View in EDS HTML Full Text PDF Full Text

Performance of the Autism Observation Scale for Infants with Community-Ascertained Infants Showing Early Signs of Autism

Saved in:
Bibliographic Details
Title: Performance of the Autism Observation Scale for Infants with Community-Ascertained Infants Showing Early Signs of Autism
Language: English
Authors: Hudry, Kristelle (ORCID 0000-0002-2752-8345), Chetcuti, Lacey, Boutrus, Maryam, Pillar, Sarah, Baker, Emma K., Dimov, Stefanie, Barbaro, Josephine (ORCID 0000-0001-7359-0097), Green, Jonathan, Whitehouse, Andrew J. O., Varcin, Kandice J.
Source: Autism: The International Journal of Research and Practice. Feb 2021 25(2):490-501.
Availability: SAGE Publications. 2455 Teller Road, Thousand Oaks, CA 91320. Tel: 800-818-7243; Tel: 805-499-9774; Fax: 800-583-2665; e-mail: journals@sagepub.com; Web site: http://sagepub.com
Peer Reviewed: Y
Page Count: 12
Publication Date: 2021
Document Type: Journal Articles
Reports - Research
Descriptors: Autism, Pervasive Developmental Disorders, Infants, At Risk Persons, Toddlers, Child Development, Siblings, Diagnostic Tests, Test Validity, Symptoms (Individual Disorders), Psychometrics, Foreign Countries, Child Behavior
Geographic Terms: Australia
Assessment and Survey Identifiers: Autism Diagnostic Observation Schedule, Vineland Adaptive Behavior Scales, Mullen Scales of Early Learning
DOI: 10.1177/1362361320965397
ISSN: 1362-3613
Abstract: Often included within 'high-risk sibling' studies, the Autism Observation Scale for Infants (AOSI) has only one independent replication study and no evaluation with community-ascertained cohorts. We administered the AOSI and established clinical measures with 103 infants (68% male) at 'high autism likelihood' on the Social Attention and Communication Surveillance - Revised (SACS-R) tool, at 9-14 months of age and again 6months later. AOSI Total scores showed adequate internal consistency and strong inter-rater agreement (live- or video-coded) and were approximately normally distributed at each visit. Modest significant associations presented between Time 1 AOSI scores and concurrent developmental/adaptive skills measures. Concurrent associations were stronger at Time 2, particularly between AOSI Total and Autism Diagnostic Observation Schedule (ADOS) Social Affect scores. AOSI scores were only moderately associated across Time 1 and 2 assessments, as were Time 1 AOSI with Time 2 ADOS scores. These data from a clinically indicated cohort broadly replicate previous AOSI validity accounts from 'high-risk sibling' studies, particularly beyond the first year. Strong inter-rater agreement indicates viable AOSI inclusion within protocols necessitating blinded evaluation (e.g. intervention trials). Moderate within-participant stability suggests that, like 'high-risk siblings,' community-ascertained infants experience variable early trajectories. [This article was co-written by the AICES Team.]
Abstractor: As Provided
Entry Date: 2021
Accession Number: EJ1285280
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwE4LdfSEQd17w_mjghThwdFAAAA4jCB3wYJKoZIhvcNAQcGoIHRMIHOAgEAMIHIBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDG5XPxLyQArBgVIuIwIBEICBmu9a9uD3PeWp2KSZ3W-dVIZ7uykbRs6eZqsLgUM-C8pkXMlSHwYxBsEurkP0jiLvTMUzL94kZ6cugb2-lTtoIyPU3_CfylxYYbk5Rt0TC9hnXmMpEevfCyEGgqRVlZOhi9a-7zwYNbj9R-NLfoj-UphD72jUievDA4b3_M8DwArW48L1u8QPnD_Si8tdKfvp8THoTeIXnNyiLos=
Text:
  Availability: 1
  Value: <anid>AN0148624091;f9d01feb.21;2021Feb12.02:02;v2.2.500</anid> <title id="AN0148624091-1">Performance of the Autism Observation Scale for Infants with community-ascertained infants showing early signs of autism </title> <p>Often included within 'high-risk sibling' studies, the Autism Observation Scale for Infants (AOSI) has only one independent replication study and no evaluation with community-ascertained cohorts. We administered the AOSI and established clinical measures with 103 infants (68% male) at 'high autism likelihood' on the Social Attention and Communication Surveillance - Revised (SACS-R) tool, at 9–14 months of age and again 6 months later. AOSI Total scores showed adequate internal consistency and strong inter-rater agreement (live- or video-coded) and were approximately normally distributed at each visit. Modest significant associations presented between Time 1 AOSI scores and concurrent developmental/adaptive skills measures. Concurrent associations were stronger at Time 2, particularly between AOSI Total and Autism Diagnostic Observation Schedule (ADOS) Social Affect scores. AOSI scores were only moderately associated across Time 1 and 2 assessments, as were Time 1 AOSI with Time 2 ADOS scores. These data from a clinically indicated cohort broadly replicate previous AOSI validity accounts from 'high-risk sibling' studies, particularly beyond the first year. Strong inter-rater agreement indicates viable AOSI inclusion within protocols necessitating blinded evaluation (e.g. intervention trials). Moderate within-participant stability suggests that, like 'high-risk siblings,' community-ascertained infants experience variable early trajectories. We investigated whether a commonly used research assessment – the Autism Observation Scale for Infants (AOSI) – accurately measures autism behaviours among infants showing early signs of autism identified within the community. The AOSI is often included in studies tracking the development of infants at increased likelihood of autism, such as the infant siblings of diagnosed children. However, the suitability of this measure has not previously been tested with community-referred infants. We administered the AOSI with infants when aged 9 to 14 months and again 6 months later. Our researchers – independent of the AOSI development team and newly trained on this measure – were able to administer the brief interactive assessment and score it accurately. The infants' AOSI scores were linked to their scores on other established and validated clinical assessments, particularly at the second visit when average age was 18 months. Stronger correspondence of AOSI and other scores at this second visit suggests early autism behaviours are better established and more consistent by 18 months of age, even though these infants showed clear enough signs of possible autism to prompt referral to our study around 12 months of age. However, the moderate association of AOSI scores over time suggests that, like infant siblings – who mostly do not develop autism – community-identified infants showing early signs may also have variable developmental pathways in early life.</p> <p>Keywords: developmental psychology; diagnosis; early signs; individual variability/heterogeneity; infants; psychometrics</p> <p>The Autism Observation Scale for Infants (AOSI; [<reflink idref="bib9" id="ref1">9</reflink>]) is a brief, semi-structured observational assessment to quantify early behaviours associated with Autism Spectrum Disorder (hereafter, <emph>autism</emph>) among infants aged 6–18 months. Behaviours are observed within interactive free-play periods and more structured activity bids, and codes capture features of visual-attention tracking, emerging social-communication and social responsiveness, and broader behavioural responses including temperamental reactivity, and sensory/motor control and atypicality. AOSIs are conventionally scored live, and 16 key items are summed for a Total score (range: 0–38), while a Number of Markers (NoM) count is computed as the number of items with <emph>non-zero</emph> score (range: 0–16).</p> <p>Developed to meet the need for an objective tool to characterise the nature and timing of emergent autism behaviours at a younger age than that addressed by other available instruments ([<reflink idref="bib6" id="ref2">6</reflink>]; [<reflink idref="bib8" id="ref3">8</reflink>], [<reflink idref="bib9" id="ref4">9</reflink>]), the AOSI has been widely incorporated in international research. This includes in prospective studies tracking the early development of 'high-risk siblings' (e.g. [<reflink idref="bib3" id="ref5">3</reflink>]) and infants in other 'risk' groups, including with pre-term birth ([<reflink idref="bib24" id="ref6">24</reflink>]), Tuberous Sclerosis Complex ([<reflink idref="bib17" id="ref7">17</reflink>]) and Fragile X Syndrome ([<reflink idref="bib12" id="ref8">12</reflink>]). To date, there has been only one independent validity study ([<reflink idref="bib11" id="ref9">11</reflink>]) and the AOSI has not been validated for clinical use ([<reflink idref="bib9" id="ref10">9</reflink>]; [<reflink idref="bib7" id="ref11">7</reflink>]).</p> <p>[<reflink idref="bib25" id="ref12">25</reflink>] provided the first report on AOSI validity in the context of a moderately sized prospective sample of 65 high-risk siblings and 23 low-risk controls. AOSI scores at 12 months, but not at 6 months, differentiated infants who did/not subsequently receive an autism diagnosis at age 24 months. Specifically, six of seven later-diagnosed children had 12 month AOSI NoM counts ⩾7, compared to only 2 of 81 infants who did not develop autism (i.e. 84% sensitivity; 98% specificity). Subsequent reports from the same group have examined data from key subgroups in greater detail. For example, in a case series of high-risk infants – each first assessed at 6 months and ultimately diagnosed with autism ~3 years – [<reflink idref="bib8" id="ref13">8</reflink>] reported 12 month AOSI Total scores ⩾9 for six of nine infants. Similarly, [<reflink idref="bib6" id="ref14">6</reflink>] found that among 14 infants meeting diagnostic criteria for autism by 3 years, 11 had AOSI Total scores exceeding this threshold at 12-month assessment. In their more detailed psychometric report on the AOSI, [<reflink idref="bib9" id="ref15">9</reflink>] indicated moderate 2-week test–retest reliability for infants aged 12 months (Total = 0.61; NoM = 0.68), while inter-rater agreement, across examiner- and live-observer scores, modest at 6 months (Total = 0.74; NoM = 0.68), was far stronger at 12 (Total = 0.93; NoM = 0.92) and 18 months (Total = 0.92; NoM = 0.90).