Training support staff to modify fluids to ... - Wiley Online Library

bs_bs_banner

Journal of Intellectual Disability Research 84

doi: 10.1111/jir.12013

volume 58 part 1 pp 84–98 january 2014

Training support staff to modify fluids to appropriate safe consistencies for adults with intellectual disabilities and dysphagia: an efficacy study D. D. Chadwick,1 J. Stubbs,2 S. Fovargue,2 D. Anderson,2 G. Stacey2 & S. Tye2 1 School of Applied Sciences, The University of Wolverhampton, Wolverhampton, UK 2 Mill Road Therapy Centre, Mill Road, Colchester, UK

Background Modifying the consistency of food and drink is a strategy commonly used in the management of dysphagia for people with intellectual disabilities (ID). People with ID often depend on others for the preparation of food and drink and therefore depend on those caregivers achieving the correct consistency to keep them safe and avoid discomfort during mealtimes. Clinical experience and prior research have demonstrated that although training can improve modification, carers often find modification difficult and potentially stressful and recommend additional support for carers. Fluid consistency is often modified through the addition of powdered thickener. This study investigates the efficacy of typical training and use of consistency guides, the Thickness Indicator Model (TIM) tubes, in helping carers to modify fluids accurately. Method A 3 ¥ 3 pre–post experimental design with a control group was employed to compare the observed accuracy of modification across three

groups and at three time points (pre-intervention baseline, immediately post-training intervention and 3–10 months post-training). Sixty-two paid carers who supported people with ID were recruited to participate in the study and each was randomly allocated to one of the three groups: a control group given written guidance only, a group who received typical training and written guidance and a group who received training, written guidance and the TIM tubes. Results & Conclusions Typical training resulted in significantly greater carer accuracy in modifying fluid consistencies when compared with written guidance alone. Use of the TIM tubes also significantly improved accuracy in the modification of drinks compared with the group who modified with the aid of written guidance alone. At 3–10-month follow-up only the group who received typical training alongside the TIM tubes were significantly more accurate than the Written Guidance group. Further research is warranted to ascertain the effectiveness of the training and the utility of the TIM tubes in improving accuracy over a longer time scale and in individuals’ usual living environments.

Correspondence: Dr Darren David Chadwick, School of Applied Sciences, The University of Wolverhampton, Wolverhampton, UK (e-mail: [email protected]).

Keywords dietary modification, dysphagia, efficacy, intellectual disabilities, training

Abstract

© 2013 The Authors. Journal of Intellectual Disability Research © 2013 John Wiley & Sons Ltd, MENCAP & IASSIDD

volume 58 part 1 january 2014

Journal of Intellectual Disability Research 85 D. D. Chadwick et al. • Training support staff to modify fluids

Background Dysphagia has been defined as ‘eating and drinking disorders which may occur in the oral, pharyngeal and oesophageal stages of deglutition. Subsumed in this definition are problems positioning food in the mouth and in oral movements, including suckling, sucking, mastication and the process of swallowing’ (Royal College of Speech and Language Therapists, Communicating Quality 3, 2006). Dysphagia is a significant problem for people with intellectual disabilities (ID) (Chadwick & Jolliffe 2009). Dysphagia puts people with ID at risk of discomfort, poor nutritional status, dehydration, aspiration, choking and reduced quality of life (Helfrich-Miller et al. 1986; Wood 1994; Kennedy et al. 1997; Kerr 1998; Aziz & Cambell-Taylor 1999; Chadwick et al. 2003, 2006; Samuels & Chadwick 2006). Many of these problems can be life threatening or can lead to other life threatening problems, for example, aspiration leading to upper respiratory infections (Helfrich-Miller et al. 1986) and choking leading to asphyxia (Dupont & Mortenson 1990; Samuels & Chadwick 2006). The National Patient Safety Agency identified dysphagia as one of the five key risk areas for people with ID (NPSA 2004) and its importance is further highlighted by the MENCAP report ‘Death by Indifference’ (MENCAP 2007) as swallowing difficulties were a contributing factor in all of the case studies contained in the report. Dysphagia is managed by a range of rehabilitative and direct and indirect compensatory strategies, dependent upon the underlying aetiology (Groher 1987). Consistency modification is one such strategy and incorporates modification of the consistency of both food and drinks; in this study we are focussing particularly on the thickening of fluids. Referring to modification of both food and drinks, Logemann (1999) stated that consistency modification should be the last compensatory strategy examined, unless other strategies, for example altering posture or using techniques to help to protect their airway, such as the effortful swallow or the Mendelsohn Maneuve (Logemann 1999), are not feasible as is often the case with adults with ID and other neurological or degenerative conditions. In an American survey of Speech and Language Pathologists, 84.6% indicated that they considered

