User testing of a Scottish Intercollegiate Guideline Network public guideline for the parents of children with autism

Abstract

Background: The Scottish Intercollegiate Guidelines Network (SIGN) is the leading national clinical guideline producer in Scotland. Improved design and dissemination of guidelines produced for the public can empower people to take an active role in self-management and shared decision-making. The public version of the guideline examined covered getting assessed and diagnosed with autism, and approaches that can help. The aim of this study was to test a public version of a guideline for the parents of children and young people with autism, implement improvements, and identify what works in making it usable and accessible. Methods: We recruited mothers from across Scotland. User testing involved formal ‘think aloud’ semi-structured interviews that guided users through the booklet. Interviews took place individually and were recorded and transcribed. Key findings were identified and themed using the honeycomb user experience model. Results: Fourteen user-testing interviews were conducted. Facilitators for usability and desirability of the guideline included the chunking of text, consistent use of colour and boxes to highlight important information. Simple language, written in a tone of partnership, helped to engage mothers. Value arose from the guidelines ability to explain the process of diagnosis and make mothers feel empowered in their relationships with healthcare professionals. There was a lack of consensus on the usefulness of rating the strength of evidence and recommendations. Conclusion: There was a marked similarity between what was important to the mothers and what has been found to be important to other groups. The involvement of service users and carers in the guidelines development was key to its credibility. One size does not fit all in presenting evidence-based recommendations to the public and it is a challenge to provide sufficient information while avoiding information overload. Recommendations and evidence levels are suitable for use in public versions, but these should be kept as simple as possible.