Efficacy of a multicomponent singing intervention on communication and psychosocial functioning in chronic aphasia: a randomized controlled crossover trial

Abstract

The ability to produce words through singing can be preserved in severe aphasia, but the benefits of group-based singing rehabilitation in aphasia are largely unknown. Our aim was to determine the efficacy of a multicomponent singing intervention on communication and speech production, emotional-social functioning and caregiver well-being in aphasia. Fifty-four patients with acquired brain injury and chronic aphasia and their family caregivers (n = 43) were recruited. Using a crossover randomized controlled trial design, participants were randomized to two groups who received a 4-month singing intervention either during the first or second half of the study in addition to standard care. The intervention comprised weekly group-based training (including choir singing and group-level melodic intonation therapy) and tablet-assisted singing training at home. At baseline, 5- and 9-month stages, patients were assessed with tests and questionnaires on communication and speech production, mood, social functioning, and quality of life and family caregivers with questionnaires on caregiver burden. All participants who participated in the baseline measurement (n = 50) were included in linear mixed model analyses. Compared with standard care, the singing intervention improved everyday communication and responsive speech production from baseline to 5-month stage, and these changes were sustained also longitudinally (baseline to 9-month stage). Additionally, the intervention enhanced patients' social participation and reduced caregiver burden. This study provides novel evidence that group-based multicomponent singing training can enhance communication and spoken language production in chronic aphasia as well as improve psychosocial wellbeing in patients and caregivers. https://www.clinicaltrials.gov, Unique identifier: NCT03501797.

Keywords: aphasia; communication; rehabilitation; singing; speech production.

Graphical abstract

Figure 1

Flowchart outlining the design and progress of the trial. LMM, linear mixed effects model; T1, timepoint 1 (baseline); T2, timepoint 2 (5-month); T3, timepoint 3 (9-month).

Figure 2

Communication and speech production results from the LMM analysis (N = 50). (A) Communication index (F = 7.082, P = 0.011). (B) Responsive speech index (F = 4.100, P = 0.049). AB received the intervention from T1 to T2, and BA received the intervention from T2 to T3. The statistical method used in this study was the LMM analysis. The bar plots (mean—SEM) show changes in test scores over the three time-points (T1–T3) presented group-wise (AB/BA). Significant Time × Group interactions are shown with solid lines and significant within-group Time main effects are shown with dashed lines. LMM, linear mixed effects model; SEM, standard error of the mean; T1, timepoint 1 (baseline); T2 timepoint 2 (5-month); T3 timepoint 3 (9-month).

Figure 3

Functional outcome and caregiver burden results from the LMM analysis (N = 50). (A) SIS participation and role function (F = 6.440, P = 0.015). (B) Caregiver burden as measured by GHQ-12 and ZBI-22 (F = 6.765, P = 0.014). AB received the intervention from T1 to T2, and BA received the intervention from T2 to T3. The statistical method used in this study was the LMM analysis. The bar plots (mean—SEM) show changes in test scores over the three time-points (T1–T3) presented group-wise (AB/BA). Significant Time × Group interactions are shown with solid lines. Time main effects are shown with dashed lines. GHQ-12, General Health Questionnaire 12; LMM, linear mixed effects model; SEM, standard error of the mean; SIS, Stroke Impact Scale; T1, timepoint 1 (baseline); T2, timepoint 2 (5-month); T3, timepoint 3 (9-month); ZBI-22, Zarit Burden Interview 22.