thickening of thin fluids to be an effective management technique for dysphagia (Garcia et al. 2005). Modifying the consistency of food and drinks has been found to delay bolus transit through the mouth and pharynx, reduce the speed of tongue peristalsis, extend the duration of pharyngeal peristalsis and prolong upper oesophageal opening, reducing avoidance of food and drinks, coughing and choking (Griggs et al. 1989; Reimers-Neils et al. 1994). However, Langmore (1999) highlights the need for more research into the use of modified consistencies to manage dysphagia in different populations. There is no strong clinical research evidence supporting modification of food and drinks as an effective management strategy specifically for adults with dysphagia and ID. However, in the general research literature, of the different dysphagia management strategies used, consistency modification has the strongest evidence supporting its efficacy (Groher 1987; Griggs et al. 1989; Reimers-Neils et al. 1994). Consistency modification is commonly used in the management of dysphagia for people with ID, with Speech and Language Therapists (SLTs) typically training caregivers to modify food and drinks to a consistency appropriate for the individual based on clinical and videofluoroscopic assessments (Chadwick et al. 2002, 2003). A recent study by Ball et al. (2012) found that 25% of adults with ID who required mealtime support had fluid consistency modified. People with ID and dysphagia who are receiving modified food and drinks have a dependence on the skills and knowledge of the caregiver and in particular, the modification of fluids using powdered thickener, relies on carers achieving the correct consistency through correct instruction and adequate guidelines (Harding & Halai 2009). It is important that fluids are thickened to the appropriate consistency specific to the individual so as to facilitate comfortable safe oral transit and deglutition and reduce the risk of aspiration. In a study of inadequate fluid intake in dysphagic acute stroke patients, Whelan (2001) found that the use of powder thickeners resulted in less than 50% of patients receiving fluids thickened to the correct consistency. Moreover, caregivers modifying consistently is dependent on appropriate training by professionals and worryingly, professionals have




been found to have difficulty being consistent when modifying liquids (Glassburn & Deem 1998). An investigation into the barriers caregivers experience in managing dysphagia for adults with ID found that caregivers often found food and drink modification one of the most challenging aspects of management. In circumstances where the person with ID had a history of choking on unmodified food or drinks caregivers in residential and day settings found modification and management stressful, often worrying about whether they had achieved the appropriate consistency (Chadwick et al. 2006). Crawford et al. (2007) in an investigation of caregiver compliance with dysphagia recommendations in day centres, found that ‘food provision’ guidelines were reported to be difficult to follow and highlighted the need for training in the area of food modification. These studies suggest that although training can increase accuracy of fluid modification (Chadwick et al. 2003) that mealtime support staff can still have difficulties achieving a safe, appropriate fluid consistency (Chadwick et al. 2006; Crawford et al. 2007), recommending additional support for caregivers supporting people with ID and dysphagia, at least until they feel fully confident modifying fluids. This investigation will add to the small number of studies focusing on dysphagia intervention by exploring the efficacy of training caregivers to modify fluids. The study was prompted by the concerns of a group of clinicians [dietitians, Speech & SLTs, Occupational Therapists (OTs) and community nurses] regarding the management of dysphagia for adults with ID living in the community. They observed that caregivers were sometimes unable to reliably and accurately modify fluids to the prescribed consistency. In considering how to overcome this difficulty the group of clinicians developed a visual tool which could serve as a comparator and a memory aid when people are modifying fluids. The Thickness Indicator Model (TIM) was developed comprising a series of fluid filled tubes, which have been used in this study. Harding & Halai (2009) identified visual demonstration as a way of enabling people to gain a clear understanding of the liquid thickness required and concluded that research needs to focus on the training methods used to support staff.

Typically, training with regard to modifying food and drinks to a safe consistency, incorporates the provision of written information describing the consistencies, discussing modification, modelling how to modify food and drinks to appropriate consistencies, watching caregivers modify meals and drinks and providing feedback on the results (Chadwick et al. 2002). This study focuses on fluid modification and will compare the efficacy of this typical model of training intervention for fluid modification with the provision of written guidance alone. The study also aims to evaluate the benefits of using a visual model (the TIM tubes) to support training and act as a comparator, thus reducing the reliance on memory when modifying fluids. Given the prior literature and clinical experience suggesting that training can improve accuracy but that additional support is required to overcome modification errors, we hypothesised that the training should improve accuracy and that the addition of the TIM tubes should enhance fluid modification accuracy further. Thus we developed the followed research questions and hypotheses prior to commencement of the study.

Research questions 1 Is there a difference between the three training groups in the degree of improvement in observed accuracy of fluid consistency modification measured immediately post-training? 2 Is there a difference between training groups in the degree of improvement in observed accuracy of fluid consistency modification measured at 3–10month follow-up post-training?

Specific hypotheses 1 Typical training will result in greater caregiver accuracy in modifying fluid consistencies compared with written guidance alone immediately posttraining intervention when adjustments are made for pre-intervention baseline scores. 2 Typical training together with use of the TIM tubes will result in greater caregiver accuracy in modifying fluid consistencies compared with both the typical training and written guidance immedi-




ately post-training, when adjustments are made for pre-intervention baseline scores. 3 The group given typical training will be more accurate at modifying fluids compared with the group given written guidance only at 3–10-month follow-up, when adjustments are made for preintervention baseline scores. 4 The group given both typical training and the TIM tubes will be more accurate in modifying fluid consistencies than both the group who received typical training and the group who received written guidance only at 3–10-month follow-up, when adjustments are made for pre-intervention baseline scores.

Methodology Study design This is an efficacy study employing a pre–post experimental design with a control group (3 ¥ 3 mixed experimental design repeated on one factor) (see Table 1). The between groups independent variable (IV1 – Group) for the study was the extent of training and guidance on consistency modification received by participants. This comprised three levels: 1 Written Guidance only (Acting as a control group);

2 Typical Training in achieving appropriate consistencies and written guidance (Typical Training); and 3 Training in achieving appropriate consistencies, written guidance and use of TIM tubes to check achieved consistencies (Typical Training & TIM). The within groups independent variable (IV2 – Time) for the study was the occasions on which the outcomes were assessed. Participants were assessed on three different occasions: (1) pre-training intervention baseline; (2) immediately post-training intervention; and (3) 3–10 months post-training intervention. The first two assessments were completed during the initial visit. The dependent variable for the study was the accuracy of modification to the appropriate consistency for the three fluid consistencies identified in the national descriptor guidance (Stages 1, 2 & 3, see Appendix 1) (British Dietetics Association/Royal College of Speech and Language Therapists 2006). A mean accuracy score for the three stages was calculated for each participant for each of the three time points. Ethical approval for the study was granted by Central Manchester Local Research Ethics Committee (Ref: 06/Q1407/242).

