Background: We explore the efficacy of a new computer therapy for sentence comprehension and production impairments in post-stroke aphasia. The intervention is based upon the theoretical framework of usage-based Construction Grammar, which has yet to be systematically applied to the management of sentence processing disorders in aphasia. Components of the intervention were trialled in two small case series, with the results of one pilot used to inform a power calculation. The aims of the study are: (1) To determine if UTILISE therapy (Unification Therapy Integrating LexIcon and SEntences) for aphasic sentence processing impairment is more effective than usual care; (2) To determine if non-invasive transcranial direct current stimulation (tDCS) enhances treatment outcomes; (3) To determine if improvements are evident in functional use of language after treatment; (4) To determine if any treatment effects are apparent after an 8-week maintenance period. Methods: We will conduct a Phase II prospective randomised control trial. The behavioural component of the study (computer therapy) is single-blinded on the primary outcome measure, while the tDCS element is double-blinded. The trial is conducted within a university setting and aims to recruit 66 participants with post-stroke aphasia from community and post-acute rehabilitation settings. After first baseline evaluation, participants are randomised on a 2:2:1:1 basis to one of four conditions: 1) immediate treatment + active tDCS; 2) immediate treatment + sham tDCS; 3) deferred treatment + active tDCS; 4) deferred treatment + sham tDCS. A second baseline assessment is completed prior to start of treatment with a gap of four weeks between baselines for the immediate groups, and eight weeks for the deferred groups. The deferred trial entry conditions act as a usual care control across their extended baseline phase. Treatment is administered in 12 x 60-minute sessions over a four-week period. Outcomes are assessed immediately at the end of intervention and after an eight week no treatment period. Participants will also undergo a structural magnetic resonance (MR) brain scan to determine lesion location and extent. The primary outcome measure is degree of connectivity (ratio of three-word combinations to total number of words) in spontaneous narrative speech (personal narratives and narration of picture cartoon series). Speech samples will be audio-recorded and subsequently transcribed and tagged by raters blind to allocation and phase. Analysis will be performed by automated analysis software (Frequency in Language Analysis Tool, FLAT; Zimmerer, Newman, Thomson, Coleman, & Varley, 2018). Secondary outcome measures are spoken sentence comprehension (Test of Reception of Grammar, TROG-2; Bishop, 2003), a study-specific story completion test (adapted from Goodglass, Gleason, Ackerman Bernholtz, & Hyde, 1972), and the Stroke and Aphasia Quality of Life Scale (SAQOL-39; Hilari, Byng, Lamping, & Smith, 2003). A shortened form of a written synonym judgement task (adapted from the Action for Dysphasic Adults Auditory Comprehension Battery, A.D.A.; Franklin, Turner, & Ellis, 1992) is administered as a probe of an untreated behaviour.Discussion: In a series of planned comparisons, we will first determine the stability of behaviour between the two baseline evaluations. We will examine the efficacy of UTILISE through comparison of change in scores between the immediate and deferred entry group, and subsequent maintenance of any behavioural gains by change between the immediate outcome and the follow-up maintenance measures. We will explore the effect of active vs. sham tDCS, and the relationship between outcomes and demographic and behavioural profiles, and lesion size and location.
Delayed treatment or deferred treatment: What’s intended? A commentary on national oocyte thaw outcomes data
