Mealtime support for adults with intellectual disabilities: Understanding an everyday activity

Background: Mealtime support has a direct bearing on the diet-related health of men and women with intellectual disabilities as well as opportunities for expressing dietary preferences. Method: Semi-structured interviews with a sample of direct support staff providing mealtime support to adults with intellectual disabilities. Results: When managing tensions between a person's dietary preferences and en suring safe and adequate nutrition and hydration, direct support staff are sensitive to a wide range of factors. These include the following: clinical advice; service users’ rights to choose; their (in)capacity to weigh up risks; how service users communicate; the constituents of a healthy diet; and a duty to protect service users' health. Conclusions: Those responsible for setting standards and regulating the care practices need to look beyond too simple ideas of choice and safety to recognize ways in which providing support at mealtimes is a complex activity with serious conse-quences for people's health and well-being.


| INTRODUC TI ON
As more adults live longer with age-related infirmities and lifelong disabilities, public concern over the quality of the care and support they receive has grown (Nelson, 2015). This paper explores the constituents of competent care through a study of the mealtime assistance that direct support staff (DSS) provide to men and women with intellectual disabilities. Mealtimes provide a powerful focal point for considering what constitutes competent care; for as well as ensuring safe and adequate nutrition and hydration, DSS should also facilitate opportunities for the expression of dietary preferences (Department of Health 2001). For men and women with intellectual disabilities, these considerations are of particular importance.
Swallowing problems (dysphagia) are prevalent in this population (Ball et al., 2012), while mealtimes may be among the few occasions where people have an opportunity to express and enjoy their preferences (Schwier & Stewart, 2005). For DSS providing mealtime assistance, these two imperatives, safety and autonomy, can, however, come into conflict, as when, for example, a person wishes to eat something that may cause them to choke or aspirate. The practical actions taken to address such dilemmas will have a direct bearing on the quality of the care and support being provided. Others' concerns over safety often result in people with intellectual disabilities having their autonomy restricted. Yet, with respect to eating and drinking, aspiration-related respiratory illnesses are the leading cause of premature deaths (Tyrer & McGrother, 2009) and of avoidable admissions to hospital (Glover & Evison, 2013). Moreover, rates of obesity and life-threatening low weight are much higher in this population controlling or restricting the diets of adults with intellectual disabilities is seen as infantilizing them (Davies, 2007). In this context, the quality of the mealtime assistance provided by DSS will have a direct bearing on both the diet-related health of men and women with intellectual disabilities (Adolfsson, Fjellström, Lewin, & Mattsson Sydner, 2012;Ptomey, Goetz, Lee, Donnelly, & Sullivan, 2013) and their well-being (Schwier & Stewart, 2005).

| ME THOD
To examine the mealtime assistance provided by DSS and to begin to consider how to assess the quality of this assistance, a sample of DSS working in the East of England (the county of Cambridgeshire and the parliamentary constituency of North East Essex) were recruited. This area is diverse, comprising variations in affluence and deprivation as well as urban and rural settings. There are an estimated 174 providers in community residential services in this region, for a total population of 5,383 persons known to specialist services, of whom 142 took part in an exploratory, prospective cohort study of adults identified as needing assistance with eating and drinking (Perez et al., 2017). DSS were eligible for inclusion if they had provided mealtime support to an adult with an intellectual disability who had taken part in an earlier study to estimate the extent and nature of such support (Ball et al., 2012). Using a consent process approved by the National Research Ethics Service (www.hra.nhs.uk), opportunity sampling, dividing contacts evenly between the county of Cambridgeshire and the parliamentary constituency of North East Essex, led to the recruitment of 19 DSS (14 females) working in 18 different community residential services. No claims can be made about the representativeness of this sample with respect to sex or experiences, in relation to the wider population of DSS. The topic guide for the semi-structured interviews explored the practicalities of ensuring safe and adequate nutrition and hydration while also respecting a person's dietary preferences. More specifically, prompts were used to ask respondents to describe the assistance a particular individual needed; the risks of not providing that assistance; whether that assistance affected the person's enjoyment of their food; and what opportunities the person had to choose their own food. The interviews were conducted in the residential services where the DSS worked, lasted between 40 and 80 minutes and were audio-recorded. These recordings were transcribed verbatim. An initial thematic coding (Braun & Clarke, 2006) of these data using NVivo revealed 36 separate incidents (across 19 interviews), in which DSS described responding to tensions between a service user's dietary preferences and interventions designed to safeguard their health.
These incidents fell into three broad themes that, not surprisingly, reflect the areas of interest reported in the research literature.
Namely, concerns over aspiration and choking (Glover & Evison, 2013;Tyrer & McGrother, 2009), undernutrition (Emerson et al., 2010) and dietary choices thought likely to cause an unhealthy gain in weight (Adolfsson, et al., 2012). From these 36 incidents, six excerpts have been chosen for detailed examination, two examples from each of the three accounts of concerns. 1 This choice is necessarily idiosyn-1 Three instances of persons with behavioural/psychological problems that disrupted orderly eating and drinking were reported, but these have been excluded from the analysis presented because they did not involve a tension between protecting a person's health and honouring their wishes.

