bs_bs_banner

Journal of Intellectual Disability Research doi: 10.1111/jir.12284

464
VOLUME 60 PART 5 pp 464–477 MAY 2016

Special Issue

The proportion of minimally verbal children with autism

spectrum disorder in a community-based early
intervention programme
V. Rose,1 D. Trembath,1 D. Keen2 & J. Paynter3
1 Menzies Health Institute, Griffith University, Brisbane, Australia
2 Autism Centre of Excellence, Griffith University, Brisbane, Australia
3 AEIOU Foundation, Brisbane, Queensland, Australia

Abstract depending on measures and definitions used, with

parent report indicating that as many as 29.4% of
Background Estimates of the proportion of children
children were not ‘naming at least three objects’
with autism spectrum disorder (ASD) who are
consistently, and 43.3% not using ‘phrases with a noun
minimally verbal vary from 25%to 35%. However,
and verb’ consistently at exit. More than half of the
there is a lack of consensus in defining minimally
children who entered the programme with minimal
verbal and few detailed reports of communication
speech exited the programme with a similar language
outcomes for these children following intervention.
profile. A small percentage of children (1.2%–4.7%)
The aim of this study was to explore how minimally
regressed in their language level over time.
verbal children have been defined and to document
Conclusions Despite advances in early intervention,
the proportion of minimally verbal children in a group
and access to services at a younger age, around a
of children with ASD receiving a community based
quarter of individuals with ASD in this study exited
early intervention programme.
early intervention with significant communication
Method A longitudinal cohort design was used to
needs. Our findings are considered in relation to the
examine the proportion of children who met criteria
literature and clinical implications, and future
for minimally verbal in 246 children with ASD when
research directions are discussed.
they entered and exited an early intervention
programme. Keywords autism spectrum disorder,
Results Overall, 26.3% of the children in this study communication, early intervention, minimally verbal,
exited the programme using ‘fewer than five non-verbal, preschool
spontaneous and functional words’ and 36.4% exited
not using ‘two word phrases’ as indicated by direct
assessment. However, our findings were mixed Introduction

Correspondence: Ms Veronica Rose, Menzies Health Institute

Autism spectrum disorder (ASD) is a
Queensland, Griffith University QLD 4222, Australia (e-mail: neurodevelopmental disorder characterised by
veronica.rose@griffithuni.edu.au). impairments in social communication development,

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
Journal of Intellectual Disability Research VOLUME 60 PART 5 MAY 2016
465
V. Rose et al. • Proportion of minimally verbal children with ASD

and restrictive and repetitive behaviour patterns 2013; Kasari et al. 2014; Norrelgen et al. 2014;
(American Psychiatric Association 2013). Signs of ASD Woynaroski et al. 2015); ‘non-verbal’ (Jonsdottir et al.
are usually observable under 3 years of age (American 2007; Norrelgen et al. 2014); and ‘low verbal’ (Yoder
Psychiatric Association 2013), and early diagnosis & Stone 2006). The common element across
provides the opportunity for children to access definitions is the specification of the maximum
intervention at an early age (Dawson 2008; Rogers & number of spontaneous functional words a child uses
Vismara 2008). While there is evidence to suggest that (ranging from 5–20 words) as observed during a
early intervention positively impacts on natural language sample and/or parent report (refer to
communication and cognitive development in the Appendix for expanded definitions). Kasari et al.
individuals with ASD, individual differences in (2013) operationalised ‘minimally verbal’ as ‘fewer
response to intervention are apparent (Trembath & than 20 functional words’ (p. 2) for research
Vivanti 2014). An important area of heterogeneity is in purposes; however, noted that in clinical contexts the
the area of spoken language (Kjelgaard & Tager- ‘exact number of words used does not matter that
Flusberg 2001; Paul et al. 2013). Early language skills much’ (, p. 2) and depends on a range of individual
are an important predictor of later outcomes, and factors. In this article, we used the Kasari et al. (2013)
children presenting with complex communication definition to identify and review relevant studies
needs are more likely to have associated intellectual relating to children who used ‘a very small repertoire
disabilities that may impact on their social learning of spoken words or fixed phrases that are used
experiences, and communication and cognitive abilities communicatively’ (Kasari et al. 2013, p. 2).
later in life (Luyster et al. 2008; Fernell et al. 2013). The second barrier Norrelgen et al. (2014)
In research, communication outcomes for children identified is the lack of empirical data reporting on
with ASD are commonly reported using total scores communication outcomes for this group of children.
derived from standardised assessments (Matson et al. Indeed, there are few, and inconsistent, reports of the
2010), including from communication (e.g. Preschool prevalence of children with ASD who are minimally
Language Scales) and social communication verbal in the population or at exit from early
measures (e.g. Communication and Symbolic intervention. Historically, it was suggested that 50%
Behaviour Scales). These overall scores may fail to of individuals with ASD did not develop functional
capture individual variability in spoken language language (Rutter 1978). However, with the advent of
outcomes (Charman et al. 2003) and may not be earlier identification and intervention, more recent
appropriate for assessing minimally verbal children estimates put the figure closer to 25%–30% (Lord
with ASD (Charman et al. 2003; Abbeduto et al. 2011; et al. 2004; Anderson et al. 2007; Norrelgen et al.
Kasari et al. 2013). Tager-Flusberg et al. (2009) 2014), yet some studies are still reporting figures of up
proposed a set of ‘spoken language benchmarks’ to 50% (Kjelgaard & Tager-Flusberg 2001; Magiati
relating to key developmental stages (i.e. first words, et al. 2011). Differences in definition, measurement,
word combinations and sentences) aimed at better sample characterisation and age make comparisons
characterising the language abilities of children with across studies difficult (Jonsdottir et al. 2007;
ASD. These benchmarks highlight the importance of Norrelgen et al. 2014).
documenting and understanding the variability in Anderson et al. (2007) conducted a longitudinal
language outcomes for children with ASD, study of expressive language skills in 206 children
particularly those who exit early intervention with ASD, Pervasive Developmental Disorder-Not
programmes with minimal spoken language. Otherwise Specified (PDD-NOS), and other
Norrelgen et al. (2014) identified two key barriers to Developmental Disabilities (DD). Children were
documenting and understanding language outcomes under 3 years of age at the beginning of the study and
for children with ASD who exit intervention were then assessed at approximately 3, 5 and 9 years of
programmes with minimal spoken language. First, age. Anderson et al. (2007) reported that 29% of the
they noted that as a group, children who use little or ASD group and less than 10% of the PDD-NOS and
no spoken language have been inconsistently defined DD children remained non-verbal [i.e. using fewer
using a variety of terms including ‘minimally verbal’ than five words daily according to the Autism
(e.g. Goods et al. 2013; Kasari et al. 2013; Paul et al. Diagnostic Interview – Revised (ADI-R)]. A key

