Article Text


Psychiatry and psychology/community

Statistics from


H. E. Miller, R. J. McCarter, A. L. Curran, L. Hunt, A. Brownsdon, P. M. Sharples, the Kids Head Injury Study Group. Frenchay Hospital, Bristol

Introduction: TBI is a major cause of hospital admission in childhood. TBI is recognised to result in attentional problems, yet surprisingly few data exist concerning their nature and determinants, especially early after TBI.

Aims: (1) To define attentional status in TBI children admitted to hospital at 1 month after injury compared with normal uninjured controls; (2) to explore the factors related to attentional status.

Methods: Prospective cohort study. TBI was classed by Glasgow Coma Score (GCS) as severe (GCS 3–8), moderate (GCS 9–12), or mild (GCS 13–15). Attentional status was assessed using the Test of Everyday Attention for Children (TEA-Ch). Psychological response was assessed by the Birleson Depression Scale (BDS), Impact of Events Scale (IES), Revised Children’s Manifest Anxiety Scale (RCMAS), and Child Behaviour Checklist (CBCL). Maternal depressive symptomatology was assessed by the Beck Depression Inventory-II (BDI-II).

Results: 88 TBI subjects (51 severe/moderate; 37 mild) and 47 controls were recruited. There was no significant difference in age between TBI (mean 13.1 years) and controls (mean 13.2 years) (p 0.877). There was a significant difference between severe/moderate TBI children and controls but not mild TBI and controls on selective attention (Map Mission, p<0.001), attentional switching (Opposite Opposite Worlds, p<0.001), and divided attention (Sky Search DT, p<0.05). There was a significant relationship between attentional status and problematic behaviour in the child (CBCL externalising score, p 0.005; CBCL total problem score, p 0.004) but not between attentional status and symptoms of depression or anxiety (BDS, p 0.36, IES, p 0.74; RCMAS, p 0.55). There was a significant correlation between the child’s attentional status and maternal depressive symptomatology (p 0.008).

Conclusion: TBI produces significant problems across a range of attentional domains. Attentional difficulties correlate with injury severity and problematic behaviour in the child, but not with depression or anxiety, in contrast to the pattern seen in adults. The positive correlation between attentional status and depressive symptomatology in the mother suggests that further research should explore the impact of improved maternal support on the child’s cognitive outcome.


C. Macpherson, M. C. Stewart, F. Alderdice, B. G. McClure. Royal Maternity Hospital, Belfast and Department of Child Health, Queen’s University, Belfast

Introduction: Parenting starts with the unborn child and increasing emphasis is now placed on recognising potential problems in the antenatal period.

Aims: To ascertain the prevalence of unwanted pregnancy among a sample of women and their partners attending for routine antenatal care and examine the associations between parental–fetal attachment and pregnancy intention.

Methods: A prospective cohort study. Structured interviews were carried out with mothers and fathers in the first and third trimesters of pregnancy. Questionnaires were designed to incorporate validated items relating to pregnancy intention and the Antenatal Emotional Attachment Questionnaire (AEAQ).1

Results: A total of 100 women and 67 men completed the AEAQ. Twenty six women wanted to be pregnant sooner, 34 to be pregnant then, 27 to be pregnant later, and 13 had unwanted pregnancies. Women with unwanted pregnancies had significantly lower attachment scores in the third trimester of pregnancy than women in the other groups. The mean global attachment score among women with unwanted pregnancies was 75.5, compared with 82.7 for women in the later group, 83.3 for those in the then group and 80.2 for those who wanted to be pregnant sooner (p<0.005). Women with unwanted pregnancies had particularly lower quality of attachment scores. Paternal–fetal attachment was not significantly related to male pregnancy intention. Comparable paternal–fetal attachment scores were significantly lower than maternal scores (p<0.0001).

Conclusion: Unwanted pregnancy is associated with poorer emotional attachment to the unborn child and should be recognised as an important potential risk factor in the antenatal period.



D. Skuse, J. Gilmour, M. Place, W. Mandy, K. Puura. Behavioural and Brain Sciences Unit, Institute of Child Health, 30 Guilford Street, London WC1N 1EH

Aims: Our objective was to evaluate, using novel dimensional measures of psychopathology in both clinical and normal populations of children, the boundaries of the autistic spectrum. We thus aimed to clarify the extent and severity of autistic symptoms in normal children and those presenting clinically with other conditions, from conduct disorder to depression.

Methods: We employed novel computerised interview and neuropsychological methods to assess miscellaneous samples of clinical and non-clinically referred children (total n 470). Most had normal intelligence (90% IQ >70). Complex validated algorithms generated dimensional scores of impairment from interview material, permitting us to evaluate the severity of autistic symptoms in autism, normal children, and in comorbid conditions.

