Music for Autism (M4A) (M4A)

Music for Autism: Binational Randomised Controlled Trial of Music Therapy Versus Play Therapy for Autistic Children

The Music for Autism (M4A) trial evaluates the neurobehavioral outcomes of a music therapy (MT) intervention, compared to a matched play therapy (PT) intervention, on social communication skills, brain connectivity and structural brain changes. In a crossover randomised controlled trial (RCT), 80 children with autism across all levels of functioning, aged 6-12 years, undergo a baseline assessment, which includes measurements of social communication, participation, functional connectivity and brain structure. Participants are then randomly allocated to a sequence of interventions (MT-PT or PT-MT) and assessments are taken before and after each intervention period. Both interventions will target common goals and follow the same structure, while at the same time allowing for flexibility in the therapists' approach. It is hypothesized that 12 weeks of intervention through MT, compared to PT, will improve social communication skills, participation, and other relevant mental health outcomes in children with autism spectrum disorder (ASD), as well as regulate resting-state functional over and under-connectivity and increase grey and white matter volume in specified regions. The investigators also expect changes in functional brain connectivity to correlate with behavioural outcome measures, specifically with improved social communication skills.

M4A will combine biomedical research with clinical outcome research to investigate if 12 weeks of intervention through MT, compared to PT, improve social communication skills, participation, family quality of life, receptive vocabulary, adaptive behavior, and symptom severity in children with ASD, and if this is accompanied by a change in resting state functional connectivity (rsFC) as well as grey and white matter volume change. Additional mental health outcomes include not only core areas of impairment, but also associated problems such as chronic stress, which is a significant issue for people with ASD; impedes learning; and can be reduced through music, both as an outcome in itself and as a mediator for other health outcomes. M4A will also investigate if clinical improvement is correlated with an increase of rsFC between auditory and striatal/fronto-motor regions as well as a decrease of rsFC between auditory and visual regions in MT compared to PT. Possible changes in grey/white matter volume will be measured by voxel-based morphometry (VBM) in a whole-brain scan before and after the interventions. Sample size and power: This study will be powered for an effect size of d=0.34. With a two-sided significance level of 5%, a sample of n=70 will be required to detect the effect with 80% power. Attrition is expected at <10%; the study will therefore recruit at least 80 participants. More specifically, the investigators expect to find a mean difference of 4.84 (SD=14.24), corresponding to an effect size d=0.34, on the primary outcome. The investigators expect the scores to be correlated within participants by r≥0.50. Sample size to achieve 80% test power was calculated in R. Treatment fidelity: All sessions will be recorded on video to help ensure and assess treatment fidelity. Fidelity will be rated by 2 raters, who are trained on a manual for assessment of treatment fidelity, on 4 different dimensions: (1.) Program adherence (number of sessions completed; number and types of activities covered, from the therapist's weekly reports); (2.) Process fidelity (delivery of the theoretical concepts of the intervention); (3.) Content fidelity (establishment of a therapeutic relationship between the participant and the therapist, measured using quality of delivery and participant responsiveness as well as the theoretical principles underlying the interventions); (4.) Programme differentiation between MT and PT ("Music was central to this activity"). Statistical analysis of behavioural outcomes will compare change from before to after each intervention within each participant. The intention-to-treat principle will be followed as applicable in a crossover trial: Participants will be analysed in the group to which they were randomised, regardless of whether they actually received the full allocated intervention. The main analysis will include all participants with valid data for both intervention periods; in addition, multiple imputation of missing outcomes will be used as a sensitivity analysis. Tests will use a two-sided 5% significance level. The two main secondary outcomes, participation and quality of life, will be Bonferroni corrected; the remaining secondary outcomes will be exploratory. Analysis software will be R. Brain connectivity of frontotemporal regions, measured as rsFC from 6 seeds, will be used as the main neuroscientific outcome. The time-series for each of the seeds will be used to generate individual participant-level maps using whole-brain general linear models at baseline and after the interventions. First-level maps will then be entered into the second-level analyses. For comparison after the intervention, the investigators will use ANCOVA with post-intervention rsFC as a dependent variable, and intervention, baseline rsFC, age, and intelligence quotient (IQ) as covariates. Z-scores of parameter estimates will be used to measure connectivity strength. Results will be reported with a 5% significance level adjusted for multiplicity by family-wise error rate. Z-statistics for each participant from the post-intervention rsFC maps will be used in a linear regression model to evaluate correlation between rsFC and behaviour change. Changes in grey and white matter volume will be assessed in a whole-brain scan using VBM, derived from the anatomical T1 image, acquired at the beginning of each fMRI scan. ROIs include the 6 seeds above as well as other areas identified in our previous review (cerebellum, superior temporal sulcus, temporo-parietal area). The investigators will use SPM12 on Matlab for standard preprocessing and analysis of VBM, and CONN for denoising and rsFC analysis.