</p> <p>In a more recent independent study, [<reflink idref="bib11" id="ref16">11</reflink>] broadly replicated this validity evidence in a UK-based sample of 54 high-risk siblings and 50 low-risk controls. AOSI assessments were completed with each infant ~7 and 14 months of age, with follow-up at ~2 and 3 years (and diagnostic outcome determined at the latter visit). AOSI Total scores showed excellent inter-rater agreement at both 7 and 14 months (⩾0.95). Compared to low-risk infants, high-risk infants subsequently diagnosed with autism had higher 14 month AOSI Total scores. Within the high-risk group, however, AOSI Total scores were only marginally higher among those infants who were vs were not later diagnosed with autism (and this effect was strengthened and statistically significant when subgroup differences in developmental/cognitive ability were controlled). Furthermore, AOSI Total scores for high-risk siblings with non-autism outcome did not differ from those of low-risk controls. Gammer et al. also examined associations between infants' AOSI Total scores and the well-established Autism Diagnostic Observation Schedule (ADOS; [<reflink idref="bib16" id="ref17">16</reflink>]); often considered the 'gold-standard' observational assessment informing autism diagnosis. AOSI Total scores at 14 months were moderately positively associated with subsequent ADOS scores (<emph>r</emph> = 0.3–0.4 against Module 1 and 2 algorithm total scores at ages 24 and 37 months). These data support the moderate predictive validity of earlier AOSI scores, from around the first birthday, for later autism symptom presentation and clinical best-estimate diagnostic classification.</p> <p>[<reflink idref="bib25" id="ref18">25</reflink>] raised the possibility that the AOSI could support future early identification efforts from as early as the first birthday. However, this group have also cautioned against assuming that findings from high-risk sibling cohorts necessarily generalise to samples of children with autism ascertained in other ways, with recent empirical data emerging to support this point (e.g. [<reflink idref="bib20" id="ref19">20</reflink>]). Dedicated investigation of the performance of the AOSI as a measure of emerging autism behaviours beyond the context of 'high-risk siblings' is therefore warranted.</p> <p>Embedded within a large pre-emptive intervention trial, we conducted an independent evaluation of AOSI validity with a sample of infants all showing early behavioural markers of autism between 9 and 14 months of age. Infants completed AOSI and other standardised, norm-referenced assessments at an initial visit shortly after referral, and again ~6 months later, with the Autism Diagnostic Observation Schedule – Toddler Module (ADOS-T; [<reflink idref="bib15" id="ref20">15</reflink>]) also administered at the latter visit. We consider psychometric properties of AOSI scores from this community-ascertained cohort assessed at mean ages 12 and 18 months, including indices of (a) scoring reliability across live and video coding protocols, (b) convergent validity with concurrently administered clinical phenotyping measures and (c) stability and predictive validity of scores from earlier versus later assessment.</p> <hd id="AN0148624091-2">Methods</hd> <p></p> <hd id="AN0148624091-3">Design and patient and public involvement</hd> <p>Data for this study came from baseline and immediate outcome assessments conducted with 103 infants recruited into a clinical trial of pre-emptive intervention for infants showing early signs of autism ([<reflink idref="bib23" id="ref21">23</reflink>]). Approval for the trial was provided by the Child and Adolescent Health Service Ethics Committee (2016008EP) and all aspects of design and execution, including interpretation of findings, included input from individuals from the autism community (i.e. parent of children with autism, practicing community professionals) who are named co-authors.</p> <p>Infants were aged 9 to 14 months at referral and eligible for the trial on the basis of showing ⩾3 of 5 behavioural autism markers from the 12 month checklist of Barbaro and colleagues' Social Attention and Communication Surveillance - Revised (SACS-R) tool ([<reflink idref="bib1" id="ref22">1</reflink>], [<reflink idref="bib2" id="ref23">2</reflink>]; [<reflink idref="bib18" id="ref24">18</reflink>]) – atypicality on three or more of eye-contact, response to name, proto-declarative pointing, other social gesture use (e.g. waving) and imitation. Other inclusion/exclusion criteria included sufficient parental English language to permit active trial participation, absence of diagnosed infant neurological/developmental comorbidity (including gestation <32 weeks), and no family intention to relocate away from the trial site/s across a 2 year follow-up period. While the same eligibility criteria applied to families recruited across two trial sites, the specific ascertainment process differed according to local community service provision ([<reflink idref="bib23" id="ref25">23</reflink>]). Briefly, in Perth, infants were identified upon engagement with the local government service for children with developmental delays, typically accessed following parental self-referral or upon community health professional advice. Newly engaged infants within the target age range were identified by a researcher who subsequently conducted telephone eligibility screening, including completion of the SACS-R. In Melbourne, potentially eligible infants were identified by community Maternal and Child Health (MCH) nurses who had received SACS-R trained by the study team and subsequently made direct referrals to the trial. Upon referral receipt, a trial researcher contacted families to conduct formal telephone eligibility screening, including completion of the SACS-R as at the Perth site.</p> <p>Each infant participant was administered the AOSI and other standardised clinical characterisation measures at a trial baseline assessment (hereafter, Time 1) and again ~6 months later (Time 2; immediate treatment end-point for those randomised to the intervention). Mean infant age was 12 months at Time 1, and 18 months at Time 2, with 94% (<emph>n</emph> = 97) retention to follow-up. [<reflink idref="bib23" id="ref26">23</reflink>] report detailed sample characterisation, including evidence of treatment vs control group similarity at baseline on three characterisation factors (i.e. age, sex, and referral markers; informing stratified randomisation) and key measures for the current study (i.e. the AOSI and other established clinical measures). Also reported by Whitehouse et al. there was no treatment effect on infants' Time 2 AOSI scores (between-group difference −0.74 (95% CI: −2.47 to 0.98)) nor on Time 2 Mullen domain scores (see Table 2 in original trial outcome report). Furthermore – and not previously reported by Whitehouse et al. – we verified the absence of any between-group difference on Time 2 ADOS-T domain algorithm scores (Social Affect: 0.10 (95% CI: −1.69 to 1.89); Restricted Repetitive Behaviour: −0.23 (95% CI: −1.00 to 0.55)) or VABS Adaptive Behaviour Composite (−2.53 (95% CI: −7.91 to 2.86); though we note Whitehouse et al. did identify an effect favouring the treatment group specific to the VABS Communication domain; 6.43 (95% CI: 1.06 to 11.81)). Given substantive similarity of treatment versus control groups, we proceeded with the current analyses treating participants as a single group.</p> <hd id="AN0148624091-4">Assessment procedure and measures</hd> <p>Written informed consent was obtained from primary caregivers at study entry. AOSI and other direct clinical assessments and parent-report measures were completed with infants and their caregivers at each visit, at a lab/clinic site (or at Time 2 home visit to support retention (<emph>n</emph> = 2)). The AOSI ([<reflink idref="bib9" id="ref27">9</reflink>]) is completed in around 15–20 min and includes semi-structured and structured elements designed to elicit a range of social-communication, engagement and other behaviours, scored across 19 operationally defined codes. Higher item-level scores indicate greater atypicality, and we followed previously published protocols to compute an AOSI Total score (sum of 16 key codes) and NoM count (number among 16 codes with non-zero scores) for each infant at each visit.</p> <p>Two well-established clinical assessments were also completed at each visit. The Mullen Scales of Early Learning (MSEL; [<reflink idref="bib19" id="ref28">19</reflink>]) is a standardised, norm-referenced examiner-administered assessment of verbal- and non-verbal developmental abilities suitable for use with children from birth through early childhood. The Vineland Adaptive Behaviour Scales–Second Edition (VABS-II; [<reflink idref="bib21" id="ref29">21</reflink>]) is an informant-report measure, suitable for use across the lifespan. The VABS-II Parent/Caregiver Form was utilised here to assess child adaptive communication, socialisation, personal independence and motor skills. Key measures retained for the current analysis were the MSEL Early Learning Composite (ELC), and VABS-II Adaptive Behaviour Composite (ABC), both Standard Scores (population <emph>M</emph> = 100; <emph>SD</emph> = 15) providing broad appraisal of child learning abilities/adaptive behaviour against age-related expectations.