The NCES results available in 1991 were less useful to that committee in assessing the possible causal relation between DPT and permanent neurologic damage. A 12-month follow-up yielded an RR of 4.7 (95 percent CI, 1.1 to 28.0; Miller et al., 1985). However, the children were too young for a well-founded assessment of their neurologic status and the analysis was incomplete—the outcome measures were not well described and control children were not assessed in the same manner as case children. In addition, the neurologic status of the children prior to the acute event might not have been studied in a rigorous manner. The results of a 10-year follow-up were also incompletely reported (Madge et al., 1990) These data have since been reported in full (Madge et al., 1993; Miller et al., 1993).
Madge, Miller, and colleagues published the results from the 10-year follow-up of the NCES (Madge et al., 1993; Miller et al., 1993); 983 of the original case children and 1,012 control children were traced (the investigators attempted to trace only one of the two original matched controls). By the time of follow-up, 190 of the original case children and 4 control children had died. Outcomes were assessed for 754 case children and 976 control children (in addition to those who had died). Follow-up assessments were conducted by contacting school teachers, school doctors, family doctors, and parents. In an attempt to minimize bias, neither physicians nor teachers were informed of the child's status as a case or a control or of the child's immunization history, (although it is possible that parents of case children with concerns about the effects of DPT might have told the physicians or teachers of the child's history, thus possibly biasing some aspects of the follow-up assessments). Areas of dysfunction detected on examination or reported on questionnaires were quite broad and included neurologic, motor, sensory, educational, behavioral, and self-care dysfunctions. Detailed descriptions of the measures can be found in the report by Madge et al. (1993).
Neurologic dysfunction was indicated by any of (a) a definite or possible neurologic disorder as judged by a physician who examined the child, (b) information from parents or physician that the child had convulsions or took anticonvulsant medication in the previous 2 years or had a diagnosis of epilepsy, or (c) a medical diagnosis of cerebral palsy or a related disorder. Detection by a physician of noticeable tremor, gross motor incoordination, fine motor incoordination, motor impersistence, muscle weakness or spasticity in one or more limbs, or abnormal tendon reflexes in one or more limbs indicated motor dysfunction. Sensory impairment was indicated if children could not detect sounds of 25 dB at 250 Hz and 20 dB at all other frequencies, or if a test of