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
Journal of Intellectual Disability Research VOLUME 60 PART 5 MAY 2016
466
V. Rose et al. • Proportion of minimally verbal children with ASD

contribution of this study was the item level reporting group of children with ASD who received a
of the assessment data that provided more detail community-based early intervention programme. The
regarding language level than reporting standardised specific aim was to analyse the proportion of children
scores alone. Additionally, expressive language level who were minimally verbal at entry and exit, using a
was indicated by the Autism Diagnostic Observation combination of parent and clinician measures.
Schedule (ADOS) module used (e.g. ADOS Module
3 ‘Complex sentences’ and ADOS Module 2
‘Sentences but not fluent’). The reporting of item
Method
level data is consistent with the approach adopted by The study was independently reviewed and approved
Tager-Flusberg et al. (2009) and provides more by the Griffith University Human Research Ethics
clarity regarding communication outcomes using Committee and the Research Advisory Group at the
standardised assessments. AEIOU Organisation. Signed informed consent was
Magiati et al. (2011) and Norrelgen et al. (2014) also obtained from parents of participating children.
reported item level data from standardised
assessments, in an attempt to characterise the spoken
Participants and setting
language outcomes for children with ASD in early
intervention studies. Magiati et al. (2011) reported Participants were 246 children [202 male (82.1%), 44
outcomes for 36 children, between two and five years female (17.9%)] who attended an ASD-specific
after they commenced community based early centre-based early intervention programme in
intensive behavioural intervention. At initial testing Australia between 2010 and 2015. The programme
(mean age 3 years 4 months) only 8% of the follows a comprehensive curriculum consistent with
participants had functional phrase speech as assessed Australian Good Practice Guidelines (Prior et al.
using the ADI-R, increasing to 36% at Time 2 (mean 2011) that uses a blend of strategies with theoretical
age 5 years 5 months). Norrelgen et al. (2014) used underpinnings in behavioural and developmental
item level analyses of the Vineland Adaptive Behavior approaches. Emphasis is on practices with emerging
Scales-II (VABS-II) to document communication or higher levels of evidence as categorised by the
outcomes for 165 children with ASD aged 4–6.5 years National Autism Centre Standards Report (National
from a population-based community sample. The Autism Center 2015), including the use of
authors reported that 25% of the participants in their Augmentative and Alternative Communication
study presented with minimal or no functional speech. systems (AAC) (Paynter et al. 2015). Individual
Through item level reporting and analyses using Education Plans are developed for each child based
standardised tests, researchers have provided a clearer on the organisation’s autism-specific curriculum
indication of spoken language level than could be (Paynter et al. 2012). Teaching occurs within the
provided by standard scores alone. Replication of context of classroom routines in an early learning
such studies are important to gather descriptions from context including circle and mat times, free play,
multiple settings, using large samples, and across small group activities and meal times (Paynter et al.
populations, to gain a detailed understanding of 2012). The programme is delivered in a
communication outcomes for children with ASD. transdisciplinary team including teachers, speech
There is also a need to move beyond item level pathologists, occupational therapists, behaviour
reporting and analyses of parent-reported measures therapists and paraprofessionals with training in early
(e.g. ADI-R, VABS-II) to include item scores from childhood in a ratio of 1:2 to 1:4 in a centre-based
clinician administered direct assessments. Using a early learning ASD-specific setting. Refer to Paynter
variety of assessments (i.e. parent/caregiver and et al. (2015) for further information on this
clinician direct assessment) is important to develop a programme. Children attended the programme 15–
comprehensive profile of a child’s communication 25 h per week with most attending five days per week
skills (Lane et al. 2013; American Speech-Language- for an average of 14 months (SD = 6.45) between
Hearing Association 2015). Therefore, the overall intake and exit assessments. Programme eligibility
objective of this study was to document the included a community-based diagnosis of ASD
proportion of children who were minimally verbal in a (DSM-IV and DSM-5) by a medical practitioner

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
Journal of Intellectual Disability Research VOLUME 60 PART 5 MAY 2016
467
V. Rose et al. • Proportion of minimally verbal children with ASD

(paediatrician or child psychiatrist) or highest score available) indicating full points were
multidisciplinary team, and chronological age achieved. The total number of points available varies
between 30 and 71 months (Paynter et al. 2015). by item. We used the following ‘Expressive Language’
Diagnosis was verified with the Social item scores to determine expressive language level:
Communication Questionnaire (SCQ) using a cut-off Item 11 [‘Says first words’, scoring (0) 0 words, (1)
of 11 as recommended in previous research with pre- Says 1 word, (2) Says 2 to 7 words, (3) Says 8 words];
school age children (Lee et al. 2007), resulting in the Item 17 [‘Uses two-word phrases’, scoring (1) Yes, (0)
exclusion of 18 participants from analyses (n = 228). No]; Item 18 [‘Picture Vocabulary’, scoring (0)
Names 0-4 pictures, (1) Names 5-10 pictures, (2)
Names 11-14 pictures, (3) Names 15-16 pictures, (4)
Measures
Names 17 pictures, (5) Names 18 ‘pictures’] and Item
Communication outcomes included (a) ASD screening 22 [‘Uses three- to four-word sentences’, scoring (1)
and diagnostic tools, (b) direct standardised assessment Yes, (0) No]. Only behaviours observed during direct
of the children’s expressive language skills and (c) assessment were given credit, and only children’s ver-
standardised parent–report assessment of children’s bal communication (i.e. no use of AAC) was scored.
expressive language skills in home and community C. The VABS-II (Sparrow et al. 2005) is a
settings. Clinician administered assessments were parent/caregiver form used to understand a child’s
conducted by staff employed by the organisation with adaptive behaviours. Items are rated as 2 = usually,
experience in assessing children with ASD, but who 1 = sometimes or partially or 0 = never (Sparrow et al.
were not involved in the daily programme. These staff 2005). We used the following ‘Talking’ item scores
included the fourth author who is an experienced as per Norrelgen et al. 2014: Item 12 [‘Names at least
psychologist, and staff under her direct supervision with three objects (for example, bottle, dog, favourite toy,
all assessors trained in test administration and engaging etc.)’] and Item 18 [“Uses phrases with a noun and
in regular supervision to ensure use of standard a verb (for example, ‘Katie stay’; ‘Go home’; etc.)”].
administration procedures as per the test manuals. For
parent–report measures, parents/caregivers were
instructed to respond based on their child’s use of
verbal communication only, excluding use of AAC.
Results
Data screening was conducted to look for possible
A1. The SCQ (Rutter et al. 2003) is a missing data; some measures were missing because of
parent/caregiver ASD screening checklist. We used differences in the assessment battery across sites
Item 1 (‘Is she/he now able to talk using short phrases (ADOS), parent/caregiver checklists not being
or sentences?’) as a measure of spoken language level. returned (usually for exit assessments), missed
A2. The ADOS (Lord et al. 2000) is a semi- assessments because of scheduling constraints and a
structured standardised assessment of communica- small number of children who left the programme
tion, social interaction, play and restricted and repet- prior to assessment; these are presented in Table 1
itive behaviours (Lord et al. 2000). We used item A1 along with the children’s chronological age at, and
of the coding scheme (‘Overall level of Non-Echoed time between, assessments.
Spoken Language’) to determine children’s language
level at entry and exit. Just under a third of the partic-
Communication outcomes based on screening and
ipants (n = 76) completed ADOS assessments at both
diagnostic tools
entry and exit because of differences in data collection
across centres and measures for documenting out- As presented in Table 2, data derived from Item 1 of
comes in this cohort. the SCQ (‘Is she/he now able to talk using short phrases
B. The Mullen Scales of Early Learning (MSEL) or sentences?’) indicated that 62.5% of children were
(Mullen 1995) is a standardised assessment of devel- not talking in phrases/sentences at entry. At exit, this
opmental level of young children (Mullen 1995). reduced to 32.1%, with a significant increase in the
Items are scored along a continuum from 0 to 5, 0 in- proportion of children using phrase speech over time,
dicating an incorrect or absent response, and 5 (or the χ2 (1) = 38.782, P < 0.001, OR = 1.95. However, the

