What is the optimum method of identifying depression in children with type 1 diabetes?

Lernmark et al, 199957462/114 eligible children from the diabetes clinic at Children’s Hospital, Stockholm
Inclusion criteria: aged 9–18 years with at least a 2-year history of type 1 diabetes
Exclusion criteria: mental retardation, non-Swedish speaking families
Median age 14 years, 25 males
Psychological function:
Depression (The Children’s Depressive Inventory)
Self-esteem (‘I think I am’ test), abilities/talents, psychological wellbeing, relations to others/family
Fear (The Fear Survey Schedule for children)
Adaptation to diabetes (Adaptation to Diabetes Scale, FIAD), monitoring, impact on daily life, attitude to diabetes, emotional difficulties, feelings toward diabetes
Metabolic control (average of all HbA1c during the previous year)
Prevalence of depression: 14.5%
Patients with more depressive symptoms had poor adaptation, poor self-esteem, poor metabolic control (p < 0.01)
Regression analysis showed adaptation to diabetes was significantly related to metabolic control and depression (p < 0.0001 and p = 0.002)
Depressive symptoms defined as CDI cut-off score of ≥ 15Cross-sectional surveyIII
Whittemore et al, 200257597 young people attending Yale Children’s Diabetes Program; November 1995 to November 2000
Inclusion criteria: aged 12–20 years, no other illness except treated hypothyroidism, on insulin for at least 1 year, recent HbA1c in range 7.2–14%, no severe hypoglycaemic events within previous 6 months
Exclusion criteria: type 2 diabetes, co- morbid chronic illness, inability to comply with protocol
Mean age 14.3 ± 2.0 years, 38% male
Relationship between depressive symptoms (The Children’s Depression Inventory) and:
Age, duration of illness
Family factors (Family Adaptability and Cohesion Scale)
Family behaviour (Diabetes Family Behaviour Scale)
Metabolic control (HbA1c)
Prevalence of depressive symptoms at entry:15.4%; 2-year follow-up: 10%
Depressive symptoms more prevalent among 14–16 year olds (25%) and those with diabetes ≥ 10 years (23%)
At entry general and diabetes-specific family function was significantly associated with depressive symptoms (p = 0.03); upon multivariate analysis only diabetes-specific factors (less warm and caring behaviour) at 2-year follow-up (p = 0.08)
At 2-year follow-up those with depressive symptoms had significantly higher HbA1c: 9.0 ± 0.85% vs. 8.3 ± 1.4% (p = 0.03)
Depressive symptoms defined as CDI score ≥ 13; analysis at entry n = 97 and at 2-year follow-up n = 57Longitudinal cohort with assessments at study entry and 2-year follow-upIIb
Kovacs et al, 199757992 sequential inpatient admissions at Children’s Hospital, Pittsburgh, recruited between 1978 to 1985, 85 were prospectively followed up for 5 years
Inclusion criteria: type 1 diabetes, aged 8–13 years
43 boys, mean age 11 years
Prevalence of Psychiatric Disorders:
Interview Schedule for Children and Adolescents (ISCA)
BDI and Hamilton Depression Rating Scales (HAM-D) used to assess maternal psychopathology
16% had a psychiatric disorder predating type 1 diabetes onset (none had depression)
42% developed at least 1 episode of a psychiatric disorder during follow-up
26% had major depressive/dysthymic disorder
Cumulative probability of any depression during 10 years after type 1 diabetes onset: 0.27 (significantly higher than all other disorders)
Maternal depression was a significant risk factor for depression among patients with type 1 diabetes (p = 0.02); regression coefficient: 0.97
Time intervals between assessments varied across casesLongitudinal cohort
Initial evaluation 2–3 weeks post-diagnosis then 1–3 visits every year
Subjects and parents interviewed
Goldston et al, 199458095 consecutive children admitted to inpatient unit, Children’s Hospital, Pittsburgh between 1978 and 1985, 5- year follow-up period
Inclusion criteria: type 1 diabetes, no other systemic illness, aged 8–13 years, within commuting distance
Exclusion criteria: mental retardation
Median age 11 years, 44 males
  1. Suicide Ideation Predictor variables: female gender, low socio-economic status, pre-existing psychiatric disorder, age, depressive, anxiety, and severity of illness at diagnosis symptoms cluster
  2. Suicide Attempt
    (The Interview Schedule for Children and Adolescents was used for both outcome measures)
  1. Within 1 year before intake: 21.1% (retrospective)
    At study intake: 29.5%
    Severity of depression significantly related to history of suicide ideation (p < 0.004)
    During follow-up (n = 85): 46%
    Those with suicide ideation were more likely be non-compliant compared with those without suicide ideation (p < 0.003)
  2. Suicide attempt during follow-up (n = 6) 6.4%, 3 had a history of earlier suicide ideation
Time intervals between assessments varied across cases, parents and children interviewed, pervasive non- compliance not defined5-year follow- up cohort (3 visits per year and then 1 assessment every 8–12 months), first research assessment 2–3 weeks after diagnosisIIb
