One session of prism adaptation training does not increase immediate engagement in occupational therapy in people with spatial neglect early after stroke: a proof-of-concept study nested within a phase II RCT

Objectives Spatial neglect, a debilitating cognitive syndrome and predictor of poor functional outcome, affects attention and awareness after stroke. Early rehabilitation is essential but neglect itself may impede participation in therapy. In a proof-of-concept study nested within an RCT, we investigated whether the oft-reported immediate effects of prism adaptation training (PAT) might enable engagement if introduced at the start of an occupational therapy session.

Methods Early after stroke we video-recorded in-patients carrying out a standardised activity in their first RCT occupational therapy session, before and after PAT (or a control therapy activity). Level of engagement was later scored by a video-rater, experienced in therapy, blind to arm allocation (intervention/control) and whether randomly presented videos were recorded pre-or post-therapy. The rater recorded engagement scores on a 100mm visual analogue scale. Treating therapists also reported, on a 3-point Likert scale, whether or not engagement changed.

Results 49 of the RCT’s 53 patients were recruited (37 PAT, 12 control), 43 of whom consented to be video-recorded. Regression analysis did not suggest improvement in engagement following one session of PAT, using the blinded expert video scoring method: mean difference (95% CI) = −0.5 (−7.4 to 6.4) mm; p=0.89). Similarly, post-hoc re-rating of engagement scores (the video-rater viewed paired pre- and post-therapy recordings but remained blind to arm allocation) excluded any material difference in engagement following PAT: mean difference (95% CI) = 1.2 (−2.5 to 4.9) mm; p=.52). Impressions of level of engagement provided by the treating occupational therapists also suggested no change: OR (95% CI) = 1.3 (0.13 to 13); p=0.81).

Conclusions Despite the need to enable neglect patients to engage in the therapy they are offered, we are confident that a single session of PAT at the start of a therapy session does not enhance immediate engagement in occupational therapy early after stroke. Our study does not address the alternative definition of engagement as a longitudinal, rapport-building process which could meaningfully be explored.