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
Journal of Intellectual Disability Research VOLUME 60 PART 5 MAY 2016
468
V. Rose et al. • Proportion of minimally verbal children with ASD

Table 1 Children’s chronological age at, and time between, assessments

Entry Exit Time between assessments

n Age (months) n Age (months) (months)

Valid Missing Median Mean SD Valid Missing Median Mean SD Mean SD

SCQ 228 0 44.00 44.18 9.44 170 58 59.50 59.29 7.51 14.39 6.61
ADOS 122 106 42.00 42.53 8.89 82 146 57.00 56.68 16.21 14.13 6.11
MSEL 226 2 44.50 44.61 9.55 176 52 60.00 59.56 7.39 14.07 6.64
VABS 224 4 44.39 44.00 9.56 172 56 60.00 59.39 8.80 13.99 6.63

Table 2 Social Communication Questionnaire (SCQ) – Item 1

Exit – ‘using phrases or sentences’

Total
No Yes n = 168

Entry – ‘using phrases or sentences’ No 31.0% (52) 31.5% (53) 62.5% (105)
Yes 1.2% (2) 36.3% (61) 37.5% (63)
Total 32.1% (54) 67.9% (114) 100.0% (168)

move towards the development of phrase speech was Module 1 = 69; Module 2 = 7; Exit: Module 1 = 48;
not uniform across the sample, with two children Module 2 = 27, missing = 1, χ2 (1) = 13.725,
(1.2% of sample) reported to be speaking in phrases at P < 0.001, OR = 1.44]. We were interested in the
entry but not at exit. percentage of children documented to be using less
Children’s scores on Item A1 of the ADOS than 5 spontaneous and functional words/word
(‘Overall level of Non-Echoed Spoken Language’) at approximations during the assessment (i.e. score of 3,
entry and exit are shown in Table 3. Of the 228 4 or 8; refer to Table 3 for definitions). Overall, 26.3%
children, 76 had ADOS assessments at both time of children (15.8% + 10.5%) used less than 5
points and were included in this analysis [Entry: functional words at their exit ADOS assessment

Table 3 Autism Diagnostic Observation Schedule (ADOS)

Exit ADOS A1
Total
Score 0 1 2 3 4/8 n = 76

Entry ADOS A1 0 32.9% (25) 1.3% (1) 0.0% (0) 0.0% (0) 0.0% (0) 34.2% (26)
1 9.2% (7) 2.6% (2) 3.9% (3) 0.0% (0) 0.0% (0) 15.8% (12)
2 6.6% (5) 2.6% (2) 1.3% (1) 1.3% (1) 0.0% (0) 11.8% (9)
3 2.6% (2) 7.9% (6) 0.0% (0) 9.2% (7) 1.3% (1) 21.1% (16)
4/8 2.6% (2) 0.0% (0) 0.0% (0) 5.3% (4) 9.2% (7) 17.1% (13)
Total 53.9% (41) 14.5% (11) 5.3% (4) 15.8% (12) 10.5% (8) 100.0% (76)

Note. 0, regular use of utterances with two or more words; 1, occasional phrases only; mostly single words; 2, recognisable single words or word
approximations only; must use at least five different words during ADOS evaluation; 3, at least one word or word approximation, but fewer than
five words used during the ADOS-2 evaluation; 4/8, no words or word approximations (depending on ADOS Module 1 or 2).

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
Journal of Intellectual Disability Research VOLUME 60 PART 5 MAY 2016
469
V. Rose et al. • Proportion of minimally verbal children with ASD

