Browse by author
Lookup NU author(s): Professor James Law,
Dr Jenna Charlton,
Professor Cristina McKean,
This is the final published version of a report that has been published in its final definitive form by Newcastle University, 2020.
For re-use rights please refer to the publisher's terms and conditions.
BACKGROUND: The gap in the cognitive development and specifically oral language skills between children from different social backgrounds is widely acknowledged. This gap is identifiable very early in life and well established by school entry and can have long-term consequences in terms of educational attainment and adult outcomes. Importantly early communication difficulties may also be indicators of a wider range of neurodevelopmental conditions. Effective interventions are available but matching the right children to the right intervention is sometimes a challenge, in part because the rate at which children’s language develops naturally varies and it can be difficult to know when to consider intervention. Central to this process is the review of child development carried out by health visitors (HV) at the 2-2½ year review.A public health approach to speech and language development has been advocated 15and the issue of developmental surveillance has attracted considerable attention in recent years. However, a formal screening programme has not been advocated because they have not met standard criteria. Rather there has been a focus on developing approaches which improve early identification of children with speech, language and communication needs, foster a conversation between practitioner and parent about a child’s needs, and equip parents with the skills needed to support their child’s development.This project was commissioned in 2018 as part of the UK government’s Social Mobility Action Plan 18 and was one of three dimensions to a programme of work delivered by Public Health England (PHE) and the Department for Education (DfE) in England. he programme of work included: * The provision of enhanced training for health visitors to help them address the needs of children with speech, language and communication needs (SLCN) and their families; * the development of guidance to Local Areas to support the development of local evidence-based Speech, Language and Communication Needs (SLCN) pathways; * The development of an early language assessment tool and intervention (the present project) designed to facilitate a conversation between practitioner and parent about the child’s communication skills. WHAT WE DID: The present study was carried out in five sites in England: Derbyshire, Middlesbrough, Newham, Wakefield and Wiltshire, between January 2019 and March 2020. The “voice” of parents and practitioners was an important element of the study and Public, Patient Involvement (PPI) groups were run in each site throughout to inform key elements of the study. The final output from the project includes both identification and intervention elements. The summary report for the project is available at https://www.gov.uk/government/publications/best-start-in-speech-language-and-communication.A handbook for the identification procedure and the early stages of the intervention are provided at https://www.gov.uk/government/publications/best-start-in-speech-language-and-communicationA new measure called the Early Language Identification Measure – Extended (ELIM-E) based on parental report and professional judgement was developed by members of the research team using evidence from the literature and with input from a number of expert groups and parent forums. It was then tested and rolled out across the five sites identified by PHE at the start of the project. The ELIM-E comprised five sections corresponding to areas that are commonly used to identify early language difficulties: language milestones, vocabulary list, family history and social risk factors, health visitor observations and parental concerns. Data collected were then used to reduce the ELIM-E down to the factors that best predicted which children were at risk of language difficulties and in need of further engagement with health visitors, ultimately resulting in a shortened version, the Early Language Identification Measure-Shortened (ELIM-S).The extended version of the measure was carried out by HVs and their skill mix teams as part of the Healthy Child Programme (HCP) 2-2 ½ year review19. The measure was carried out alongside the Ages and Stages Questionnaire-3rd Edition (ASQ-3), which is a population measure of child development currently used as part of all 2 -2½ year reviews by HVs in England 20. In addition, children involved in the study were then assessed by a speech and language therapist using a “gold standard” language measure called the Preschool Language Scale-UK 5th Edition (PLS-5)21, to ascertain where the child’s skills lay relative to a predetermined threshold on the PLS-5. The threshold derived from the literature 22 23 was set at the tenth percentile, which indicates that it would pick up only those children whose language scores fell in the bottom ten percent of the population.Matters to consider in relation to identifying need include the risk of missing something important, (false negatives) or identifying something that is unimportant, (false positives). The performance of a tool can be summarised in terms of sensitivity (the proportion of true positive cases identified) and specificity (the proportion of true negative cases identified). It is often possible to vary the threshold at which a screening tool triggers a definitive assessment. If we choose a low threshold, we increase the chance of identifying all true positive cases; in other words, sensitivity will be increased. However, there is a risk that this will raise the number of false positives and specificity