Training & guidance Descriptions of the components of the different types of training undertaken by the participants are

Table 1 Design for the fluid consistency modification efficacy study

IV1

Group 1: National guidance Group 2: National guidance document only (Written document & Typical Guidance – Control Training Group)

Group 3: National guidance document,Typical Training & TIM tubes

IV2 Prior to training Accuracy assessment for (Pre-intervention Baseline – All groups all three consistencies could access written guidance)

Accuracy assessment for all Accuracy assessment for all three consistencies three consistencies

Random assignment to one of the three training groups Post-training 3–10 months post-training

Accuracy assessment for all three consistencies Accuracy assessment for all three consistencies

Accuracy assessment for all Accuracy assessment for all three consistencies three consistencies Accuracy assessment for all Accuracy assessment for all three consistencies three consistencies

IV, independent variable.




presented below. Water was the fluid thickened to all consistencies by all intervention groups, 600 ml of water in each case. The same starch based commercial food thickening agent was used throughout the study. Trainers checked that they were consistent in delivering the training procedures among themselves prior to data collection and a guidance document detailing the training process was used by trainers throughout the study to help ensure that training presentation was consistent. Written guidance The written guidance given was based on the National Descriptors for Texture Modification in Adults (BDA/RCSLT 2002). The fluid consistencies identified in this document are thin, naturally thick, and stage 1, 2 and 3 thickened liquids, full descriptions can be seen in Appendix 1. The written guidance for the participants detailed the equipment in front of each participant, gave instructions on how to thicken a drink, and descriptions of the stage 1, 2 and 3 fluid consistencies. Thin and naturally thick consistencies were omitted as no modification is required to achieve these. The written guidance for the participants asked them to inform the trainer when they believed the desired consistency had been achieved. Training details Typical training incorporated a brief explanation of the reasons for modifying fluid consistency, modelling how to modify to the three consistencies, and observation and feedback on each participant’s initial attempt at modifying the fluid to each of these consistencies. Training was delivered by one of five clinicians, each with more than 10 years experience in dysphagia management, including two SLTs, one OT and two Dieticians. Before conducting the training, the clinicians devised a document detailing the training process at each stage. This gave a brief description of why fluids might be modified and gave examples of words the clinician could use to describe the different consistencies to ensure that they were consistent in the terminology they used during the training process. Detailed protocols were developed for each stage of the study to ensure each clinician followed the same procedure.

Consistency indicator tubes (TIM tubes) The TIM tubes were developed to help to convey to people supporting those with dysphagia how a drink should appear when thickened to a prescribed consistency. They provide a visual cue that can be compared with the target fluid. They were developed in response to the clinical problem that written guidance does not convey the rate of flow of fluids to new carers, and because the clinicians had observed that carers were often unable to reliably and accurately modify fluids to the prescribed consistency. To overcome this difficulty a prototype visual aid was developed. This was a crate type structure with a wooden base and wooden lid containing five similar jam jars in a line. The jars contained fluids of varying thickness replicating the stages of thickening as described by The National Descriptors. This was trialled with a small team of carers who were finding it difficult to thicken the drinks of the person for whom they were caring; they found the visual aid helpful. The idea of a visual aid was developed further and fabricated with the support of the Health Innovations East team who enabled the production of the TIM tubes, The TIM tubes contain approximations of the national fluid descriptors (see Table 2 and Fig. 1). Each consistency is a different colour and is housed in a 135-mm long and 35-mm diameter sealed cylindrical tube.

Participants, sampling, randomisation The participants were derived from populations of non-health professional caregivers within community-based ID support services in North East and Mid Essex in the east of England. Managers at all care homes and day centres registered for

Table 2 Fluid descriptor contents, colour and viscosity of the TIM (Thickness Indicator Model) tubes

Naturally thick Stage 1 Stage 2 Stage 3

Viscosity (centistokes)

Fluid colour

200 500 5 000 30 000

Yellow Green Red Blue




Figure 1 TIM (Thickness Indicator Model) tubes showing consistencies from right to left – naturally thick, Stage 1, Stage 2, Stage 3.

adults with ID were contacted and asked to provide names of staff working within their organisation. Inclusion criteria for the study were that participants were support staff who had no prior experience of modifying fluid consistencies or working with people with dysphagia. This resulted in a sampling frame of 171 support staff. This list of names was randomised electronically and the first 60 people from the list created were sent a letter and information sheet which was followed by a telephone interview to determine if they were eligible and willing to participate. Information about the rationale for the study and how much time would need to be committed was given. Piloting of the data collection procedure among the clinicians indicated that approximately 1 h of involvement would be required for each visit. Once they had consented to participate an appointment was arranged. Any participants who did not wish to take part or who did not meet the inclusion criteria were not included in the study and the next person on the list was then contacted. Of the 171 names initially given 82 were either excluded from the study, did not wish to participate, or were not able to participate. Reasons for exclusion included, previous experience of modifying fluid consistencies and being about to change jobs or retire. Those who were unwilling or unable to take part did not always explain why but reasons that were given included lack of transport to get to

