6 hours

Research Report

Integrated speech and phonological awareness intervention for pre-school children with Down syndrome

Anne Katherine van Bysterveldt†, Gail Gillon‡ and Susan Foster-Cohen§

†Department of Communication Disorders, University of Canterbury, Christchurch, New Zealand ‡College of Education, University of Canterbury, Christchurch, New Zealand §The Champion Centre, Burwood Hospital, Christchurch, New Zealand

(Received 14 October 2008; accepted 27 April 2009)

Abstract

Background: Children with Down syndrome experience difficulty with both spoken and written language acquisition, however controlled intervention studies to improve these difficulties are rare and have typically focused on improving one language domain. Aims: To investigate the effectiveness of an integrated intervention approach on the speech, letter knowledge, and phonological awareness development of ten pre-school children with Down syndrome aged between 4;4 and 5;5. Methods & Procedures: A multiple single-subject design was used to evaluate treatment effectiveness. Baseline and intervention measures for speech and pre- and post-intervention measures for letter knowledge and phonological awareness were compared. The intervention comprised three components: a parent-implemented home programme; centre-based speech–language therapy sessions, and ‘Learning through Computer’ sessions with a total intervention time of 20 hours over 18 weeks. Letter knowledge and phonological awareness activities were linked to each child’s speech targets. Outcomes & Results: Results indicated significant treatment effects on speech measures for all ten participants. Six of the ten participants showed increases on letter knowledge and nine showed increased awareness of initial phonemes in words but responses were not above binomial chance level (that is, 70% correct) for phonological awareness tasks. Individual results are presented and implications for parents and therapists are discussed. Conclusions and Implications: The findings of this study suggest an intervention approach that integrates speech, letter knowledge, and phonological awareness targets is effective in remediating speech error patterns at the single- word level in young children with Down syndrome. Phonological awareness and letter knowledge appeared to be stimulated through the intervention, but significant improvement above chance levels on untrained phonological awareness tasks was not evident. Follow-up investigation is necessary to determine longer-term outcomes.

Keywords: Down syndrome, speech intervention, phonological awareness intervention.

What this paper adds Children with Down syndrome experience difficulty with both spoken and written language acquisition, exhibiting persistent speech difficulties and poor underlying skills critical for early reading development. Intervention research to address these deficits is scarce and previous studies have typically focused on remediating one language domain. The present study advances our knowledge through investigating the effectiveness of an integrated intervention approach which aims to improve speech, letter knowledge, and phonological awareness development in pre-school children with Down syndrome. This study suggests an intervention which includes a letter knowledge and phonological awareness focus is effective in remediating speech errors at the single-word level in pre-school children with Down syndrome and may stimulate letter knowledge and early phonological awareness in this population.

International Journal of Language & Communication Disorders ISSN 1368-2822 print/ISSN 1460-6984 online q 2010 Royal College of Speech & Language Therapists

http://www.informahealthcare.com DOI: 10.3109/13682820903003514

Address correspondence to: Anne Katherine van Bysterveldt, Department of Communication Disorders, University of Canterbury, Private Bag 4800, Christchurch, New Zealand; e-mail: [email protected]

INT. J. LANG. COMM. DIS., MAY–JUNE 2010, VOL. 45, NO. 3, 320–335

Introduction

Individuals with Down syndrome (DS) experience difficulty with both spoken and written language acquisition. They demonstrate poorer verbal skills compared with non-verbal skills and perform more poorly on language measures and speech production measures than do peers with other cognitive impair- ments matched for mental age (Smith and Stoel- Gammon 1983, Roberts et al. 2008). These findings lead to the conclusion that individuals with DS present with speech and language impairment in addition to cognitive impairment. Children with concurrent speech and language impairment are known to be at high risk for reading difficulties (for example, Catts et al. 2001) which suggests that reading development in children with DS may be particularly compromised.

Reading achievement is a valid goal for young children with DS and is associated with superior spoken language abilities (Cardoso-Martins et al. 2008), however research investigating variables that influence early reading acquisition is limited. Recent research suggests children with DS perform poorly on phonological awareness (for example, Snowling et al. 2002) — an area critical for early reading success.

Controlled intervention studies to improve spoken or written language development in children with DS are rare and have typically focused on improving only one language domain. However, given these children’s concurrent speech and language deficits and weakness in the underlying skills critical for reading, an integrated intervention approach that may lead to improvements in both spoken and written language is intuitively appealing. An integrated intervention aims to simultaneously facilitate the development of different skills, in this instance speech production, letter name and sound knowledge, and phonological awareness. The current study sought to investigate the impact of such an intervention on the speech, letter knowledge, and phonological awareness development of pre-school children with DS.

Speech difficulties in children with DS

An integrated intervention that will simultaneously improve speech articulation in addition to stimulating phonological awareness is important, given the well documented speech production difficulties in children with DS. Although early vocalizations and babble appear similar to those produced by their peers with typical development, phonological development and early word production are substantially delayed in children with DS (Smith and Stoel-Gammon 1983). The speech of children with DS is reported to contain the same type of error patterns and is qualitatively

similar to that of typically developing younger children (Smith and Stoel-Gammon 1983), however, slower speech development in individuals with DS results in a widening gap across time.

As well as delay, Dodd et al. (1994) report characteristics of phonological disorder, including significantly more atypical speech error patterns and greater inconsistency in the speech errors of individuals with DS compared with mental age-matched peers. Given the phonological speech error patterns observed in children with DS, a phonological approach to intervention where children’s conscious awareness of the sounds structure of the spoken word is enhanced, in addition to practice in articulating correct speech patterns, may be effective in improving the articulation of words.

Interventions to enhance speech production

A number of interventions have focused on physical and motor based activities to improve the speech production of children with DS. For example, in a radical approach to try to improve speech intelligibility, children with DS may undergo tongue reduction surgery. Parsons et al. (1987) compared the articulation errors of 18 children with DS pre-, post- and 6 months after tongue reduction surgery and found no significant differences across the three measures, nor between the surgery group and a non-surgery contrast group.

Non-speech oral motor exercises and treatments have been recommended for children with DS to increase physical awareness and strength of the oral and facial muscles used during speech (Kumin 2006). However, Barnes et al. (2006) reported a disassociation between oral motor structure and function, with children with DS demonstrating better oral function skills than speech function skills, supporting the hypothesis that remediating oral-motor deficits will not improve speech production. Reviews of the literature investigating the use of non-speech oral motor exercises indicated that the evidence does not support their effectiveness in the treatment of speech disorders (Lof and Watson 2008), nonetheless, their use is still widely reported by clinicians working with children with speech sound disorders, including children with DS (Lof and Watson 2008). Thus, there is an urgent need to explore other types of interventions such as phonologically based interven- tions that may hold more promise to improving speech difficulties in children with DS.

