, 84% (22 mothers) believed that cally whether they considered their child of the greatest benefit was being reassured that normal weight and if not, what they felt the there was no underlying organic cause. Sixteen degree of underweight might be. per cent (four mothers) remained worried that Mann-Whimey U tests were carried out to 'something had been missed'. Some mothers compare the index and comparison mothers' felt that being asked to attend for follow up eating disorder examination scores. The paired t appointments at the hospital undermined the test was used to compare change between refer- previous reassurance given. ral and discharge measures of the childrens' weight deficit for height. MOTHERS ATTITUDES TO CHILD'S WEIGHT AND SHAPE Results Despite each child's weight being at the third CHARACTERISTICS OF THE SAMPLE centile or below, 58% (15 mothers) believed Of the 26 children identified, 10 were boys, 16 their child to be of normal weight or just were girls and their mean (SD) age was 3-8 slightly underweight. Thirty eight per cent (10 (2 0) years (range 0-9-9-6). The mean age of mothers) perceived their child's shape as the index mothers was 30 1 (4 0) years (range normal, with a further 34% (nine mothers) 24 3-38 8), while that of the comparison stating that their child was only slightly thin. group was 29-7 (3 5) years (range 24-1-38-0). Thirty eight per cent (10 mothers) described There was no significant difference between their child being a somewhat finicky eater, and

the body mass index of the two groups (mean a further 23% (six mothers) reported that their http://adc.bmj.com/ (SD) 23-78 (3 30) in the index group v 23-87 child was extremely finicky. A wide range of (4.26) in the comparison group). foods were reported as being refused, particu- All birth weights were at or above the 10th larly foods that had a 'coarse' texture. As a centile (mean 25th centile, range lOth-75th result of their child's finicky eating, mothers centile). From birth to the referral, all demon- described mealtimes being mostly tense, as strated deteriorating growth and had crossed they 'battled' over food intake with their child.

major centile lines. At referral, the mean Fifty per cent (13 mothers) stated that they on September 30, 2021 by guest. Protected copyright. weight deficit for height was 22-3% (range restricted their child's intake of 'sweet' foods 100/o-32%). All children were investigated and such as sweets, chocolate, and biscuits; 30% none were found to have an organic cause for (eight mothers) admitted to restricting other their failure to thrive. types of food (for example, fried foods, meat, nuts), as they considered them to be unhealthly and possibly fattening. After refer- EATING HABITS AND ATTITUDES OF THE ral, 54% (14 mothers) attempted to increase MOTHERS their child's food intake by having more fre- None of the mothers in either group fulfilled quent meals and increasing the quantity of DSM-III-R diagnostic criteria for anorexia food given. nervosa or bulimia nervosa. However, the index group scored significantly higher on the restraint subscale than the comparison group MANAGEMENT (index mean 1 78, comparison mean 0-56, All families received a combination of dietary p<0001). The restraint subscale assesses advice, together with feeding and psychologi- attempts to restrict food intake to influence cal management, on an outpatient basis and a shape or weight. There was no significant dif- few cases required social services involvement. ference between the two groups on the other None were admitted to hospital for treatment. subscales. The parents were encouraged to discuss in detail the difficulties they were having in feeding their child and how this made them MOTHER'S ATITIUDE TO THE PAEDIATRIC feel. This provided an opportunity for staff to REFERRAL explore the parents own attitudes to food, Sixty nine per cent (18 mothers) reported their shape, and weight and to examine further how predominant feelings were those of worry and the parent's attitudes influenced adversely the 236 McCann, Stein, Fairburn, Dunger

management of their child's feeding difficul- relatively small sample size, the cross sectional ties. Depending on the nature and severity of nature of the study, and the lack of compara-