compared with 38.2% at entry (21.1% + 17.1%), with Table 6 shows the percentage of children who
a statistically significant increase in language level accurately labelled 0, 5–10, 11–14, 15–16, 17 and 18
over time, χ2 (16) = 84.153, P < .001, OR = 1.45. pictures correctly during the Picture Vocabulary task
(Item 18) of the MSEL. As shown, the majority of
children were not yet labelling pictures at entry (67.1%).
Communication outcomes based on standardised Of the 67.1% of children who were not labelling pictures
direct assessment at entry, 16.8% labelled 18 pictures at exit, while 37%
Four items of the MSEL Expressive Language Scale still could not label any pictures at exit, representing a
were used to examine changes in the children’s significant increase in language ability from entry to
spoken language skills. As per the assessment exit, χ2 (20) = 61.532, P < 0.001, OR = 1.76.
protocol adopted, only behaviours observed during As presented in Table 7, 68.6% of children were
assessment were scored. Table 4 presents the not ‘using three-to-four word sentences’ (Item 22) at
percentages of children who were using 0 words, 1 entry, reducing to 40.6% at exit, indicating a
word, 2–7 words and 8 or more words as scored on statistically significant decrease in proportion over
Item 11 of the MSEL (‘says first words’). The time, χ2 (1) = 44.583, P < 0.001, OR = 1.63.
proportion of children using 0 or only 1 word
reduced from 32.8% (31% + 1.8%) at entry, to 22.8%
Communication outcomes based on standardised
(20.5% + 2.3%) at exit, representing a significant
parent report
increase in language level over time, χ2 (16)
= 84.153, P < 0.001, OR = 1.43. Two items on the VABS-II Talking Domain were
As presented in Table 5, 57.2% of children were not used to examine changes in the children’s spoken
using ‘two word phrases’ (Item 17) at entry, reducing language skills as indicated by parent report. We were
to 36.4% at exit, with a significant increase in interested in the percentage of children documented
language level between entry and exit, χ2 (1) to ‘name(s) at least three objects’ as indicated by Item
= 49.128, P < 0.001, OR = 1.57. 12 of the VABS-II, and not consistently use ‘phrases

Table 4 Mullen Scales of Early Learning (MSEL) – Expressive Language Item 11

Exit – ‘Says first words’

Total
Score 0 words 1 word 2-7 words 8 words n = 171

Entry – ‘Says first words’ 0 words 15.8% (27) 1.8% (3) 3.5% (6) 9.9% (17) 31.0% (53)
1 word .6% (1) 0.0% (0) 0.0% (0) 1.2% (2) 1.8% (3)
2–7 words 2.3% (4) .6% (1) 3.5% (6) 8.2% (14) 14.6% (25)
8 words 1.8% (3) 0.0% (0) 4.1% (7) 46.8% (80) 52.6% (90)
Total 20.5% (35) 2.3% (4) 11.1% (19) 66.1% (113) 100.0% (171)

Table 5 Mullen Scales of Early Learning (MSEL) – Expressive Language Item 17

Exit – ‘Uses two-word phrases’

Total
No Yes n = 173

Entry – ‘Uses two-word phrases’ No 33.5% (58) 23.7% (41) 57.2% (99)
Yes 2.9% (5) 39.9% (69) 42.8% (74)
Total 36.4% (63) 63.6% (110) 100.0% (173)

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
Journal of Intellectual Disability Research VOLUME 60 PART 5 MAY 2016
470
V. Rose et al. • Proportion of minimally verbal children with ASD

Table 6 Mullen Scales of Early Learning (MSEL) – Expressive Language Item 18

Exit – Picture vocabulary

0–4 5–10 11–14 15–16 17 18 Total

pictures pictures pictures pictures pictures pictures n = 173

Entry – 0–4 pictures 37.0% (64) 5.8% (10) 6.4% (11) 1.2% (2) 0.0% (0) 16.8% (29) 67.1% (116)
Picture 5–10 pictures .6% (1) .6% (1) 1.7% (3) 0.6% (1) 0.0% (0) 3.5% (6) 6.9% (12)
Vocabulary 11–14 pictures 0.0% (0) 0.0% (0) .6% (1) 0.0% (0) 0.0% (0) 5.2% (9) 5.8% (10)
15–16 pictures 0.0% (0) 0.0% (0) 0.0% (0) 0.0% (0) 0.0% (0) .6% (1) .6% (1)
17 pictures 0.0% (0) 0.0% (0) 0.0% (0) 0.0% (0) 0.0% (0) .6% (1) .6% (1)
18 pictures .6% (1) 1.2% (2) 1.2% (2) .6% (1) 0.0% (0) 15.6% (27) 19.1% (33)
Total 38.2% (66) 7.5% (13) 9.8% (17) 2.3% (4) 0.0% (0) 42.2% (73) 100.0% (173)

Table 7 Mullen Scales of Early Learning (MSEL) – Expressive Language Item 22

Exit – ‘Uses three- to four- word sentences’

Total
No Yes n = 175

Entry – ‘Uses three- to four- word sentences’ No 40.6% (71) 28.0% (49) 68.6% (120)
Yes 1.7% (3) 29.7% (52) 31.4% (55)
Total 42.3% (74) 57.7% (101) 100.0% (175)

with a noun and a verb’ as indicated by Item 18. We than three words’ (p.3). As shown in Table 10, 53.5%
did not give full credit to children who were scored a of children were using fewer than three words at entry,
‘1’ indicating they ‘sometimes or partially’ displayed this and a further 20.3% were using more than three words
skill (i.e. score 0 or 1 rated as minimally verbal). As but rarely two word phrases. At exit, 29.4% of children
shown in Table 8, 52.4% of children were not were still using fewer than three words, and a further
consistently naming at least three objects (Item 12) at 15% were using three words, but not yet using two
entry, reducing to 29.5% of children at exit, with a word phrases, representing a significant increase in
significant increase in language level over time, χ2 (4) language level from entry to exit, χ2 (4) = 65.322,
= 61.009, P < 0.001, OR = 1.78. P < 0.001, OR = 1.66 (collapsed non-verbal and
Table 9 shows the number of children reported to minimally verbal).
be using ‘phrases with a noun and a verb’ (Item 18) as
indicated by the VABS-II. In total, 73.5% of children
Comparing parent and clinician ratings
were not consistently using two word phrases at entry,
reducing to 43.2% of children at exit, with a To examine the extent to which different measures of
significant increase in language level between entry children’s spoken language abilities yielded similar
and exit, χ2 (4) = 62.55, P < 0.001, OR = 1.70. results, we compared the scores for each child on
We computed scores by combining responses relevant items on the parent-completed SCQ (Item 1)
across these two items (12 and 18) to allow and VABS-II (Item 18), with those on the clinician-
comparison across studies using Norrelgen et al.’s administered MSEL (Item 17). To allow for
(2014) classification of minimally verbal as ‘using at comparisons across measures, and to ensure
least three words but never or only sometimes/partially consistency with our previous analyses, we
two-word phrases’, and non-verbal as ‘using fewer categorised children who received a ‘2’ on the VABS-

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
Journal of Intellectual Disability Research VOLUME 60 PART 5 MAY 2016
471
V. Rose et al. • Proportion of minimally verbal children with ASD

nd
Table 8 The Vineland Adaptive Behavior Scales – 2 Edition (VABS-II) – Talking Item 12

Exit – ‘name(s) at least three objects’