Phonological intervention approaches have been extensively used with young children with spoken language impairment in the absence of DS (Hesketh et al. 2000, Gillon 2005). However, only two phonological intervention approaches to improving

Integrated intervention for children with DS 321

speech difficulties in children with DS have been documented (Cholmain 1994, Dodd et al. 1994). Cholmain’s study included six children with DS aged 4;1–5;6 with language ages of 1;3–2;8. The interven- tion was implemented via a modified cycles approach (Hodson and Paden 1991) to remediate phonological patterns, and was delivered using amplification with a focus on contrastive versus homophonic production of the targeted patterns in words, and a home-based programme of speech targets practice and joint story reading. Despite previous stability in the 3–12 months before intervention, response to the intervention was rapid with all the children demonstrating positive change in the first two weeks of the intervention.

Nine children with DS aged between two and 6 years took part in the Dodd et al. (1994) study which used a core vocabulary approach to reduce non- developmental errors and variability in the children’s speech. Parents were trained to provide corrective feedback to their child who was required to produce a consistent pronunciation of each target word, with no non-developmental errors present. Fewer errors and inconsistencies were reported post-intervention. These two studies demonstrate a phonological intervention approach can be effective for young children with DS and that parents can play a key role in modelling target production and providing corrective feedback to improve their child’s speech. The current study extends this early work through explicitly linking goals to stimulate areas of phonological development that are critical for reading success, with goals to stimulate correct phonological production.

Integrated phonological awareness intervention

Integrated phonological awareness intervention which incorporated phonological awareness, letter knowledge and speech production goals has been shown to be effective for young children with speech disorders and normal cognitive ability (Gillon 2005, McNeill et al. 2009). Intervention to improve phonological awareness in children with DS has been the focus of a number of recent studies. Reported findings included improve- ment in reading skills, alphabet knowledge, and phonological awareness (van Bysterveldt et al. 2006, Goetz et al. 2008), however the effects of the interventions on speech production were not examined. Recent evidence suggests that achieving a precise phonological representation of a word is particularly demanding for children with DS ( Jarrold et al. 2009). Improving explicit awareness of the underlying representation of a spoken word and a child’s ability to consciously access this representation through phoneme awareness tasks (such as identifying the initial sound in a word or segmenting a word into

phonemes), may improve the child’s speech production of the target word.

There is increasing research into the role of phonological awareness in reading for children with DS, and a positive relationship between phonological awareness and reading for individuals with DS has been reported by researchers (Cupples and Iacono 2000, Snowling et al. 2002, Goetz et al. 2008). For example, Cupples and Iacono (2000) examined the relationship between phonological awareness and reading skills in young children with DS aged 6 to 10 years and found better phonological awareness skills were associated with better reading of both real and non-words.

Roch and Jarrold (2008) compared real word and non-word reading ability in individuals with DS. They found that based on the relationship between these skills in the typically developing control group, non- word reading skills of participants with DS were poorer than would be predicted from their real word reading ability. However, reported correlation coefficients between non-word reading and composite phonologi- cal awareness scores were equivalent between the two groups. Thus, while real and non-word reading appeared differently related, non-word reading and phonological awareness skills were similarly related in both groups. Therefore, the research appears to suggest that while individuals with DS do exhibit strength in visual reading strategies relative to other reading sills, phonological awareness skills also play a role in supporting reading in this population.

Letter knowledge and phonological awareness

The importance of phonological awareness and letter knowledge in early reading and spelling acquisition is now well established, with both skills strongly predictive of later reading outcomes (Wagner et al. 1994). Letter name and sound knowledge are distinct but overlapping skills, with letter name knowledge typically preceding and predictive of letter sound knowledge and both skills associated with reading- related measures (McBride-Chang 1999). Letter knowledge and phonological awareness are also positively associated and reciprocal gains in these skills have been demonstrated when their instruction has been integrated and explicitly linked (Oudeans 2003). Joint book reading, a key element of the home literacy environment (for example, Bus et al. 1995) is an important medium through which specific early literacy skills such as phonological awareness and letter knowledge may be targeted (Justice and Ezell 2000, van Bysterveldt et al. 2006).

Given the potential of phonological intervention and the important role of parents in supporting such an approach to improve speech in children with DS,

322 Anne Katherine van Bysterveldt et al.

as well as the success of an integrated phonological awareness approach for children with other types of speech and language impairment, the current study sought to investigate the impact of an intervention approach which combines all these components, on the speech and phonological awareness development of pre- school children with DS. Specifically it was hypoth- esized that the experimental integrated phonological awareness intervention would improve the following skills:

. Speech production accuracy in trained and untrained speech targets.

. Letter name and letter sound knowledge.

. Phonological awareness skills on untrained phoneme level tasks.

Method

Research design

A multiple single-subject repeated measures (duplicated AB) design was used to evaluate treatment effectiveness.

Participant selection process

Ten children (five girls and five boys) with DS who ranged in age from 4 years 4 months to 5 years 5 months (mean ! 4 years 11 months, standard deviation (SD) ! 4.08 months) participated in the study. The participants who were from a range of socio- economic backgrounds were recruited from a group of thirteen 4- and 5-year-old children enrolled in a transition to school programme for children with DS at a specialist early intervention centre. Throughout the intervention participants attended weekly sessions at the centre, in small groups of up to six children where they received an individual programme based on the national curriculum. Each child received services from a team of six specialists/therapists which included a physiotherapist or occupational therapist, a cognitive therapist, a speech–language therapist, a music therapist, an early childhood teacher and a computer specialist who they saw individually and sequentially throughout the morning.

Following parent consent, criteria for inclusion in the intervention study were (1) a diagnosis of Down syndrome with no known current major medical conditions or additional developmental disabilities, (2) enrolment in the centre’s transition to school clinic, (3) standard New Zealand English as a first and only language and (4) no additional speech therapy for the duration of the intervention. These criteria excluded three children from the study, one with a serious medical condition and two who intended to leave the centre before the completion of the intervention period.

Procedure

Standardized and experimental measures were used to assess the speech, expressive and receptive language, phonological awareness, letter knowledge and hearing of participants. Formal assessment of participants’ mental age was incompatible with the philosophy of the early intervention centre. The lead researcher (a qualified speech–language therapist) administered all the experimental tasks. Two senior student Speech– Language Therapists assisted in the administration of the standardized assessments under clinical supervision procedures.

Participants received audiological assessment twice during the intervention period. Seven participants had some degree of hearing loss ranging from slight to moderate. Average hearing thresholds across all frequencies tested, both ears (if tested) and both times (if tested) ranged from 17.5 to 39.37 dB. Six of the ten participants wore corrective glasses and one child was scheduled to have cataract surgery at the completion of the intervention.

Participants were assessed on the following measures of receptive and expressive language:

Standardized assessments:

. Peabody Picture Vocabulary Test — III (PPVT- III) (Dunn and Dunn 1997). This receptive vocabulary test requires the child to point to one of four pictures named by the examiner. The assessment was administered and scored according to the examiner manual. As standard score floor effects were apparent, both raw scores and standard scores are reported.

. Pre-School Language Scale — Fourth Edition (Australian Language Adaptation) (PLS-4) (Zimmerman et al. 2002). This norm-referenced test assesses the receptive (Auditory Comprehen- sion) and expressive (Expressive Communication) language of children aged from birth to 6 years 11 months. The assessment was administered and scored according to the examiner manual. Standard score floor effects were apparent for the majority of children therefore a Total Language Score (TLS) presented as a language age score is also reported.