</p> <p>Finally, examiners also administered the ADOS-T with each infant at Time 2, for a measure of early autism symptoms that has been developed and clinically validated for use with children from 12 months of age (i.e. not the case for all of our cohort at Time 1 who ranged upwards of 9 months). Like the AOSI, the ADOS-T is a play-based assessment combining more structured presses with less structured free-play time to elicit a range of social-communication, engagement, and other behaviours which are scored against operationally defined codes, among which some contribute to algorithms for which thresholds of autism concern have been delineated. Unlike the AOSI, the ADOS-T assumes some independent mobility (i.e. to walk a few steps independently) and non-verbal mental age ⩾12 months ([<reflink idref="bib15" id="ref30">15</reflink>]). While this was not always the case for our infants at Time 2, we nevertheless completed ADOS-T administration with all infants and used published protocols to compute ADOS-T algorithm Total scores for Social Affect and Restricted Repetitive Behaviour domains and classified each infant within an ADOS-T 'Range of Concern' (i.e. little-to-no, mild-to-moderate, moderate-to-severe concern; [<reflink idref="bib15" id="ref31">15</reflink>]).</p> <hd id="AN0148624091-5">Behavioural coding procedures</hd> <p>Researchers responsible for conducting assessments for this trial completed formal training with the AOSI developers and local ADOS trainers and demonstrated high administration standard and coding agreement (i.e. three consecutive practice assessments administered to fidelity and with ⩾80% consensus coding agreement with the AOSI developers) before beginning to assess infants in this study. All assessments were videotaped to permit scoring from footage.</p> <hd id="AN0148624091-6">AOSI</hd> <p>At Time 1, we followed the convention of making live ratings of infant behaviour during AOSI assessments but also had each AOSI video-rated by the administering researcher for evaluation of within-rater/cross-format agreement (i.e. live vs video codes). Furthermore, 22 tapes (21.4%) were scored by the alternate-site researcher for evaluation of inter-rater agreement. At Time 2, an adjusted procedure was followed to ensure all AOSIs were blind-coded (i.e. given the context of the larger intervention trial; and pre-specification of AOSI as the primary outcome measure). While the administering researcher continued to make live ratings, all Time 2 AOSIs were then scored from video by the alternate-site researcher (except in three cases where technical error resulted in loss of footage). A subset of 20 tapes (20.6%) was then scored by the administering researcher for evaluation of inter-rater agreement on Time 2 assessments.</p> <hd id="AN0148624091-7">ADOS-T</hd> <p>The protocol for coding ADOS-T assessments mirrored that of Time 2 AOSI coding, with all scoring conducted from video by a research-reliable examiner at the alternate site to guarantee blindness to randomisation group and other data collected for the given infant. Approximately 10 times per year across the trial data collection period, a local ADOS trainer conducted a coding check with the team of researchers responsible for ADOS-T administration and scoring, with the group reviewing a recently filmed assessment and meeting via teleconference to discuss and agree on codes by consensus. Item-level data were recorded for each of these consensus meetings and demonstrated that the four key researchers responsible for scoring trial assessments maintained an average 80% agreement across 16 ADOS-T tapes reviewed in this way.</p> <hd id="AN0148624091-8">Results</hd> <p></p> <hd id="AN0148624091-9">Sample characterisation</hd> <p>As outlined above, infants were recruited on the basis of presenting three or more (of five) behavioural autism markers according to the SACS-R 12-month checklist ([<reflink idref="bib2" id="ref32">2</reflink>]; [<reflink idref="bib18" id="ref33">18</reflink>]). Approximately equal thirds of the sample presented with three (<emph>n</emph> = 32), four (<emph>n</emph> = 34) and five (<emph>n</emph> = 37) markers at study entry. As shown in Table 1, the infants were aged 9–16 months at Time 1 assessment and 15–23 months at Time 2. For 14 infants (13.5%) born pre-term (between 32 and 37 weeks' gestation), all subsequent reference is to an <emph>adjusted age</emph> computed to correct for prematurity (i.e. 40 minus gestational age in weeks). Approximately two-thirds of infants were male (<emph>n</emph> = 70), and one-third were first-born/only children (<emph>n</emph> = 41). Among the two-thirds with older sibling/s, parents reported an autism diagnosis in 20 cases (and an additional three further cases of infants with a diagnosed cousin). Hence, the cohort comprised clinically indicated infants predominantly <emph>without</emph> a family history of autism (<emph>n</emph> = 80; 78%).</p> <p>Graph</p> <p>Table 1. Characterisation of infant cohort at two assessment timepoints, with descriptive separation by sex.</p> <p> <ephtml> <table><colgroup><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th /><th align="left" colspan="2">Full Sample</th><th align="left" colspan="2">Males</th><th align="left" colspan="2">Females</th></tr><tr><th /><th align="left">N</th><th align="left">Mean (<italic>SD</italic>) Range</th><th align="left">N</th><th align="left">Mean (<italic>SD</italic>), range</th><th align="left">N</th><th align="left">Mean (<italic>SD</italic>) Range</th></tr></thead><tbody><tr><td colspan="7">Time 1 (Baseline)</td></tr><tr><td> Infant age (months)</td><td>103</td><td>12.4 (2.0) 9–16</td><td>70</td><td>12.4 (1.9) 9–16</td><td>33</td><td>12.4 (2.2) 9–16</td></tr><tr><td> MSEL ELC SS</td><td>102</td><td>87.3 (16.2) 53–129</td><td>70</td><td>85.6 (15.1) 53–128</td><td>32</td><td>90.8 (18.2) 56–129</td></tr><tr><td> VABS ABC SS</td><td>92</td><td>83.5 (11.2) 58–110</td><td>62</td><td>84.2 (11.3) 58–110</td><td>30</td><td>81.9 (10.9) 63–109</td></tr><tr><td colspan="7">Time 2 (Follow-up)</td></tr><tr><td> Infant age (months)</td><td>97</td><td>18.6 (2.1) 15–23</td><td>66</td><td>18.6 (2.1) 15–23</td><td>31</td><td>18.6 (2.2) 15–22</td></tr><tr><td> MSEL ELC SS</td><td>96</td><td>86.5 (16.8) 49–124</td><td>66</td><td>85.2 (16.1) 50–124</td><td>30</td><td>89.3 (18.3) 49–123</td></tr><tr><td> VABS ABC SS</td><td>83</td><td>91.1 (12.3) 63–120</td><td>57</td><td>90.9 (12.2) 64–120</td><td>26</td><td>91.6 (12.7) 63–109</td></tr><tr><td> ADOS-T SA Total</td><td>96</td><td>7.4 (4.4) 0–18</td><td>66</td><td>8.0 (4.3) 0–18</td><td>30</td><td>6.2 (4.4) 0–18</td></tr><tr><td> ADOS-T RRB Total</td><td>96</td><td>2.6 (1.9) 0–8</td><td>66</td><td>2.80 (2.0) 0–8</td><td>30</td><td>2.2 (1.5) 0–5</td></tr></tbody></table> </ephtml> </p> <p>1 <emph>Note.</emph> MSEL = Mullen Scales of Early Learning ([<reflink idref="bib19" id="ref34">19</reflink>]); ELC = Early Learning Composite; SS = Standard Score. VABS = Vineland Adaptive Behaviour Scales ([<reflink idref="bib21" id="ref35">21</reflink>]); ABC = Adaptive Behaviour Composite. ADOS-T = Autism Diagnostic Observation Schedule – Toddler Module ([<reflink idref="bib15" id="ref36">15</reflink>]); SA = Social Affect; RRB = Restricted Repetitive Behaviours. Parameters computed following adjustment of values for three outlier cases (all Vineland Adaptive Behaviour Scales).</p> <p>Primary caregivers were predominantly (96%) biological mothers (3% fathers, 1% other guardian), of mean age 34.3 years, and generally well educated (70% families with one/both parents having completed tertiary/graduate education). Annual household income varied, with around half of families (48.5%) reporting high income >AUD104K (18% reporting income between AUD104K and AUD78K; and 23% reporting income ⩽AUD78K (<emph>n</emph> = 10.5% missing)).</p> <p>Table 1 also shows descriptive data for the infants' composite scores on the standardised, norm-referenced clinical assessments completed at both visits – the MSEL and VABS – as well for the Time 2 ADOS-T assessments. Group mean-level developmental abilities and adaptive skills were somewhat below the level expected given the infants' ages at each visit, but with substantial individual variability. Average Time 2 ADOS scores were within the mild-to-moderate range for Social Affect symptoms and approaching the moderate-to-severe range for Restricted Repetitive Behaviour symptoms. Again, however, substantial heterogeneity was apparent, including ADOS-T scores, spanning almost the full possible range.</p> <hd id="AN0148624091-10">Reliability and concurrent convergent validity of AOSI scores</hd> <p>Table 2 shows descriptive data for the two AOSI metrics – Total score and NoM – computed from both live- and video-ratings at each visit. Reliability indices presented in Table 3 – Cronbach's alpha for internal consistency, and single measures ICCs for inter-rater agreement (two-way mixed effect models for absolute agreement) – indicated variable internal consistency but good-to-excellent intra-rater (i.e. cross-modality) and inter-rater reliability. Paired-samples <emph>t</emph>-tests on Time 1 scores revealed small but statistically significant differences between ratings made by the administering researchers from live versus video observation – for both the Total score, <emph>t</emph>(<reflink idref="bib102" id="ref37">102</reflink>) = 2.89, <emph>p</emph> = 0.005, <emph>d</emph> = 0.29, and NoM, <emph>t</emph>(<reflink idref="bib102" id="ref38">102</reflink>) = 2.30, <emph>p</emph> = 0.024, <emph>d</emph> = 0.23. That is, video-coding yielded slightly higher scores (i.e. indicating greater number of autism risk behaviours) than did live coding, when conducted by the same assessor. No such difference presented between video ratings of the administering researcher and alternate-site coder (<emph>p</emph> ⩾ 0.559), nor across AOSI scores made by any rater across any format at Time 2 (<emph>p</emph> ⩾ 0.107).