Total
No Yes n = 187

Entry – ‘name(s) at least three objects’ No 51.0% (50) 49.0% (48) 52.4% (98)
Yes 5.6% (5) 94.4% (84) 47.6% (89)
Total 29.4% (55) 70.6% (132) 100% (187)

nd
Table 9 The Vineland Adaptive Behavior Scales – 2 Edition (VABS-II) – Talking Item 18

Exit – using ‘phrases with a noun and a verb’

Total
No Yes n = 185

Entry – using ‘phrases with a noun and a verb’ No 58.1% (79) 41.9% (57) 73.5% (136)
Yes 2.0% (1) 98.0% (48) 26.5% (49)
Total 43.2% (80) 56.8% (105) 100% (185)

II as having speech, and ‘0’ (‘never’) or ‘1’ (‘sometimes/ received a community-based early intervention
partially’) as not. We combined scores from both programme. Overall, 26.3% of children exited the
entry and exit assessments and found 77.8% programme with fewer than five spoken
agreement across all three measures [49.8% of the words/phrases used spontaneously and
time all measures agreed ‘yes’ (to speech) and 50.2% communicatively as indicated by the ADOS (Kasari
agreed ‘no’]. We also found inconsistent agreement et al. 2013). We found that while just under half of the
across the three measures (i.e. two agreed, one did children in our study progressed from having minimal
not) on 22.2% of occasions. In these cases, the speech at entry (e.g. ‘says first words’ and ‘uses
assessment which was different (i.e. different to phrases with a noun and a verb’) to some speech (e.g.
response on both other measures) was SCQ in 23.9%, using phrases/sentences on the MSEL and SCQ, and
MSEL in 42.3% and VABS in 33.8%. using more than five spontaneous and functional
words/word approximations on the ADOS) at exit,
more than half of the children who entered the
Discussion
programme with a ‘very small repertoire of spoken
We aimed to investigate the proportion of children words’, exited the programme with similar language
who were minimally verbal in a group of children who levels. Consistent with previous research, we also

Table 10 Minimally verbal status using combined VABS-II Item 12 and Item 18

Exit – Verbal status

Total
Non-verbal Minimally verbal Phrase speech n = 187

Entry–- Verbal status Non-verbal 26.7% (50) 10.2% (19) 16.6% (31) 53.5% (100)
Minimally verbal 2.1% (4) 4.3% (8) 13.9% (26) 20.3% (38)
Phrase speech 0.5% (1) 0.5% (1) 25.1% (47) 26.2% (49)
Total 29.4% (55) 15.0% (28) 55.6% (104) 100.0% (187)

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
Journal of Intellectual Disability Research VOLUME 60 PART 5 MAY 2016
472
V. Rose et al. • Proportion of minimally verbal children with ASD

found that a small percentage of children (1.2%–4.7% (2014), who documented that 25% of their
depending on item used) regressed in their language participants were described as being non- or
use, although our figure was less than previously minimally verbal at 6.5 years as indicated by the
reported (Lord et al. 2004). However, previous VABS-II. These children were, on average,
studies have also included loss of vocalisations and approximately 18 months older than our sample.
fluctuating word loss in their reports, which was However, our finding that 29.4% of children in our
beyond the scope of our analysis (Lord et al. 2004). study met Norrelgen et al. (2014) criteria for non-
Differences in defining minimally verbal, measures verbal and 15% met their criteria for minimally verbal
used to operationalise this, age of participants and were considerably higher. In the Norrelgen et al.
sampling approaches all impact on outcome (2014) and our study, children had received
comparison across studies (Jonsdottir et al. 2007; approximately one to two years of intervention. When
Norrelgen et al. 2014). Tager-Flusberg et al. (2009) considering change in proportion over time, it is
and Kasari et al. (2013) highlighted the need to look at important to note that children were on average
finer-grained approaches to measuring 4.5 years of age in both studies and that magnitude of
communication outcomes in order to provide specific change may be greater had the studies included
detail regarding overall language level. In addition, it children who commenced intervention earlier.
is important that our definition of minimally verbal An important contribution of the current study was
aligns with our measurement. Consistent with the use of item level analysis from direct assessment, in
previous studies (Anderson et al. 2007; Magiati et al. addition to parent/caregiver report. We did find,
2011; Norrelgen et al. 2014), we took this approach by however, relatively high reliability between
looking at individual test scores, rather than reporting parent/caregiver report and clinician administered
overall standardised scores. We found significant assessments, indicating that all methods are useful for
changes in language level over time as assessed by measuring and documenting language level. This is
relevant items from diagnostic and screenings tools consistent with previous research in adaptive behaviour
(SCQ and ADOS). In addition, we found significant comparing caregiver and parent ratings on the VABS-II
changes in language when reporting on specific items (Lane et al. 2013). However, it is important to note subtle
from direct assessments (MSEL). While no measure differences between measures (e.g. ‘says first words’ on
can replace the gold standard of language sampling MSEL versus ‘names at least three objects’ on VABS-II),
and analysis of spoken language in natural which could lead to differences in scoring, depending on
environments to assess language use (Kasari et al. the child’s use of nouns in his or her early spoken
2013), standardised assessments offer a consistent vocabulary. Furthermore, we propose that reporting
testing context, and item level reporting as used in item scores, rather than overall standard scores or age
this study (e.g. SCQ, MSEL and VABS-II) is readily equivalents, yielded a clearer picture of children’s spoken
available and interpretable by caregivers, clinicians language levels with regards to the number of words used
and educators. Based on our definition of minimally functionally, and the complexity of their language use
verbal and outcome measures used, we found that (i.e. single words and phrases/sentences).
while there was a significant change in language level While there are inherent problems with the sole
following early intervention, a substantial number of reliance on standardised assessments to document
children remained without spontaneous and communication outcomes for children with minimal
functional language at exit of the programme, and a spoken language (e.g. placing demands on children to
small percentage regressed in their spoken language. understand and respond to questioning and, floor
Our finding that 26.3% of the children in our study effects because of difficulties performing under strict
did not develop functional language (ADOS) was standardised conditions) (Abbeduto et al. 2011; Kasari
slightly lower than reported in some earlier studies et al. 2013), our results suggest that item level analysis
(Kjelgaard & Tager-Flusberg 2001; Anderson et al. can provide further detail regarding specific language
2007; Magiati et al. 2011). These may be attributed to level, and we recommend this fine-grained data be
differences in definition, method, participant age and reported in intervention studies (Anderson et al. 2007;
other sample characteristics. Our findings from the Magiati et al. 2011). This would help us understand
ADOS are consistent with those of Norrelgen et al. the individual variability in communication outcomes