Participants’ speech production was initially assessed 6 weeks before the start of the intervention using the following measures:

Hodson Assessment of Phonological Patterns — Third Edition (HAPP-3) (Hodson 2004). This is a single word articulation test comprising 50 single and multi syllabic words elicited by naming manipulatives and line drawings. For all speech data, where a spontaneous response could not be elicited by the picture or stimulus item, a response was elicited

Integrated intervention for children with DS 323

following delayed imitation. Speech data were recorded using a high-quality digital voice recording device (Belkin F8E462). All responses were transcribed via broad transcription. These samples were analysed using Computer Profiling (PROPH; Long and Fey 2005). Initial assessment data for measures of speech and language are reported in Table 1.

The following measures were administered pre- and post-intervention to all ten participants:

Letter knowledge and phonological awareness experimental measures

. Letter name and letter sound knowledge tasks. Letter name and sound knowledge was assessed using the Gillon Preschool Phonology and Letter Knowledge probes (Gillon 2005). These probes comprise a lower-case presentation of all 26 letters presented six at a time on an A4-sized grid. The child is required to point to the letter name or sound the assessor names, for example ‘Which one says ssss?’, or ‘Show me the letter t’. To reduce the effects of a position response pattern, a correct response was credited when the appropriate letter was identified on two occasions. The phonological awareness and letter knowledge tasks did not require a verbal response.

. Initial phoneme identity (Gillon 2005). In this task the therapist names a large colourful picture of an animal and draws the child’s attention to the first sound of that word. The child is then required to identify this as the initial sound in a word from a choice of three pictures presented and named by the examiner. Corrective feedback is given if required. The test comprised two training items and ten test items.

. Initial phoneme identity with words (Gillon and McNeill 2007). In this task the therapist presents a large letter and draws the child’s attention to the sound it makes. The child is then required

to identify this as the initial sound in a word and match this to one of three cards consisting of a picture with the word written underneath, presented and named by the examiner. Corrective feedback is given if required. The test comprised one training item and twelve test items.

(Phonological awareness measures may be freely downloaded from: http://www.education.canterbury. ac.nz/people/gillon/resources.shtml).

Intervention target selection

For each participant, four speech targets for interven- tion and a control target were selected based on the initial speech assessment data (Table 2). Each speech target was then assessed with twelve speech cards to determine the percentage usage of dominant phonolo- gical error patterns in the target words presented.

Pre- and post-intervention measures

Twelve speech cards consisting of pictures with words of a similar phonological structure were prepared for each speech target (that is, 48 cards per child). Six were randomly selected to be the trained items and included in the intervention activities with the remaining six assessed only as a generalization measure. Correct production of the target phoneme in the appropriate word position was credited correct; however non- targeted phonemes in the word may have been in error. For example a child whose speech target was the correct articulation of final /p/ would be credited as correct with the production of the word ‘cup’ as /tLp/. An additional set of twelve speech cards was prepared as a control measure for each child. This set was not included in the intervention activities and was assessed only. The cards were modelled on the format used by Gillon and McNeill (2007). (Examples of these speech

Table 1. Participants’ assessment data

P1 P2 P3 P4 P5 P6 P7 P8 P9 P10

Child’s age 4;11 5;0 5;05 4;10 5;02 5;05 4;04 4;09 5;02 4;08 Gender F M F F M F F M M M PPVT-III RSa 42 3 36 13 18 28 19 39 28 31 PPVT-III SSb 81 40* 68 48* 53 60* 65 80* 62 73 PLS-4:ACc 66 50 50 50 50 50 60 74 51 73 PLS-4:ECd 62 50 53 50 50 55 50 79 52 77 PLS-4:TLSe 3;4 1;5 2;8* 1;8* 2;2 2;5* 2;1 3;3* 3;0 3;1 PCC-Rf 66.7 36.2* 72.1 22.4* 43.8 43.5* 38.7 76.1 53.2 53.3 PVCg 97.8 92.0* 89.1 93.3* 86.4 84.6* 100 91.5 85.2 93.5 Total wordsh 39 25 39 26 21 26 14 50 47 46

Note. Age (years; months). P, participant. aPeabody Picture Vocabulary Test — III Raw Score. bPeabody Picture Vocabulary Test — III Standard Score. cPreschool Language Scale — 4 Auditory Comprehension Standard Score. dPreschool Language Scale — 4 Expressive Communication Standard Score. ePreschool Language Scale — 4 Total Language Score is language age in years; months. fPer cent consonants correct — Revised. gPer cent vowels correct. hTotal number of words analysed in the sample using computerized profiling (PROPH; Long and Fey 2005). *Unwillingness to participate in the assessment.

324 Anne Katherine van Bysterveldt et al.

cards are freely available from: http://www.education. canterbury.ac.nz/people/gillon/resources.shtml).

Participants were assessed on all their target and control speech cards (60 trials) on five occasions throughout the study. The first two administrations completed one week prior and immediately prior to the start of the intervention, comprised the final two baseline measures pre-intervention. Intervention phase assessments were completed at the end of cycle one and the start and end of cycle two. Participants’ correct scores on speech targets ranged between zero and 16.6% correct at the two pre-intervention assessments, with all participants scoring zero per cent correct on control targets pre-intervention.

Intervention

Intervention procedures

The experimental integrated intervention implemented in this study comprised the following three com- ponents:

. A parent-implemented home programme. Before the intervention, parents attended an information and training evening following the format detailed in van Bysterveldt et al. (2006) whereby parents were trained to use print referencing techniques at home to bring their child’s attention to targeted letters and sounds during joint story book reading. The intervention was implemented four times per week for 10 minutes per session throughout the 18 week intervention period (that is, 12 hours in total). Parents’ implementation of the print referencing tech- niques was monitored on a weekly basis during the intervention.

. Speech therapy (SLT) sessions.

. Learning through computer (LTC). Participants attended the early intervention centre one morning per week where they saw six different therapists in 20-minute sessions throughout the morning. In keeping with the service delivery model of the early intervention centre, the LTC sessions supported the SLT goals for the child,

Table 2. Examples of participants’ dominant phonological error patterns present in target words, percentage usage and speech sound targets

P PCC-R Examples of dominant phonological error patterns Usage (%) Target sounds selected (initial position unless stated)

1 40.8 Initial cluster reduction 84 /fl/ /gr/ /sp/ /l/ Gliding 91

2 8.3a Final consonant deletion 100 /k/ /m/ /t/ /f/ Deletion of initial consonants 28 Context sensitive voicing 50 Deletion of fricatives 75

3 45.0 Cluster reduction 64 /kr/ /sp/ /sw/ /t R /

Stopping 64 4 18.9 Early stopping 100 /t/ final /k/ final /n/

Final consonant deletion 69 final /p/ Deletion of final stops 68 Deletion of final nasals 100

5 18.8 Cluster reduction 100 /l/ /f/ /sp/ /h/ Deletion of initial h 91 Liquid deletion 60 Deletion of initial fricatives 50

6 26.8 Final consonant deletion 65 /v/ final /m/ final /p/ Later stopping 67 final /d/ Deletion of final stops 67

7 11.6 Final consonant deletion 78 /p/ /n/ /t/ /b/ all final Deletion of final stops 82 Deletion of final nasals 100

8 54.3 Later stopping 95 /kr/ /tr/ /d3/ /t R /

Cluster simplification 96 9 44.1 Later stopping 71 /l/ /v/ /t

R / final /k/

Gliding 33 Liquid deletion 25 Deletion/substitution of final k 56

10 54.0 S nasal cluster reduction 100 /h/ /v/ /sn/ and/sm/ Substitutions/distortions of fricatives 50 /t

R /

Distortions/substitutions of affricates 58 Deletion/substitutions/distortions of initial h 60

Note: aSeverely restricted sample.