</p> <p>Graph</p> <p>Table 2. Descriptive and internal consistency statistics for live- and video-rated AOSI scores at Time 1 and 2.</p> <p> <ephtml> <table><colgroup><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th /><th align="left" colspan="6">Ratings by administering researcher</th><th align="left" colspan="3" rowspan="2">Alternate-site ratings video</th></tr><tr><th /><th align="left" colspan="3">Live</th><th align="left" colspan="3">Video</th></tr><tr><th /><th align="left">n</th><th align="left">M (<italic>SD</italic>) Range</th><th align="left"><italic>α</italic></th><th align="left">n</th><th align="left">M (<italic>SD</italic>) Range</th><th align="left"><italic>α</italic></th><th align="left">n</th><th align="left">M (<italic>SD</italic>) Range</th><th align="left"><italic>α</italic></th></tr></thead><tbody><tr><td colspan="10"><bold>Time 1</bold></td></tr><tr><td>Total Score</td><td>103</td><td>8.90 (4.3) 1–28</td><td>0.61</td><td>103</td><td>9.49 (4.2) 1–26</td><td>0.57</td><td>22</td><td>10.5 (3.5) 4–19</td><td>−</td></tr><tr><td>NoM</td><td>103</td><td>5.97 (2.4) 1–13</td><td>0.50</td><td>103</td><td>6.28 (2.3) 1–13</td><td>0.45</td><td>22</td><td>6.77 (1.7) 4–11</td><td>−</td></tr><tr><td colspan="10"><bold>Time 2</bold></td></tr><tr><td>Total Score</td><td>97</td><td>8.60 (4.7) 1–24</td><td>0.64</td><td>20</td><td>8.70 (5.3) 2–24</td><td>−</td><td>94</td><td>9.32 (4.7) 1–24</td><td>.67</td></tr><tr><td>NoM</td><td>97</td><td>5.63 (2.6) 1–13</td><td>0.57</td><td>20</td><td>5.65 (2.9) 1–13</td><td>−</td><td>94</td><td>6.11 (2.7) 1–15</td><td>.65</td></tr></tbody></table> </ephtml> </p> <p>2 <emph>Note.</emph> NoM = Number of Markers.</p> <p>Graph</p> <p>Table 3. Intra-Class Correlation coefficients indexing agreement across various approaches to AOSI coding.</p> <p> <ephtml> <table><colgroup><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="left">Administering researcher</th><th align="left" colspan="2">Live ratings</th><th align="left" colspan="2">Alternate-site (video) ratings</th></tr><tr><th align="left">Video ratings</th><th align="left">n (%)</th><th align="left">ICC</th><th align="left">n (%)</th><th align="left">ICC</th></tr></thead><tbody><tr><td colspan="5"><bold>Time 1</bold></td></tr><tr><td> Total score</td><td>103 (100%)</td><td>0.88</td><td>22 (~20%)</td><td>0.83</td></tr><tr><td> Number of markers</td><td>103 (100%)</td><td>0.82</td><td>22 (~20%)</td><td>0.73</td></tr><tr><td colspan="5"><bold>Time 2</bold></td></tr><tr><td> Total score</td><td>20 (~20%)</td><td>0.92</td><td>20 (~20%)</td><td>0.88</td></tr><tr><td> Number of markers</td><td>20 (~20%)</td><td>0.91</td><td>20 (~20%)</td><td>0.85</td></tr></tbody></table> </ephtml> </p> <p>3 <emph>Note.</emph> ICC = Intra-Class Correlation.</p> <p>Figure 1 shows distributions of AOSI Total scores and NoM derived from the primary data retained for all subsequent analysis – video ratings made at Time 1 by the administering researcher and at Time 2 by the alternate-site researcher. Again, each measure presented substantial variability with scores broadly normally distributed at each timepoint, albeit with two cases of Total scores presenting as high outliers (different infants; one at each visit) which were adjusted by re-assigning a value one point above the nearest non-outlier score (following [<reflink idref="bib22" id="ref39">22</reflink>]).</p> <p>Graph: Figure 1. Distribution (before adjustment of outliers) of infants' AOSI Total scores and Number of Markers at Time 1 (M age = 12 months) and Time 2 (M age = 18 months) with skew and kurtosis data computed (computed following adjustment of outliers; SE = Standard Error).</p> <p>Table 4 shows concurrent Time 1 associations among AOSI scores and other measures of developmental and adaptive skills, and the number of SACS autism markers at referral. Associations were in the expected directions, with significant (albeit weak) correlations, such that more AOSI behavioural signs of autism were associated with lower MSEL developmental skills and VABS adaptive behaviour. Infants rated with more signs of autism during AOSI assessment also had greater number of parent-reported SACS-R autism markers at referral (i.e. significant difference between infants with 4/5 atypical social-communication markers vs only 3: Cohen's <emph>d</emph> effect sizes; AOSI Total Score = 0.72; NoM = 0.66).</p> <p>Graph</p> <p>Table 4. Association of Time 1 AOSI scores with scores from concurrent standardised assessments, and with number of SACS autism markers reported at referral into the study.</p> <p> <ephtml> <table><colgroup><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="left" rowspan="2">Assessment</th><th align="left" rowspan="2">Measure</th><th align="left" colspan="2">Time 1 AOSI</th></tr><tr><th align="left">Total score</th><th align="left">Number of markers</th></tr></thead><tbody><tr><td>MSEL</td><td>Early Learning Composite</td><td><italic>r</italic> = −0.24; <italic>p</italic> = 0.016</td><td><italic>r</italic> = −0.27; <italic>p</italic> = 0.006</td></tr><tr><td>VABS</td><td>Adaptive Behaviour Composite</td><td><italic>r</italic> = −0.24; <italic>p</italic> = 0.020</td><td><italic>r</italic> = −0.27; <italic>p</italic> = 0.010</td></tr><tr><td rowspan="4">SACS no. markers</td><td>3</td><td><italic>M</italic> = 7.41 (<italic>SD</italic> = 3.25)</td><td><italic>M</italic> = 5.22 (<italic>SD</italic> = 2.03)</td></tr><tr><td>4</td><td><italic>M</italic> = 10.03 (<italic>SD</italic> = 4.17)</td><td><italic>M</italic> = 6.65 (<italic>SD</italic> = 2.39)</td></tr><tr><td>5</td><td><italic>M</italic> = 10.65 (<italic>SD</italic> = 3.98)</td><td><italic>M</italic> = 6.86 (<italic>SD</italic> = 2.19)</td></tr><tr><td>Omnibus Significance Test</td><td><italic>F</italic>(2,102) = 6.73, <italic>p</italic> = 0.002</td><td><italic>F</italic>(2,103) = 5.46, <italic>p</italic> = 0.006</td></tr></tbody></table> </ephtml> </p> <p>4 <emph>Note.</emph> AOSI = Autism Observation Scale for Infants; MSEL = Mullen Scales of Early Learning; VABS = Vineland Adaptive Behaviour Scales; SACS = Social Attention and Communication Surveillance.</p> <p>Table 5 shows concurrent associations among Time 2 data; between AOSI scores and other standardised measures, including the ADOS-T assessment of autism symptoms. Associations were again in the expected directions, with stronger effects suggested here than at Time 1 – moderate negative correlation of Time 2 AOSI behavioural signs of autism with concurrent MSEL developmental skills and VABS adaptive behaviour, and strong positive association with concurrent ADOS-T autism symptoms (Total Algorithm score). When examined separately, AOSI scores were strongly associated with Social Affect, but only moderately associated with Restricted/Repetitive Behaviour, algorithm totals. Furthermore, evaluation as a function of categorical ADOS-T 'ranges of concern' indicated significantly higher AOSI scores among infants in the moderate-to-severe range vs mild-to-moderate/little-to-no concern ranges (Cohen's <emph>d</emph>s; AOSI Total Score = 1.36; NoM = 1.10).</p> <p>Graph</p> <p>Table 5. Association of Time 2 AOSI scores with scores from concurrent standardised assessments of developmental skills (MSEL), adaptive behaviour (VABS), and autism symptom presentation (ADOS-T).</p> <p> <ephtml> <table><colgroup><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="left" rowspan="2">Assessment</th><th align="left" rowspan="2">Measure</th><th align="left" colspan="2">Time 2 AOSI</th></tr><tr><th align="left">Total Score</th><th align="left">Number of Markers</th></tr></thead><tbody><tr><td>MSEL</td><td>Early Learning Composite</td><td><italic>r</italic> = −0.37; <italic>p</italic> < 0.001</td><td><italic>r</italic> = −0.36; <italic>p</italic> < 0.001</td></tr><tr><td>VABS</td><td>Adaptive Behaviour Composite</td><td><italic>r</italic> = −0.43; <italic>p</italic> < 0.001</td><td><italic>r</italic> = −0.41; <italic>p</italic> < 0.001</td></tr><tr><td>VABS</td><td>Adaptive Behaviour Composite</td><td><italic>r</italic> = −0.43; <italic>p</italic> < 0.001</td><td><italic>r</italic> = −0.41; <italic>p</italic> < 0.001</td></tr><tr><td rowspan="3">ADOS-T Algorithm Scores</td><td>Total Algorithm Score</td><td><italic>r</italic> = 0.63; <italic>p</italic> < 0.001</td><td><italic>r</italic> = 0.56; <italic>p</italic> < 0.001</td></tr><tr><td>Social Affect Domain</td><td><italic>r</italic> = 0.62; <italic>p</italic> < 0.001</td><td><italic>r</italic> = 0.57; <italic>p</italic> < 0.001</td></tr><tr><td>Restricted/Repetitive Domain</td><td><italic>r</italic> = 0.40; <italic>p</italic> < 0.001</td><td><italic>r</italic> = 0.34; <italic>p</italic> = 0.001</td></tr><tr><td rowspan="4">ADOS-T Ranges of Concern</td><td>Little/None</td><td><italic>M</italic> = 7.36 (<italic>SD</italic> = 3.70)</td><td><italic>M</italic> = 5.06 (<italic>SD</italic> = 2.42)</td></tr><tr><td>Mild/Moderate</td><td><italic>M</italic> = 8.65 (<italic>SD</italic> = 3.45)</td><td><italic>M</italic> = 5.91 (<italic>SD</italic> = 2.39)</td></tr><tr><td>Moderate/Severe</td><td><italic>M</italic> = 13.71 (<italic>SD</italic> = 4.36)</td><td><italic>M</italic> = 8.33 (<italic>SD</italic> = 2.33)</td></tr><tr><td>Omnibus Significance Test</td><td><italic>F</italic>(2,93) = 22.38, <italic>p</italic> < 0.001</td><td><italic>F</italic>(2,93) = 14.96, <italic>p</italic> < 0.001</td></tr></tbody></table> </ephtml> </p> <p>5 <emph>Note.