Feasibility study of the National Autistic Society EarlyBird parent support programme

Sample characterisation

Demographic information on parental age, ethnic background, marital status and employment was obtained from parents at baseline. The ADOS-2 (Lord et al., 2012) was conducted during the baseline assessment to measure children’s autism characteristics (communication, social interaction, play, and restricted and repetitive behaviour). If the child had been diagnosed with autism within the previous 12 months, the ADOS-2 was not administered and scores were obtained from the diagnosing team. This was the case for 10 children and scores were received from the diagnosing team for five of these children. One clinical service did not complete ADOS assessments as part of their diagnostic procedures, so for three of the four children diagnosed by this service we conducted an ADOS-2 assessment at post-intervention. Of the 15 children who received an ADOS assessment either as part of the study or by their diagnosing clinical team, a module 1 ADOS-2 assessment was done with 11 children and module 2 assessments were completed with four children. As some diagnostic teams used the ADOS-G (Autism Diagnostic Observation Schedule–Generic; the earlier version of the tool; Lord et al., 2000) to assess autism severity, the algorithm scores are reported in the results with higher scores indicating more autism characteristics.

Parent-reported child autism characteristics were also measured at baseline using the Social Communication Questionnaire–Lifetime Version (SCQ; Rutter, Bailey, & Lord, 2003) and the Social Responsive Scale–second edition (SRS-2; Constantino & Gruber, 2012). The SCQ consists of 40 yes-no items that ask about the presence of autism traits. Scores range from 0 to 40 and higher scores indicate greater autism severity. Scores greater than or equal to 15 signify a possible autism spectrum condition. The SRS-2 is a 65-item questionnaire measuring the severity of autism by tapping into four aspects of social behaviour (receptive, cognitive, expressive, and motivational) and preoccupations. Items are rated on a 4-point scale ranging from ‘not true’ (1) to ‘almost always true’ (4). Total scores on the SRS-2 range from 65 to 260 with higher scores indicating greater autism severity. Scores of 76 or more suggest a clinical diagnosis of autism. The pre-school version of the SRS-2 (2½–4½ years) was used as deemed to be more appropriate for families in the current study.

An assessment of the child’s cognitive ability was conducted at baseline using the Mullen Scales of Early Learning (MSEL; Mullen, 1995). The MSEL measures fine and gross motor skills, visual reception, and expressive and receptive language. It provides T-scores for all domains in addition to a standard composite score (M = 100, SD = 15).

Child outcome measures

Adaptive behaviour was measured using the Vineland Adaptive Behaviour Scales–second edition (VABS-2; Sparrow, Cicchetti, & Balla, 2005), a semi-structured interview conducted with parents at baseline and post-intervention. It provides age equivalent and standardised scores for four domains of adaptive behaviour, including communication, socialisation, daily living and motor skills, together with an adaptive behaviour composite score (M = 100, SD = 15). If the child could remain by himself or herself in the assessment room, parental interviews were conducted simultaneously in a different room with a different researcher. Wherever possible, the same researcher conducted both time-point interviews. Increases in adaptive behaviour were expected after intervention.

Child emotional and behaviour problems were measured using the parent-report Strengths and Difficulties Questionnaire (SDQ; Goodman, 1997). The SDQ comprises 25 items that measure emotional, conduct and peer problems as well as hyperactivity and prosocial behaviour. Each of the five subscales consists of five items rated on a 3-point scale ranging from ‘not true’ (0) to ‘certainly true’ (2). Higher scores indicate more problems or prosocial behaviour, and a Total Difficulties score (0–40) is derived by adding together scores on all the subscales except prosocial behaviour. Scores of 16 or higher indicate clinically significant emotional or behavioural difficulties. Either the 3- to 4-year-old or 4- to 16-year-old versions were administered according to the child’s chronological age. We expected that the intervention would reduce child emotional and behaviour problems, reflected by lower scores on the SDQ at post-intervention.

Parent outcome measures

The Autism Parent Questionnaire (APQ; Anderson et al., 2006) was developed as part of an evaluation of EarlyBird in New Zealand to assess the effectiveness of the intervention. It consists of 27 items that are rated on a 6-point Likert-type scale ranging from ‘not true at all’ (1) to ‘definitely true’ (6). Greater autism knowledge is denoted by higher scores and it was anticipated that autism knowledge would increase following receipt of the intervention.