Integrated intervention for children with DS 325

thus the experimental intervention was delivered during the SLT and LTC sessions.

The intervention was delivered in two, 6 weekly cycles separated by a 6-week break (that is, 4 hours total SLT and 4 hours total LTC). SLT sessions characteristically included four 5-minute activities as this was compatible with the children’s attention span and ability to engage with the task. Sessions began with a letter knowledge activity, followed by one or two phoneme matching tasks. Each session also typically contained at least one game where letter name, letter sound and phoneme matching were integrated into the one activity. A description of the intervention activities is provided in Appendix A.

The SLT sessions integrated speech goals with phonological awareness and letter knowledge goals and was implemented by the lead researcher, focusing on two speech targets per session. For example if the child’s speech target was the correct articulation of final /p/ in CVC words then the letter p was used in letter sound knowledge activities, and phoneme matching tasks included CVC words ending with /p/. The lead researcher provided a correct model and feedback during all activities. Children were encouraged to engage in tasks that required the articulation of their target speech words, letter name and letter sound knowledge tasks and phoneme matching tasks based on the structure described in Gillon and McNeill (2007) and adapted by the researcher for use with pre-school children with DS. LTC sessions comprised these phonological awareness and letter knowledge tasks adapted for presentation on a computer to pre-school children with DS.

Travel issues and participants’ health influenced attendance at the centre, with participants attending an average of 75% of centre based sessions (approximately 6 hours) equating to a total mean intervention time of 18 hours 8 minutes (SD ! 1 hour 18 minutes).

Reliability

Twenty per cent of speech assessment and speech probe data were re-transcribed by an independent SLT. Point- by-point analysis showed mean agreement of 92.1% with a range of 87.7–95.0% for assessment data and mean agreement of 98.9% with a range of 83.3–100% agreement for speech card probe data. Any differences were resolved by consensus after repeated listenings.

Treatment fidelity

Print referencing sessions

Parents were videotaped administering the print referencing techniques and received structured feedback

two to four times during the intervention period. Additionally, a trained independent reviewer reviewed 23 video-taped sessions and recorded that nine of the ten parents consistently and accurately implemented the intervention. One parent who did not consistently use all three key parts of the technique during the session received extra training and corrective feedback however the implementation remained inconsistent.

Parents completed a weekly report and although all parents reported abandoning sessions due to non-compliance, they all completed four 10-minute sessions of print referencing per week for the full 18-week period. Sessions were largely presented by mothers; 100% in six families, 90% in three families and 70% in one family. Seven reviewed sessions (32%) were affected by or abandoned due to child behaviours including screaming, grabbing the book, running away or refusing to listen to the story.

Speech therapy sessions

Thirty-nine randomly selected intervention sessions (just over 41% of the total sessions) which included at least two sessions from each participant were evaluated by a trained independent reviewer. A minimum of twelve productions by the researcher of each of the three intervention elements: letter name instruction, letter sound instruction, and phoneme matching was required, with 79.4% (31 sessions) adhering to the treatment fidelity protocol. Additionally the reviewer noted the child’s overall compliance and engagement within the session and the occurrence of interaction breakdowns. The reviewer identified 28 different challenging behaviours such as biting, pinching, shouting, throwing toys, hiding, damaging equipment etc. Interaction breakdowns affected 25 of the 39 sessions reviewed (just over 64%) with 90 separate challenging behavioural events recorded, ranging from one to ten per session (mean ! 2.3, SD ! 2.79) which prevented one or more elements being completed in approximately 18% (7) of the sessions. Four participants averaged less than one event per session, with the remaining six participants averaging between one and seven (mean ! 3.6, SD ! 2.96) challenging behavioural events per session. No pattern was observed between the presentation of the different activities and the occurrence of interaction breakdowns. The presence of additional visiting specialists impacted the presentation of one element in the remaining session.

The wide variation in participants’ speech abilities and the impact of behaviour and motivation meant minimum target production across participants was not appropriate. Average production across participants ranged from 0.5 to 29 for letter names, from 0.3 to 22

326 Anne Katherine van Bysterveldt et al.

for letter sounds and from five to 32 for whole words, with considerable within-participant variability across sessions. Corrective feedback, modelling or cueing by bringing the child’s attention to the visual support (that is, letter cards or speech cards) and phoneme– grapheme connections, was required from the researcher on at least 50% of occasions when the child’s letter name, letter sound or target word production was in error. All reviewed sessions adhered to the corrective feedback standard.

Learning through computer sessions

The computer specialist’s clinical notes from a randomly selected 32 (just over 34%) sessions were reviewed by the researcher. Sessions were required to contain at least two of the activities developed by the researcher and the computer specialist known to include letter names, letter sounds and phoneme matching. All the reviewed sessions adhered to the protocol. Interaction breakdowns which resulted in disengagement with the task were recorded in seven (just over 21%) of the reviewed sessions.

Results

Scoring

Where the assessment was unable to be completed due to a participant’s inability to understand the task or an unwillingness to attempt or complete the task, a score of zero was given for that item and the next item was presented. Where participants did not respond during the speech sound assessment, stimulus cards were placed on the bottom of the pile and represented later in the assessment. A further non-response resulted in a zero score for that item.

Speech production accuracy

No statistical difference was found between partici- pants’ performances on trained and untrained target speech items ( p $ 0.1) therefore these data were combined for all analyses. The two standard deviation band method (Portney and Watkins 2009) was used to determine whether a statistical difference existed between scores within the baseline and intervention phases. The mean of the baseline phase and a two standard deviation band are plotted across both phases with intervention data points outside the banded area indicative of significant change (Portney and Watkins 2009). A linear regression line was also calculated depicting the line which best describes the relation- ship between speech production performance and time during the baseline phase, and providing

an estimate of predicted speech performance during the intervention phase.

Additional analysis was undertaken using the statistical process control (SPC) model described in Portney and Watkins (2009) to control for the contribution of time-series data to the probability of Type I error (false-positive), to verify the existence of stable baseline performances and to determine whether significant change has occurred between baseline and intervention phases. A moving range score (X-mR) based on the mean variability between adjacent baseline data points is calculated to determine the upper and lower control limits (UCL and LCL). Data within these control limits of common cause variation indicate a stable baseline. Data points outside these limits comprise special cause variation which is indicative of significant change.