</emph> MSEL = Mullen Scales of Early Learning; VABS = Vineland Adaptive Behaviour Scales; AOSI = Autism Observation Scale for Infants; ADOS-T = Autism Diagnostic Observation Schedule – Toddler Module.</p> <hd id="AN0148624091-11">Stability of AOSI scores and predictive validity against ADOS-T</hd> <p>While distributions of AOSI scores at Time 1 and 2 were similar, both centred around mean Total = 9 and NoM = 6, only moderate stability was apparent (<emph>r</emph> = 0.44, <emph>p</emph> < 0.001 (both metrics)).</p> <p>As shown in Table 6, continuous AOSI scores at Time 1 were significantly, moderately associated with ADOS-T Total Algorithm scores at Time 2. When separated by domain, however, this association was maintained for Social Affect but not Restricted/Repetitive Behaviour. Again, evaluation of continuous Time 1 AOSI scores as a function of Time 2 categorical ADOS-T 'ranges of concern' indicated a significant association, such that infants with higher Time 1 AOSI scores were more likely to be classified in the moderate-to-severe range of concern 6 months later, rather than mild-to-moderate/little-to-no ranges of concern (Cohen's <emph>d</emph>s; AOSI Total = 0.61; NoM = 0.56).</p> <p>Graph</p> <p>Table 6. Association of Time 1 AOSI scores with Time 2 continuous and categorical autism symptom measures.</p> <p> <ephtml> <table><colgroup><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="left" rowspan="2">Time 2 Assessment</th><th align="left" rowspan="2">Measure</th><th align="left" colspan="2">Time 1 AOSI</th></tr><tr><th align="left">Total score</th><th align="left">Number of markers</th></tr></thead><tbody><tr><td rowspan="3">ADOS-T Algorithm Scores</td><td>Total Algorithm Score</td><td><italic>r</italic> = 0.41; <italic>p</italic> < 0.001</td><td><italic>r</italic> = 0.40; <italic>p</italic> < 0.001</td></tr><tr><td>Social Affect Domain</td><td><italic>r</italic> = 0.44; <italic>p</italic> < 0.001</td><td><italic>r</italic> = 0.43; <italic>p</italic> < 0.001</td></tr><tr><td>Restricted/Repetitive Domain</td><td><italic>r</italic> = 0.18; <italic>p</italic><italic>=</italic> 0.080</td><td><italic>r</italic> = 0.18; <italic>p</italic> = 0.074</td></tr><tr><td rowspan="4">ADOS-T Ranges of Concern</td><td>Little/None</td><td><italic>M</italic> = 8.25 (<italic>SD</italic> = 3.52)</td><td><italic>M</italic> = 5.67 (<italic>SD</italic> = 2.17)</td></tr><tr><td>Mild/Moderate</td><td><italic>M</italic> = 9.09 (<italic>SD</italic> = 2.83)</td><td><italic>M</italic> = 6.04 (<italic>SD</italic> = 1.69)</td></tr><tr><td>Moderate/Severe</td><td><italic>M</italic> = 11.84 (<italic>SD</italic> = 4.39)</td><td><italic>M</italic> = 7.60 (<italic>SD</italic> = 2.22)</td></tr><tr><td>Omnibus Significance Test</td><td><italic>F</italic>(2,95) = 8.16, <italic>p</italic> = 0.001</td><td><italic>F</italic>(2,95) = 7.28, <italic>p</italic> = 0.001</td></tr></tbody></table> </ephtml> </p> <p>6 <emph>Note.</emph> AOSI = Autism Diagnostic Observation Schedule; ADOS-T = Autism Diagnostic Observation Schedule – Toddler Module.</p> <hd id="AN0148624091-12">Discussion</hd> <p>We conducted an independent evaluation of psychometric performance of the AOSI, representing the first evaluation of this measure with a community-ascertained sample of infants showing behavioural autism markers. Our data suggest the AOSI can be reliably coded (both live and from video) by newly trained examiners, and validly quantifies early autism behaviours in a community sample, similarly to previous reports among high-risk infant siblings.</p> <hd id="AN0148624091-13">Instrument validity for quantifying early autism behaviours</hd> <p>Following formal training with the instrument developers, our team achieved and maintained reliable scoring of AOSI assessments. This was reflected in adequate- to excellent inter-rater agreement on AOSI Total Score and NoM metrics for subsets of double-coded tapes at each visit, yielding reliability coefficients slightly lower than those reported by [<reflink idref="bib9" id="ref40">9</reflink>] and [<reflink idref="bib11" id="ref41">11</reflink>] for assessments with infants at mean age of 12 months, but similar levels for assessments conducted around the age of 18 months. Other indicators of reliability, including internal consistency estimates and item-level inter-rater agreement, were lower, as discussed below. We evaluated the consistency of scores derived from video- versus standard live-rating protocols, finding good to excellent within-rater/cross-modality agreement. For Time 1 (but not Time 2) assessments, our scores derived from video-rating were slightly but significantly higher than those from live-rating, regardless of whether scoring was by the researcher who had administered the assessment or researcher at the alternate site. There are two possible reasons for this slight discrepancy. First, our researchers had less experience with the AOSI during the period of Time 1 assessments, so video review may have facilitated the identification of subtle features among young infants that were missed during live coding. Alternatively, quality of the social atmosphere and/or children's subtle social-affective displays might have been more challenging to discern retrospectively from video, particularly at Time 1 when infants' behavioural repertoire would have been more limited. Notwithstanding statistical significance, the small effect sizes for live- vs video-rating differences – equating to 0.5 AOSI Total score and 0.3 NoM – are unlikely to be clinically meaningful.</p> <p>Other parts of our analysis support previous accounts of the AOSI as a valid measure of behaviours associated with emerging autism ([<reflink idref="bib6" id="ref42">6</reflink>]; [<reflink idref="bib9" id="ref43">9</reflink>]; [<reflink idref="bib11" id="ref44">11</reflink>]), extending this evidence to the context of infants identified with early signs from the community (rather than high-risk siblings). It will be important to follow these infants into early childhood, to determine the ultimate predictive validity of AOSI scores in infancy for later diagnostic outcome. However, associations observed between our data collected at mean age 12 and 18 months were in the expected directions. That is, AOSI scores were significantly associated with scores from established clinical phenotyping measures – the MSEL, VABS, and ADOS-T – concurrently and also prospectively across this 6-month period. The pattern of relative magnitude of associations across timepoints and measures also reflected the greater/lesser conceptual similarity of constructs assessed. That is, the increasing magnitude of associations among Time 2 assessments (vs concurrent Time 1 assessments) is consistent with the notion that autism behaviours emerge and consolidate with age during this developmental period. AOSI scores were <emph>negatively</emph> associated with developmental level and adaptive behaviour at both timepoints, but with smaller effect sizes suggested at Time 1 than Time 2.</p> <p>The strongest associations observed in our dataset were between Time 2 scores from the brief (15–20 min) AOSI and longer (45 min) ADOS-T. To our knowledge, this study provides the first <emph>direct</emph> comparison of AOSI scores against ADOS-T. Previous studies have included [<reflink idref="bib25" id="ref45">25</reflink>] and [<reflink idref="bib11" id="ref46">11</reflink>] reports of association between children's earlier AOSI scores and later classification or continuous scores on the basis of ADOS Module 1/2. [<reflink idref="bib6" id="ref47">6</reflink>] examined the relative predictive value of item scores on each measure, taken concurrently at 18 months, for subsequent diagnostic outcome, but did not report the concurrent convergence of AOSI and ADOS Module 1 summary scores. Each of these analyses drew on data collected before publication of the new ADOS Toddler Module, suitable for use from the age of 12 months ([<reflink idref="bib15" id="ref48">15</reflink>]). We included the ADOS-T within our study protocol at Time 2 only, as we planned to recruit and conduct Time 1 assessments with infants prior to their first birthdays, from as young as 9 months. We observed convergence of Time 2 AOSI scores against concurrent ADOS-T total algorithm scores and categorical 'ranges of concern', such that infants categorised at moderate/severe concern had substantially higher concurrent AOSI scores. Similarly, we observed convergent validity for Time 1 AOSI scores against (a) concurrent categorical SACS autism markers and (b) prospective ADOS-T categorical 'ranges of concern' and continuous algorithm total scores at Time 2. However, some specificity was suggested in that AOSI scores were more strongly associated with ADOS-T symptoms in the Social Affect rather than Restricted/Repetitive Behaviour domain, both concurrently at Time 2 and prospectively for Time 1 AOSI predictive of Time 2 ADOS-T.