Telehealth Administration of Narrative and Procedural Discourse: A UK and US Comparison of Traumatic Brain Injury and Matched Controls

Background: Impaired discourse production is commonly reported for individuals with traumatic brain injury (TBI). Discourse deficits can negatively impact community integration, return to employment and quality of life. COVID-19 restrictions have reduced in-person assessment services for people with communication impairments. Advances in telehealth may help speech and language therapists (SLTs) to assess monologic discourse more systematically and improve access to services for patients who may find it difficult to attend in-person. Aims: To examine the feasibility of telehealth administration of narrative and procedural discourse tasks with individuals with TBI and matched controls. Methods & Procedures: A total of 20 individuals with TBI and 20 healthy controls, aged 18-55 years, were directly recruited from the UK and indirectly recruited from the US. For participants with TBI, time post-injury was at least 3 months with no diagnosis of aphasia. Control participants were matched for sex and as closely as possible for age. Feasibility of measures was based upon the time to administer both narrative tasks, the report of any technological problems, and participant feed. Discourse samples were transcribed verbatim and analysed using story grammar analysis (for narrative discourse) and identification of propositions (for procedural discourse). Interrater reliability was calculated using percentage agreement for 50% of the data. Non-parametric analyses were used to analyse the performance of the two groups. Outcomes & Results: Narrative and procedural discourse samples were collected via telehealth in approximately 10 min with no reported technical difficulties or complaints from any participants. For narrative discourse performance, there were significant differences for the TBI and control groups for measures of complete episodes (p < 0.001) and missing episodes (p = 0.005). No significant group differences were noted for any of the procedural discourse measures. Conclusions & Implications: Results support the feasibility of collecting discourse samples via telehealth. Although the participants' discourse performance distinguished the TBI and control groups on the narrative task, no differences between the groups were noted for the procedural task. The narrative discourse task may have been more difficult than the procedural task, or video cue support reduced the cognitive load of the procedural task. This finding suggests the use of more complex procedural tasks without video cue support may be needed.