Results: Social communication and language skills were highly correlated throughout the sample (r2 0.84; p<0.001). Severe impairment was found in many children whose clinical diagnosis was not autism, but primarily, for example, conduct disorder, ADHD, and Tourette syndrome. Stereotyped/repetitive behaviours correlated weakly with social/language disability (r2 0.38), and did not correlate with IQ (r2 0.00).

Conclusions: Significant autistic features of behaviour (primarily those impairing social relationships and language) are often found in association with common psychiatric disorders, in children of normal intelligence. The clinical management of many psychiatrically disturbed children demands these comorbid features of their condition are recognised and appropriately treated.


D. Nicholls, C. Edwards, L. Cooke, R. Viner, J. Wardle. Institute of Child Health and University College London

Aims: The best evidence for successful treatment of childhood obesity comes from the family based behavioural treatment (FBBT) of Epstein et al. However, studies have been limited to self selected families in one treatment setting in the USA. We evaluated a pilot study of an adaptation of FBBT for obese children from a range of ethnic and social backgrounds in the UK.

Methods: Four pilot treatment groups of 8–9 children received 12 weekly sessions. FBBT is based on learning theory and utilises behaviour modification principles to increase healthy eating and activity behaviours. At least one parent attended with each child. Weight and height were measured at assessment and week 12. Fasting insulin and glucose were measured before commencement. As there is uncertainty regarding expression of overweight status, we expressed outcome as change in BMI and BMI z score and weight for height (percentage of BMI median centile for age, sex, and height).

Results: Thirty three children (23 female, 10 male) aged 8 to 13 years entered and 27 completed. Twenty two were white, 6 black, 3 Asian, and 2 of mixed ethnicity. Mean BMI z score at assessment was 3.4 (SD 0.6). Mean change in weight for height for completers was a loss of 9% (8) of overweight. Mean loss of BMI in the whole group was 1.25 kg/m2 (1.2) with mean loss of BMI z score of 0.2 (0.16). Mean loss of BMI in boys was 1.6 (0.7) compared with 1.1 (1.4) in girls (not significant). Loss of BMI was not related to age, initial BMI, or maternal BMI. There was an inverse relationship between fasting insulin and reduction in BMI z score (r −0.4, p 0.04).

Conclusions: These data confirm that loss of BMI can be achieved through an FBBT programme in children from diverse ethnic and social backgrounds. Preliminary evidence suggests that further work is needed to examine whether genetic or metabolic factors may influence potential for weight loss. Further studies of efficacy and effectiveness are required.


D. W. Beverley, B. Wright, B. Ashby, J. Jordan, C. Willia. York Hospital and Lime Trees Child and Adolescent Psychiatry Unit

Introduction: The diagnosis and treatment of children with chronic fatigue syndrome (CFS) is controversial. We have conducted a pilot randomised controlled trial in 13 children (age 8–16; 5 of whom were boys) to determine optimum management and the feasibility of undertaking a randomised trial.

Methods: A paediatrician saw the children, and the diagnosis of CFS according to the Oxford criteria was made. They were randomly assigned to an active rehabilitation programme with family based CBT (7 patients), or to “pacing” (activity within tolerance only). Therapists from Lime Tress child and adolescent mental health service (CAMHS) undertook baseline assessments, and supervised treatment according to manuals. The children were offered appointments to see the therapists at weekly intervals initially, decreasing to every 4 weeks as therapy progressed. Throughout the year’s therapy the children were offered appointments to see a paediatrician every 12 weeks. At 1 year repeat assessments were made.

Results: The children in the active rehabilitation programme had significantly improved Global Health Questionnaire scores. (4.2 at start, 2.2 at end) compared with pacing (4.2 start, 4.0 end) (p<0.01). A measure of physical activity as assessed by walking and climbing stairs was also significantly improved (p<05). There were also improvements in school attendance at the end of treatment (p 0.057) and at 6 months post treatment (p<0.05), suggesting therapy may have ongoing benefits. There were trends towards improvement for the Young Persons Ability scores, fatigue scores, depression scores, and 4 other physical activity scores though not all of these scores reached a significance of (p<0.05).

Conclusion: Active rehabilitation with family based CBT is a successful treatment for children with CFS. It is possible to conduct randomised controlled trials in CFS, and a larger study would confirm these findings.


B. Mulley, M. C. J. Rudolf, J. Chalkley. Belmont House Growth and Nutrition Team, Belmont House, East Leeds PCT, Leeds; Avon and Wiltshire Mental Health Partnership, NHS Trust, Bath

Aim: It is often hard to audit the outcome for families referred when a child has psychosocial or emotional/behavioural problems. Failure to thrive and eating difficulties are particular examples where families report that communication between themselves and health professionals is poor. Personal questionnaires were developed in the 1960s as a clinical and research tool for use in adult mental health. We decided to explore their use in auditing our work on children with failure to thrive and/or eating difficulties.