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
Journal of Intellectual Disability Research VOLUME 60 PART 5 MAY 2016
473
V. Rose et al. • Proportion of minimally verbal children with ASD

and provide detailed insight into specific language future research would include population level data
level and use. Furthermore, using multi-informant through health or education records; however, such
measures also provides information regarding data are not currently available in Australia; thus, our
individual language level across settings and contexts approach yields an estimate based on available data.
(Abbeduto et al. 2011). A further limitation of this study is the focus on
The assessment tools reported in our study are assessing spoken language only. As mentioned, the
commonly used in intervention research involving programme the children accessed supported the use
children with ASD (Trembath et al. 2016) and have of AAC, and it is therefore possible that some children
been designed for use in both clinical and research who used AAC in the programme may have
settings. Accordingly, we advocate for the consistent functional communication in other modalities, which
use of tools such as these, where possible, across was not captured in the standardised assessments. As
settings to allow for tracking and comparison of the parents in our study were instructed only to
children’s outcomes. However, we acknowledge that complete reports on communication based on spoken
the cost of administering these tools and expertise language, it is possible that communication conveyed
required likely act as barriers to their widespread use via other modalities was not captured. The proportion
in community based early intervention services. of children who were able to use phrases or functional
Fortunately, service providers routinely document communication using AAC is an important question,
children’s development in rich detail through but one that could not be answered in this study.
assessment reports, file notes, staff-parent Thus, in future research, collection of data on the use
correspondence, children’s work portfolios and of AAC including functional communication would
educational planning documents. Increasingly, this be of value in characterising the full spectrum of
information is being generated and stored communication outcomes.
electronically, creating the opportunity for sharing
and analysis of de-identified data. The challenge is to
Future research directions
develop and adopt a set of minimum standards for the
collection and reporting of spoken language Our findings point to the need for a consistent
information (e.g. spoken language benchmarks; definition and protocol for assessing communication
Tager-flusberg et al. 2009), which can be accurately outcomes in children with ASD with little or no
interpreted and easily applied by clinicians, parents functional speech, for example language sampling to
and educators working to support the learning and gain further insight into an individual’s overall
development of children with ASD. communication repertoire (Kasari et al. 2013). We
The findings of our study need to be considered encourage clinicians and researchers to specify
with respect to several limitations. As with previous communication outcomes and report individual
studies, population sampling, the intervention outcomes in their data. We stress the need to pay
programme children receive and the outcome closer attention to how we talk about this group of
measures employed impact on the proportion of individuals presenting with little or no functional
children found to be minimally verbal. It is likely that speech and how we operationalise this for research
some high functioning children will not yet be purposes. Ultimately, we advocate movement
diagnosed by preschool age (Anderson et al. 2007; towards a uniform definition for research purposes,
Centers for Disease Control and Prevention 2014), including the reporting of standard measures. Given
and therefore may not have been included in our that standardised assessments such as those used in
sample. Likewise, even if diagnosed, these children this study are frequently used in research, they
may not present for early intervention because of provide an immediate avenue through which
displaying higher levels of adaptive behaviour. That researchers can report finer detail, and authors of
said, the ASD-specific centre-based early intervention previously published research can revisit their data
programme caters for children with a range of abilities and publish their findings so that we can benefit from
and is the single largest ASD early intervention information already collected. In addition, given the
provider in the Queensland region, suggesting that it use of AAC with individuals with complex
likely captures a broad range of abilities. Ideally, communication needs, it is important to report on

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
Journal of Intellectual Disability Research VOLUME 60 PART 5 MAY 2016
474
V. Rose et al. • Proportion of minimally verbal children with ASD

AAC use in addition to spoken language (e.g. number Acknowledgements

of spontaneous words/phrases/sentences used
We warmly thank the children, parents and
functionally using alternate communication systems).
caregivers, and staff whose participation in the early
As there is increasing awareness of, and evidence
intervention programme and study made this research
supporting the use of AAC with individuals with ASD
possible. In addition, our sincerest thanks go to Peggy
and minimal speech, it is important that we recognise
Wong at AEIOU for her countless hours spent
and acknowledge the contribution of these systems
entering the data for the current study.
for enhancing the communicative competence for
individuals with little or no functional speech, by
reporting on the use of these programmes as part of an References
individual’s overall communication repertoire (Goods Abbeduto L., Kover S. T. & McDuffie A. (2011) Studying
et al. 2013; Kasari et al. 2014). Finally, in addition to the language development of children with intellectual
analysing how children with ASD communicate (i.e. disabilities. In: Research Methods in Child Language (ed.
spoken language, AAC and combination of E. Hoff). Wiley-Blackwell, Hoboken.
modalities), future research should also evaluate why American Psychiatric Association (2013) Diagnostic and
children with ASD communicate and how effectively Statistical Manual of Mental Disorders, 5th edn. American
Psychiatric Association, Washington, DC.
and functionally they use their communication skills
American Speech-Language-Hearing Association (2015) Autism:
to interact and learn from their environment. overview [Online]. Available: http://www.asha.org/PRPSpecific
Topic.aspx?folderid=8589935303&section=Overview.
Anderson D. K., Lord C., Risi S., DiLavore P. S., Shulman
C., Thurrn A. et al. (2007) Patterns of growth in verbal
Conclusion abilities among children with autism spectrum disorder.
Journal of Consulting and Clinical Psychology 75, 594–604.
Of the participants in our study who entered the
Centers for Disease Control and Prevention (2014) Prevalence of
programme as minimally verbal, more than half
autism spectrum disorder among children aged 8 years—autism and
continued to present with significant communication developmental disabilities monitoring network, 11 Sites, United
needs at exit of the programme with regards to spoken States, 2010. Available: http://www.cdc.gov/mmwr/preview/
language. Nevertheless, of those who did progress from mmwrhtml/ss6302a1.htm (accessed 1 September 2015).
initially a very small repertoire of spoken words or fixed Charman T., Drew A., Baird C. & Baird G. (2003)
phrases to having some speech, a number of individuals Measuring early language development in preschool
children with autism spectrum disorder using the
noticeably progressed in their language skills,
MacArthur Communicative Development Inventory
highlighting considerable differences in language (Infant Form). Journal of Child Language 30, 213–36.
trajectories irrespective of initial language level at entry. Dawson G. (2008) Early behavioral intervention, brain
The findings of this research indicate that standardised plasticity, and the prevention of autism spectrum disorder.
assessments can be used to document communication Development and Psychopathology 20, 775–803.
outcomes for children with ASD with little or no Fenson L., Dale P., Reznick J., Thal D., Bates E., Hartung J. et al.
functional speech, but that more attention should be (2003) MacArthur communicative development inventories: User’s
given to particular items administered rather than solely guide and technical manual, Paul H. Brookes, Baltimore, MD.
Fenson L., Marchman V A., Thal D J., Dale P S., Reznick J
standard scores. We encourage other researchers to
S. & Bates E. (2007) MacArthur-bates communicative
report on item level scores in order to contribute to our development inventories: User’s guide and technical manual,
understanding of the communication profiles of 2nd edn. (ed. P. H. Brookes) Baltimore, MD.
individuals with ASD and their response to Fernell E., Eriksson M. A. & Gillberg C. (2013) Early
intervention. In addition to reporting spoken language diagnosis of autism and impact on prognosis: a narrative
outcomes for individuals with ASD, we need to review. Clinical Epidemiology 5, 33–43.
recognise the importance of documenting AAC use. Goods K., Ishijima E., Chang Y. C. & Kasari C. (2013)
While spoken language constitutes an important part of Preschool based JASPER intervention in minimally verbal
children with autism: pilot RCT. Journal of Autism and
the communication process, it is important that we are Developmental Disorders 43, 1050–6.
also evaluating other important communication skills Jonsdottir S. L., Saemundsen E., Asmundsdottir G.,
including, non-verbal and augmentative and alternative Hjartardottir D., Asgeirsdottir B. B., Smaradottir H. H.
communication. et al. (2007) Follow-up of children diagnosed with

