Objectives

• To review the clinical practice and research evidence for the interventions, outcomes measured and tools used to measure these outcomes for EDSD in young children with neurodisability.
• To identify the subgroups of children for whom there is the most robust evidence on intervention success/failure.
• To investigate the extent to which interventions have been defined and manualised to facilitate replication.

Methods

We searched for literature pertaining to any intervention that aimed to improve EDSD for children with neurodisability. This was a mapping review rather than a systematic review to establish an estimate of the effectiveness or assess the quality of the evidence. Nonetheless, the approach taken to searching and screening was rigorous and consistent with that used in a systematic review.

Inclusion criteria

Literature was included in the mapping review if it met the following criteria:

• Population – children (aged 0–8 years) with any type of non-progressive neurodisability who had EDSD. The following conditions were excluded: cystic fibrosis, gastro-oesophageal reflux and structural abnormalities [e.g. cleft lip and palate, and CHARGE (coloboma, heart defects, choanal atresia, growth retardation, genital abnormalities and ear abnormalities) syndrome]. Children who had rumination (i.e. persistent regurgitation, re-chewing, re-swallowing or vomiting of previously eaten foods), eating disorders (unless specifically about food avoidance/restrictions not related to a desire for thinness) or problem behaviour at mealtimes that was not related to eating were also excluded. Studies were included if any of the participants were aged 0–8 years.
• Intervention – any intervention to improve eating, drinking and swallowing that can be delivered by parents to their children aged 0–8 years. The following interventions were excluded: Pharmacological, Dietary or Nutritional interventions, Gastrostomy and Oral appliances. Interventions that focused on speech development or improvement and the swallowing of tablets were also excluded.
• Comparator – any other intervention for eating, drinking and swallowing or mealtime behaviour, any intervention described as ‘treatment as usual’ or no intervention.
• Outcome – any outcome pertaining to food intake, behaviour, health, well-being or acceptability.
• Study design – systematic reviews of interventions and any controlled or non-controlled study of intervention effects or acceptability. Editorial/commentary/opinion articles were excluded.
• Limitations – manuscripts written in English and published from January 1985 to October 2017.

Searches were designed by an information specialist in collaboration with the project team. The search strategy was designed on MEDLINE [via Ovid^Ⓡ (Wolters Kluwer, Alphen aan den Rijn, the Netherlands)] using thesaurus headings and title and abstract keywords, and translated as appropriate to the following databases: Cumulative Index to Nursing and Allied Health Literature (CINAHL) database [via EBSCOhost (EBSCO Information Services, Ipswich, MA, USA)], PsycINFO (via Ovid), Web of Science™ (WoS; Clarivate Analytics, Philadelphia, PA, USA), EMBASE™ (Elsevier, Amsterdam, the Netherlands) (via Ovid), Education Resources Information Center (ERIC) (via EBSCOhost), Cochrane Database of Systematic Reviews [via Wiley Online Library (John Wiley & Sons, Inc., Hoboken, NJ, USA)], Cochrane Central Register of Controlled Trials (CENTRAL) (via Wiley Online Library), The Speech Pathology Database for Best Interventions and Treatment Efficacy (speechBITE; The University of Sydney Lidcombe, NSW, Australia) (www.speechbite.com) and Occupational Therapy Systematic Evaluation of Evidence (OTseeker; www.otseeker.com). This search was run between 5 October and 17 October 2017. Full details of the search strategies are presented in Appendix 4. Two researchers (HT and LP) independently screened titles and abstracts to identify studies meeting the inclusion criteria. The full texts of potentially eligible articles were retrieved and assessed independently against inclusion criteria by two researchers (HT and LP or HM). Where there were discrepancies in these processes, a third person from the review team was consulted and a consensus was reached. One researcher (HT) extracted the data and classified each study; LP checked the data extraction and coding.