</p> <hd id="AN0148624091-14">Behavioural presentation among community-ascertained infants</hd> <p>Our primary aim was to conduct an independent replication test of the AOSI performance as a measure of early autism behaviours among a clinically indicated (vs high-risk sibling) cohort. However, our data also provide insights into the early manifestation of autism behaviours among infants identified from the community in whom the condition may be emerging. In this cohort, the spread of AOSI Total scores and NoM was broadly normally distributed at each timepoint. While single infants presented as high outlier cases at each visit, there was evidence of neither ceiling nor floor effects. In other datasets, AOSI score distributions have been positively skewed – including data from high-risk sibling cohorts (e.g. [<reflink idref="bib11" id="ref49">11</reflink>]) but also from a study reporting scores from AOSI assessments conducted with infants identified 'at-risk' following Level 1 community-based screening (i.e. with the First Year Inventory (FYI); [<reflink idref="bib4" id="ref50">4</reflink>]). While our planned follow-up of this cohort into early childhood will allow us to examine the predictive validity of early AOSI scores for later diagnostic outcome, the heterogeneity of scores evident both AOSI and ADOS-T assessments within this community-ascertained group also presents a unique opportunity for the prospective examination of development among infants spanning the full range of subclinical- through clinical-level presentation of early autism behaviours. Following-up this heterogeneous cohort using analytic approaches that exploit rather than constrain variability raises the prospect of examining whether developmental trajectories and outcomes vary as a function of the degree and manifestation of <emph>specific</emph> autism symptoms. Such a dimensional approach is espoused within the broader field of psychiatry ([<reflink idref="bib13" id="ref51">13</reflink>]) and has promising potential to refine our understanding of clinical phenomena, such as autism.</p> <p>The apparent similarity of the Time 1 and 2 distributions of AOSI scores for the cohort masks the presence of only moderate association between <emph>individuals'</emph> scores at each timepoint, reflecting modest stability across this 6-month follow-up period. [<reflink idref="bib9" id="ref52">9</reflink>] reported only modest 2-week test–retest reliability for AOSI scores for infants aged around 12 months, and other groups have noted similar within-participant variability in the overt presentation of early-life autism behaviours; including among high-risk siblings (Tony Charman, personal communication October 16, 2019, based on data reported in publication by [<reflink idref="bib11" id="ref53">11</reflink>]) and infants born pre-term ([<reflink idref="bib24" id="ref54">24</reflink>]). Rather than raising concerns about fitness-for-purpose of the AOSI as a measure of early autism behaviours, these modest stability data may serve to highlight early phenotypic variability as a genuine feature of the autism prodrome. Even among infants with no systematic family history of autism – those in the community for whom some concern has been identified, triggering referral – around the first birthday, emerging symptoms vary from assessment to assessment, reinforcing the need for routine developmental surveillance, rather than single timepoint screening, and comprehensive developmental assessment to inform diagnostic decision making (e.g. see [<reflink idref="bib1" id="ref55">1</reflink>]).</p> <p>In contrast to sound intra- and inter-rater reliability, Cronbach's alpha coefficients suggested only modest internal consistency reliability for the current AOSI data. That is, while our raters agreed on the overall degree of behavioural atypicality shown by infants within this brief assessment, the extent to which item scores were coherent was less than usually deemed acceptable for a unitary scale (i.e. below the conventional ⩾0.70 standard; e.g. [<reflink idref="bib10" id="ref56">10</reflink>]). Internal consistency data are not provided in previous reports on AOSI performance (e.g. papers by [<reflink idref="bib9" id="ref57">9</reflink>]; [<reflink idref="bib11" id="ref58">11</reflink>]) so it is unclear whether the current modest data reflect something about our novel application of the AOSI or that the measure itself is not capturing a unitary construct. Indeed, at face value, AOSI items span different domains including social-communication skills, behavioural atypicalities, temperament, attention/motor control, such that low alphas for the test of a unitary scale may be unsurprising. However, it is also possible that internal consistency may have been affected by our specific use of this measure, as an independent and newly trained group, and/or its application to community-identified infants showing possible emerging autism as opposed to 'high-risk siblings.'</p> <p>Our data align with the body of evidence that early autism behaviours present with greater clarity and consistency in the second year of life. Alongside the stronger magnitude of associations observed between Time 2 AOSI and other concurrent measures (vs Time 1 AOSI and other concurrent measures, or prospective Time 1 to 2 associations), our assessors achieved consistently higher inter-rater agreement at the second assessment (although this may also have reflected their greater experience with the assessment, at this later timepoint) at which point internal consistency metrics were also higher and approaching <emph>α</emph> = 0.70. Again, however, planned follow-up of this cohort into early childhood will be important to inform (a) the ultimate predictive validity of early AOSI scores for subsequent diagnosis and (b) the developmental course of emerging autism behaviours/symptoms alongside other phenotypic features.</p> <p>Developers of the AOSI have explicitly noted that the assessment is not yet validated for clinical use (e.g. [<reflink idref="bib7" id="ref59">7</reflink>]), and there is growing recognition of the need for researchers to demonstrate, empirically, to what extent inferences drawn from 'high risk sibling' cohorts can be generalised to the broader population of infants developing autism. Recent research suggests that children with autism who represent simplex cases, have clinical profiles that differ somewhat from multiplex cases (e.g. [<reflink idref="bib5" id="ref60">5</reflink>]). High-risk siblings who develop autism necessarily represent multiplex cases and are the later-born children within families where there is already some experience of raising an autistic child (including the journey to receiving a diagnosis for that child). Comparing high-risk siblings with autism vs children diagnosed with autism following community ascertainment, [<reflink idref="bib20" id="ref61">20</reflink>] demonstrated higher ADOS Social Affect domain (but not Restricted/Repetitive Behaviour) symptom scores among the latter group at ages 2 and 3 years. Our study is the first to have included the AOSI within a protocol of the prospective study of community-ascertained infants, in such a way as to allow examination not only of the presentation of early autism behaviours according to this assessment but also their apparent stability/variability in early life.</p> <hd id="AN0148624091-15">Limitations and future directions</hd> <p>As a potential limitation of the current study, we cannot claim absolute blindness of T2 AOSI and ADOS-T scoring for the whole cohort. In the context of the larger trial, our scoring protocol was established to guarantee these data would be coded blind to infants' treatment allocation. The result was that for almost half of children, the same researcher coded both T2 AOSI and ADOS-T footage for a given infant, so ADOS-T ratings could plausibly have been influenced by knowledge of the child's AOSI score. We note, however, that AOSI coding was conducted gradually over the course of trial data collection, while ADOS-T coding was mostly completed 'in bulk' at the completion of all T2 data collection; so, it seems unlikely a researcher would have remembered a child's AOSI score when subsequently making ADOS-T ratings given the substantial volume of assessments to be coded and temporal separation of the AOSI and ADOS-T coding work.</p> <p>Another potential criticism of our study design is the wide range of child ages represented at each of our assessment timepoints, which were also overlapping (i.e. 9–16 and 15–23 months). While much child development research – including studies prospectively tracking the development of high-risk siblings – typically closely constrains child ages at assessment timepoints, our design might be considered to more closely mirror what occurs in community practice. That is, children might present for assessment around, but not closely constrained to, particular age-related milestones. Further, the correlational nature of our design/analyses means that variable child age could be accounted for in analyses, and indeed we conducted checks to confirm no substantive age-related influence/moderation on the data we have presented here within each assessment timepoint. Nevertheless, as we have already considered above, achieving a better understanding of how and when early autism behaviours emerge in early life is a key continuing future direction for the field.</p> <p>Finally, we have already considered the modest internal consistency reliability observed for AOSI scores and another clear avenue for future research is to address the reasons for these scores falling below the conventional level accepted for psychometric robustness. Investigating modifications to AOSI scoring at item- or summary-level, for example, could bring changes to the measure while also affording insights into the nature of the construct/s captured during this promising brief assessment of behaviours relevant to the autism prodrome.