The Parental Sense of Competence (PSOC; Johnston & Mash, 1989) scale comprises 17 items measuring parenting satisfaction and parenting efficacy. Items are rated on a 6-point scale, ranging from ‘strongly disagree’ (1) to ‘strongly agree’ (6) with higher scores reflecting greater competence. Following EarlyBird, we expected parenting satisfaction and efficacy to increase.

Parental stress was measured using the Parental Stress Index–Short Form (PSI-SF; Abidin, 2012). This consists of 36 items rated on a 5-point scale, ranging from ‘strongly disagree’ (1) to ‘strongly agree’ (5). It measures parental distress, negative parent–child interactions and perception of a difficult child. Higher scores indicate greater parental stress (maximum score = 180) and it was anticipated that parenting stress would be lower after intervention.

A measure of positive parental wellbeing was obtained using the Warwick-Edinburgh Mental Well-Being Scale (WEMWBS; NHS Health Scotland, University of Warwick, & University of Edinburgh, 2006). The WEMWBS consists of 14 positively phrased items rated on a 5-point scale, ranging from ‘none of the time’ (1) to ‘all of the time’ (5), with higher scores indicting greater positive wellbeing (maximum score = 70). Positive wellbeing was predicted to increase following EarlyBird.

Intervention measures

For all families, rates of attendance were obtained from the EarlyBird facilitators. As there is no standard intervention fidelity measure of EarlyBird, a bespoke measure of provider-level intervention fidelity was designed for the study based on EarlyBird guidance for best practice. The resulting measure focuses on nine domains considered to be important for the delivery of EarlyBird: (1) knowledge is shared using jargon-free language, (2) parents are encouraged to problem solve themselves, (3) parents are encouraged to get to know one another, (4) taught strategies are personalised, (5) the facilitator creates an informal and relaxed atmosphere, (6) successes and progress are acknowledged, (7) small-group activities are conducted in pairs, (8) suggested timings are adhered with, and (9) the overall structure of the intervention is followed. Criteria for scoring each domain were based on the content of the certified materials from the EarlyBird providers. Fidelity was measured by the same researcher, who attended the sessions and coded them live. Each domain was rated as either present (1) or absent (0) resulting in a score from 0–9 for each session. Partially present domains (those which did not fulfil the full criteria defined for the domain) were assigned a score of 0.5. Seven of the EarlyBird programmes were assessed using this measure during one of their sessions (i.e. three during session 6, two during session 7, and two during session 8). The fidelity measure was not used to rate three of the programmes as they ended before the measure was developed.

Post-intervention interviews

Semi-structured interviews were conducted to explore parents’ and facilitators’ views of the intervention and perceived impacts. Parents were asked about their overall impression of the intervention, aspects they liked and suggestions for improvement, their views on the practical aspects and process of the intervention (e.g. the number of sessions, views on home visits), and about any differences the intervention had made. Facilitators were asked about recruitment, their views on the intervention and impacts of the course, and about the materials and practicalities of delivering EarlyBird.

The informants were (1) parents who attended EarlyBird (n = 6), representing views from four of the five different services delivering the intervention; (2) parents who were involved in the feasibility study (n = 3) and (3) facilitators who delivered EarlyBird from all five participating services (n = 10, two participated in a joint interview). Interviews were conducted once the intervention had finished; either at the participant’s home, the researcher’s workplace, where the EarlyBird programme was delivered, or over the phone. All interviews were conducted by the same researcher and audio-recorded.

The interviews were then transcribed and analysed independently by another member of the research team using an inductive thematic analysis approach based on grounded theory methods (Palinkas, 2014). After multiple readings of the transcripts, a coding scheme was developed based on identified themes and applied to the raw data. The framework method (Gale, Heath, Cameron, Rashid, & Redwood, 2013) was used to help reduce, code and display data for interpretation. A matrix displaying the summarised data was developed to facilitate analysis across and within participants. Identified key themes were then grouped together into conceptually related overarching themes. The parent and facilitator interviews were analysed using two separate thematic analyses. There was considerable overlap in themes that were identified from the interviews with parents and facilitators so in order to obtain a more comprehensive understanding and triangulate the findings, the themes across different participant groups are presented together (Patton, 1999). The interpretation of themes was checked by the researcher who conducted the interviews to ensure that the analysis adequately captured the interviews.

Reasons why parents took part in the study

Three parents were interviewed at post-intervention about their experiences of taking part in the feasibility study. All mentioned that receiving a report about their child’s functioning was a key reason for participating. Another key reason for participation was to help evaluate EarlyBird and identify areas for improvement so other parents could benefit in the future.