Effect sizes and confidence intervals were generated for seven participants. Three participants achieved zero scores at all points during the baseline phase, thus neither SPC analysis nor effect size analysis was appropriate. A standardized mean difference statistic was calculated using the procedure described in Shadish et al. (2008). This measure has been developed in an effort to provide a mechanism to compare the effectiveness of single-case designs with between- group designs. This procedure yielded a g-statistic of g ! 2.59 (Hedges adjustment for sample size d ! 2.48) which is comparable with those reported by Shadish et al. (2008) in their meta analysis of single-case experimental designs.

The speech production gains of one child (participant 4) analysed via the two standard deviation band method and the statistical process control method are presented in Figures 1 and 2, respectively. A summary of change scores, significance and effect sizes for target speech sounds for all participants is presented in Table 3.

Correct production of target sounds ranged from zero to 10.41% at pre-intervention (mean ! 2.91, SD ! 3.42) and from 20.8% to 70.8% post- intervention (mean ! 51.42, SD ! 16.54) indicat- ing gains of between 20.8% and 66.64% (mean ! 48.51, SD ! 15). A paired samples t-test performed on the grouped data revealed significant differences between the pre- and post-intervention scores (t(9) ! 10.02, p , 0.001). Per cent consonants correct (PCC-R) and per cent vowels correct (PVC) and percentage change for all sounds in participants’ speech target words are presented in Table 4. The PCC-R metric, which scores distortions (both clinical and non- clinical) as correct, is recommended by Shriberg et al. (1997) as the measure which is most appropriate to compare the speech of individuals with diverse speech and age profiles. Post-intervention gains ranged from

Integrated intervention for children with DS 327

22.5 to 53.8 for PCC-R and from 5.3 to 46.4 for PVC. All participants scored zero per cent correct on control sounds pre-intervention and six participants remained at zero per cent correct scores post-intervention with four participants demonstrating a small improvement in the production of their control sounds post- intervention. All participants showed greater change on target sounds than control sounds, with average

difference scores for target sounds ranging from 2.5 to eight out of twelve and for control sounds from zero to three out of twelve. Spearman rank-order correlations revealed no significant relationships between percentage change on target phonemes and pre-intervention measures of PPVT (r ! 20.05, p ! 0.86), hearing thresholds (r ! 0.41, p ! 0.21) or chronological age (r ! 0.18, p ! 0.58). Analysis revealed no relationship

Figure 1. Baseline and intervention phase performance for speech sound measures for Participant 4 using 2 Standard Deviation Band analysis. Performance in the intervention phase must be above the shaded 2 standard deviation band to demonstrate significant improvement (in this example, speech sound measures pre-intervention were mean ! 2.45, SD ! 1.85).

Figure 2. Baseline and intervention phase performance for speech sound measures for participant 4 using Statistical Process Control analysis. Performance in the intervention phase must be above the upper control limit (UCL) to demonstrate significant improvement (in this example the UCL ! 7.98 and the lower control limit (LCL) ! 23.0776.

328 Anne Katherine van Bysterveldt et al.

between gains on speech targets and gains on letter knowledge (r ! 20.33, p ! 0.38), initial phoneme identity (IPI) (r ! 0.20, p ! 0.53) or initial phoneme identity with words (IPIW) (r ! 20.45, p ! 0.17).

Letter name knowledge

Letter name knowledge was assessed pre- and post- intervention with results presented in Figure 3. At pre- intervention two participants had complete or near complete letter name knowledge. Of the remaining eight participants, five showed improvement in scores at post- intervention, with large gains demonstrated by one participant. Three participants who were unable to complete the task at both testing times were assigned a score of zero. Responses included closing the assessment book, pointing to several letters at once and providing no response. No significant relationships were found between gain scores on letter name knowledge and pre- intervention measures of PPVT (r ! 0.03, p ! 0.88), hearing thresholds (r ! 0.33, p ! 0.38) or chronological age(r ! 20.18,p ! 0.61).Duetoceilingeffectsthedata from two participants with high scores at pre-intervention were excluded from this analysis of change scores. Spearman rank-order correlations revealed pre-interven- tionscoresofletternameknowledgewerenotsignificantly correlated with hearing thresholds (r ! 20.48, p ! 0.13) or chronological age (r ! 0.03, p ! 0.89), howeverthecorrelationwithPPVTscoresapproachedthe level of significance (r ! 0.55, p ! 0.06). Analysis also revealed significant and near significant relationships between gain scores on the letter name knowledge task and gain scores on the IPI (r ! 0.86, p , 0.001) and IPIW tasks (r ! 0.65, p ! 0.05).

Letter sound knowledge

Results of the pre- and post-intervention assessment of letter sound knowledge are presented in Figure 4.

At pre-intervention two participants demonstrated high scores on the letter sound knowledge task with one other participant able to identify one letter sound correctly. Post-intervention two of these participants showed some increases in scores but no additional children were able to demonstrate any letter sound knowledge. Five participants at pre-intervention and four participants at post-intervention were unable to complete the task and were assigned a score of zero.

Letter name knowledge was more advanced than letter sound knowledge at both testing times, and appeared to be prerequisite for letter sound knowledge in the three children able to demonstrate this knowledge, with no child able to demonstrate knowledge of a letter’s sound without knowledge of that letter’s name.

Initial phoneme identity

Pre- and post-intervention assessment results are presented in Figure 5. Three participants at pre- intervention and nine participants at post-intervention achieved scores in the Initial Phoneme Identity (IPI)

Table 3. Summary of change and effect sizes in target speech sounds

Effect size

P Baseline Intervention 2 SD significant? SPC significant? Cohen’s d 295% CI "95% CI

1 0.52 (1.04) 13.88 (13.22) Yes Yes 1.42 20.24 3.09 2 0 14.90 (23.12) Yes Yes n.a. n.a. n.a. 3 0 61.77 (8.66) Yes Yes n.a. n.a. n.a. 4 2.45 (1.85) 45.11 (17.30) Yes Yes 3.46 1.11 5.82 5 1.47 (2.94) 15.26 (5.22) Yes Yes 3.25 0.98 5.51 6 1.04 (2.08) 52.07 (22.52) Yes Yes 3.19 0.94 5.43 7 3.85 (2.70) 38.88 (15.91) Yes Yes 3.42 1.05 5.75 8 0 22.77 (15.73) Yes Yes n.a. n.a. n.a. 9 11.40 (3.50) 53.11 (14.47) Yes Yes 3.96 1.40 6.52

10 4.68 (3.55) 43.74 (16.66) Yes Yes 3.24 0.97 5.50

Notes: Baseline and intervention scores are the mean and standard deviation (SD) of the baseline and intervention scores. P, participant; 2 SD significant?, significant using the two SD band method; SPC significant?, significant using the Statistical Process Control method; CI, confidence intervals.