appointments, ill health, and feeling that participation was not relevant to their role. By the end of the recruitment process all 89 suitable participants had been contacted of which 16 had failed to attend appointments and three had left their employment as support staff prior to completing the data collection for the study. The remaining 70 people all participated fully but eight sets of data were incomplete and excluded from the final analysis, leaving a final sample size of 62 participants, 22 in the Written Guidance only group and 20 in each of the two intervention groups. Each participant was randomly assigned to one of the three training groups during their first appointment. Random assignment occurred through the use of 62 identical, sealed envelopes. Each envelope contained a piece of paper indicating which of the three groups the participant should be assigned to and this was not opened until after the preintervention baseline assessment had been completed. Block randomisation for each batch of 15 participants was conducted to ensure similar numbers in each of the three groups in case of premature finishing of the study and also to control for temporal extraneous variables. This meant that after each batch of 15 participants were allocated there were an additional five participants in each of the three groups. Forty-three women and 19 men completed the study. They ranged from 19 to 64 years of age




(Mean = 41.26, SD = 12.84), the mean ages of participants in the three training groups were 43.14 (SD = 12.53) for the Written Guidance only group, 38.95 (SD = 11.80) for the Typical Training group and 41.50 (SD = 12.84) for the Typical Training and TIM tube training group. Most participants had English (87.1%) as their first language with eight participants citing English as a second language. Only three of the participants said they found written information difficult to understand and the instructions were then read to them.

Data collection At the initial appointment participants were told that the study was concerned with modification of fluids and no further information was given at this stage. Following this background information about the age, gender and first language was collected. Participants were then given the written guidance and asked to modify 600 ml of water to each of the three target consistencies which were then assessed for accuracy (as described below). Participants were then randomly assigned to one of the three groups by selecting a sealed envelope from a box. Once allocated to the intervention group participants were told why fluids might be thickened and were then either trained (Groups 2 and 3) or asked to read the modification guidance again (Group 1). Following each type of training intervention participants were asked to modify water to the three consistencies (Stage 1, 2 & 3). While they were modifying the fluids all participants could access the written guidance. As noted earlier Groups 2 and 3, who received typical training, had an initial attempt at modifying the fluids and received feedback on this initial attempt. Group 3, were additionally shown the TIM tubes and told that this was the consistency that they were being asked to achieve; they could use the TIM tubes as a reference tool while they were modifying the fluids to each stage. The time taken for training and the time taken by participants to thicken to each consistency were both recorded. At the end of the initial appointment the time, date and venue for their next assessment was arranged with participants. During reassessment participants were reminded that this was a study looking at thickening drinks. Assessors checked that participants had had no practice at modifying fluids

in the interim period. All participants were provided with written instructions, Group 3 were also given the TIM tubes to use. The participants were asked to modify the fluid (water) to the three consistency stages. All participants were given second appointments within an initially selected 3–6-month time frame. However, the time between the initial appointment and the follow-up actually ranged between 3 and 10 months (Mean = 4.15 months, SD = 1.55), though most occurred between 3 and 6 months (93.5%). Reasons for failure to meet the 3–6-month target included illness, maternity leave and repeated failure to attend. There was no significant difference in the number of months between the first and second visit between the three groups [Kruskall–Wallis1 c2(2) = 3.53, P = 0.17; Mean Number of months for Written Guidance only = 3.94 (SD = 1.23), Typical training = 4.41 (SD = 1.99), Typical Training & TIM tubes = 3.9 (SD = 1.24)]. Letters reminding the participant of their next appointment were sent and a reminder phone call made 1 week before the second appointment. Training and testing took place on four sites across North East and Mid Essex and was conducted under similar environmental conditions. At each time of testing, the trainer and blind assessor recorded the temperatures of the room in which the liquid was mixed and the room where the blind assessment of accuracy was carried out. The temperatures of the fluids before modification, after modification and before assessment were also recorded. Each research clinician acted as both trainer and blind assessor during the course of the research and was responsible for training participants from each of the three training groups.

Accuracy measurement Prior to commencing the study, the clinicians involved had a number of meetings where they spent time modifying water to the three different consistencies and checking them against each other and the TIM tubes. This ensured that they all felt confident that they could identify all the consistencies by observing them and could modify fluids consistently and accurately to those consistencies. 1 Kruskall–Wallis was conducted as variances were unequal across the groups in the anova analysis (Levene(2,59) = 3.92, P = 0.025).




blind observed participant accuracy data [calculated by averaging the accuracy data for each of the three modification stages (1, 2 & 3)] was analysed to answer the research hypotheses using ancova. This analysis was used to examine the change in accuracy scores between the different groups at the time post-training intervention time points while controlling the pre-training intervention score. Because of the a priori nature of the specified hypotheses planned contrasts (unpaired t-test) were used to test them. Data were analysed using IBM spss Statistics for Windows version 19.0 (IBM Corporation 2010).