the problem, parents and their child were seen tive data on the children in the comparison Arch Dis Child: first published as 10.1136/adc.70.3.234 on 1 March 1994. Downloaded from monthly and the duration of the intervention group, limits the conclusions that can be ranged from three months to two years. drawn. However, were further studies to impli- Parents of the children of 'finicky' eaters cate a causal role for maternal eating habits found that clarification of their child's feeding and attitudes in the genesis of non-organic difficulties, together with the support and failure to thrive, this would have important advice regarding firm, consistent management implications for management. of their child, led to an increase in energy Given the study's findings, further research intake and a general improvement in mealtime is indicated on the eating habits and attitudes behaviour. of the mothers of children with non-organic In five cases, assessment of the patient's failure to thrive. From a clinical perspective, energy intake from weekly recordings revealed assessment of parents' eating habits and atti- that their intake was grossly inadequate. In tudes would seem warranted when assessing four of these cases, psychosocial difficulties such children. When abnormalities are found, were evident, including marked parenting diffi- management needs to address the parents' eat- culties, relationship problems between parents, ing habits and attitudes. depression maternal and anxiety. In one of Jacinta McCann was supported by a Wellcome Trust research the five cases, both parents were vegans. They registrar post, Alan Stein is a Wellcome Trust lecturer, and subsequently Christopher Fairburn, is a Wellcome Trust senior lecturer. placed their child on a vegan The community study from which the comparison data were diet, which proved nutritionally deficient. drawn was funded by a programme grant from the Wellcome These five cases appeared to respond best Trust. We are grateful to the families who participated in the to study and to Dr Harvey Marcovitch who assisted in the recruit- treatment. ment of the sample. All the children put on weight and, at discharge from the outpatient clinic, their Berwick DM, Levy JC, Kleinerman R. Failure to thrive: diagnostic yield or hospitalisation. Arch Dis Child 1982; mean centile was at the 5th centile (range 57:347-51. 3rd-25th centile), with their mean weight 2 Hannaway PJ. Failure to thrive. A study of 100 infants and children. Clin Pediatr (Phila) 1970; 9: 96-9. deficit for height being 11 2% (range 3 Patton RG, Gardner L. Growth failure in maternal depriva- 30/6-20%), which was significantly improved tion. Springfield, Illinios: Charles C Thomas, 1963. compared with referral measures 4 Fischoff J, Whitten C, Pettit M. A psychiatric study of (t=4-35, mothers of infants with growth failure secondary to mater- p<0oOOl). nal deprivation. JPediatr 1971; 79: 209-15. 5 Homer C, Ludwig S. Categorisation of aetiology of failure to thrive. AmJDis Child 1978; 132: 967-9. 6 Russell Davis D, Apley J, Fill G, Grimwaldi C. Diet and Discussion retarded growth. BMJ 1978; i: 539-42. 7 Whitten CF, Pettit MG, Fischoff J. Evidence that growth The main finding of this study is that, although failure from maternal deprivation is secondary to under- mothers of children with non-organic eating. JAMA 1969; 209: 1675-82. failure to 8 Skuse DH. Non-organic failure to thrive: a reappraisal. Arch thrive were not suffering from clinical eating Dis Child 1985; 60: 173-8. http://adc.bmj.com/ disorders, 9 Dagnelie PC, Staveren WA van, Vergote FJVRA, Burema J. they did score significantly higher on Nutritional status of infants aged 4 to 18 months on the restraint subscale of the eating disorder macrobiotic diets and matched control infants: a popula- tion-based mixed-longitudinal study. Eur J Clin Nutr examination. Another significant finding was 1989; 43: 325-38. that, despite their children having been 10 Dagnelie PC, Staveren WA van, KlaverenJD van, Burema J. referred to paediatricians for poor weight Do children on macrobiotic diets show catch-up growth? A gain, population-based cross-sectional study in children aged half the mothers reported restricting moder- 0-8 years. EurJtClin Nutr 1988; 42: 1007-16. 11 Pugliese MT, Weyman-Daum M, Moses N, Lifshitz F. ately (that is, more than four days per week) Parental health beliefs as a cause of non-organic failure to on September 30, 2021 by guest. Protected copyright. their child's intake of sweet foods, and 30% thrive. Pediatrics 1987; 80: 175-81. 12 Van Weizel-Meijler G, Wit JM. The offspring of mothers described restricting foods considered to be with anorexia nervosa: a high-risk group for under fattening or unhealthy. These findings are nutrition and stunting? Eur J Pediatr 1989; 149: 130-5. 13 Stein A, Fairburn C. Children of mothers with bulimia ner- striking as one would expect that the mothers vosa. BMJ7 1989; 229: 777-8. of children with non-organic failure to thrive 14 Lacey JH, Smith G. Bulimia nervosa: the impact of pregnancy on mother and baby. BrJ7Psychiatry 1987; 150: would be attempting to increase their child's 777-81. food intake rather than restrict it. These results 15 Brinch M, Isager T, Tolstrup K. Anorexia nervosa and motherhood: reproduction pattern and mothering could be explained as arising from the belief behaviour of 50 women. Acta Psychiatr Scand 1988; 77: that sweet or fattening foods are unhealthy and 611-7. 16 Tanner JM, Whitehouse RH, Takaishi M. Standards from therefore need to be restricted. Another possi- birth to maturity for height, weight, height velocity, and bility is that the high level of dietary restraint in weight velocity: British children, 1965. Part I. Arch Dis Child 1966; 41: 454-71. the mother led her to restrict the child's food 17 Tanner JM, Whitehouse RH, Takaishi M. Standards from intake as well. birth to maturity for height, weight, height velocity, and weight velocity: British children, 1965. Part II. Arch Dis The findings of this study raise the question Child 1966; 41: 613-35. of whether a causal link exists between the 18 Cooper Z, Fairburn C. The eating disorder examination: a semi-structured interview for the assessment of the index mothers' own dietary restraint and the specific psychopathology of eating disorders. International poor growth of their child. Clearly, the J7ournal of Eating Disorders 1987; 6: 1-8.