Parental attitudes to a telehealth parent coaching intervention for autism spectrum disorder

Introduction

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterised by impairments in social-communication and repetitive and stereotyped behaviours, affecting 1% of the population. ¹ Children with ASD require early intervention. However their needs are often unmet by health services, both in the United States ² and in Europe, ³ with serious short- and long-term consequences in terms of their functioning, quality of life and familial stress.

Since treatment delivery costs, scarcity of trained professionals and home-to-clinic travel time are among the main barriers to accessing specialist autism services, particularly in rural areas, ⁴ telehealth can play a key role in providing timely and competent support to a larger number of families than those who can access traditional services. Parent training and parent coaching, effective in the traditional face-to-face mode of delivery, ⁵ are beginning to be tested in telehealth modality. The few existing studies report promising initial findings, but are limited by the small sample sizes (respectively nine, two, three and six subjects^6–9). Moreover, it remains to be ascertained whether these groups of parents, recruited through research institutions, are actually representative of the general population, and consequently whether the implementation of telehealth models is feasible in typical clinical settings. A telehealth model can lead to significant cost savings compared with a travel care model. ¹⁰ However, it is crucial to identify and address potential barriers to participation to ensure that the potential beneficiaries are engaged and clinical outcomes are maximised. In the field of telepsychiatry and telepsychology, feasibility studies have generally looked at barriers to implementation from the perspective of providers, ¹¹ but there is a need to understand parents’ attitudes towards telehealth, particularly among those living in rural areas of middle- or low-income countries, as they are those who face the most significant challenges in accessing services.

From the users’ perspective, acceptance of telehealth is typically associated with lower age, higher education and prior exposure to computer technology. As for the effect of human–technology interaction factors, such as computer anxiety or computer self-efficacy, there is mixed evidence due to a lack of comparability of measures or constructs across studies. ¹² The patients’ satisfaction with telemedicine is generally high, ¹³ and prospective users, including patients with physical and/or behavioural chronic conditions,^14–16 often report a positive attitude towards telehealth. While this evidence suggests the potential acceptability and feasibility of telehealth parent coaching for ASD, the limitations of existing studies (none of which were conducted in the community), combined with the characteristics of the syndrome, regarded a very impairing neurodevelopmental disorder, ¹⁷ requires the investigation of additional, ASD-specific factors.

A potentially relevant factor is that of parental self-efficacy, often compromised ¹⁸ due to the very high levels of stress entailed in raising a child with autism, which involves the management of persisting core^19,20 and associated^21,22 symptoms. A telehealth intervention places an increased responsibility on the parent to carry out psycho-educational techniques that can only be described or demonstrated but not fully performed by the therapist as would normally occur in a face-to-face intervention. It is therefore crucial to examine the contribution of self-efficacy alongside other well-known correlates of parents’ attitudes to telehealth interventions.

The present study examines the potential feasibility of tele-delivered parent coaching in a rural area of Italy through a survey of parental attitudes towards this type of intervention. Specifically, the study aims to (a) report on the prevalence of instrumental barriers that preclude participation in a telehealth parent coaching intervention, and (b) identify the determinants of willingness to enrol in such an intervention in the absence of instrumental barriers.

Method

Results

Factors associated with interest in telehealth

A logistic regression was conducted on the subsample of parents who had the instrumental means to receive the intervention potentially (n = 36; 85%). The outcome variable was interest in telehealth (‘yes’, n = 26; ‘no’, n = 10). The predictors in the model were parental self-efficacy, parental ability in Internet use, previous experience of video calling, parental educational level, time required to travel from home to the centre, parental satisfaction with the ASD clinic, and the child’s level of expressive verbal ability. The model’s chi-square statistic was significant (χ²[7]) = 19.680, p = 0.006). The Hosmer and Lemeshow’s goodness-of-fit test was not significant (χ²[8] = 7.087, p = 0.568), indicating a well-fitting model; the Nagelskerke R² indicated a moderately strong relationship between the predictors and the prediction (R²= 0.67). The model correctly classified 79% of cases. Parents with better Internet skills and those who reported higher levels of satisfaction with the ASD clinic were more likely to be willing to enrol. None of the other predictors was significantly associated with propensity to enrol in a tele-delivered intervention (Table 1).

OPEN IN VIEWER

Table 1.

Factors associated to willingness to enrol in a tele-delivered intervention.

	OR (p)	95% CI
Educational level	4.07 (0.346)	0.22–75.47
Internet skills	2.10 (0.016)	1.15–3.85
Experience with video calling (yes)	2.34 (0.453)	0.25–21.48
Satisfaction with ASD clinic	5.92 (0.035)	1.13–30.88
Travel time to ASD clinic	1.02 (0.640)	0.94–1.11
Parental self-efficacy	1.30 (0.065)	0.98–1.70
Child’s expressive ability	1.10 (0.711)	0.66–1.82

Note: Predictors significant at p < 0.05 have been marked in bold.

OR: odds ratio; CI: confidence intervals; ASD: autism spectrum disorder.