crossover randomised controlled trial, music therapy, autism spectrum disorder, functional connectivity, functional magnetic resonance imaging (fMRI), social communication, grey and white matter

M4A is designed as an assessor blinded crossover RCT, comparing music therapy (MT) to a structurally matched play therapy (PT) intervention. Following the baseline assessment, a concealed random allocation to a sequence of interventions (MT-PT or PT-MT) will be implemented by one researcher who does not have contact with participants. Each intervention has a duration of 3 months and includes a 3 month washout period between interventions. Baseline assessments will be conducted before each intervention period.

To ensure blinding, the primary outcome measure will be completed by special educators who are familiar with the child (rather than their parents). Success of blinding (for all outcomes) will be verified at the last follow-up.

This sequence of interventions begins with 12 weeks of music therapy intervention, followed by a 3 month washout period and concluding with 12 weeks of play therapy intervention. Both interventions will consist of 12 weekly one-on-one sessions, 45 minutes each, conducted in the same setting by a licensed music therapist, in accordance with an intervention manual. Using a theoretically motivated approach, both interventions will target similar domains: creating a shared experience, building meaningful relationships, fostering self-expression. A varied set of activities combining therapist- and child-led interactions will target common goals: multisensory integration, verbal and social communication, emotion regulation, turn-taking, social appropriateness, and interaction. In both interventions, children can choose 4 activities per session using a visual schedule.

This sequence of interventions begins with 12 weeks of play therapy intervention, followed by a 3 month washout period and concluding with 12 weeks of music therapy intervention. Both interventions will consist of 12 weekly one-on-one sessions, 45 minutes each, conducted in the same setting by a licensed music therapist, in accordance with an intervention manual. Using a theoretically motivated approach, both interventions will target similar domains: creating a shared experience, building meaningful relationships, fostering self-expression. A varied set of activities combining therapist- and child-led interactions will target common goals: multisensory integration, verbal and social communication, emotion regulation, turn-taking, social appropriateness, interaction. In both interventions, children can choose 4 activities per session using a visual schedule.

Music therapy will use rhythmic cues, music instruments (piano, drums, djembe, xylophone, harmonica), songs, and stories accompanied by songs or musical instruments to target common goals.

Play therapy is designed as a play-based active comparison condition to control for factors such as support, therapist attention, positive expectancies, and emotional engagement. It will use verbal interaction, toys (Lego, finger puppets, Play Doh, puzzles), and the same stories as in MT, but without a musical component, to target common goals.

The Children's Communication Checklist-2 is a measure designed to assess the communication skills of children. The caregiver-report scale measures aspects of pragmatic communication with 70 items across 10 subdomains. The standard general communication composite standard score will be used as a measure of a child's general pragmatics and communication ability. Scores on the general composite of the Children's Communication Checklist-2 have a mean of 100 (SD=15). Higher scores indicate better social-communication skills.

Brain connectivity of frontotemporal regions will be measured as resting state functional connectivity (rsFC) from 6 seeds, will be used as the main neuroscientific outcome. Seeds will be anatomically defined regions of interest (ROIs) in Montreal Neurological Institute space for the left and right Heschl's gyrus, inferior frontal gyrus, and temporal pole.