Accuracy was measured by observation, of each consistency at each time point. Assessment of the dependent variables was conducted by specialist SLTs, OTs and Dieticians. The researchers conducting the training made observations about the accuracy of the modified fluid and rated it on a Likert scale of -4 to +4 with 0 being the target consistency, -4 being too thin and +4 being too thick (see Table 3). For analysis purposes minus scores were recoded so that -4 became +4 and meant inaccurate (too thick/thin) with 0 meaning accurate. The modified fluids were then taken to a second researcher in another room who also observed the modified fluids and rated them on the same scale. This second blind assessment of accuracy was the measurement used in the analysis. Having two independent observational raters of accuracy of fluid modification for all participants at all consistency stages allowed assessment of reliability of the observed accuracy ratings via analysis of their inter-rater reliability. Intra-class correlation was used to determine inter-rater reliability (Shrout & Fleiss 1979); coefficients ranged between 0.71 and 0.90 (n = 61), comparing these with the thresholds laid out by Cohen (1988) coefficients above 0.5 indicate a high level of agreement between the clinician observers.

Results Pre-intervention equivalence of the three groups A series of analyses were conducted to determine if there were any salient differences between the three groups.

Background characteristics of the participants With regard to demographic characteristics, no significant differences were found between the three groups in age (F2,59 = 0.55, P = 0.59) and gender (c2(2) = 3.57, P = 0.17) with, as expected, more women in all groups.

Statistical analysis

Temperature of the rooms

Following screening and preliminary analyses to determine pre-intervention equivalence of groups,

As temperature can influence fluid viscosity, it was decided to check the temperature of the room, for

Table 3 The scale and descriptors from the assessment sheet for modification to the three stages

Observational rating scale & associated descriptors Stage

-4

Stage 1

Water

Stage 2

Naturally thick Stage 1 or thinner

Stage 3

-3

-2 Naturally thick Stage 1 Stage 2

-1

0 Target consistency Target consistency Target consistency

+1

+2

+3

+4

Stage 2

Stage 3

Stage 3

Thicker than stage 3 Practically Solid

Thicker than stage 3

Note: Following initial piloting of the rating scales, raters felt that the increments described were not sufficiently detailed to adequately reflect the range of possible consistencies to which the fluids could be modified. To address this concern additional increments were introduced to allow for intermediate ratings, e.g. a Stage 1 fluid that was thinner than naturally thick but not as thin as water would be rated as -3.




and to drink more safely (see Appendix 1 for full descriptions). Accuracy was measured for all three stages for all participants. The pattern of findings across the stages supports observations made by the clinician authors that Stage 3 is more difficult to accurately achieve. Despite this, because of the similarity in the pattern of changes in accuracy score across the three stages with training (Table 4), the accuracy scores across the three stages were pooled for group comparisons.

each participant, prior to data collection. There were no significant differences across groups in the temperature of the room where fluid consistency modification and training took place at any of the three time points: (i) Pre: F2,59 = 0.37, P = 0.69, Written Guidance mean = 21.82 (SD = 1.65), Typical Training mean = 21.40 (SD = 1.98), Typical Training & TIM tube mean = 21.40 (SD = 1.88); (2) Post: F2,59 = 0.03, P = 0.97, Written Guidance mean = 21.68 (SD = 1.59), Typical Training mean = 21.55 (SD = 2.01), Typical Training & TIM tube mean = 21.60 (SD = 1.54); (3) 3–10 months: F2,59 = 1.26, P = 0.29), Written Guidance mean = 22.36 (SD = 1.86), Typical Training mean = 21.50 (SD = 1.40), Typical Training & TIM tube mean = 21.85 (SD = 2.01). In addition, there were no significant changes in temperature for the three groups over time (F1,116 = 1.475, P = 0.23).

Group differences in modification accuracy immediately following training

Observed accuracy prior to training Prior to training no significant differences were found in observed accuracy of modification between the three groups Stage 1 (F2,59 = 0.70, P = 0.50), Stage 2 (F2,59 = 2.11, P = 0.13) & Stage 3 (F2,59 = 1.12, P = 0.33) (see Table 4 for Means & SD). These findings indicate that there were no significant differences in accuracy, demographics or room temperature between the groups prior to training. Different stages of fluid consistency have been used to help people with dysphagia manage fluids

As can be seen in Table 4 and Fig. 2, accuracy increased following all training types but this increase was most marked for the two intervention groups involving Typical Training. Accuracy reduced at 3–10-month follow-up, but not to the original pre-intervention level. The pattern of means and the plot follow the predicted hypotheses regarding greater accuracy immediately post intervention for the Typical Training and Typical Training and TIM tubes groups compared with the group that received Written Guidance alone. To control for any preintervention differences between the groups that may affect the outcome of the analysis, ancova was carried out, with pre-training intervention baseline measurements incorporated as covariates. This is in line with the advice of Vickers (2001), who stated that ancova has the greatest power in analysis of controlled trials over percentage change and change

Table 4 Mean observed accuracy scores across time, training group & stage

Overall

Stage 1

Stage 2

Stage 3

Time

Training

n

Mean

SD

Mean

SD

Mean

SD

Mean

SD

Prior to training (Pre-intervention baseline)

Written Guidance Typical Training Typical Training & TIM tubes Written Guidance Typical Training Typical Training & TIM tubes Written Guidance Typical Training Typical Training & TIM tubes

22 20 20 22 20 20 22 20 20

2.58 2.43 2.08 2.12 1.23 0.77 2.14 1.70 1.21

0.80 1.09 0.98 1.00 0.91 0.48 1.18 0.91 0.91

3.09 2.70 2.75 2.86 1.10 0.50 2.59 1.70 1.25

0.92 1.53 1.02 1.17 0.91 0.51 1.14 1.08 1.45

2.23 1.85 1.50 1.86 0.95 0.30 1.82 1.40 0.90

1.15 0.99 1.28 1.17 0.95 0.47 1.33 0.94 0.97

1.77 1.85 1.40 1.45 0.80 0.55 1.77 1.40 0.95

1.10 1.53 1.47 1.18 0.83 0.51 1.48 1.47 0.83

Post-training

3–10 months post-training

Note: Potential accuracy scores range from 0 (accurate) to 4 (inaccurate), with higher values indicating less accuracy.