Reasons for refusing telehealth

A qualitative analysis was conducted with NVivo software to explore the responses of parents who were not willing to enrol in the tele-delivered parent coaching (n = 10). Themes were generated to categorise similar responses (Table 2). The refusal to enrol in the intervention was primarily due to a generic preference towards face-to-face interaction (n = 6), qualified by the majority of parents with further statements. These included reporting difficulties in using computers or the Internet (n = 3), the fear of not being able to communicate effectively with the clinician (n = 3) and the expectation that distance interaction would be lacking in warmth (n = 2). Two participants also expressed the opinion that a change in their habits would have not been welcome (one parent qualified it as anxiety provoking), and/or that adding a telehealth intervention would not have been necessary for them. One participant expressed concerns over the security of the Internet connection in terms of privacy.

OPEN IN VIEWER

Table 2.

Reasons for refusing telehealth.

	Subject 1	Subject 2	Subject 3	Subject 4	Subject 5	Subject 6	Subject 7	Subject 8	Subject 9	Subject 10	Frequency
Better in person	×		×		×	×	×			×	6
Use of computer/Internet		×			×						3
Not being able to communicate			×	×							3
Lack of warmth					×					×	2
Don’t want to change habits								×		×	2
Not necessary						×					2
Not secure connection		×		×			×				1
Anxiety provoking			×								1

Discussion

A timely identification of barriers to participation ahead of implementation plans of novel intervention practices is key to planning the allocation of resources effectively and successfully bridging the research-to-practice gap in treatment science. ²⁹

The main barrier to accessing telehealth lies, by definition, in the impossibility to receive the intervention due to a lack of Internet access, reported in our sample by 16% of respondents (in line with the Italian national data ²⁵ : 13%). It is notable that parents not accessing the Internet at home reported significantly worse expressive abilities in their children. Further research with larger samples should specifically target parents without Internet access to establish whether other forms of telehealth (such as telephone consultation) may be a more appropriate way to engage these hard-to-reach families.

Among those participants who were equipped to receive telehealth, 26 (72%) were willing to enrol in a tele-delivered parent coaching. Satisfaction with quality of services, a well-known correlate of better compliance, ³⁰ was the strongest predictor of willingness to enrol in the intervention. It is important to note that this did not reflect a difference in the amount of services received. All other factors being equal, parents with worse self-rated Internet skills were less likely to be willing to enrol in the intervention. All parents reported regularly using the Internet at home, which means they had at least some basic knowledge of the devices necessary for Internet browsing. This indicates that basic strategies such as the provision of tutorials on how to download and operate free video-calling software may be critical to increase the uptake of telehealth.

In this survey, the interviewer gave an accurate description of how the telehealth coaching would be implemented. However, a live demonstration was not performed, and we cannot exclude that this could have influenced the propensity to enrol in the intervention. Distance from health-care services is typically associated with reduced access to services. In our sample, the mean time needed to reach the ASD clinic was almost double the time typically needed to reach the nearest non-specialist health-care facility in rural areas of Italy (as reported in the latest WHO Health Survey ²⁶ : 15 minutes). However, parents living close to the clinic were as likely to be willing to enrol in the intervention as those living farther away were. This may indicate that different factors, other than cost and burden of transport, may play a role in parental interest in telehealth. A possible explanatory factor may be that telehealth reduces additional hidden costs of service use (such as childcare or time off work) which apply to all families.

The child’s level of verbal ability was not a significant predictor of parental choice of enrolment, suggesting that telehealth may be recommended to parents regardless of their children’s level of impairment. Parental self-efficacy levels were not significantly associated with choice of enrolling in the tele-delivered parent coaching. However, despite not reaching statistical significance (p = 0.065), the direction of the effect indicates a tendency for parents with lower self-efficacy to be less likely to be interested in enrolling in the intervention. This trend should be explored in further studies and, if confirmed, could be due to a number of different factors. One possible explanation may be that in an Internet delivered coaching model, parents have a stronger sense of being responsible for the outcome of the intervention than in the traditional face-to-face interaction, where the therapist is physically present and can directly intervene, if necessary. Future research should further explore whether this important dimension of parenting is associated with treatment choice, as it could potentially undermine the uptake of parent coaching among those who need it most, such as stressed and fatigued parents, who also tend to have lower self-efficacy levels. ³¹ The qualitative thematic analysis supports the above-reported results, highlighting the concerns for both technical and relational aspects of video calling as main factors driving the choice of not opting for telehealth.

The present study is the first contribution to the investigation of barriers to participation in tele-delivered parent coaching for ASD and, as such, can inform the planning of future feasibility studies. The study has a number of strengths, including the fact that we were able to reach 91% of the target population, which makes the results valid and meaningful for the local population, and the use of a mixed methodology to characterise barriers to the adoption of telehealth better. The study will enable the set-up of a randomised wait-list controlled trial of telehealth parent coaching which will build on the findings of existing studies to provide further evidence of its efficacy (or lack thereof) in a larger sample, representative of the local population. A preliminary test run of the intervention in this population showed good feasibility and acceptability. ³²

However, the study is limited by the small sample size and the fact that recruitment was carried out through one single clinic, which reduced the power of the statistical analysis and the generalisability of the findings. This is particularly relevant, as the factors considered (e.g. access to the Internet, attitudes towards technology) may vary regionally or culturally. Further research is needed to establish whether there are similar patterns of propensity to using telehealth in this population across rural areas.