RLI was evaluated in the first randomised controlled trial of an educational intervention designed for children with Down syndrome, and one of the largest of any study of children with Down syndrome to date - funded by the UK Big Lottery Fund. The study was conducted by researchers at Down Syndrome Education International and at The Centre for Reading and Language at the University of York.
Research looking at the developmental disabilities associated with Down syndrome has advanced steadily over the past 40 years. In some areas, we now understand enough about the learning needs of children with Down syndrome to design interventions to better meet these needs. Evidence-based approaches to early intervention and education for children with Down syndrome are clearly welcome. However, robust evaluations of interventions are critical if we are to be sure that they really do deliver the benefits we expect.
RLI was evaluated in a large randomised controlled trial (RCT). Randomised controlled trials offer a high standard of evidence by evaluating outcomes for participants who are similar to one another at the outset and are randomly assigned to one of two groups - usually an intervention (or 'treatment') group and a control (or 'no treatment' or placebo) group. This study design can give us greater confidence that any differences in outcomes that we observe between the two groups are due to the presence (or absence) of the intervention being evaluated. In our study, the 'treatment' was the reading and language intervention. Our control condition was 'teaching as usual' - the typical reading and language instruction provided to children with Down syndrome included in UK mainstream primary schools with one-to-one support. After the first phase of the study, the initial control group also received the intervention and we continued to evaluate the progress both groups made.
The study was divided into three 20-week phases (see Table 1): the RCT phase (A), the supported intervention phase (B) and the independent intervention phase (C).
Throughout phases A and B, when first the initial intervention group (1) and then the waiting control group (2) started the reading and language programme, the Teaching Assistants delivering the intervention were fully supported by the research team. The Teaching Assistants followed a scripted intervention, were contacted by telephone or email every two weeks, and were observed once each term.
In phase C, the Teaching Assistants took the principles of the intervention and created their own intervention content (within the same session structure), and were not routinely contacted by the research team.
|Phase||Weeks||Initial intervention group (1)||Waiting control group (2)|
|A||Randomised controlled intervention||1 to 20||Daily intervention (supported)||Regular classroom education|
|B||Supported intervention||21 to 40||Daily intervention (supported)||Daily intervention (supported)|
|C||Independent intervention||41 to 60||Daily intervention (independent)||Daily intervention (independent)|
We recruited 57 children with Down syndrome living in two areas of the UK (Yorkshire and Hampshire) to participate in the study. All of the children were attending mainstream primary (elementary) schools and receiving one-to-one support from a Teaching Assistant. A total of 50 schools participated in the study. Any child who was attending and expected to remain in primary school throughout the duration of the study could participate. We excluded no one on the basis of their existing skill levels, and therefore enrolled participants who were broadly representative of the wide range of abilities seen among children with Down syndrome.
Table 2 shows the average scores on receptive and expressive vocabulary and letter-sound knowledge measures at the start of the study (t1), and the range of scores (lowest to highest). The scores are expressed as age equivalents, in the format of years:months.
At the beginning of the project, the average age of the children was 6 years and 11 months, ranging from 5 years 2 months to 10 years. The measures illustrate the extent of their receptive and expressive language delay at this time. The average word reading score is not included in the table as 21 children could not read any words (reading age below 57 months) and 3 children were at ceiling on the standardised measure (reading ages above 92 months).
The children were randomly assigned to two groups after initial assessments. The two groups did not differ on any language or literacy measures.
Of the 57 children recruited, 54 remained in the study through two school years and were assessed at each time point although 4 of these children discontinued the intervention on moving schools during the study. Two other children were lost to the study as they moved schools and one child refused to participate in testing and the school withdrew.
The Teaching Assistants
The intervention was delivered by Teaching Assistants who were already working with the children. We trained over 70 Teaching Assistants, all of whom were female and aged between 20 and 54 years. The Teaching Assistants possessed a wide range of knowledge and experience, with qualifications ranging from GCSE to degree level. Some had been Teaching Assistants for a long time (up to 24 years), while others had been a Teaching Assistant for a year or less. Their experience working with children with Down syndrome was also diverse: some had supported the child they were currently working with for up to 5 years, whereas others had only just started working with them. For 75% of the Teaching Assistants participating, their current post was their first experience of working with a child with Down syndrome. The number of training courses specifically relating to Down syndrome specific that the Teaching Assistants had attended prior to our training ranged from 0 to 6.
At the start of the study, the Teaching Assistants working children in both the intervention and waiting control group were provided with a two-day training course about meeting the educational needs of children with Down syndrome in inclusive classrooms.
Prior to starting their respective intervention periods, the Teaching Assistants were also provided with a two-day training course on the intervention itself. (For the initial intervention group this was at the start of first 20 week phase, and for the waiting control group this was at the start of the second 20 week phase.) The Teaching Assistants were also given written guidance, assessment and teaching resources when starting their intervention periods. They were given pre-prepared language strand session plans. They were also provided with a 'refresher' day after the first 10 weeks of each supported intervention period. During the initial intervention phase (A) the intervention group received termly visits and a telephone call or email every two weeks to offer advice and answer any questions arising. During the second (supported) intervention phase (B), both groups received termly visits and a telephone call every two weeks.
Delivering the intervention
The schools decided when to deliver the intervention each day. For some children it replaced their usual class literacy instruction but for others it was an additional literacy session. For some children the intervention was delivered as one 40-50 minute session, for others the reading and language strands were delivered at different times during the school day. Children were not always supported by the same TA all day and so in some schools one TA delivered the reading strand and another TA delivered the language strand. In other words we took a flexible approach and the schools worked out what was going to be the best way for them to deliver the programme.
We assessed the children at four points throughout the project: before the first intervention phase A (time 1, t1), at the end of phase A (t2), at the end of phase B (t3) and at the end of phase C (t4).
We assessed a range of language and literacy skills:
- Single word reading (reading of regular and irregular words in isolation)
- Letter-sound knowledge (testing of letter and digraph knowledge)
- Phoneme blending (blending of sounds e.g. knowing that b-e-d makes bed)
- Nonword reading (reading of nonsense words e.g. hic)
- Phonetic spelling (spelling simple words - e.g. dog - phonetically)
- Taught expressive vocabulary (the ability to use words taught during the intervention)
- Taught receptive vocabulary (the ability to understand words taught during the intervention)
- Standard expressive vocabulary (the ability to use words not trained in the intervention)
- Standard receptive vocabulary (the ability to understand words not trained in the intervention)
- Expressive grammar (appropriate use of grammar in speech e.g. using different tenses)
- Expressive information (reporting relevant information when describing pictures)