Figure 2 Mean observed accuracy of modification of fluids consistency for the three training groups over time (higher values indicate lower accuracy; error bars represent standard deviations).

from pre-intervention baseline. Because of the a priori nature of the specified hypotheses, planned contrasts were used to discern differences between the groups. This first ancova addressed hypotheses 1 and 2 and aimed to determine if there were any differences in fluid modification accuracy between the three groups immediately following training, while controlling for pre-intervention baseline differences. Levene’s test for equality of error variances was not significant (F2,59 = 3.02, P = 0.58). The ancova comparing the mean accuracy scores for all three training groups collected immediately post-training

were compared showed a significant main effect for Training group (F2,59 = 12.488, P < 0.001, r = 0.30). Using Cohen (1988) as a benchmark for effect sizes, this represents a medium effect size (0.25). Planned contrasts revealed that the Training & TIM (Mean difference = -1.223, P < 0.001) and the Typical Training (Mean difference = -0.850, P = 0.001) groups were significantly more accurate at modifying the fluids than the Written Guidance only group. Observed accuracy scores for the two Training groups did not significantly differ from each other (Mean difference = 0.373, P = 0.15) post intervention.




Group differences in modification accuracy at 3–10-month follow-up To determine if there was a difference between training groups in mean observed accuracy score for fluid consistency modification measured at 3–10month follow-up (hypothesis 3), an ancova was used, again with pre-intervention baseline scores entered as covariates. This comparison also proved significant [F2,59 = 3.32, P = 0.02, r = 0.31, representing a medium effect size (Cohen 1988)]. Levene’s test of equality of error variance was non-significant (F2,59 = 2.399, P = 1.00). Planned contrasts revealed that accuracy was significantly greater for the Typical Training and TIM tube group compared with the Written Guidance alone group (Mean difference = -0.754, P = 0.02). The Typical Training group did not significantly differ from the Typical Training and TIM tube group (Mean difference = 0.365, P = 0.25) or Written Guidance group (Mean difference = -0.388, P = 0.21) in their accuracy scores at 3–10-month follow-up.

Percentage change in accuracy To investigate the degree of change for the different training types percentage change was calculated from the group mean accuracy scores (see Table 4). In the Written Guidance only group accuracy increased between the first and second modification assessments by 17.83%, and remained relatively stable at 3–10 months changing by 0.94%. For the Typical Training group accuracy increased by 49.38% following training. At 3–10 months there was a 38.21% drop in accuracy, though accuracy was still 30.04% greater than it was pre-training. Finally, for the Typical Training and TIM tube group accuracy increased by 62.98% following training though this group also saw the greatest percentage drop in accuracy (57.14%). Despite this at 3–10-month follow-up accuracy of fluid modification for this group remained 41.82% better than it was prior to training.

Discussion In this study we aimed to evaluate the efficacy of typical fluid consistency modification training and of the training combined with the use of the TIM

tubes by comparing these two forms of intervention with the provision of written guidance alone posttraining and at 3–10-month follow-up. We investigated four hypotheses and the results were analysed using mean observed accuracy scores across time, intervention group (see Table 4). Accuracy improved following all types of training and then reduced at the 3–10-month follow-up but not to the original pre-intervention level. The increase and subsequent decline were more marked for the two interventions involving training and the greatest observed accuracy was among the Typical Training and TIM tube group. With regard to the specific hypotheses, the results supported hypothesis 1 and partially supported hypothesis 2. We found, in line with hypotheses 1 and 2 that the two types of Training (Typical Training alone & Typical Training and the TIM tubes) resulted in significantly greater accuracy in fluid modification than written guidance alone immediately post-training. The Typical Training and Typical Training plus TIM tube groups did not significantly differ in accuracy immediately post intervention; therefore this aspect of hypothesis 2 was not supported. In line with hypothesis 4, Typical training and using the TIM tubes together did result in greater accuracy post-training and at 3–10 months than Written Guidance and Typical Training alone. However, this increase in accuracy was only statistically significant compared with the Written Guidance alone group so this hypothesis was only partially supported. The difference hypothesised between the TIM tube and Typical Training group and the Typical Training group at 3–10-month follow-up was not found, thus we cannot accept this aspect of hypothesis 4. Though more accurate, the accuracy of the Typical Training group was not significantly greater than that of the Written Guidance group at 3–10-month follow-up and so hypothesis 3 is not supported here. Thus hypothesis 1 was supported, hypotheses 2 and 4 were partially supported and hypothesis 3 was not supported. The non-significant difference between the two groups involving Typical Training may be because of the relatively small sample size for this study. It may also indicate that the TIM tubes may have more utility as a reference tool to aid caregivers at the point of thickening fluids