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
Journal of Intellectual Disability Research VOLUME 60 PART 5 MAY 2016
475
V. Rose et al. • Proportion of minimally verbal children with ASD

pervasive developmental disorders: stability and change autism-specific, early intervention programme.
during the preschool years. Journal of Autism and International Journal of Special Education 30, 1–7.
Developmental Disorders 37, 1361–74. Paynter J., Scott J., Beamish W., Duhig M. & Heussler H. S.
Kasari C., Brady N., Lord C. & Tager-Flusberg H. (2013) (2012) A pilot study of the effects of an australian centre-
Assessing the minimally verbal school-aged child with based early intervention program for children with autism.
autism spectrum disorder. Autism Research 6, 479–93. The Open Pediatric Medicine Journal 6, 7–14. DOI: 10.2174/
Kasari C., Kaiser A., Goods K., Nietleld J., Mathy P., Landa 1874309901206010007.
R. et al. (2014) Communication interventions for minimally Paul R., Campbell D., Gilbert K. & Tsiouri I. (2013)
verbal children with autism: a sequential multiple Comparing spoken language treatments for minimally
assignment randomized trial. Journal of the American verbal preschoolers with autism spectrum disorders. Journal
Academy of Child & Adolescnt Psychiatry 53, 635–46. of Autism and Developmental Disorders 43, 418–31.
Kjelgaard M. M. & Tager-Flusberg H. (2001) An investigation Prior M., Roberts J. M. A., Rodger S., Williams K. & Sutherland
of language impairment in autism: implications for genetic R. (2011) A Review of the Research to Identify the Most Effective
subgroups. Language and cognitive processes 16, 287–308. Models of Practice in Early Intervention of Children with Autism
Lane B. R., Paynter J. & Sharman R. (2013) Parent and Spectrum Disorders. Department of Families, Housing,
teacher ratings of adaptive and challenging behaviours in Community Services and Indigneous Affairs, Australia.
young children with autism spectrum disorders. Research Rogers S. J. & Vismara L. A. (2008) Evidence-based
in Autism Spectrum Disorders 7, 1196–203. comprehensive treatments for early autism. Journal of
Lee L. C., David A. B., Rusyniak J., Landa R. & Clinical Child and Adolescent Psychology 37, 8–38.
Newschaffer C. J. (2007) Performance of the Social Rutter M. (1978) Diagnosis and definition of childhood autism.
Communication Questionnaire in children receiving Journal of Autism and Childhood Schizophrenia 8, 139–61.
preschool special education services. Research in Autism Rutter M., Bailey A. & Lord C. (2003) The Social
Spectrum Disorders 1, 126–38. Communication Questionnaire. Western Psychological
Lord C., Risi S., Lambrecht L., Cook E. H. J., Leventhal B. Services, Los Angeles.
L., DiLavore P. C. et al. (2000) The autism diagnostic Sparrow S. S., Cicchetti D. V. & Balla D. A. (2005) Vineland
observation schedule-generic: a standard measure of social and TM
Adaptive Behavior Scales, Second Edition (Vineland -II).
communication deficits associated with the spectrum of autism. AGS Publishing, Circle Pines, MN.
Journal of Autism and Developmental Disorders 30, 205–23.
Tager-flusberg H., Rogers S., Cooper J., Landa R., Lord C.,
Lord C., Shulman C. & DiLavore P. (2004) Regression and Paul R. et al. (2009) Defining spoken language
word loss in autistic spectrum disorders. Journal of Child benchmarks and selecting measures of expressive language
Psychology and Psychiatry 45, 936–55. development for young children with autism spectrums
Luyster R. J., Kadlec M. B., Carter A. & Tager-Flusberg H. disorders. Journal of Speech, Language, and Hearing
(2008) Language assessment and development in toddlers Research 52, 643–52.
with autism spectrum disorders. Journal of Autism and Trembath D. & Vivanti G. (2014) Problematic but
Developmental Disorders 38, 1426–38. predictive: individual differences in children with autism
Magiati I., Moss J., Charman T. & Howlin P. (2011) spectrum disorders. International Journal of Speech-
Patterns of change in children with autism spectrum Language Pathology 16, 57–60.
disorders who received community based comprehensive Trembath D., Westerveld M. & Shellshear L. (2016) Assessing
interventions in their pre-school years: a seven year follow- spoken language outcomes in children with ASD: A systematic
up study. Research in Autism Spectrum Disorders 5, 1016–27. review. Current Developmental Disorder Reports 3, 33–45.
Matson J. L., Mahan S., Kozlowski A. M. & Shoemaker M. Woynaroski T., Yoder P. & Watson L. R. (2015) Atypical
(2010) Developmental milestones in toddlers with autistic cross-modal profiles and longitudinal associations
disorder, pervasive developmental disorder—not between vocabulary scores in initially minimally verbal
otherwise specified and atypical development. children with ASD. Autism Research Epub ahead of print.
Developmental Neurorehabilitation 13, 239–47. DOI: 10.1002/aur.1516.
Mullen E. (1995) Mullen scales of early learning. American Yoder P. & Stone W. L. (2006) Randomized comparison of
Guidance Service Inc, Circle Pines, MN. two communication interventions for preschoolers with
National Autism Center (2015) Findings and Conclusions: autism spectrum disorders. Journal of Consulting and
National Standards Project, Phase 2. National Autism Clinical Psychology 74, 426–35.
Center, Randolph, MA. Yoder P., Watson L. R. & Lambert W. (2014) Value-added
Norrelgen F., Fernell E., Eriksson M., Hedvall Å., Persson predictors of expressive and receptive language growth in
C., Sjolin M. et al. (2014) Children with autism spectrum initially nonverbal preschoolers with autism spectrum
disorders who do not develop phrase speech in the disorders. Journal of Autism and Developmental Disorders
preschool years. Autism 19, 934–43. 45, 1254–70.
Paynter J., Riley E. P., Beamish W., Scott J. G. & Heussler
H. S. (2015) Brief report: an evaluation of an australian Accepted 15 March 2016