Results

Our searches identified 5790 references; following sifting on title and abstract, we retrieved 492 full texts, of which 147 fitted the inclusion criteria (Figure 4). Fifteen of the papers identified through the updates of the published systematic reviews of interventions (those from the Marshall et al.⁵ and NICE⁶ updates) were also found in the mapping review, including nine single-case experimental design studies replicated across fewer than four participants.

FIGURE 4

The PRISMA flow chart of mapping review.

Study participants ranged in age from < 1 year to 31 years, with many of the studies including participants outside the age range that we defined as ‘young children’ (i.e. ≥ 9 years). In most cases, the results for our target group of young children (aged 0–8 years) could not be disaggregated. The interventions reported across the studies were grouped as addressing physical, mixed or non-physical factors affecting eating, drinking and swallowing: 27 studies addressed physical EDSD, 53 non-physical EDSD and 66 mixed EDSD. Most interventions directly targeted EDSD, such as Modifications (Environment, Equipment, Food or drink, Placement of food and Positioning), improved mealtime communication (Enhancing communication strategies, Responding to the child’s cues for feeding and Pace of feeding) and desensitisation strategies (Graded exposure to foods or textures, and Oral and sensory desensitisation). Other interventions that did not directly target EDSD included Psychological support for child and parent and Self-feeding. Teaching techniques (Prompting and Reinforcement) were referred to frequently in the teaching of any of these interventions. There was a range of outcomes measured across these studies including Swallowing function, Chest health, Amount of food eaten, Eating efficiency, Oral motor function, Number (percentage) of bites, Variety of food consumed, Mealtime behaviour, Self-feeding, Food acceptance and Amount of liquid consumed. Further details of the included studies are presented in Table 3.

TABLE 3

Summary of evidence from the mapping review.

The totals for the number of studies and included participants do not include the systematic or literature reviews to prevent double counting and due to the reviews including a large number of studies that did not meet the criteria for inclusion.

The majority of studies described multicomponent interventions; for example, an intervention might ensure that the children were in a safe position to eat and drink (Positioning), were fed textures that they could swallow easily (Modifying food or drink) and received praise for swallowing (Reinforcement). The frequency of individual interventions studies included in the mapping review is shown in Figure 5. The mapping process enabled us to disaggregate multicomponent interventions to explore a number of questions, including (1) which individual interventions were more frequently provided together as a multicomponent intervention; (2) the difference in the frequency of interventions between participants with physical and mixed EDSD and participants with non-physical EDSD; and (3) the number of participants in whom each intervention had been assessed.

FIGURE 5

Number of studies and included participants by intervention and type of EDSD.

We identified 18 individual interventions, most of which had been assessed within a multicomponent intervention that included participants with physical, mixed and non-physical EDSD (see Figure 5). In Figure 5, we have presented the frequency of assessment of each intervention, based on the number of studies, alongside the total number of participants in those studies. The frequency count is based on the number of studies reporting the primary outcome. Based on the number of studies, the most common individual interventions considered across the populations were Reinforcement (109 studies, 554 participants) and Prompting (97 studies, 393 participants). However, these individual interventions are teaching techniques to support the delivery of specific EDSD interventions. Beyond the teaching interventions/techniques, the most commonly assessed interventions for children with physical or mixed EDSD were Modification of food or drink (33 studies, 519 participants), Positioning (22 studies, 456 participants), Modifying equipment (19 studies, 194 participants) and Oral motor exercises (17 studies, 498 participants). The most commonly assessed interventions for children with non-physical EDSD were Modification of food or drink (26 studies, 104 participants), followed by Visual supports (19 studies, 120 participants), Food desensitisation (18 studies, 124 participants) and Scheduling of meals (17 studies, 139 participants). Psychological support for parents (three studies, 44 participants) and Responding to a child’s cues for feeding (three studies, 72 participants) were assessed only for participants with physical or mixed EDSD; however, Psychological support for the child (one study, 11 participants) was found in an intervention assessing only children with non-physical EDSD. Although Psychological support for parents and the child and Responding to a child’s cues for feeding were included in studies infrequently (i.e. evaluated in fewer than five studies), a large number of individual interventions were seen in more than 10 studies of both children with physical or mixed EDSD and children with non-physical EDSD. Figure 5 illustrates the significant overlap in the individual interventions being considered for children with physical or mixed and non-physical EDSD. Only three interventions were considered in only one of the populations: Responding to a child’s cues for feedings (three studies) and Psychological support for the parents (three studies) were considered in only a non-physical population, and Psychological support for the child (one study) was considered in only children with physical and mixed EDSD. These three individual interventions were also the least frequently considered.

