Describing caregiver-child outcomes and caregiver perceptions of non-specialist caregiver coaching for autistic children in South Africa

Dawood, Zahra Ismail

Describing caregiver-child outcomes and caregiver perceptions of non-specialist caregiver coaching for autistic children in South Africa

Thesis / Dissertation

2023

Publisher

University of Cape Town

Department

Department of Pathology

Faculty

Faculty of Health Sciences

Abstract

Background: There is limited access to caregiver coaching using naturalistic developmental behavioural interventions (NDBI). Caregiver use of NDBI strategies can support growth in a child's joint attention skills and engagement, which in turn may promote child language acquisition. Caregiver's access to coaching may be increased using a cascaded task-sharing approach where coaching is delivered by non- specialists. When interventions are adapted to new contexts, implementation outcomes such as acceptability, appropriateness and feasibility need to be tracked alongside behavioural changes. In this proof of principle study, carried out in Cape Town, South Africa, caregiver-child behavioural outcomes and caregiver perceptions of the strategies that were taught in a task-sharing coaching intervention delivered by non-specialists are described. First, the Joint Engagement Rating Inventory (JERI) was used to assess changes in child behaviours, caregiver behaviours, and dyadic engagement in 10 caregivers and their children, in response to 12, 1-hour NDBI- informed caregiver coaching sessions. Then 10 another set of caregivers were interviewed to understand their outcomes and perceptions of the acceptability, appropriateness, and feasibility of the caregiver coaching strategies. Methods: 20 caregiver-child dyads (caregiver: ≥18 years old; child: DSM-5 diagnosis of autism spectrum disorder, 18-72 months, African or Coloured (mixed race)) received 12, 1-hour NDBI-informed coaching sessions. While delivery modality differed due to COVID-related restrictions (10 received in-person coaching and 10 received telehealth coaching), the coaching session content remained the same. The JERI was used to capture child joint engagement and social communication skills, caregiver ability to support and extend their child's abilities, and fluency of the dyadic interaction for 10 caregiver-child dyads. Summary statistics (median and interquartile range) were presented graphically for each participant (in-person coaching group). An additional 10 caregivers, in the telehealth coaching group, completed qualitative interviews to elicit perceptions of coaching strategies. Interviews were audio-recorded, transcribed verbatim and analysed using directed content analysis focusing on the acceptability, appropriateness, and feasibility of the coaching strategies. Results: There was a general increase in JERI caregiver items. The median for caregiver strategies including scaffolding and following in on child's focus increased from pre-intervention (Mdn=5) to post-intervention (Mdn=6). In addition, an increase was also seen in caregiver language facilitation and communicative temptations, pre- (Mdn=4) and post-intervention (Mdn=5). Children spent a similar amount of time in object (Mdn=2) and joint engagement (Mdn=6) states pre- and post- intervention. Children's responsiveness to their caregivers pre- and post-intervention was similar (Mdn=4.5) however, an increase in attention to caregiver was noted pre- (Mdn=3.5) and post-intervention (Mdn=4). The greatest increase was seen in child expressive language pre- (Mdn=2) and post-intervention (Mdn=4.5). In the qualitative interviews, caregivers reported that the strategies they learnt were acceptable and met their approval. They reported that the intervention was appropriate, as the strategies worked for them and helped to improve their child's social interaction, communication skills and behaviour. Lastly, caregivers reported that the strategies taught to them were feasible, and with sufficient time to practice the strategies, they became easier to incorporate into their daily routine. Caregivers also stated that due to the intervention they gained a greater understanding of their child, they noticed an increase in their own confidence, and indicated they were motivated to continue to use session strategies after the intervention was complete. Conclusion: The JERI can be used to detect changes in caregiver and child behaviours, as well as dyadic interaction following the NDBI coaching intervention in a low-resource South African context. Caregivers reported that the cascaded task- sharing NDBI approach was acceptable, appropriate, and feasible and they were able to integrate the strategies into their daily routines. Although further research is needed, these findings indicate that NDBI can be implemented in a low-resource South African context by caregivers after being coached by non-specialists.