CHOLESTEROL SCREENING OF CHILDREN AT HIGH-RISK - BEHAVIORAL AND PSYCHOLOGICAL EFFECTS

Objective: To assess the behavioural and psychosocial effects of scree ning asymptomatic children at high risk for hyperlipidemia. Design: Ob servational study involving prospective longitudinal and cross-section al portions. Setting: Two tertiary care pediatric lipid clinics in Mon treal. Subjects: Longitudinal portion: all children aged 4 to 17 years who presented for screening at the lipid clinics between April 1990 a nd June 1991. Of the 56 eligible children 52 (93%) (and their mothers) agreed to participate, 34 with hyperlipidemia (case subjects) and 18 without hyperlipidemia (control subjects). Thirty-five children (67%) completed 3 assessments over 12 months. Cross-sectional portion: all c hildren aged 4 to 17 years in whom hyperlipidemia had been diagnosed 2 to 5 years earlier at one of the lipid clinics. Of the 58 eligible ch ildren 48 (83%) (and their mothers) participated. Outcome measures: Fo r children, mean scores on the Child Behavior Checklist (Behavior Prob lems subscale) (CBCL), the Children's Depression Inventory (CDI) and t he State-Trait Anxiety Inventory for Children (STAIC); for mothers, me an scores on the Beck Depression Inventory (BDI) and the State-Trait A nxiety Inventory (STAI). Results: in the longitudinal portion of the s tudy, there was no significant difference between the case and control subjects in the mean CDI or STAIC scores at 1 or 12 months. At 1 mont h after diagnosis the case subjects in the longitudinal portion had a significantly higher mean CBCL score than the children in the cross-se ctional component (p = 0.01). With the control group as the reference group, the adjusted odds ratios for a high CBCL score (greater than 62 ) for the case subjects were 15.5 (95% confidence interval [CI] 2.4 to 99.8) at 1 month and 15.8 (95% CI 1.1 to 223.4) at 12 months. The cor responding values for the children in the cross-sectional component we re 1.3 (95% CI 0.3 to 6.2) and 5.0 (95% CI 0.5 to 50.9). Conclusions: The observed behavioural problems in children with a recent diagnosis of hyperlipidemia were independent of other risk factors, such as age and sex of child and mother's age and BDI score. Our results suggest t hat identification of hyperlipidemia in children may have harmful psyc hological effects in the families involved. This evidence strengthens arguments for the exclusion of cholesterol measurement from the period ic health examination of children at moderately high risk.