Report Viewer

Exploratory: Cerebral Palsy (CP)

The original GAS uses a 5 point scale. Some authors have attempted to use a 3 and 7 point scale. Requires extensive therapist training and experience to set valid, accurate goals. There is no administration or scoring manual and thus the GAS is not administered or scored in a consistent manner among therapists or facilities. Reliability (therapst's judgement of the impact of the intervention) and validity (ability of therapist to set realistic, accurate goals) may be difficult to establish. Recommend extensive training; utilization of therapists with at least one year of experience, use of independent raters to score outcomes may reduce bias, and supplement this measure with a standardized and objective tool.

Strengths/Weaknesses: The GAS is a flexible, individualized tool that can be used across diagnoses, age groups, and problem areas. It uses a client-centered approach to goal setting which has been shown to improve outcomes in children with CP. Validity and reliability issues are reported in the literature (therapist's ability to accurately predict future performance; need for adequate training to achieve inter-rater reliability for scale development; need for independent raters to reduce bias). It is time consuming to develop goals.

 

Psychometric Properties: It is an individualized, patient generated scale and the content may not be the same from patient to patient. The psychometric properties for pediatric rehabilitation reported in a critical review (Steenbeek, 2007) are reported here (samples included 65 infants 3–30 months with motor delays, and 41 children ages 2–7 years with unilateral CP). The authors used the following definitions:  

Content reliability: Correlation in terms of textual content of GAS scales between two independent scale developers.

Interrater reliability: Correlation between the scores of two independent raters on the same GAS scale.

Content validity: Value of the textual content of GAS scales in relation to the relevance to the subject.

Concurrent validity: Correlation between the outcome in terms of GAS and other appropriate measures.

• Internal consistency: Not reported

• Interrater reliability: Kappa coefficient of 0.89  

• Concurrent validity: Low correlations with PDMS-2 and COPM

• Content validity: (relevance of the goal to the subject): 77%–88% of therapists' goals met criteria for content validity in a random sample of 10 goals

• Sensitivity to change: The studies included in the critical review concluded that sensitivity to change depends upon whether the goals and levels of attainment represent clinically important changes. Two studies reported t-scores >50 (better than expected outcome). A study comparing the effect of BTX with a control found large GAS effect sizes.

• Discriminative validity: GAS t-scores were able to discriminate between infants who were more or less motorically delayed. The GAS was able to detect changes within and between two groups

 

Overall, Steenbeek et al. (2007) concluded that the GAS appears to be a promising tool to measure change in heterogeneous populations with a variety of treatment goals. Reliability is largely unknown, and training to ensure reliability is necessary, as are descriptions of training should be reported in publications. Also, the validity of the GAS is ambiguous: concurrent validity is low, however its individualized approach allows a therapist to measure exactly what needs to be measured.