will drop, meaning that some children without language problems will be referred for specialist assessment and treatment. This has practical implications for how services respond to language problems identified through such procedures; the majority of screening instruments lack sufficient sensitivity and specificity to identify child language problems at the individual level at reasonable cost. Indeed from the reviews to which reference was made above the key issue for most developmental conditions is sensitivity – the tests used commonly miss too many children in need of support. In this study, the emphasis was on optimising sensitivity – not missing children with SLCN, and on initiating a conversation with parents about what would best meet the needs of their child, thus managing the needs of the child without necessarily referring to specialist services.Underlying the process of identifying the right children is the delivery of both universal and targeted interventions, to promote robust language development to be offered to children and families at the 2-2½ year review. Thus, within the programme, the ELIM-S measure identifies the children and the intervention element then informs the support that is offered to them. Study methods were based on the most recent guidance regarding best practice in complex intervention design and behaviour change interventions and involved extensive stakeholder involvement and co-design. WHAT WE FOUND: The measure The data from 894 children were collected using the ELIM-E and of these 403 also received the PLS-5. The sample had representation in all the Income Deprivation Affecting Children Indices (IDACI) deciles, although there was a slight skew to the more disadvantaged end of the distribution. Each ELIM-E item was split into a binary variable and a single score given for each section. Different combinations of the sections were compared with the PLS-5 threshold. The priority was the sensitivity of the measure; its ability to correctly identify those children with SLCN (true positive). Alongside sensitivity we also measure specificity; the ability of the test to correctly identify those children without SLCN (true negative). Low specificity leads to children being over-identified. Data suggested that each section of the ELIM-E had some discriminatory power, but this varied considerably. The observation section gave the highest sensitivity and the vocabulary list section gave the highest specificity. These two sections taken in combination produced a sensitivity of 0.94 and specificity 0.65. By contrast, against the same criteria the ASQ-3 has good specificity of 0.93 but a relatively low sensitivity of 0.64. Thus out of 403 children seen on the combination of the practitioner observation and the word list, only six children with language difficulties were not picked up by the observation and/or the vocabulary list combination. The proportion of children over identified is higher. 108 children out of 306 were false positives. However the key here is the conversation that follows the ELIM-S that allows the practitioner to integrate their knowledge of the child and the family with the views of the parent to identify those most likely to need further engagement, and to equip parents with the skills needed to support their child’s development. Our proposal is therefore that these two sections (practitioner observation and the vocabulary list) be retained in the revised and shortened version of the ELIM.Parents’ responses to a survey carried out after their child had been seen at the 2-2½ year review suggested that the majority found the ELIM-E to be acceptable. A small number of parents reported difficulties with access to the HV, with the advice they were given, and with the interaction with their child. Parents participating in the telephone interviews had a broader range of views. For them acceptability was influenced by communication with the HV, convenience and ease of the review, the perceived expertise of the professional and the relationship that the HV established with the parent and with their child. From the perspective of the HV, the acceptability of the ELIM-E was related to the clarity of the rationale for its items, the interface between the timing of the review and related services such as speech and language therapy, alongside the potential of ELIM-E to support their decision-making and facilitate constructive conversations with parents. HVs felt that successful delivery of the ELIM-E was related to appropriate and sustainable training and practicalities such as the location of the review and the familiarity with the child and family not related to the tool itself. For HVs, the management of the conversation with the parent was crucial to the success of the review. Given the high sensitivity and lower specificity, the effective management of the needs of the child and parent is critical. The identification of the need is only ever the first stage and the resultant conversation needs to help the practitioner and parent consider other contributing factors, such as parental concern, behavioural and attention issues, whether the child speaks more than one language etc, and, to work with parents to determine the most appropriate level of support or intervention based on a continuum of need. Of course, this will also include decisions on whether there is good evidence that the child needs to be referred to child development or speech and language therapy services. The key issue is that the practitioner must draw upon their own knowledge and expertise to determine the most appropriate means of supporting child and parent, and we anticipate that the needs of most children can be managed by the health visitor team working to equip parents with the skills needed to support their child’s