</p> <hd id="AN0148624091-16">Summary</hd> <p>We undertook an independent evaluation of psychometric properties of the AOSI as a brief, semi-structured assessment of early autism behaviours, extending from previous work to apply this tool in the assessment of infants in whom autism may be emerging, identified from the community from 9 months of age. Our extension of the standard protocol, to show high-level reliability of ratings made from video review, supports viable use of the AOSI within protocols necessitating blinded evaluation (e.g. as a primary outcome measure within pre-emptive intervention trials). Moreover, concurrent and predictive associations of AOSI scores with other established clinical phenotyping measures supports the validity of this tool. Modest but adequate internal consistency reliability of AOSI scores, and their stability over time suggests there is more to be understood about the nature (i.e. dimensional structure) and course of emerging autism behaviours in community-ascertained samples. However, results from this study show parallels with previous data from 'high-risk siblings,' suggesting marked variability in the early presentation and course of emerging autism symptoms ([<reflink idref="bib14" id="ref62">14</reflink>]). We are currently following this cohort into early childhood, with further assessments planned at ~24 and 36 months of age, at which point we will determine clinical best-estimate diagnostic outcome, permitting further evaluation of the predictive validity of early AOSI scores. The observed heterogeneity (and normal distribution) of early AOSI scores means follow-up of this cohort may also yield insights into development, cutting across conventional diagnostic boundaries, spanning the clinical-to-subclinical/broader autism phenotype.</p> <p>We thank the infants and caregivers for their time commitment to participate in this study and the following colleagues for their contributions by way of conduct of activities for the larger trial beyond specific data collection and analysis reported here: Gail Alvares, Catherine Bent, Alena Clark, Matthew Cooper, Emma Davidson, Jane Doyle, Megan Grant, Catherine Rowbottam, Carol Taylor. The AICES Team (alphabetical order): Cheryl Dissanayake, Teresa Iacono, Murray Maybery, Michelle Renton, Nancy Sadka, Leonie Segal, Vicky Slonims, Scott Wakeling, Ming Wai Wan, John Wray.</p> <ref id="AN0148624091-17"> <title> References </title> <blist> <bibl id="bib1" idref="ref22" type="bt">1</bibl> <bibtext> Barbaro J., Dissanayake C. (2010). Prospective identification of autism spectrum disorders in infancy and toddlerhood using developmental surveillance: The social attention and communication study. Journal of Developmental and Behavioral Pediatrics, 31, 376–385.</bibtext> </blist> <blist> <bibl id="bib2" idref="ref23" type="bt">2</bibl> <bibtext> Barbaro J., Dissanayake C. (2013). Early markers of autism spectrum disorders in infants and toddlers prospectively identified in the Social Attention and Communication Study. Autism, 17, 64–86.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref5" type="bt">3</bibl> <bibtext> Bedford R., Gliga T., Hendry A., Jones E., Pasco G., Charman T., Johnson M. (2019). Infant regulatory function acts as a protective factor for later traits of autism spectrum disorder and attention deficit/hyperactivity disorder but not callous unemotional traits. Journal of Neurodevelopmental Disorders, 11, 14.</bibtext> </blist> <blist> <bibl id="bib4" idref="ref50" type="bt">4</bibl> <bibtext> Ben-Sasson A., Carter A. (2012). The application of the first year inventory for ASD screening in Israel. Journal of Autism and Developmental Disorders, 42, 1906–1916.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref60" type="bt">5</bibl> <bibtext> Berends D., Dissanayake C., Lawson L. (2019). Differences in cognition and behaviour in multiplex and simplex Autism: Does prior experience raising a child with autism matter? Journal of Autism and Developmental Disorders, 49, 3401–3411.</bibtext> </blist> <blist> <bibl id="bib6" idref="ref2" type="bt">6</bibl> <bibtext> Brian J., Bryson S., Garon N., Roberts W., Smith I., Szatmari P., Zwaigenbaum L. (2008). Clinical assessment of autism in high-risk 18-month-olds. Autism, 12, 433–456.</bibtext> </blist> <blist> <bibl id="bib7" idref="ref11" type="bt">7</bibl> <bibtext> Bryson S., Zwaigenbaum L. (2014). Autism observation scale for infants. In Patel V., Preedy V., Martin C. (Eds.), Comprehensive guide to autism (pp. 299–310). Springer.</bibtext> </blist> <blist> <bibl id="bib8" idref="ref3" type="bt">8</bibl> <bibtext> Bryson S., Zwaigenbaum L., Brian J., Roberts W., Szatmari P., Rombough V., McDermott C. (2007). A prospective case series of high-risk infants who developed autism. Journal of Autism and Developmental Disorders, 37, 12–24.</bibtext> </blist> <blist> <bibl id="bib9" idref="ref1" type="bt">9</bibl> <bibtext> Bryson S., Zwaigenbaum L., McDermott C., Rombough V., Brian J. (2008). The autism observation scale for infants: Scale development and reliability data. Journal of Autism and Developmental Disorders, 38, 731–738.</bibtext> </blist> <blist> <bibtext> Cortina J. (1993). What is coefficient alpha? An examination of theory and applications. Journal of Applied Psychology, 78, 98–104.</bibtext> </blist> <blist> <bibtext> Gammer I., Bedford R., Elsabbagh M., Garwood H., Pasco G., Tucker L.... BASIS Team. (2015). Behavioural markers for autism in infancy: Scores on the Autism Observational Scale for Infants in a prospective study of at-risk siblings. Infant Behaviour and Development, 38, 107–115.</bibtext> </blist> <blist> <bibtext> Hogan A., Caravella K., Ezell J., Rague L., Hills K., Roberts J. (2017). Autism spectrum disorder symptoms in infants with fragile X syndrome: A prospective case series. Journal of Autism and Developmental Disorders, 47, 1628–1644.</bibtext> </blist> <blist> <bibtext> Insel T., Cuthbert B., Garvey M., Heinssen R., Pine D., Quinn K. (2010). Research Domain Criteria (RDoC): Toward a new classification framework for research on mental disorders. American Journal of Psychiatry, 167, 748–751.</bibtext> </blist> <blist> <bibtext> Jones E., Gliga T., Bedford R., Charman T., Johnson M. (2014). Developmental pathways to autism: A review of prospective studies of infants at risk. Neuroscience and Biobehavioural Reviews, 39, 1–33.</bibtext> </blist> <blist> <bibtext> Lord C., Luyster R., Gotham K., Guthrie W. (2012). Autism Diagnostic Observation Schedule, 2nd Edition (ADOS-2) manual (Part II): Toddler module. Western Psychological Services.</bibtext> </blist> <blist> <bibtext> Lord C., Risi S., Lambrecht L., Cook E., Leventhal B., DiLavore P.,... Rutter M. (2000). The Autism Diagnostic Observation Schedule–Generic: A standard measure of social and communication deficits associated with the spectrum of autism. Journal of Autism and Developmental Disorders, 30, 205–223.</bibtext> </blist> <blist> <bibtext> McDonald N., Varcin K., Bhatt R., Wu J., Sahin M., Nelson C., Jeste S. (2017). Early autism symptoms in infants with tuberous sclerosis complex. Autism Research, 10, 1981–1990.</bibtext> </blist> <blist> <bibtext> Mozolic-Staunton B., Donelly M., Yoxall J., Barbaro J. (2020). Early detection for better outcomes: Universal developmental surveillance for autism across health and early childhood education settings. Research in Autism Spectrum Disorders, 71, 101496. https://doi.org/10.1016/j.rasd.2019.101496</bibtext> </blist> <blist> <bibtext> Mullen E. (1995). Mullen scales of early learning. American Guidance Service.</bibtext> </blist> <blist> <bibtext> Sacrey L., Zwaigenbaum L., Szatmari P., Bryson S., Georgiades S., Brian J.,... Elsabbagh M. (2017). Brief report: Characteristics of preschool children with ASD vary by ascertainment. Journal of Autism and Developmental Disorders, 47, 1542–1550.</bibtext> </blist> <blist> <bibtext> Sparrow S., Cicchetti D., Balla D. (2005). The Vineland Adaptive Behavior Scales, 2nd Edition (VABS-II). NCS Pearson Inc.</bibtext> </blist> <blist> <bibtext> Tabachnick B., Fidell L. (2013). Using multivariate statistics (6th ed.). Pearson.</bibtext> </blist> <blist> <bibtext> Whitehouse A., Varcin K., Alvares G., Barbaro J., Bent C., Boutrus M.,... Hudry K. (2019). Pre-emptive intervention versus treatment as usual for infants showing early behavioural risk signs of autism spectrum disorder: A single-blind, randomised controlled trial. Lancet Child and Adolescent Health, 3, 605–615.</bibtext> </blist> <blist> <bibtext> Yaari M., Yitzhak N., Harel A., Friedlander E., Bar-Oz B., Eventov-Friedman S.,...Yirmiya N. (2016). Stability of early risk assessment for autism spectrum disorder in preterm infants. Autism, 20, 856–867.</bibtext> </blist> <blist> <bibtext> Zwaigenbaum L., Bryson S., Rogers S., Roberts W., Brian J., Szatmari P. (2005). Behavioral manifestations of autism in the first year of life. International Journal of Developmental Neuroscience, 23, 143–152.</bibtext> </blist> </ref> <ref id="AN0148624091-18"> <title> Footnotes </title> <blist> <bibtext> The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.</bibtext> </blist> <blist> <bibtext> The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The larger study is funded by grants from the La Trobe University Understanding Disease Research Focus Area, Cooperative Research Centre for Living with Autism, Western Australian Children's Research Fund, and Angela Wright Bennett Foundation. AJOW is funded by a Senior Research Fellowship from the National Health and Medical Research Council. The funders and trial sponsor have had no role in the study design, manuscript drafting or decision to submit the manuscript for publication.