Communication Confidence, Aphasia Severity and Well-Being: A Cross-Sectional and Longitudinal Study in People With Post-Stroke Aphasia.

The study was exploratory and small: 27 participants at baseline and 18 at follow-up, so the findings may not apply to other people with chronic aphasia. There was no control group, so increases in general confidence and well-being could be due to natural variation, regression to the mean, response shift, selective attrition, or taking part in the study. Five participants were redirected to a separate therapy study before follow-up, which may have changed who remained in the longitudinal analysis. People who reported clinical depression or chronic anxiety were excluded, so the sample does not represent many stroke survivors who have these conditions. The sample did not include anyone classified as severe aphasia on the WAB-R, although it included mild to very severe aphasia. Time since stroke varied widely, from 6 to 363 months, so the follow-up period may not correspond to a single stage of recovery. Recruitment was through stroke groups and self-referral, and the online format required internet access, video-conferencing familiarity, and enough language ability to complete the measures, which may have selected for more confident or able participants. Education, marital or family relationships, support networks, and opportunities to practise functional communication were not measured, but the paper says they likely influenced confidence. The paper states that no established Minimally Important Clinical Differences exist for the CaSM and ONS-4, so score changes may not be clinically meaningful. Correlations cannot show that one factor causes another.

This research was supported by a PhD scholarship from King Saud bin Abdulaziz University for Health Sciences through the Saudi Arabian Cultural Bureau in London, and the paper names the Saudi Arabian Cultural Bureau as the funder.

Do not read the increase in general confidence and well-being as proof that a therapy caused it. There was no control group, and the authors say measurement error, regression to the mean, response shift, selective attrition, or study participation could explain the change.