Completeness and suitability of measures

Table 3 summarises the characteristics of the sample at baseline. The high completion rates (15–16/17) of these assessments indicate that they were appropriate and feasible to administer. Ten children (56% of the sample) had been diagnosed within the previous 12 months so local diagnostic teams were asked to provide ADOS scores. In one service, ADOS assessments were not conducted as part of diagnosis, so these were completed at post-intervention for three of the four families. For the remaining child, ADOS scores were not received from the diagnostic team prior to termination of the study and the research team missed the opportunity to administer the assessment at post-intervention. For children who completed the ADOS, scores around or above the cut-off indicated the appropriateness of the assessment for this group. High scores on the SCQ and SRS-2 measures of parent-reported autism characteristics also suggest that the families invited to take part in EarlyBird were suitable recipients.

OPEN IN VIEWER

Table 3.

Child characterisation measures.

Measure	M	SD	n
MSEL Early Learning Composite Standard Score	60.18	22.31	17
MSEL Visual Reception T Score	27.47	15.85	17
MSEL Fine Motor T Score	25.41	13.21	17
MSEL Expressive Language T Score	23.94	9.98	17
MSEL Receptive Language T Score	27.00	15.90	17
ADOS a Social Affect Raw Total	9.47	4.26	15
ADOS a Restricted and Repetitive Behaviours Raw Total	13.20	5.66	15
SCQ Total	21.06	5.09	16
SRS-2 Total	75.19	11.28	16

Higher ADOS, SCQ and SRS-2 scores indicate more severe autism symptoms. MSEL: Mullen Scales of Early Learning; SCQ: Social Communication Questionnaire; SRS-2: Social Responsiveness Scale.

a

ADOS: Autism Diagnostic Observation Schedule, scores are provided from different modules according to verbal ability and chronological age and different scoring forms (i.e. ADOS-G and ADOS-2).

Table 4 displays completion rates for each parent and child outcome measure and the pre- and post-intervention scores and effect sizes (Cohen’s d) for families who completed measures at both time points. Retention of families for post-intervention assessments was high with the majority (15/17 families, 88.2%) completing some of the measures at post-intervention. The two families who did not complete post-intervention assessments had dropped out of the EarlyBird intervention and were unable to be contacted.

OPEN IN VIEWER

Table 4.

Completion rates and group Ms and SDs at pre- and post-intervention scores on child- and parent-related outcome measures.

Measure	Completion rates		Group Ms and SDs for paired sample at pre- and post-intervention
	Pre(N = 17)	Post(N = 15)	Pre(N = 11–12)		Post(N = 11–12)		d a	Direction of effect
	n	n	M	SD	M	SD
Child-related outcomes
VABS Adaptive Behaviour Composite Standard Score	14	15	65.58	11.16	70.00	15.82	0.33	↑ b
VABS Communication Standard Score	14	15	66.83	22.09	71.67	25.26	0.20	↑ b
VABS Daily Living Skills Standard Score	14	15	67.67	14.32	67.00	12.42	−0.05	–
VABS Socialisation Standard Score	14	15	64.83	6.67	68.83	10.48	0.47	↑ b
VABS Motor Skills Standard Score	14	15	73.92	11.17	81.92	14.68	0.62	↑ b
SDQ Total Difficulties	14	13	30.36	3.30	30.36	3.98	0.00	–
SDQ Emotional Problems	14	13	7.36	1.91	8.00	2.15	−0.31	↑
SDQ Conduct Problems	14	13	6.91	2.43	7.00	1.41	−0.05	–
SDQ Hyperactivity	14	13	8.55	1.81	8.64	2.25	−0.04	–
SDQ Peer Problems	14	13	7.55	1.75	6.73	1.79	0.46	↓ b
SDQ Prosocial Behaviour	14	13	7.36	1.63	7.73	1.56	0.23	↑ b
Parent-related outcomes
APQ Total	14	12	114.36	18.33	121.55	15.17	0.43	↑ b
PSOC Efficacy Total	14	13	28.55	5.09	30.91	7.38	0.37	↑ b
PSOC Satisfaction Total	14	13	32.45	8.79	33.91	8.19	0.17	–
PSI-SF Total Stress	14	13	102.82	16.17	100.64	16.84	0.13	–
WEMWBS Total	14	13	45.73	9.26	48.45	8.76	0.30	↑ b

n = 14 at baseline due to missing data on individual questionnaires. ↑ and ↓ denote increase and decrease, respectively, in scores from pre- to post-intervention. For the VABS, APQ, PSOC and WEMWBS, higher scores indicate more positive outcomes and for the SDQ and PSI-SF, lower scores indicate more positive outcomes. VABS: Vineland Adaptive Behaviour Scales; SDQ: Strengths and Difficulties Questionnaire; APQ: Autism Parent Questionnaire; PSOC: Parental Sense of Competence; PSI-SF: Parental Stress Index–Short Form; WEMWBS: Warwick-Edinburgh Mental Well-Being Scale.

