The Child Behaviour Checklist (CBCL) is the most commonly used questionnaire to assess parent-reported emotional and behavioural problems in children and adolescents. In two of the 120 questions in the 1991 version of the questionnaire, on which most published studies are based, parents are asked to report if their child suffers from allergy or asthma. These two questions were, as were all items, included on the assumption that they would help predict referrals to child psychiatric services. This did not hold true and the answers to the two items were eliminated in summation of scores (1).
The present study was carried out to determine if the item on presence of child's asthma can be used together with the other questions to study the epidemiology of emotional and behavioural problems in childhood asthma. The Swedish 1992–93 normative sample of 1308 6–16-year olds, described in detail elsewhere, was used (2). Data from 1200 children (579 boys and 621 girls, mean age 11.2 years) with complete records from the two questions on allergy and asthma and the 118 emotional and behavioural problem items were examined.
The distribution of problem scores is given as mean and 95% confidence intervals (CI) of mean, prevalence rates as proportions with 95% CI. Eighty-one parents rated either 1 (somewhat true) or 2 (very true) for the presence of asthma, resulting in a 6-month prevalence of 5.7% (95% CI 4.5–7.1%). Parents of 925 children reported neither allergy nor asthma. In view of very small gender differences for problem scores, figures for boys and girls were collapsed.
In this first analysis using the allergy and asthma questions of the CBCL, the 6-month asthma prevalence rate using the CBCL questionnaire was low compared with the 12-month asthma prevalence of 13.6% (95% CI 12.1–15.1%) reported in a postal questionnaire in 1995 by parents of 7–16-year olds in one Swedish area (3) and the 9.3% (95% CI 8.1–10.6%) reported in 1996–97 by parents of 11-year olds in another Swedish area (4).
Parents were asked to rate their children's problems on a 0–2 scale, 0 = no or never during the previous 6 months. Two main groups of problems were identified: internalising (withdrawn, somatizing and anxious/depressed) and externalizing (delinquent and aggressive). As seen in the Table 1, mean problem scores were slightly higher in children with asthma than in healthy children. However, CI of means overlapped for the two groups. The somewhat increased figures were much lower than those reported in a meta-analysis of clinical samples of children with asthma (5). In a hospital asthma clinic sample of 59 7–9-year old Swedish children, data collected in 1997–98, mean total problem scores were 25.9 (95% CI 20.4–31.4), mean internalizing problem scores 7.7 (95% CI 5.8–9.6) and mean externalizing scores 10.0 (95% CI 7.8–12.1). These figures were way above those of healthy children (6).
|Children with neither asthma nor allergy (n = 925)||Children with asthma (n = 81)|
|Total problems||13.9 (13.1–14.7)||16.9 (13.5–20.3)|
|Internalizing problems||3.8 (3.5–4.1)||4.8 (3.6–6.0)|
|Externalizing problems||5.5 (5.2–5.9)||6.4 (4.8–8.1)|
It is concluded that the CBCL scores on presence of asthma are not sufficiently valid. The questionnaire should therefore not be used as a single source to study the epidemiology of emotional and behavioural symptoms of the illness. However, the questionnaire is very rich in information on parents’ perceptions of their children's emotional and behavioural problems and large numbers of children from many countries have been screened. These data ought to be used in the creation of empirically derived models of which features of asthma contribute to poor psychological adjustment and how social conditions can add to emotional and behavioural problems of children with asthma.