Screening for childhood disabilities in Jamaica: the use of ten question screen in Clarendon - abstract

ABSTRACT

No screening test for childhood disability in Jamaica has previously been formally evaluated. The present report describes the validity of the Ten Question Screen (TQ) used by community workers in Jamaican component of the International Epidemiology Study of Childhood Disability, done in the parish of Clarendon. It was previously reported by this study that approximately nine per cent of children aged 2-9 years had mild, moderate or severe disabilities (one or more of motor, visual, hearing speech or cognitive disabilities, or fits). In Clarendon, 5,461 children were screened, using the TQ and a sub-sample of 993 was subsequently assessed by a doctor and a psychologist for the presence of disabilities. The TQ proved to be an acceptable, reliable instrument and was relatively quick and easy to administer. Sensitivity was 41 percent overall, and 64 percent for moderate/severe disability. Specificity was high, at 87 percent and 86 percent, respectively. Due to the low prevalence, the false-positive rates were high, 75 percent and 90 percent, respectively. False-negative rates varied from 2 to 12 percent in the different age/sex groups, with overall rates of 6 percent for all disabilities, combined and 1 percent in serious disabilities. The lack of sensitivity was mainly due to failure to identify disabilities , especially mild cognitive disabilities, in older boys. The unacceptably high rate of false positives on the screening test indicates the need for a second level of screening to decrease the false-positive rates (increase the positive predictive value) of the test. The reasons for the high rate of false-positive results in using the Ten Question Screen and the possibility of reducing thisby using probes and an additional questionnaire were also investigated. In the 522 false-positive cases, analyses were performed of concordance of responses for three different categories of interviewer, of other disabilities not among the ones being detected, impairments, handicaps and the possibility of the parent being excessively worried about the child's health. The use of probes in TQ positive cases and the Activities of Daily Living Questionnaire (ADLQ) were tested to determine whether false positives could be reduced. Forty-four per cent of respondents in false-positive cases changed their response from positive to the community worker to negative to the doctor. Forty-eight per cent of false-positive cases had other disabilities or impairments and 4 percent had developmental handicaps. Worry was slightly more common among mothers of false-positive cases. Use of the probes and ADLQ reduced the false-positive rate (AU)