The Child and Adolescent Scale of Participation is a caregiver-report scale consisting of 20 ordinal-scaled items across four subdomains (home participation, community participation, school participation, home and community living activities), that assesses participation. The 20 items are rated on a four-point scale: 1=Unable to participate, 2=Very limited, 3=Somewhat limited, 4=Age expected / Full participation. Higher scores mean a better outcome, indicating more participation in general. Minimum-maximum total scores: 25-100.

The 25 item scale is used to measure several aspects of families' perceived satisfaction in terms of quality of family life under five domains: Family Interaction, Parenting, Emotional Well-being, Physical / Material Well-being, and Disability-Related Support. The scale uses satisfaction as the primary response format. Participants are asked to rate their levels of satisfaction with certain statements on a 5-point scale, where 1 = very dissatisfied, 3 = neither satisfied nor dissatisfied, and 5 = very satisfied. Higher scores mean a better outcome as they indicate more satisfaction with the families' quality of life. Minimum-maximum total scores: 25-125.

The Peabody Picture Vocabulary Test- 4th edition is a measure used to assess receptive (hearing) vocabulary through 228 items (each with a spoken word + 4 pictures). Raw scores are calculated, then converted to standard score equivalents using tables from the manual. The converted scores have a mean of 100 and a standard deviation of 15. Scores from 70 to 85 are considered moderately low, and scores less than 70 are extremely low. Scores from 85 to 115 are considered average. Scores from 115 to 130 are considered moderately high and scores greater than 130 are extremely high. Higher scores indicate better receptive vocabulary.

Symptom severity will be assessed through the Social Responsiveness Scale, a 65-item rating scale measuring deficits in social behavior associated with autism. Caregivers rate on a 4-point Likert scale how often a statement has described a child's behavior over the past six months (1=not true to 4=almost always true). Adding all items generates a total score, with higher scores indicating greater severity of social difficulties and other behaviors associated with autism spectrum disorder. Additionally, the Social Responsiveness Scale yields five subscales: social awareness, social communication, social motivation, social cognition, and restricted and repetitive behaviors. Adding the items within each subscale yields a total score for each construct. Minimum- Maximum total score: 65-195.

The Vineland Adaptive Behavior Scales is a standardized assessment tool that utilizes semi-structured interview to measure adaptive behavior and support the diagnosis of intellectual and developmental disabilities, autism, and developmental delays. The maladaptive behaviours subdomain of the scale is used to identify the presence of behavior problems such as challenging internalizing and externalizing behaviours in children up to age 18. The scale is administered as a semi-structured interview to an informant who knows the child well. VABS v-scale scores have a mean of 15, (SD=3). Three categories are used to convey the degree of maladaptive behaviour in an individual corresponding to v-scale scores- Below 18: Average, 18-20: Elevated, 21-24: Clinically significant.

Chronic stress will be measured through hair cortisol concentration in the scalp-nearest 3cm segment, reflecting cumulative cortisol secretion over the past 3 months.

Changes in grey and white matter volume will be assessed in a whole-brain scan using voxel-based morphometry (VBM), derived from the anatomical T1 image, acquired at the beginning of each fMRI scan. ROIs include the 6 seedS: left and right Heschl's gyrus, inferior frontal gyrus, and temporal pole), as well as cerebellum, superior temporal sulcus, and temporo-parietal area.

Inclusion Criteria: Meet diagnostic criteria as specified in Diagnostic and Statistical Manual of Mental Disorders, Fifth edition (DSM-V) and officially diagnosed with autism spectrum disorder (ASD) by a licensed clinical professional using standardised diagnostic tools (Autism Diagnostic Observation Schedule (ADOS), Autism Diagnostic Interview Revised (ADI-R)). Exclusion Criteria: Recent or current music therapy Metallic or electronic implants

De-identified clinical and neuroimaging data will be made accessible for re-use by other researchers via platforms such as ENIGMA. We also plan to publish the primary outcome IPD on a public repository such as NSD (www.nsd.no)

De-identified clinical data will be stored in a publicly available repository (Open Science Foundation, https://osf.io/).

Geretsegger M, Fusar-Poli L, Elefant C, Mossler KA, Vitale G, Gold C. Music therapy for autistic people. Cochrane Database Syst Rev. 2022 May 9;5(5):CD004381. doi: 10.1002/14651858.CD004381.pub4.