development. The aim is certainly not to just increase referrals to other services but to ensure that the child receives the right level of support for their speech and language development. Some children will respond to targeted intervention and may return to the universal level, while others may go on to have a more persistent need and require specialist services. The prerequisite to this however is always the conversation with parents and taking the parents’ views into consideration, i.e. shared decision-making. The Intervention The team synthesised child language intervention research evidence with expert knowledge, practitioner expertise and parent/caregiver views and preferences. We found practitioners (health visitors and members of the health visitor team) have an appetite and enthusiasm to promote children’s speech, language and communication development, but were not sure precisely how to work with families to deliver the most appropriate and acceptable intervention. Parents/caregivers wanted to be supported, to be proactive and agentive for their child as soon as possible. Based on stakeholder (parent and practitioner) preferences and intervention evidence, an intervention model was developed to support families to increase their use of responsive interaction behaviours within their daily routines and in contexts tailored to individual family circumstances. For equitable intervention delivery, we found we must not only create a need led proportionate model but also a tailored one, considering the specific barriers and enablers for each family. Potential barriers and enablers to the behaviour change across families were identified and a method devised for tailoring interventions accordingly. Communication between practitioner and parent/caregiver was identified as vital to success: language which invites partnership, dialogue and shared decision-making is essential.An intervention model was co-designed through iterative workshops, which is acceptable, practicable and equitable to the stakeholder participants. The resulting tiered intervention model, designed to be universal in reach and personalised in response is described. Differing pathways (levels of service), the steps through the intervention and the content, procedures, and materials are described. It is important to stress that while there are certainly preliminary indicators for how practitioners should respond to the findings of the ELIM-S and the face value of this approach has been demonstrated, further development is required for the testing of this approach to the intervention, including the production of the intervention materials, (videos, shared-goal setting tools, invitation letters etc.) and the linked practitioner training programme,prior to piloting and evaluation. CONCLUSIONS: The shortened Early Language Identification Measure (ELIM-S) has the potential to be a powerful tool in identifying all children with SLCN at the 2-2½ year review. However, as with all such brief measures of child development, it cannot stand on its own and it is imperative that it is closely associated with the conversation between health visitors and their teams and the parent; there is a need to conceptualise the ELIM-S as a part of a wider intervention. The study has demonstrated an appetite amongst practitioners for this focus on SLCN and shown the importance of the practitioner-parent relationship. And it is clear that parents want to be listened to. The trust that comes from this relationship is critical to the shared decision-making that, in turn, is fundamental to the guidance that is offered to parents. Underpinning this is appropriate levels of training for those involved in identifying and working with children, and the importance of health visitors and early years practitioners working together to make sure that they are monitoring children’s assessment and development effectively, working to equip parents with the skills needed to support their child’s development and providing interventions in collaboration with parents.Our preferred solution is a three-step process whereby practitioner observation and vocabulary act as a starting point for identification, which is underpinned by a preliminary exploration of parental concern. The practitioner observation and vocabulary list (Step 1) then lead into further exploration with the parent about the areas of greatest concern (Step 2), whether or not further referral to other local provisions such as speech and language therapy services is appropriate, signposting to relevant materials etc. Finally practitioner and parent agree and review intervention goals together (Step 3) in the context of parental capability and motivation and the opportunities available to them. As indicated further development and evaluation of the whole programme with the detailed intervention component is needed.Clearly the 2-2½ year review process is one which involves a great many people – the parents and children, the health visitors and their teams, the early years practitioners in the settings where a proportion of the children attend, and the speech and language therapists to whom the children with the more marked difficulties will be referred, but who are also likely to be instrumental in supporting the other members of the team. Inevitably there are challenges in ensuring that all the members of the team share goals, expertise and expectations around guidance so that the parents feel that the services are working with them in this all important review process.
Author(s): Law J, Charlton J, McKean C, Watson R, Roulstone S, Holme C, Gilroy V, Wilson P, Rush R
Publication type: Report
Publication status: Published
Online publication date: 06/11/2020
Acceptance date: 03/11/2020
Institution: Newcastle University
Place Published: Newcastle upon Tyne