The Family Life Questionnaire: The Development of An Autism-Specific Measure of Family Life Functioning

Background: The involvement of service users in the development of the design and outcome measures in treatment trials is a current priority in the field¹. There is currently no specific measure of family quality of life for use in autism treatment trials or studies.  We undertook to develop such a user nominated measure in the context of the MRC Preschool Autism Communication Trial (PACT), a large preschool treatment trial in autism funded by the UK Medical Research Council.  The Trial, with its RCT methodology and large well defined sample, provided an optimal context for the development and initial evaluation of such a measure.
Objectives: To explore the internal and external validity of the Family Life Questionnaire. To ascertain this measure’s sensitivity to change in a treatment trial.

Methods: Development of measure.
Initial focus groups were undertaken with families of young children with autism (none of whom was involved in the PACT trial).  Groups were asked to nominate areas of child and family functioning that would be considered key markers of whether a pre-school autism treatment had been effective. From the focus group analysis  78 change markers were submitted to a national web-based consultation in collaboration with the UK National Autistic Society. The web consultation weighted each item for importance and clarity.  The resulting refined questionnaire consisted of the 56 most highly nominated items in three domains a) experience of parenting their child; b) quality of family life; c) child development, social functioning and behaviour.
Data collection.
The Family Life Questionnaire (FLQ) was administered at baseline and 13 month endpoint to the 152 families involved in the PACT RCT. 
Analysis plan. 
A) Confirmatory factor analysis to explore the construct validity and assess whether the latent variable structure tapped by the measure fits the three domains previously specified; exploratory factor analysis added as necessary to explore further the latent factor structure.  B) External validity investigated through relationships with the other trial measures, particularly the Vineland Adaptive Behaviour Scales (Parent Survey Form) and other baseline measures including demographic characteristics and child autism severity. C) The FLQ data is analysed for sensitivity to change between baseline and endpoint against the primary and secondary outcomes of the RCT.

Results: The data are complete. Results will be presented in the panel on each area of the pre-specified analysis plan as above. Conclusions: We comment on the method used to generate this measure and its empirical validity. We discuss its potential usefulness within future autism treatment trials.
Reference
1. Chalmers, I., & Glasziou P. (2009) Avoidable waste in the production and reporting of research evidence. The Lancet, Vol. 374 No. 9683 pp 86-89