Rasch analysis finds limits in Swedish autism genetic counselling scale

A Rasch-based psychometric study found structural limitations in both a new Swedish autism-specific genetic counselling outcome measure and the existing Swedish general version. The researchers created the first Swedish autism-specific adaptation of the modified 24-item Genetic Counselling Outcome Scale, but the full instruments showed multidimensionality, disordered thresholds, local item dependence and invariance problems. The study, reported in BMC Psychology , combined adaptation with a Rasch analysis of measurement properties.

Investigators adapted the English modified Genetic Counselling Outcome Scale, known as mGCOS-24, into a Swedish autism-specific version. They then evaluated that instrument alongside the Swedish general GCOS-24. It was the first Rasch-based evaluation of a GCOS-24 or mGCOS-24 conducted in Swedish.

The analysis found that the Swedish autism-specific mGCOS-24 contained identifiable subscales with acceptable measurement properties. This was a narrower finding than evidence that the complete adapted scale had satisfactory properties. Although the researchers could isolate some components that met the analysis criteria, the overall instrument continued to exhibit the structural challenges identified in the study.

Applying the same subscale structure to the Swedish general GCOS-24 did not resolve that instrument’s broader limitations. The comparison therefore does not show that the general scale can be transformed into an appropriate autism-specific measure merely by reorganising its items. The study instead identified differences between the structure found in the adapted autism instrument and the limitations that remained in the general version.

These findings concern the measurement tools rather than the effectiveness of genetic counselling itself. The research did not establish whether genetic counselling improved health outcomes, nor did it test whether either outcome scale improved clinical decisions. The supported conclusion is more limited: the Swedish autism-specific mGCOS-24 provides a basis for further measurement work, while its current structure still requires refinement.

For autistic populations, the authors said the evidence points toward both opportunities to improve existing outcome measures and a possible need for new or more substantially adapted tools. That conclusion reflects the mixture observed in the study: some subscales of the Swedish autism-specific instrument had acceptable measurement properties, yet neither that instrument nor the general GCOS-24 was free from the reported structural problems. The work therefore presents an adapted autism-specific scale as a starting point for refinement rather than a finished, universally suitable measure of genetic counselling outcomes.

Subject of Research: Psychometric evaluation of Swedish genetic counselling outcome measures for autistic populations

Article Title: Assessing Swedish genetic counselling outcome measures for autism and general use: Rasch findings highlight the need for improved measures

Article References: Nordstrand, M., Falk, S. F., Johansson, M., Pestoff, R., & Tammimies, K. (2026). Assessing Swedish genetic counselling outcome measures for autism and general use: Rasch findings highlight the need for improved measures. BMC Psychology, 14(1), Article 1385. https://doi.org/10.1186/s40359-026-05427-z

Image Credits: AI Generated

DOI: 10.1186/s40359-026-05427-z

Keywords: genetic counselling, autism, Rasch analysis, outcome measures, psychometrics, Sweden, mGCOS-24, measurement invariance, multidimensionality, clinical research, Assessing, Swedish