rather than being a training aid. This warrants further study but would mean that caregivers supporting people with fluid modification needs will need one in the kitchen at all times. There are very few studies looking at dysphagia in people with an intellectual disability and fewer still investigating the efficacy and effectiveness of clinical interventions for this population. This study demonstrates the efficacy of training in improving caregiver accuracy in modifying liquids. This links to previous studies which have recognised the importance of training to enable caregivers to follow dysphagia management plans accurately (Chadwick et al. 2002, 2003, 2006; Crawford et al. 2007). The two groups that received training were significantly more accurate at modifying fluids immediately post-training compared with the Written Guidance group. However, there was a decline in accuracy from post-training to 3–10-month follow-up. In clinical practice fluid modification would be carried out routinely as prescribed for individual patients. However, it was a requirement of the study that caregivers had not worked with people with dysphagia previously. As a result they did not practice modifying fluids during the intervening 3–10 months between the first and second appointments. The decline in accuracy after 3–10 months is not therefore unexpected as there had been no opportunity for knowledge to be embedded through practice. If carers had been using their training daily, one would expect better maintenance of accuracy. However, it may be that training effects wear off over time irrespective of use, this is a question for future research. An effectiveness study would capture these long-term gains in modification accuracy by comparing carers utilising their fluid modification training over time with a control group of carers who are not. Despite the decline in accuracy, the group trained using the TIM tubes were the most accurate at modifying fluids at 3–10-month follow-up being 41.82% more accurate than they were prior to training. This may suggest that use of the TIM tubes could reduce the amount of top-up or repeat training sessions required by caregivers. The utility of the TIM tubes could also be generalised to other client groups requiring fluid modification as part of their health care, including other groups who com-

monly present with dysphagia, for example those who have had a stroke, Parkinson’s disease and older adults. Attempts were made to ensure consistency across trainers and raters, in their ability to modify and to identify specific consistencies, prior to beginning the study. However, during the study no checks of the consistency accuracy of the modelled liquids prepared by the trainers was conducted during the study. This is an oversight considering the finding that professionals are not always consistent in their attempts to modify liquids (Glassburn & Deem 1998), and should be addressed into future effectiveness investigations. The use of a viscometer in future studies would alleviate concerns about professional rater accuracy. At present there are no agreed viscosity measurements for different consistencies of fluid modification. The National Descriptors were developed in response to requests from SLTs and dietitians to provide a common language to describe consistency modification as clearly as possible (BDA/RCSLT 2002). Although food descriptors have been updated, the fluid descriptors have not been updated since 2002 and many practitioners have developed their own terminology in an attempt to clarify their recommendations. In the absence of any other guidelines, the descriptors were used as the basis for the development of the TIM tubes. The National Descriptors were also used within the study as part of the training and the written information given to participants. The TIM tubes provide the first visual representation of different fluid consistencies and this study has shown that the presence of a visual aid can improve modification accuracy. The use of the TIM tubes also paves the way for subsequent avenues of research to promote effective support for people with dysphagia. An effectiveness study in a clinical setting would be beneficial to confirm the effect of training and TIM tubes under more ecologically valid circumstances. This investigation into the efficacy of training should therefore be viewed as a precursor to an effectiveness study to determine the effectiveness of training and the use of TIM tubes in the field. Given the qualified success of training and of the TIM tubes in improving accuracy of modification of fluids subsequent effectiveness




research is warranted. As noted above the TIM tubes could also be used with other client groups, a future effectiveness study could encompass a broader group of care givers supporting people with dysphagia rather than simply being restricted to those supporting people with ID. There are many clinical implications from this study. First, dysphagia training appears to be efficacious in enabling carers to more accurately modify fluids. The combination of training and use of TIM tubes increased the accuracy of consistency modification and, if used more widely, may result in improved compliance with prescribed dysphagia management strategies. Incorrectly thickened drinks can have negative implications for patient health and well-being. Being able to achieve the correct consistency at the first attempt will limit wastage of both time and thickening product. With the presence of a TIM tube, any visiting health professional can monitor the accuracy of the consistency offered to the patient so that inaccuracies can be identified and rectified quickly, thus potentially improving patient safety. By giving the patient ownership of the TIM tube we could also increase their involvement and control over their own care. The TIM tubes may also enable caregivers to more accurately cascade the information about individual management strategies for people requiring modification of liquids which may also help caregivers to more consistently modify fluids accurately over time. This is particularly important in situations where people are reliant on multiple caregivers and where frequent staff changes occur. Improved ability for caregivers to cascade training to one another may also help to reduce the frequency of training required to be provided by professionals.

Conclusions This study has demonstrated the efficacy of typical training and the use of TIM tubes in helping caregivers to accurately modify fluids to accurate safe consistencies. The TIM tubes appear to be a useful tool supporting accurate long term management of dysphagia. Further effectiveness work is needed to confirm their utility.

Acknowledgements The authors would like to thank all the people who took part in this study. We would also like to acknowledge the contribution of Dr Jackie Ord, Research and Development Manager and Annie Hammerton, Research and Development Administrator from North East Essex Learning Disability Service, the Research Institute for Health & Social Change at Manchester Metropolitan University, the managers of registered care homes and day centres for people with ID who helped us identify potential participants for the study, Dr Seamus Hegarty for reading through a draft, and Health Enterprise East for the invaluable support they have all provided throughout the course of this investigation. This project was supported by Department of Health Support For Science funding.