Methods: At the first appointment, a personal questionnaire was developed by working with the parent to document their individual concerns in eight areas related to eating difficulties and failure to thrive. At the following visit the parent verified the questionnaire and rated each concern on a four point scale. The questionnaire was then repeated at subsequent visits. An anonymous postal survey was sent out at follow up to assess parents’ views about the process.

Results: Thirty four personal questionnaires were completed. In 30 the overall score was in the high, clinically concerning range. This dropped so only 18 had high scores at next administration and 4 by the latest. A total of 348 concerns were itemised. Over time there was a 75% reduction in concerns relating to mealtime stress and a 50% fall in concerns about type of food eaten. The anonymous postal survey (22/30 returned) indicated a positive response to the use of personal questionnaires. Parents reported that the process of compiling the questionnaire made them feel listened to and increased their own clarity about what the concerns were. They felt that it meant that their own specific concerns were focused upon, and improvement over time became evident to them as well as the health professional.

Conclusions: Personal questionnaires present a new way of working with families with psychosocial and emotional/behavioural issues. It is appreciated by parents and seems to become a therapeutic help in itself. It was also useful in auditing the intervention and may have application in research too.


C. Scott, P. Ramanaryan, J. Britto. Health Management Faculty, Business School and Department of Paediatrics, St Mary’s, Imperial College London

Introduction: The cost of clinical negligence is rising exponentially and the number of GP claims has doubled since 1994. There are 269 million GP consultations per year in the UK, of which 45 million are for children under 16 years of age and 22 million for children under 5 years of age. This study looks at the extent of paediatric clinical negligence claims in primary care and the effect of using ISABEL clinical support system to improve outcome.

Methods: A multivariate quantitative descriptive study was conducted using data provided by the Medical Protection Society (MPS). 1000 consecutive alleged clinical negligence claims over a 27 month period, from July 1996 to October 1998, were analysed, of which 151 were paediatric cases. The cases were allocated misadventure codes and 88 cases with a code of failed/delayed diagnosis were analysed. The original assessment by the MPS of the cases as to whether the GPs involved in the individual cases were at fault provided the control. Symptom lists were produced and input into ISABEL.

Results: The cases covered a wide number of paediatric specialities. Eight specific conditions provided 59% of the cases. There were two demographic peaks in the age of the children under 1 year and 17 years of age. The average number of GP consultations was five and the average number of GPs involved in each case was three. In 69% of the cases ISABEL provided the correct outcome. In 17% of cases the GP was thought to have acted appropriately (p<0.001) Therefore, in 52% of cases ISABEL could have altered the outcome of the diagnosis.

Conclusions: ISABEL clinical decision tool may significantly improve the rate of correct diagnosis in paediatric cases in primary care, particularly after the second consultation. This study is intended as a pilot study for further quantitative and qualitative analysis of paediatric negligence claims within primary care.


P. M. Barnes, P. Upton, A. Maddocks, C. Eiser, L. Price, C. Killa. Swansea NHS Trust, University of Wales–Swansea, University of Sheffield

Aims: To develop and validate a measurement system to assess the health related quality of life (HRQL) of children in local authority care.

Methods: Following a review of research and interviews with children, foster carers, paediatricians, and other child care professionals a condition specific self report questionnaire was developed for children aged 8–18 years in local authority care, along with a parallel proxy report for foster carers. These questionnaires were administered on two separate occasions along with an established generic HRQL measure to a sample of children in local authority care and their carers (n 70). The questionnaires’ psychometric properties were then tested as described below.

Results: Factor analysis suggested a single scale of 19 items. Selected items reached the recommended item total correlations of 0.4 or above, establishing the scale’s internal consistency. Cronbach’s α of 0.87 for self report and 0.91 for proxy report measures confirmed the scale’s internal reliability. Scale scores showed moderate but significant correlation with emotional functioning subscales on a generic HRQL measure (Pearson’s r 0.661 and 0.605, p<0.05 for self report and proxy report, respectively) and with an independent rating of mental health (Pearson’s r 0.483 and 0.464, p<0.05 for self report and proxy report, respectively) establishing construct validity. Moderate concordance between the self report and proxy report confirmed that the two forms were measuring the same construct. Reproducibility was acceptable for both reports (intraclass correlation 0.685 and 0.651, p<0.05 for self report and proxy report, respectively).

Conclusions: This measure offers the advantages of brevity, ease of administration and interpretation, a sound psychometric basis, and clinical validity. Finally, the availability of parallel forms for child and carer allows a complete assessment of the HRQL of the child in local authority care. This instrument will provide a much needed outcome measure for use both in routine assessment and in research with this group of children.

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.