Lawler, 199057816 young people recruited from physicians and newspaper advertisements in Oklahoma
Aged 15–18 years, with type 1 diabetes > 1 year, two non-diabetic parental figures present, 10 males
Family Emotional Health
Beck Depression Inventory
Family Inventory of Life Events and Changes
Social support
Social support positively correlated with Family Emotional Health (r = 0.46, p < 0.05)
Depression was positively correlated with diabetic control (r = 0.51, p < 0.05)
2 subjects scored mild and severe depression
10/16 experienced moderate to high stress
Thernlund et al, 199658176 children from 5 paediatric clinics in Sweden
Mean age 8.6 ± 3.9 years, 38 boys
Reactions assessed during first 3 weeksFamily crisis
Psychological adjustment
Grief and anxiety of children were less marked than that of parents (0.05)
Among children age ≥ 6 years: distress increased the odds of poor metabolic control (OR 1.3, p < 0.01)
Related to maternal stress and reaction
Factor analysis, follow-up at 10 monthsIIb
Viner et al, 199658243 children and young people and their mothers, outpatient clinic, Brisbane, Australia
Mean age 10.2 ± 3.16 years, 42% male
Metabolic control (HbA1c)
Family Inventory of Life Events (FILE) (measure of stress as perceived by mother)
Social support (perceived by mother)
Increased HbA1c at the time of FILE questionnaire associated with increased stress (rs = 0.554, p < 0.001)
Mean HbA1c associated with FILE (rs = 0.563, p < 0.001)
Family social support not directly related to HbA1c, but high support buffered the effects of family life stress. The means of the two mean HbA1c groups stratified by support level were not statistically significant at the at 1% level (p ≥ 0.01)
No significant evidence of association between admission rate or glycated Hb measures of control with family stressCross-sectionalIII
Hazell et al, 2002586Children and young people without type 1 diabetes treated for depression
Aged 6–18 years
Tricyclic antidepressants
  1. Failure to recover
  2. Change in depression checklist scores
  1. Overall improvement OR 0.84, 95% CI 0.56 to 1.25
    Adolescent OR 0.85, 95% CI 0.54 to 1.34
    Child OR 0.69, 95% CI 0.25 to 1.89
  2. Overall change in depression checklist scores −0.312 (95% CI 0.62 to −0.01)
    Adolescent −0.469 (−0.922 to −0.016)
    Child 0.147 (−0.343 to 0.638)
Patterns of co-morbidities likely to have differed across studies
13 trials found, n = 506
Intention to treat analysis
Systematic reviewIa
Harrington et al, 1998588Review article of studies addressing 3 domains of cognitive behaviour therapy (CBT)CBTImprovement in:
Depressive symptoms
Depressive disorder
As a Family intervention
9 controlled studies in children recruited from school, 6 trials provided quantitative evidence and 4 of those showed CBT significantly superior to no treatment
6 RCTs with a variety of sample populations, a meta-analysis showed significant improvement in CBT group
Limitations of existing research: poor quality trials, depression is main focus of outcome assessment (co-morbid conditions?), follow-up intervals rarely exceed a yearReview of the literature –RCTsIV
Harrington et al, 19985876 RCTs, children aged 8–19 years with depressive disorder of moderate severityCognitive behaviour therapy (CBT)
inactive interventions (waiting list, relaxing training, art exercises)
Remission ratesHigher in the CBT group (129/208, 62%) vs. comparison (61/168, 36%)
Pooled OR 3.2 (95% CI 1.9 to 5.2)
Most studies based on mild cases of depressionSystematic review/meta-analysisIa

From: Evidence tables

Cover of Type 1 Diabetes
Type 1 Diabetes: Diagnosis and Management of Type 1 Diabetes in Children and Young People.
NICE Clinical Guidelines, No. 15.2.
National Collaborating Centre for Women's and Children's Health (UK).
London: RCOG Press; 2004 Sep.
Copyright © 2004, National Collaborating Centre for Women’s and Children’s Health.

Apart from any fair dealing for the purposes of research or private study, criticism or review, as permitted under the Copyright, Designs and Patents Act, 1988, no part of this publication may be reproduced, stored or transmitted in any form or by any means, without the prior written permission of the publisher or, in the case of reprographic reproduction, in accordance with the terms of licences issued by the Copyright Licensing Agency in the UK. Enquiries concerning reproduction outside the terms stated here should be sent to the publisher at the UK address printed on this page.

The use of registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant laws and regulations and therefore for general use.

The rights of National Collaborating Centre for Women’s and Children’s Health to be identified as Author of this work have been asserted by them in accordance with the Copyright, Designs and Patents Act, 1988.

PubMed Health. A service of the National Library of Medicine, National Institutes of Health.