A range of study designs, from those providing the highest level of evidence (systematic reviews of RCTs) to those providing the lowest (case studies), were used to evaluate the interventions (Figure 6 and Table 3). Appendix 5 shows the study designs used to evaluate each intervention. Figure 6 shows that a large number of studies had designs that are widely considered to be less robust and, therefore, more prone to bias, such as case studies and before-and-after studies. In total we identified 147 studies, 121 of which were before-and-after studies, case studies, literature reviews or single-case experimental designs. We also identified 12 RCT/quasi-experimental design studies and 14 systematic reviews. The amount of evidence included in these reviews was variable and they provided no robust conclusions regarding the optimal multicomponent intervention.

FIGURE 6

Types of study design and numbers of participants by causes of EDSD.

Outcomes

The studies in the review measured 24 different outcomes (as shown in Table 3), with most studies measuring multiple outcomes.

Measures

The studies used 33 published protocols/measures to assess change in the outcomes, as shown in Table 3. A total of 25 studies used published protocols/measures, with the remaining studies using bespoke measures that the authors had developed specifically for use in their study. These bespoke measures lacked evidence of reliability, validity or responsiveness to change.

Summary of mapping review

The mapping review collated a wide range of research evidence. The scope and purpose of the mapping review was not to assess the quality of the individual studies but rather to explore and understand the frequency and level of evidence for each of the individual interventions. The aim was to present a full picture of the interventions that have already been developed and/or evaluated, to ensure that all of the potential interventions were considered in the later stages of this work. The level of evidence found suggests that there are several studies evaluating some of the individual interventions identified; however, owing to the design of the studies it is likely to be low-level evidence, given that there is a lack of RCTs. The mapping review also highlighted the significant overlaps in the interventions delivered to children with physical or mixed and non-physical EDSD in these studies.

Strengths and limitations of the mapping review

The mapping review has elicited the number of published studies evaluating each individual intervention type, the study designs used and the number of participants with physical or mixed and non-physical EDSD included. Considering the number of studies identified, it was necessary to take a pragmatic approach to data extraction, ensuring that only the necessary information was retrieved. Furthermore, we have not explored the data beyond the scope of the question that we set out to address. There are many other questions that this literature base might support answering; however, these were out of scope of this research.

Patient and public involvement in the mapping review

Parent co-investigators checked the intervention and condition terms that were included in the searches for the mapping review. They also discussed the search findings with members of the research team, to help ensure that all relevant studies were being retrieved. The PAG considered the summaries of the findings from the mapping review alongside findings from the updates of the three published systematic reviews of interventions and national survey. The PAG commented on a pictorial summary of the identified interventions and outcomes to aid discussion in the second round of focus groups (see Chapter 9) and the stakeholder consultation workshops (see Chapter 11) (see Figures 9 and 10).

How did the mapping review inform the next step?

The mapping review provided information regarding the evidence base for interventions aimed at improving EDSD in children with neurodisability, the outcomes measured and the tools used to measure those outcomes. This information informed the list of interventions and outcomes included in the national survey (see Chapter 7), the discussions within the second round of focus groups (see Chapter 9) and the searches for papers examining relevant outcome measurement tools for the measurement properties review (see Chapter 6).