a

Cohen’s ds are based on those with pre- and post-intervention scores (n = 12 for VABS and n = 11 for all other measures).

b

Effect in the predicted direction.

Completion rates of the questionnaires were slightly lower than the direct assessments (the lowest completion rate was 12/17 for the APQ, 70.6%). One parent failed to return the questionnaires at post-intervention due to a lack of time. Of the three parents who were involved in the feasibility study and completed post-intervention interviews, two reported that they felt the number of questionnaires they were asked to complete was burdensome. One other family withdrew from the study before starting the assessments as the questionnaires were considered too intrusive.

The Vineland interviews had high levels of completion. Group mean adaptive behaviour scores were higher at post-intervention but in some cases substantial changes coincided with starting school. SDQ scores indicated that most children displayed behaviours that may challenge, such as high levels of emotional and conduct problems as well as hyperactivity. Except for emotional problems, SDQ scores for the group were lower at post-intervention suggesting that the measure may be sensitive to change.

With regard to parent measures, one parent refused to complete the APQ stating that the rating scale was difficult to understand and they disliked the wording of the questions. Scores for the PSOC Efficacy subscale, the PSI and the WEMWBS were in the expected range and appeared to be sensitive to change; for example, group mean parenting efficacy and satisfaction scores were higher after attending EarlyBird.

Attendance

Parents attended an average of 6.6 out of the 8 sessions (median = 7, range = 2–8; n = 16) indicating that the intervention and number of sessions were acceptable to families.

Intervention fidelity

Intervention fidelity was moderate to high across the seven EarlyBird programmes that were evaluated using the measure (median = 7 out of 9 domains, range = 4–8). However, intervention fidelity varied across the different courses, with one course obtaining a total fidelity rating of four out of the nine domains suggesting that some sessions were not being delivered according to the manual (e.g. some topics were shortened, skipped or swapped for topics the facilitators considered more relevant for families, such as specific tips for toileting or feeding).

Views on the intervention

Ten key themes emerged from the interviews conducted at post-intervention with parents and facilitators which were then grouped into two overarching themes. They covered the following: (1) positive aspects of the intervention and (2) challenges to the delivery of EarlyBird. Descriptions and quotes to illustrate the themes are presented in Table 5. Pseudonyms are used to ensure individuals’ identities remain confidential.

OPEN IN VIEWER

Table 5.

Themes that emerged from post-intervention interviews with parents and facilitators.