References Aziz S. J. & Cambell-Taylor I. (1999) Neglect and abuse associated with undernutrition in long-term care in North America: causes and solutions. Journal of Elder Abuse & Neglect 10, 91–117. Ball S. L., Panter S. G., Redley M., Proctor C. A., Byrne K., Clare I. C. H. et al. (2012) The extent and nature of need for mealtime support among adults with intellectual disabilities. Journal of Intellectual Disability Research 56, 382–401. BDA/RCSLT (2002) National descriptors for texture modification in adults. Available at: http:// www.bda.uk.com/publications/statements/ NationalDescriptorsTextureModificationAdults.pdf (retrieved 1 October 2010). Chadwick D. D. & Jolliffe J. (2009) A descriptive investigation of dysphagia in adults with intellectual disabilities. Journal of Intellectual Disability Research 53, 29–43. Chadwick D. D., Jolliffe J. & Goldbart J. (2002) Carer knowledge of dysphagia management strategies. International Journal of Language & Communication Disorders 37, 345–58. Chadwick D. D., Jolliffe J. & Goldbart J. (2003) Adherence to eating and drinking guidelines for adults with intellectual disabilities and dysphagia. American Journal on Mental Retardation 108, 202–11. Chadwick D. D., Jolliffe J., Goldbart J. & Burton M. H. (2006) Barriers to caregiver compliance with eating and drinking recommendations for adults with intellectual disabilities and dysphagia. Journal of Applied Research in Intellectual Disabilities 19, 153–63.




Cohen J.(1988) Statistical Power Analysis for the Behavioral Sciences, 2nd edn. Lawrence Erlbaum Associates, Hillsdale, NJ. Crawford H., Leslie P.& Drinnan M.(2007) Compliance with dysphagia recommendations by carers of adults with intellectual impairment. Dysphagia 22, 326–34. Dupont A.& Mortenson P. B.(1990) Avoidable death in a cohort of severely mentally retarded. In: Key Issues in Mental Retardation Research, 1st edn. (ed. W. I. Fraser), pp. 28–34. Routledge, London. Garcia J., Chambers E.& Milander M.(2005) Thickened liquids: practice patterns of speech–language pathologists. American Journal of Speech-Language Pathology 14, 4–13. Glassburn D. L.& Deem J. F.(1998) Thickener Viscosity in dysphagia management: variability among speechlanguage pathologists. Dysphagia 13, 218–22. Griggs C. A., Jones P. M.& Lee R. E.(1989) Videofluoroscopic investigation of feeding disorders of children with multiple handicap. Developmental Medicine and Child Neurology 31, 303–8. Groher M. E.(1987) Bolus management and aspiration pneumonia in patients with pseudobulbar dysphagia. Dysphagia 1, 215–16. Harding C.& Halai V.(2009) Providing dysphagia training for carers of children who have profound and multiple learning disabilities. The British Journal of Developmental Disabilities 55, 33–47. Helfrich-Miller K. R., Rector K. L.& Straka J. A.(1986) Dysphagia: it’s treatment in the profoundly retarded patient with cerebral palsy. Archives of Physical and Medical Rehabilitation 67, 520–5. IBM Corporation (2010) IBM SPSS Statistics for Windows, Ver. 19.0. IBM Corp., Armonk, NY. Kennedy M., McCombie L., Dawes P., McConnnell K. N.& Dunnigan M. G.(1997) Nutritional support for patients with intellectual disabilities and nutrition/

dysphagia disorders in community care. Journal of Intellectual Disability Research 41, 430–6. Kerr M.(1998) Primary health care and health gain for people with a learning disability. Tizard Learning Disability Review 3, 6–14. Langmore S. E.(1999) Issues in the management of dysphagia. Folia Phoniatrica et Logopaedia 51, 220–30. Logemann J. A.(1999) Behavioural management for oropharangeal dysphagia. Folia Phoniatrica et Logopaedia 51, 199–212. MENCAP (2007) Death by Indifference. MENCAP, London. National Patient Safety Agency (2004) Understanding the Patient Safety Issues for People with Learning Disabilities. NPSA, London. Reimers-Neils L., Logemann J. A.& Larson C.(1994) Viscosity effects on EMG activity in normal swallow. Dysphagia 9, 101–6. Royal College of Speech and Language Therapists (2006) Communicating Quality 3. RCSLT, London. Samuels R.& Chadwick D. D.(2006) Predictors of asphyxiation risk in adults with intellectual disability and dysphagia. Journal of Intellectual Disability Research 50, 362–70. Shrout P. E.& Fleiss J. L.(1979) Intraclass correlations: uses in assessing rater reliability. Psychological Bulletin 2, 420–8. Vickers A. J.(2001) The use of percentage change from baseline as an outcome in a controlled trial is statistically inefficient: a simulation study. Biomed Central Medical Research Methodology 1, 6. Whelan K.(2001) Inadequate fluid intakes in dysphagic acute stroke. Clinical Nutrition 20, 423–8. Wood T.(1994) Weight status of a group of adults with learning disabilities. British Journal of Learning Disabilities 22, 97–9.

Accepted 5 November 2012




Appendix 1 Fluid guide from BDA/RCSLT (2002) national descriptors (version 1) Texture

Description of fluid texture

Fluid example

Thin fluid

Still water

Water,Tea, Coffee without milk, diluted squash, spirits, wine Full cream milk, cream liqueurs, Complan, Build up (made to instructions), Nutriment, commercial sip feeds

Naturally thick fluid Product leaves a coating on an empty glass

Thickened fluid Stage 1

Stage 2

Stage 3

Fluid to which a commercial thickener has been added to thicken consistency. • Can be drunk through a straw • Can be drunk from a cup if advised or preferred • Leaves a thin coat on the back of spoon • Cannot be drunk through a straw • Can be drunk from a cup • Leaves a thick coat on the back of spoon • Cannot be drunk through a straw • Cannot be drunk from a cup • Needs to be taken with a spoon