Table 4. PCC-R and PVC of all sounds in participants’ speech target words

PCC-R PVC

P Pre- Post- Difference Pre- Post- Difference

1 40.8 64.9 "24.1 83.7 91.5 "7.8 2 8.3a 62.1 "53.8 28.6a 75.0 "46.4 3 45.0 82.4 "37.4 89.7 97.9 "8.2 4 18.9 68.1 "49.2 65.2 80.4 "15.2 5 18.8 41.9 "23.1 77.8 89.7 "11.9 6 26.8 75.3 "48.5 52.4 95.0 "42.6 7 11.6 57.3 "45.7 26.5 60.0 "33.5 8 54.3 76.8 "22.5 91.1 100 "8.9 9 44.1 73.4 "29.3 68.2 93.1 "24.9

10 54.0 78.0 "24.0 82.6 97.9 "15.3 Notes: P, participant; PCC-R, per cent consonants correct — revised; PVC, per cent vowels correct. aSeverely restricted sample characterized by non-speech sounds.

Integrated intervention for children with DS 329

Figure 3. Letter name knowledge pre- and post-intervention.

Figure 4. Letter sound knowledge pre- and post-intervention.

Figure 5. Initial phoneme identity pre- and post-intervention.

330 Anne Katherine van Bysterveldt et al.

task however all scores were below chance level as determined by the binomial test (Portney and Watkins 2009), with a score of 7/10 or higher required for a statistically significant result ( p , 0.05). Three partici- pants at pre-intervention and one participant at post- intervention were unable to complete the task and were assigned a score of zero. Spearman rank-order correlations revealed IPI scores were significantly correlated with PPVT scores pre-intervention (r ! 0.79, p ! 0.003) and approaching significance post-intervention (r ! 0.61, p ! 0.05). A strong negative correlation was found between IPI scores and hearing thresholds at pre-intervention (r ! 20.81, p ! 0.001) but not at post-intervention (r ! 20.08, p ! 0.81) nor with chronological age pre- (r ! 20.47, p ! 0.16) or post-intervention (r ! 20.49, p ! 0.13). No significant relationships were found between IPI gain scores and pre-intervention measures of PPVT (r ! 20.10, p ! 0.75), hearing thresholds (r ! 0.54, p ! 0.08) or chronological age (r ! 0.08, p ! 0.81).

Initial phoneme identity with words

Pre- and post-intervention assessment results are presented in Figure 6. Four participants at pre- intervention and eight participants at post-intervention achieved scores in the Initial Phoneme Identity with Words (IPIW) task, however all scores were below chance level as determined by the binomial test (Portney and Watkins 2009) with a score of 8/12 or higher required for a statistically significant result (p , 0.05). Four participants at pre-intervention and two participants at post-intervention were unable to complete the task and were assigned a score of zero. Spearman rank-order correlations revealed no significant relationships between scores on IPIW and PPVT pre- (r ! 0.07, p ! 0.83) and post- intervention (r ! 0.47, p ! 0.16), hearing thresholds

pre- (r ! 0.14, p ! 0.68) and post-intervention (r ! 20.29, p ! 0.38) or chronological age pre- (r ! 20.33, p ! 0.34) and post-intervention (r ! 20.35, p ! 0.29) nor between IPIW gains scores and PPVT (r ! 0.44, p ! 0.18), hearing thresholds (r ! 20.33, p ! 0.32) or chronological age (r ! 20.16, p ! 0.63).

Discussion

This study used a multiple single-subject design to investigate the effectiveness of an integrated phonolo- gical awareness intervention for ten children with DS aged 4 and 5 years. The intervention aimed simultaneously to facilitate the development of speech production, letter name and sound knowledge, and phonological awareness. The intervention included the following three components:

. Parent-implemented print referencing during joint story reading.

. Letter knowledge and phoneme awareness activities conducted by the speech–language therapist (SLT) in a play-based format.

. Letter knowledge and phoneme awareness activities conducted by the computer specialist (CS).

The intervention was implemented by the SLT and CS at an early intervention centre during two 20-minute sessions per week, in two 6-week therapy blocks separated by a 6-week break (that is, 8 hours in total). The parents implemented the print referencing component in four 10-minute sessions per week across the 18-week intervention period (approximately 12 hours in total).

The first hypothesis stated that the research intervention would improve speech production accu- racy in the trained and untrained words featuring each

Figure 6. Initial phoneme identity with words pre- and post-intervention.

Integrated intervention for children with DS 331

child’s target speech sounds. The results from the intervention supported this hypothesis. All ten participants demonstrated statistically significant improvement in production accuracy on both trained and untrained words.

There was, however, considerable variability between the speech production skills of the participants pre- and post-intervention and in their response to intervention. This variability, however, could not be accounted for by participants’ receptive language scores, chronological age or hearing thresholds. The interven- tion appeared to be effective irrespective of whether the targeted speech errors were evident in developmentally earlier or later resolving speech error patterns or whether the participants used single speech error patterns or multiple speech error patterns at once. Although speech targets were in activities which also integrated letter name and sound knowledge and phoneme matching, analysis revealed no relationship between gains on speech targets and gains on letter knowledge, IPI or IPIW.

The study findings indicated that significant improvement in the speech articulation of single words can be achieved after a short intervention, that is, approximately 20 hours over 18 weeks (intervention time mean ! 18 hours 8 minutes, standard deviation (SD) ! 1 hour 13 minutes) in children whose rate of speech development is slow and whose performance has remained stable over the preceding two months. The improvement in speech production by the children with DS is consistent with the findings of Cholmain (1994) and Dodd et al. (1994) who reported participants made considerable speech gains over a relatively short period of time, despite stable perform- ance pre-intervention.

Grunwell (1990) described four different types of generalization of speech skills: Lexical, Socioenviron- mental, Syntactic, and Phonological. Lexical generaliz- ation refers to use of the remediated pattern in untrained words and was observed in all participants with children performing equally well on the trained and untrained words. Phonological generalization is observed when the remediated pattern is demonstrated with other phonemes in the same sound class.

Participant 4 in this study provides an example of phonological generalization and use of multiple patterns. Targeted speech patterns for participant 4 included voiced/voiceless contrasts presented in words with initial /t/, many of which were of CVC structure. Additionally, some words presented to target final consonant deletion included initial /k/, and these were typically voiced to /g/ or /d/. Analysis across time demonstrated remediation of the final consonant deletion pattern in the articulation of single words and establishment of voicing contrasts within the same

word, for example, for the target word /k ^ p/ addressing deletion of final stops, participant 4’s production progressed as follows; /d‰/ ! /d ^ p/ ! /k ^ p/ and for the target word /tæp/ addressing voicing contrasts the progression was similar /dæ/ ! /tæ/ ! / dæp/ ! /tæp/. Other improvements in participants’ speech that were not the focus of a targeted pattern, such as velar fronting in the example above, are evidenced by increased per cent consonant correct (PCC-R) and per cent vowels correct (PVC) scores from all sounds in target words.

The second hypothesis was that the research intervention would improve participants’ letter name and letter sound knowledge. This hypothesis was partially supported. Letter sound knowledge proved difficult for most participants. The intervention appeared only to stimulate further letter sound knowledge in two participants who entered the study already able to demonstrate some letter sound knowl- edge. The intervention had more effect on teaching participants letter names. Six of the ten participants demonstrated that they knew more letter names at the end of the intervention compared with pre-intervention.