Overarching theme	Description	Who by	Quote
Theme
Positive aspects of the intervention
Increased parental knowledge and skills	Positive views on the intervention content and materials emerged. EarlyBird was seen to extend parents’ understanding of autism and their child’s needs. Benefits included being more effective in their communication with their child and adapting activities for their child’s needs resulting in improved parent–child relationships. Improvements in their strategies for dealing with challenging behaviour by planning and avoiding triggers were described.	Parents and facilitators	‘The program was excellent because it gave me much more insight into autism and all the different spectrums’. – Parent‘Before I started I must admit I used to be pulling out my hair and screaming at my child but that was purely because I didn’t understand him at all. So when he started to scream and his behaviour problems kicked in I never dreamed of looking on the settings or actions of what triggered him off’. – Parent
Parents as advocates	Parents were more able to advocate for their child and their needs, appearing more confident.	Facilitators	‘[talking about changes in parents] … Confidence to talk to family members and friends and the school. Parents are empowered. They feel they’ve got a little knowledge base now’. – Facilitator
Improvements in parental stress and wellbeing	Reductions in parental stress and improvements in wellbeing were described.	Parents and facilitators	‘Some of the families look a lot happier. They appear more relaxed and feel they have a network around them. They feel much [more] in tune with their own child after doing EarlyBird. Before the programme all they think is about the behaviour and the things they can’t do, whereas after the programme they are more in tune about what the children can do’. – Facilitator‘I’m a lot calmer and I don’t scream at him anymore. More in control’. – Parent
Supportive environment for parents	The group-based nature of EarlyBird enabled parents to meet others with young children with autism, creating a sense of belonging by being able to relate to others and enhancing motivation to implement strategies. Home visits were seen to compliment the group sessions and facilitators were viewed as skillful and approachable.	Parents and facilitators	‘[talking about aspects they liked about the intervention]. . .When we as parents share our own experiences. Everyone has kids with the same condition and it’s a sense of belonging. You know you’re not the only one going through this. You can have help. What Sarah did was encouraging us to put everything we’ve learn into practice and telling us not to give up’. – Parent
Changes in children	Changes in children included playing with others and initiating interaction more, and being more co-operative and easier to manage. Perceived changes in children varied in size.	Parents	‘We can play together now and he’ll come to me by himself. If I’m playing with his brother, he’ll join us spontaneously. He likes playing with his brother now’. – Parent
Challenges to the delivery of EarlyBird
Relevance of content for families	Due to the heterogeneity of autism, the relevance of content varied depending on the needs of the participating families. Information wasn’t as helpful or relevant to parents who had prior knowledge of autism, although appeared to act as a useful reminder.	Parents and facilitators	‘For me the PECS stuff wasn’t useful because my child’s verbal, but obviously it’s a very wide spectrum. I don’t know in the future if it may be better to group parents in terms of having a non-verbal group and a verbal group’. – Parent‘Some of the information presented wasn’t relevant and stuff that I’d already gone through. It was nice to get a reminder but it was very basic’. – Parent
Deviations from the content and structure of EarlyBird	Content or examples used were adapted to make them more relevant to the needs of participating families. Deviations from the structure of the interventions occurred to enhance parental experience of the intervention and maintain engagement.	Facilitators	‘I’ll suggest things they can start doing before we do the session [on behaviour] because it’s last session and that’s a long wait when they are having difficulties’. – Facilitator
Additional content	Additional content covering managing behaviour, theory of mind, and developing resilience (e.g. developing friendships, social skills), as well as helping parents care for themselves and their families, and being able to evaluate evidence for alternative therapies was seen to improve the intervention.	Parents	‘More on helping to make friends and developing theory of mind. Social skills training would have been really helpful. The social stories were helpful and some of the books but there need to be more. We’re working on theory of mind with my child now and there are things that you can do’. – Parent
Ideal group size	Having too many families resulted in insufficient time for discussion and was seen to reduce any benefits of the intervention. Groups of between 6 and 10 were perceived as being large enough for discussion as well as manageable and cost-effective.	Parents and facilitators	‘Eight is probably a good number. I think that six is almost too small because it’s a huge resource in terms of our time. … The whole day for six families and then you have one or 2 that drop out and you’re down to 4? I think it’s very hard to justify. 12 is too many. It doesn’t give the parents enough time to really talk and to go through the iceberg [an EarlyBird strategy] themselves’. – Facilitator
Lack of time	A lack of time to implement strategies likely impacts on outcomes.	Parents	‘I’m a single mum and I was trying to work as well and I don’t have much time for it. Now I’ve moved [in] with my parents and I can claim benefits so I can concentrate more on Louie’. – Parent

Parents and facilitators talked about a number of positive aspects of the intervention (see Table 5 for themes). The content of the EarlyBird programme was viewed as informative and improvements in parental knowledge and skills were reported by parents and facilitators, with parents now feeling they had more confidence to advocate for their child. Improvements in parental stress and wellbeing were also described. The mode of intervention delivery (mix of group sessions and home visits) was also viewed as favourable, by creating a supportive environment for parents to share experiences in addition to opportunities to practice intervention content in naturalistic environments with support from experienced facilitators. Changes in children’s communication skills and behaviour were also mentioned, and improvements were generally attributed to EarlyBird along with other therapies the parent and child had received.

Both parents and facilitators also talked about various challenges to the delivery of EarlyBird. These were coded into five themes (see Table 5). Some of the content appeared less relevant for some families and additional content was suggested by some parents. Indeed, it appeared that facilitators occasionally deviated from the manual to make content more relevant for the specific needs of the participating families or to ensure that parents remained engaged throughout the intervention and enhance their experience. The size of the group was an important factor for both parents and facilitators, who wanted to ensure there was sufficient time for discussion, and that the delivery of such interventions was manageable and cost-effective. Groups that were too large were considered to have a negative impact on the effects of the intervention by limiting opportunities for parents to share their experiences.