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
John Wiley & Sons Ltd
476

Appendix: Minimally verbal terminology and definitions

Author/s Title Definition Relevant outcome measure used (if applicable)

(Goods et al. 2013) Preschool based JASPER intervention in ‘Minimally verbal’ ‘used less than 10 The Reynell Developmental Language

John Wiley & Sons Ltd

minimally verbal children with autism: spontaneous, functional and communicative Scales – Verbal Comprehension and
pilot RCT words by parent and teacher report and during Expressive Language mean scores. MSEL –
the baseline or entry assessments’ (p. 1051) Receptive and Expressive Language mean scores.
(Jonsdottir et al. 2007) Follow-up of children diagnosed with ‘non-verbal’ ADI-R – classify language based on A1 ‘overall
pervasive developmental disorders: level of language’ ‘(1) fewer than five words
stability and change during the preschool total and/or no use of speech on a daily basis;
Journal of Intellectual Disability Research

years (2) words but no phrases; no functional use of

three word phrases in spontaneous, echoed
or stereotyped speech, but use of speech on a
daily basis with at least five different words in
the last months; (3) phrases; functional use of
spontaneous, echoed or stereotyped language
that, on a daily basis, involves phrases of three
words or more that at least sometimes include
a verb and is comprehensible to other people.
Categories 1 and 2 define non-verbal status,
V. Rose et al. • Proportion of minimally verbal children with ASD

while category 3 defines verbal status’ (p. 1365).

(Kasari et al. 2013) Assessing the minimally verbal school- ‘Minimally verbal’ ‘Although from a clinical/ Not applicable. However discussion regarding
aged child with autism spectrum disorder educational perspective, the exact number best ways for capturing language level in
of words used does not matter that much, minimally verbal children with ASD.
researchers may want to impose a quantitative
definition for this population (e.g. fewer than
20 functional words)’ (p. 2).
(Kasari et al. 2014) Communication interventions for minimally ‘Minimally verbal’ ‘fewer than 20 spontaneous Transcription of Natural Language Sample –
verbal children with autism: a sequential different words used during the 20 minutes both spoken and Speech Generator Device-
multiple assignment randomized trial NLS [natural language sample]’ (p. 637) produced spontaneous utterances were
transcribed and coded, and mode noted.
Peabody Picture Vocabulary Test (PPVT-4)
VOLUME

Test of Early Language Development

60

(Norrelgen et al. 2014) Children with autism spectrum disorders ‘Minimally verbal’ ‘using at least three words VABS-II- Expressive Language age equivalent
PART

who do not develop phrase speech in the but never or only sometimes/partially two-word scores, as well as individual items from the
preschool years phrases and an expressive age equivalent ‘Talking’ domain to classify language level.
corresponding to below 24 months’ (p. 3)‘Non-
verbal’‘using fewer than three-words and an

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
5 MAY 2016

(Continues)
477

(Continued)
(Continued)

Author/s Title Definition Relevant outcome measure used (if applicable)

expressive age equivalent corresponding to

below 15 months’ (p. 3)

John Wiley & Sons Ltd

(Paul et al. 2013) Comparing spoken language treatments ‘Minimally verbal’ ‘spontaneous expressive Communication and Symbolic Behavior Scales –
for minimally verbal pre-schoolers with vocabulary by parent report of fewer than Caregiver QuestionnaireCommunication and
autism spectrum disorders 15 words’ (p. 420). Symbolic Behavior Scales-Behavioral Observation
VABS-II – expressive language age equivalent.
MSEL – non-verbal mental age and visual reception
composite scores.
(Woynaroski et al. 2015) Atypical cross-modal profiles and longitudinal ‘Minimally verbal’ and ‘preverbal’ ‘(c) were MacArthur-Bates Communicative Development
Journal of Intellectual Disability Research

associations between vocabulary scores in reported to say no more than 20 different Inventories: Words and Gestures Checklist
initially minimally verbal children with ASD words according to a parent report on the (MCDI) - Used raw scores to enable maximum
MacArthur-Bates Communicative Development sample size and appropriate imputation, and age
Inventories: Words and Gestures Checklist equivalency scores.
(MCDI) [Fenson et al., 2003]; and (d) produced
no more than five different word roots during
a 15-min language sample’ (p. 3)
(Yoder & Stone 2006) A randomized comparison of the effect of ‘Non-verbal’ or ‘low verbal’‘Fewer than 20 Semi structured free-play with examiner (SFPE)
two prelinguistic communication interventions different words used cumulatively during three Developmental
on the acquisition of spoken communication in communication samples’ (p. 700). Play Assessment Turn-taking assessment
V. Rose et al. • Proportion of minimally verbal children with ASD

preschoolers with ASD.

(Yoder et al. 2014) Value-added predictors of expressive and ‘Non-verbal’ ‘(c) were reported to say no more MacArthur Communicative Development
receptive language growth in initially non-verbal than 20 different words according to parent Inventory-Words and Gestures Form (MCDI) –
pre-schoolers with autism spectrum disorders. report on the MacArthur Bates Communicative Raw scoresCommunication and Symbolic Behavior
Development Inventories: Words and Gestures Scales – Developmental Profile Behavior Sample
checklist (Fenson et al. 2007); and (d) produced (CSBS) – weighted raw scoresUnstructured
no more than five different word roots during a Communication Sample
15-minute language sample’ (p. 3).
60 VOLUME
PART

© 2016 MENCAP and International Association of the Scientific Study of Intellectual and Developmental Disabilities and
5 MAY 2016