Participants with the strongest letter knowledge at post-intervention were three of the four participants who had language ages of at least 3 years suggesting a stronger language foundation may facilitate the acquisition of alphabet knowledge. A pattern of learning letter names before letter sounds is consistent with findings for children without DS (McBride-Chang 1999). Analysis of the videotaped intervention sessions showed many participants were able to identify correctly one or two letter names and sounds that were the focus of their intervention session. However, these skills were evident in therapy sessions only and did not generalize to the assessment session when the activity was presented in a different context. Difficulty in transferring improvement skills from a therapy context to an assessment context is common in children with high learning needs and may suggest the knowledge has not yet consolidated or further scaffolding and increased therapy time is required (Roberts et al. 2008).

The third hypothesis examined whether the research intervention would improve phonological awareness skills on untrained phoneme level tasks. The data partially supported this hypothesis. Perform- ance improved post-intervention compared with pre- intervention for most participants but their improved performance did not meet the conservative binomial level of 70% correct at post-intervention when untrained phonological awareness items were intro- duced. This suggests phonological awareness was being stimulated during the intervention period, but

332 Anne Katherine van Bysterveldt et al.

participants had not reached mastery of identifying initial sounds in words and therefore could not demonstrate the transference of knowledge to novel items.

Phonological awareness in young children typically proceeds along a developmental continuum of increas- ing ability and stability (Lonigan et al. 1998) throughout the preschool years. Lonigan et al. (1998) reported some measurable, though inconsistent, phonological awareness skills in children as young as 2 or 3 years of age and more consolidated phonological awareness skills in children aged 4 years and over. Participants’ receptive language abilities, as determined by their PPVT scores, were associated with both pre- and post-intervention measures of letter name and sound knowledge and IPI, but not with change scores. However, gain scores on letter name knowledge and both phonological awareness tasks were related, which suggests that the presentation of these activities in an integrated format may facilitate the reciprocal gains reported by Oudeans (2003). Consistent with the findings of Lonigan et al. (1998) it is plausible that some participants in the current study did demonstrate a pattern of emerging but unstable phonological awareness. Follow-up assessment of letter knowledge and phonological awareness skills is necessary to investigate any potential longer-term influence of the intervention on these skills. Follow-up assessment of speech measures is also required to determine whether treatment gains were maintained.

Clinical implications

Speech production errors are common in children with DS and these errors often persist into adulthood. However, empirically based interventions to improve speech production in this population are rare. Children with DS also exhibit weakness in the underlying skills of phonological awareness and letter knowledge tasks that are critical for early reading success. The findings of this study suggest an intervention approach which integrates speech, letter knowledge, and phonological awareness targets was effective in remediating speech errors for pre-school children with DS and may potentially stimulate letter knowledge and phonological awareness skills in the future. Speech gains were demonstrated by all children regardless of language age, however the greatest gains on letter knowledge and phonological awareness were demonstrated by children who had language ages of at least 3 years. The relationships between gains in letter knowledge and phonological awareness support the appropriateness of an approach in which instruction in these skills is integrated. In light of the persistent nature of speech difficulties in individuals with DS and the superior language abilities

associated with reading in this population, evidence based interventions which can combine several treatment goals may provide a valuable alternative to traditional therapy techniques which aim to improve only one language domain.

Acknowledgements

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

References

BARNES, E., ROBERTS, J., MIRRETT, P., SIDERIS, J. and MISENHEIMER, J., 2006, A comparison of oral structure and oral-motor function in young males with Fragile X syndrome and Down syndrome. Journal of Speech, Language and Hearing Research, 49, 903–917.

BUS, A. G., VANIJZENDOORN, M. H. and PELLEGRINI, A. D., 1995, Joint book reading makes for success in learning to read: a meta- analysis on intergenerational transmission of literacy. Review of Educational Research, 65, 1–21.

CARDOSO-MARTINS, C., PETERSON, R., OLSON, R. and PENNINGTON, B., 2008, Component reading skills in Down syndrome. Reading and Writing, 21, 277–292.

CATTS, H., FEY, M., ZHANG, X. and TOMBLIN, B., 2001, Estimating the risk of future reading difficulties in kindergarten children: a research-based model and its clinical implementation. Language, Speech and Hearing Services in Schools, 32, 38–51.

CHOLMAIN, C. N., 1994, Working on phonology with young children with Down syndrome — a pilot study. Journal of Clinical Speech and Language Studies, 1, 14–35.

CUPPLES, L. and IACONO, T., 2000, Phonological awareness and oral reading skill in children with Down syndrome. Journal of Speech, Language, and Hearing Research, 43, 595.

DODD, B., MCCORMACK, P. and WOODYATT, G., 1994, Evaluation of an intervention program: Relation between children’s phonology and parents’ communicative behavior. American Journal on Mental Retardation, 98, 632–645.

DUNN, L. M. and DUNN, L. M., 1997, Peabody Picture Vocabulary Test — III (Circle Pines, MN: American Guidance Service).

GIBBON, F. and GRUNWELL, P., 1990, Specific developmental language learning disabilities. In P. GRUNWELL (ed), Developmental speech disorders:clinical issues and practical implications (Edinborough; New York: Churchill Livingstone), 135–161.

GILLON, G., 2005, Facilitating phoneme awareness development in 3- and 4–year-old children with speech impairment. Language, Speech and Hearing Services in Schools, 36, 308–324.

GILLON, G. T. and MCNEILL, B. C., 2007, Integrated Phonological Awareness: An Intervention Program for Preschool Children with Speech–Language Impairment (Christchurch: Depart- ment of Communication Disorders, University of Canter- bury) (available at: http://www.education.canterbury.ac.nz/ people/gillon/resources.shtml).

GOETZ, K., HULME, C., BRIGSTOCKE, S., CARROLL, J. M., NASIR, L. and SNOWLING, M., 2008, Training reading and phoneme awareness skills in children with Down syndrome. Reading and Writing, 21, 395–412.

HESKETH, A., ADAMS, C., NIGHTINGALE, C. and HALL, R., 2000, Phonological awareness therapy and articulatory training approaches for children with phonological disorders:

Integrated intervention for children with DS 333

A comparative study. International Journal of Language and Communication Disorders, 35, 337–354.

HODSON, B. and PADEN, E., 1991, Targeting intelligible speech: A phonological approach to remediation, 2nd edn (Austin, TX: Pro-Ed).

HODSON, B., 2004, Hodson Assessment of Phonological Patterns — Third Edition HAPP-3 (Austin, TX: PRO-ED).

JARROLD, C., THORN, A. and STEPHENS, E., 2009, The relationships among verbal-short term memory, phonological awareness, and new word learning: Evidence from typical development and Down syndrome. Journal of Experimental Child Psychology, 102, 196–218.

JUSTICE, L. M. and EZELL, H. K., 2000, Enhancing children’s print and word awareness through home-based parent interven- tion. American Journal of Speech-Language Pathology, 9, 257–269.

KUMIN, L., 2006, Speech intelligibility and childhood verbal apraxia of speech in children with Down syndrome. Down Syndrome Research and Practice, 10, 10–22.

LOF, G. L. and WATSON, M. M., 2008, A nationwide survey of nonspeech oral motor exercise use: implications for evidence- based practice. Language, Speech and Hearing Services in Schools, 39, 392–407.

