Reliability of observational measures of the Movement Assessment of Infants.

This study was conducted to examine the reliability of the Movement Assessment of Infants (MAI), a recently published neuromotor assessment tool. Interobserver and test-retest reliability data were collected on 27 full-term and 26 preterm 4-month-old infants. Reliability coefficients (Pearson r) were calculated for both the total-risk scores and the section-risk scores on the MAI. The total-risk score was calculated by summing the questionable or abnormal ratings on each of the 65 test items. For each of the four sections of the test, tone, primitive reflexes, automatic reactions, and volitional movement, an individual section-risk score was computed in a similar manner. Fair reliability was demonstrated for the total-risk scores (interobserver: r = .72; test-retest: r = .76). Section-risk score coefficients yielded a wide range of values for both interobserver and test-retest reliabilities (poor to good reliability). These measures provide needed technical data for therapists using this test and will assist the authors of the MAI in their attempts to improve the clinical validity of this assessment tool.