</bibtext> </blist> <blist> <bibtext> Kristelle Hudry</bibtext> </blist> <blist> <bibtext>Graph</bibtext> </blist> <blist> <bibtext>https://orcid.org/0000-0002-2752-8345 Josephine Barbaro</bibtext> </blist> <blist> <bibtext>Graph https://orcid.org/0000-0001-7359-0097</bibtext> </blist> </ref> <aug> <p>By Kristelle Hudry; Lacey Chetcuti; Maryam Boutrus; Sarah Pillar; Emma K Baker; Stefanie Dimov; Josephine Barbaro; Jonathan Green; Andrew JO Whitehouse and Kandice J Varcin</p> <p>Reported by Author; Author; Author; Author; Author; Author; Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib24" firstref="ref6"></nolink> <nolink nlid="nl2" bibid="bib17" firstref="ref7"></nolink> <nolink nlid="nl3" bibid="bib12" firstref="ref8"></nolink> <nolink nlid="nl4" bibid="bib11" firstref="ref9"></nolink> <nolink nlid="nl5" bibid="bib25" firstref="ref12"></nolink> <nolink nlid="nl6" bibid="bib16" firstref="ref17"></nolink> <nolink nlid="nl7" bibid="bib20" firstref="ref19"></nolink> <nolink nlid="nl8" bibid="bib15" firstref="ref20"></nolink> <nolink nlid="nl9" bibid="bib23" firstref="ref21"></nolink> <nolink nlid="nl10" bibid="bib18" firstref="ref24"></nolink> <nolink nlid="nl11" bibid="bib19" firstref="ref28"></nolink> <nolink nlid="nl12" bibid="bib21" firstref="ref29"></nolink> <nolink nlid="nl13" bibid="bib102" firstref="ref37"></nolink> <nolink nlid="nl14" bibid="bib22" firstref="ref39"></nolink> <nolink nlid="nl15" bibid="bib13" firstref="ref51"></nolink> <nolink nlid="nl16" bibid="bib10" firstref="ref56"></nolink> <nolink nlid="nl17" bibid="bib14" firstref="ref62"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1285280
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Performance of the Autism Observation Scale for Infants with Community-Ascertained Infants Showing Early Signs of Autism
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Hudry%2C+Kristelle%22">Hudry, Kristelle</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0002-2752-8345">0000-0002-2752-8345</externalLink>)<br /><searchLink fieldCode="AR" term="%22Chetcuti%2C+Lacey%22">Chetcuti, Lacey</searchLink><br /><searchLink fieldCode="AR" term="%22Boutrus%2C+Maryam%22">Boutrus, Maryam</searchLink><br /><searchLink fieldCode="AR" term="%22Pillar%2C+Sarah%22">Pillar, Sarah</searchLink><br /><searchLink fieldCode="AR" term="%22Baker%2C+Emma+K%2E%22">Baker, Emma K.</searchLink><br /><searchLink fieldCode="AR" term="%22Dimov%2C+Stefanie%22">Dimov, Stefanie</searchLink><br /><searchLink fieldCode="AR" term="%22Barbaro%2C+Josephine%22">Barbaro, Josephine</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0001-7359-0097">0000-0001-7359-0097</externalLink>)<br /><searchLink fieldCode="AR" term="%22Green%2C+Jonathan%22">Green, Jonathan</searchLink><br /><searchLink fieldCode="AR" term="%22Whitehouse%2C+Andrew+J%2E+O%2E%22">Whitehouse, Andrew J. O.</searchLink><br /><searchLink fieldCode="AR" term="%22Varcin%2C+Kandice+J%2E%22">Varcin, Kandice J.</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Autism%3A+The+International+Journal+of+Research+and+Practice%22"><i>Autism: The International Journal of Research and Practice</i></searchLink>. Feb 2021 25(2):490-501.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: SAGE Publications. 2455 Teller Road, Thousand Oaks, CA 91320. Tel: 800-818-7243; Tel: 805-499-9774; Fax: 800-583-2665; e-mail: journals@sagepub.com; Web site: http://sagepub.com
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 12
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2021
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Autism%22">Autism</searchLink><br /><searchLink fieldCode="DE" term="%22Pervasive+Developmental+Disorders%22">Pervasive Developmental Disorders</searchLink><br /><searchLink fieldCode="DE" term="%22Infants%22">Infants</searchLink><br /><searchLink fieldCode="DE" term="%22At+Risk+Persons%22">At Risk Persons</searchLink><br /><searchLink fieldCode="DE" term="%22Toddlers%22">Toddlers</searchLink><br /><searchLink fieldCode="DE" term="%22Child+Development%22">Child Development</searchLink><br /><searchLink fieldCode="DE" term="%22Siblings%22">Siblings</searchLink><br /><searchLink fieldCode="DE" term="%22Diagnostic+Tests%22">Diagnostic Tests</searchLink><br /><searchLink fieldCode="DE" term="%22Test+Validity%22">Test Validity</searchLink><br /><searchLink fieldCode="DE" term="%22Symptoms+%28Individual+Disorders%29%22">Symptoms (Individual Disorders)</searchLink><br /><searchLink fieldCode="DE" term="%22Psychometrics%22">Psychometrics</searchLink><br /><searchLink fieldCode="DE" term="%22Foreign+Countries%22">Foreign Countries</searchLink><br /><searchLink fieldCode="DE" term="%22Child+Behavior%22">Child Behavior</searchLink>
– Name: Subject
  Label: Geographic Terms
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Australia%22">Australia</searchLink>
– Name: SubjectThesaurus
  Label: Assessment and Survey Identifiers
  Group: Su
  Data: <searchLink fieldCode="SU" term="%22Autism+Diagnostic+Observation+Schedule%22">Autism Diagnostic Observation Schedule</searchLink><br /><searchLink fieldCode="SU" term="%22Vineland+Adaptive+Behavior+Scales%22">Vineland Adaptive Behavior Scales</searchLink><br /><searchLink fieldCode="SU" term="%22Mullen+Scales+of+Early+Learning%22">Mullen Scales of Early Learning</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1177/1362361320965397
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 1362-3613
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Often included within 'high-risk sibling' studies, the Autism Observation Scale for Infants (AOSI) has only one independent replication study and no evaluation with community-ascertained cohorts. We administered the AOSI and established clinical measures with 103 infants (68% male) at 'high autism likelihood' on the Social Attention and Communication Surveillance - Revised (SACS-R) tool, at 9-14 months of age and again 6months later. AOSI Total scores showed adequate internal consistency and strong inter-rater agreement (live- or video-coded) and were approximately normally distributed at each visit. Modest significant associations presented between Time 1 AOSI scores and concurrent developmental/adaptive skills measures. Concurrent associations were stronger at Time 2, particularly between AOSI Total and Autism Diagnostic Observation Schedule (ADOS) Social Affect scores. AOSI scores were only moderately associated across Time 1 and 2 assessments, as were Time 1 AOSI with Time 2 ADOS scores. These data from a clinically indicated cohort broadly replicate previous AOSI validity accounts from 'high-risk sibling' studies, particularly beyond the first year. Strong inter-rater agreement indicates viable AOSI inclusion within protocols necessitating blinded evaluation (e.g. intervention trials). Moderate within-participant stability suggests that, like 'high-risk siblings,' community-ascertained infants experience variable early trajectories. [This article was co-written by the AICES Team.]
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2021
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1285280
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1285280
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1177/1362361320965397
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 12
        StartPage: 490
    Subjects:
      – SubjectFull: Autism
        Type: general
      – SubjectFull: Pervasive Developmental Disorders
        Type: general
      – SubjectFull: Infants
        Type: general
      – SubjectFull: At Risk Persons
        Type: general
      – SubjectFull: Toddlers
        Type: general
      – SubjectFull: Child Development
        Type: general
      – SubjectFull: Siblings
        Type: general
      – SubjectFull: Diagnostic Tests
        Type: general
      – SubjectFull: Test Validity
        Type: general
      – SubjectFull: Symptoms (Individual Disorders)
        Type: general
      – SubjectFull: Psychometrics
        Type: general
      – SubjectFull: Foreign Countries
        Type: general
      – SubjectFull: Child Behavior
        Type: general
      – SubjectFull: Australia
        Type: general
      – SubjectFull: Autism Diagnostic Observation Schedule
        Type: general
      – SubjectFull: Vineland Adaptive Behavior Scales
        Type: general
      – SubjectFull: Mullen Scales of Early Learning
        Type: general
    Titles:
      – TitleFull: Performance of the Autism Observation Scale for Infants with Community-Ascertained Infants Showing Early Signs of Autism
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Hudry, Kristelle
      – PersonEntity:
          Name:
            NameFull: Chetcuti, Lacey
      – PersonEntity:
          Name:
            NameFull: Boutrus, Maryam
      – PersonEntity:
          Name:
            NameFull: Pillar, Sarah
      – PersonEntity:
          Name:
            NameFull: Baker, Emma K.
      – PersonEntity:
          Name:
            NameFull: Dimov, Stefanie
      – PersonEntity:
          Name:
            NameFull: Barbaro, Josephine
      – PersonEntity:
          Name:
            NameFull: Green, Jonathan
      – PersonEntity:
          Name:
            NameFull: Whitehouse, Andrew J. O.
      – PersonEntity:
          Name:
            NameFull: Varcin, Kandice J.
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 02
              Type: published
              Y: 2021
          Identifiers:
            – Type: issn-print
              Value: 1362-3613
          Numbering:
            – Type: volume
              Value: 25
            – Type: issue
              Value: 2
          Titles:
            – TitleFull: Autism: The International Journal of Research and Practice
              Type: main
ResultId 1