LONG, S. and FEY, M. E., 2005, Computerized Profiling (Version 9.7.0) (Cleveland, OH: Case Western Reserve University).

LONIGAN, C. J., BURGESS, S. R., ANTHONY, J. L. and BARKER, T. A., 1998, Development of phonological sensitivity in 2- to 5-year-old children. Journal of Educational Psychology, 90, 294–311.

MCBRIDE-CHANG, C., 1999, The ABCs of the ABCs: the development of letter-name and letter-sound knowledge. Merrill-Palmer Quarterly, 45, 285–308.

MCNEILL, B. C., GILLON, G. T. and DODD, B., in press, 2009, The effectiveness of an integrated phonological awareness approach for children with childhood apraxia of speech (CAS). Child Language Teaching and Therapy.

OUDEANS, M. K., 2003, Integration of letter-sound correspondences and phonological awareness skills of blending and segment- ing: a pilot study examining the effects of instructional sequence on word reading for kindergarten children with low phonological awareness. Learning Disability Quarterly, 26, 258–308.

PARSONS, C. L., IACONO, T. A. and ROZNER, L., 1987, Effect of tongue reduction on articulation in children with Down syndrome. American Journal of Mental Deficiency, 91, 328–332.

PORTNEY, L. G. and WATKINS, M. P., 2009, Foundations of Clinical Research: Applications to Practice (Upper Saddle River, NJ: Pearson Education).

ROBERTS, J. E., CHAPMAN, R. S., MARTIN, G. E. and MOSKOWITZ, L., 2008, Language of preschool and school-age children with Down syndrome and fragile X syndrome. (pp 77–115) In J. E. Roberts, R. S. Chapman and S. F. Warren (eds), Speech and Language Development and Intervention in Down Syndrome and Fragile X Syndrome (Baltimore, MA: Paul H. Brookes).

ROCH, M. and JARROLD, C., 2008, A comparison between word and nonword reading in Down syndrome: the role of phonological awareness. Journal of Communication Disorders, 41, 305–318.

SHADISH, W. R., RINDSKOPF, D. M. and HEDGES, L. V., 2008, The state of the science in the meta-analysis of single-case experimental designs. Evidence-Based Communication Assessment and Intervention, 2, 188–196.

SHRIBERG, L. D., AUSTIN, D., LEWIS, B. A., MCSWEENEY, J. L. and WILSON, D. L., 1997, The percentage of consonants correct (PCC) metric: extensions and reliability data. Journal of Speech Language and Hearing Research, 40, 708–722.

SMITH, B. L. and STOEL-GAMMON, C., 1983, A longitudinal study of the development of stop consonant production in normal and Down’s syndrome children. Journal of Speech and Hearing Disorders, 48, 114–118.

SNOWLING, M. J., HULME, C. and MERCER, R. C., 2002, A deficit in rime awareness in children with Down syndrome. Reading and Writing, 15, 471–495.

VAN BYSTERVELDT, A. K., GILLON, G. T. and MORAN, C., 2006, Enhancing phonological awareness and letter knowledge in preschool children with Down syndrome. International Journal of Disability, Development and Education, 53, 301–329.

WAGNER, R. K., TORGESEN, J. K. and RASHOTTE, C. A., 1994, The development of reading-related phonological processing abilities: new evidence of bi-directional causality from a latent variable longitudinal study. Developmental Psychology, 30, 73–87.

ZIMMERMAN, I. L., STEINER, V. G. and POND, R. E., 2002, Preschool Language Scale — Fourth Edition — Australian Language Adaptation (PLS-4) (San Antonio, TX: Psychological).

334 Anne Katherine van Bysterveldt et al.

Appendix A: Letter name and sound knowledge example

The speech-language therapist (SLT) placed poster-sized cards of the two targets being targeted (for example, l and k) on the floor and hid six to ten small cards of each letter around the room in obvious locations, for example, under a cushion for a ‘hide and seek’-type format and the SLT and child took turns to turn them over. The SLT said the letter name and sound and the child matched it to the corresponding large letter card. The child was encouraged to say the letter name and sound, with any production reinforced and corrected by the SLT. The child then placed a mechanical toy on the large letter card and activated it to jump:

SLT: What have you found?

SLT: Ooh, you’ve found t.

SLT: t says /t/. Can you see t, /t/ anywhere?

Child: There. (child points)

SLT: Oh, good matching, you found . . . ? (prompting for the target letter name)

Child: t.

SLT: You’re right, and it says . . . ? (prompting for the target letter sound)

Child: /t/.

SLT: Well done, you matched t, /t/.

Child places small letter card onto corresponding large letter.

SLT: Let’s bounce Tigger on t.

Phoneme identity and matching example

The SLT placed poster-sized cards of the two targets being taught (for example, l and k) on the floor and two poster-sized ‘bingo’-type boards on the floor beside the corresponding letter target. The bingo boards displayed six target words with the word written underneath each picture in large bold font. An identical set was cut up for the child to match. For example, where the target was to address final consonant deletion of /k/, the bingo board included the words sock, hook, work, beak, wink, and book. These were placed face down and the SLT and child took turns to turn them over. If the child spontaneously said the word, this was reinforced by the SLT. Corrective feedback was provided if required including drawing the child’s attention to the written text. Where the child did not spontaneously say the word, it was said by the therapist and the child matched it to the corresponding large letter. Once the words had been matched to the target letter they were matched to the corresponding picture on the bingo board.

SLT: Let’s see what we’ve got. You choose one.

Child: (child chooses a card and names it) tap.

SLT: You’re right, that says tap and there’s the letter t that says /t/. (points to the text).

SLT: See, it starts with t. (points to the large letter).

SLT: Let’s check, is it the same? (compares to large letter).

Child: Yes.

SLT: You’re right, now we need to find the other picture of . . . ? (prompts child to respond)

Child: tap.

SLT: You’re right, tap.

Child: (places card on board) tap.

SLT: Good matching.

For some children, two bingo boards were too challenging or distracting so the activity was modified to present one target at a time.

Combined activity

SLT placed a poster-sized sheet on the floor with a racetrack circuit on it. The track was divided into squares on which were placed six to ten small letter cards for each of the two target letters/sounds. Two piles of word cards, each pile comprising the six target words for each sound, were placed in the middle of the track. The child, parent and SLT took turns throwing dice and moving a toy car around the track. When a person landed on a small letter card a corresponding word card was chosen. Parents were encouraged to provide feedback to the child following the SLT’s model:

SLT: Your turn to throw the dice.

Child throws dice.

SLT: Oh you got a . . . ?

Child: Three.

SLT: Great, let’s count, one, two, three. You landed on . . . ? (prompting for the target letter name)

Child: l.

SLT: You did, you landed on l, and it makes the /l/ sound. Now choose a word that starts with l. Which pile is it on?

Child: (chooses a card) lamp.

SLT: Good talking, you chose lamp. It starts with l that makes the /l/ sound. Let’s say it together; lamp. Show Mum what you got.

Child: lamp.

Mum: What does it start with?

Child: l.

Mum: Yes, l says /l/. Good job!

Integrated intervention for children with DS 335