Sonja Mensch

134 Chapter 7 and can prevent disagreement and “bias” as opposed to testing in standardized test conditions (Dobson et al., 2012). This choice also has the added benefit of no burdening the child by having to put it through extra testing. A relevant issue of the construct and criteria of Movakic is that is we have to face subjectivity due to the application of manual support (“hands-on”). Although inter-rater reliability of Movakic is found very high, objectifying the extent of “hands- on” remains debatable and personal. We stress again that performing physical activities independently is problematic in the target group and as a result, direct support of other persons is necessary in all activities of daily life (Putten, Bossink, Frans, Houwen, & Vlaskamp., 2016). Agreement between observers of the definition of extent of “hands- on” support is therefore very important to reliably use the Movakic instrument. To improve agreement, “hands-on” was described in the user’s manual of Movakic and during the required training considerable time was spent discussing the uniform use of “hands-on”. We feel reading the manual and participating in the training is needed to maximize agreement between the physiotherapists. The absence of a gold standard, applicable to children with SMD, was of methodological importance for the validity study, leading to a compromise in testing the construct validity by comparing motor ability scores and Visual Analogue Scale (VAS) (Reips & Funke., 2008) scores. It would have been better to compare Movakic scores with those of relevant parts of existing tools. However, first of all other tools do not allow for hands-on support rendering them unusable for comparison. Secondly, the necessity of repeated physical assessments would also have meant a higher burden for the participating children. So we decided that using the VAS was the most optimal solution. In order to evaluate responsiveness of Movakic to clinically relevant changes during the follow-up period of 18 months we asked participating physiotherapist to report events in the life of the child. A strong point of the design of this study was its close relation with usual care in these children. These children go through a lot of life events that might influence motor abilities. The expectation was that changes in Movakic scores were either results of physiotherapeutic intervention, surgery or of other relevant interventions, or the result of comorbidity or other influences. It remains difficult to establish the impact of a certain physiotherapeutic intervention in this vulnerable group of children when these children are subject to concurring illness or recovering from a medical intervention at the same time. All these influences will either have a positive or negative impact on motor ability. However, their individual share in the net effect on motor ability is hard to assess. In order to work around the complexity of analysing individual impact of events, we chose to compare Movakic score changes in two datasets: one containing 3-month intervals with events and one without events.