Does physiotherapeutic intervention affect motor outcome in high-risk infants? An approach combining a randomized controlled trial and process evaluation.

AIM The aim of this study was to examine the effects of intervention in infants at risk of developmental disorders on motor outcome, as measured by the Infant Motor Profile (IMP) and using the combined approach of a randomized controlled trial and process evaluation. METHOD At a corrected age of 3 months, 46 infants (20 males, 26 females) recruited from the neonatal intensive care unit at the University Medical Centre Groningen (median birthweight 1210 g, range 585-4750 g; median gestational age 30 wks, range 25-40 wks) were included on the basis of definitely abnormal general movements. Exclusion criteria were severe congenital disorders and insufficient understanding of the Dutch language. The infants were assigned to either the family-centred COPing with and CAring for Infants with Special Needs (COPCA) intervention group (n=21; 9 males, 12 females) or the traditional infant physiotherapy (TIP) intervention group (n=25; 11 males, 14 females) for a period of 3 months. Three infants assigned to the TIP group (one male, two females) did not receive physiotherapy. IMP scores were measured by blinded assessors at 3, 4, 5, 6, and 18 months. At each age, the infants were neurologically examined. Physiotherapeutic sessions at 4 and 6 months were videotaped. Quantified physiotherapeutic actions were correlated with IMP scores at 6 and 18 months. RESULTS The IMP scores of both the COPCA and TIP groups before, during, and after the intervention did not differ. Some physiotherapeutic actions were associated with IMP outcomes; the associations differed for infants who developed cerebral palsy (n=10) and those who did not (n=33). INTERPRETATION At randomized controlled trial level, the scores of both the TIP and COPCA groups did not differ in effect on motor outcome, as measured with the IMP. The analysis of physiotherapeutic actions revealed associations between these actions and IMP outcomes. However, the small sample size of this study precludes pertinent conclusions.