| 3
Published for the British Institute of Learning Disabilities REDLEY cratic but was based on a judgement about the chosen excerpts being "fuller" than others, in the sense that the DSS concerned describe, even if succinctly, not only their actions but also their thinking behind those actions. Consequently, the analysis presented below is not comprehensive in terms of detailing all predictable events; rather, it serves to demonstrate some of the ways in which DSS are approaching tensions between promoting safety and autonomy at mealtimes.
That said, there are long-standing concerns over the use of interview data as a reliable means for accessing a respondent's actions and associated thoughts (Cicourel, 1964). This is because what is reported in research interviews is rooted in the contingencies of an interaction between interviewer and interviewee (Mishler, 1991).
One way to address this is to see spoken utterances as actions, in which speakers pursue particular strategic purposes (Potter, 1996).
So, while interview data do not offer a window onto past events and thoughts, it can tell us something about how these DSS are defining "good practice," assuming, that is, that these respondents used the interviews as an opportunity to project an image of themselves as "competent" support workers (Silverman, 2001). It is this rhetorically constructed image of care practices that is used to explore and consider the constituents of good care.
Each excerpt is introduced with a description of the speaker: employment status (support worker, senior support worker or house manager) and the total number of years spent supporting people with intellectual disabilities. In addition, where a specific service user is discussed, a brief description of that person is included: age; severity of intellectual disability (mild, moderate or severe); and level of communication (verbal; limited communication; no communication). Pseudonyms are used throughout.

| FINDING S
The six chosen data excerpts are presented under the three headings identified above: risks of choking and aspiration; undernutrition through refusals to eat; and unhealthy choices.

| Risks of choking and aspiration
Mandy, with 2 years' experience of support work, reports that Mr Graham Davies "loves his toast" notwithstanding the risk of choking.
Hearing about this risk prompts the interviewer to enquire about the difficulties of denying him something he loves (see Box 1, Excerpt 1).
Mandy responds by describing what she styles as a "compromise." The advice from the SLT is formulated as discretionary ("try") rather than categorical ("don't"); the sense of loss resulting from not eating toast is graded as comparable to being denied chocolate; and Mr Davies' right to eat what he likes is asserted, even while acknowledging that he "often chokes." Mandy's account combines respecting service user choice with a responsibility to protect Mr Davies from harm, although it is debatable whether her "compromise" would win widespread approval, given that Mr Davies is reportedly choking on his toast. Excerpt 2 (see Box 1) concerns Rosa, a senior support worker with 4 years' experience, who has also received advice from a objections prior to determining her capacity: the first principle of the Act is that capacity be assumed unless it is established otherwise.
Excerpts 1 and 2 differ with respect to the actions that Mandy and Rosa take following advice from a SLT. Mandy's "compromise" ensures that Mr Davies continues to have the toast he purportedly avoid offering a choice and then "having to say"no" to those kinds of things [unhealthy foods]," she and colleagues offer Ms Lee a choice between several preselected meals. Debbie styles this an "informed choice." Although contrary to how this term is usually understood, it is Debbie and her co-workers, not Ms Lee, who are appraised of the risks and benefits of choosing some foods over others.  Bruce's validation of these observations, Daniel states that Mr Bruce sees takeaways as a "treat," that he has the capacity to "appreciate that that's not ideal [takeaway food] every single day", and that "he likes too many other foods." Showing that she understands the implications of Daniel's remarks, the interviewer says, "so, it's not a real problem [choosing to eat takeaways every day]." Daniel responds by again seeking Mr Bruce's endorsement, "I think we can safely say that would never be an issue, hey Jono," and receives the reply: "No."

Box 2 Refusals to Eat and the Risk of Undernutrition
Although Daniel invokes a "duty of care," with the implication that he might have to place limits on what Mr Bruce can choose, he neither describes the attributes of that duty nor its practical application.
Rather, he makes a number of observations as to why Mr Bruce is unlikely to choose to eat takeaways every day. In Excerpts 5 and 6, the respondents confront the possibility of service users making unhealthy dietary choices. For Debbie, this entails restricting Ms Lee to choosing from a range of preselected "healthy options." Daniel, by contrast, constructs an image of Mr Bruce as someone who is unlikely to make consistency unhealthy choices. So, while it is Daniel who cites a duty of care, he, unlike Debbie, does not appear to be exercising that duty.
What, more generally, can these six excerpts tell us about how DSS might be defining good practice with respect to mealtime support?

| D ISCUSS I ON
These six excerpts, while certainly not exhaustive of the phenomenon, illustrate ways in which DSS understand competent care when a person's wishes and choices conflict with the imperative to provide safe and adequate nutrition and hydration. These excerpts are seen to focus on three specific risks: choking and aspiration; undernutrition; and unhealthy eating. The excerpts demonstrate DSS are invoking a range of contextual factors: advice from clinicians (Excerpts 1 and 2); the right of service users to choose (Excerpts 1, 2, 3, 5 and 6); a service user's preferences (Excerpts 1, 3 and 6); the loss a service user might experience if their wishes are not respected (Excerpts 1 and 5); the (in)capacity of a service user to weigh up potential risks (Excerpts 2 and 6); how a service user communicates choice (Excerpt 3); the importance of ensuring a service user eats (Excerpts 3 and 4); the constituents of a healthy diet (Excerpts 5 and 6); and a responsibility or duty to protect a service user's health (Excerpts 1, 2 and 6) that might involve respecting a person's autonomy (Excerpt 1) or overriding an expressed preference in order to ensure a person's safety (Excerpt 2). Some courses of action seem to promote safe and adequate nutrition and hydration over choice (Excerpts 2 and 5), while others seem potentially riskier (Excerpts 1 and 3); others to balance competing imperatives (Excerpts 1 and 6) or sidestep the issue of choice altogether (Excerpt 4). The excerpts also illustrate how providing mealtime assistance can encompass a wide range of activities, such as supervising a service user in case of choking (Excerpt 1); denying a person something they want (Excerpt 2); chopping up and blending food (Excerpt 3); repeatedly presenting and re-presenting a meal until it is eaten (Excerpt 4); preparing three or four different meals (Excerpt 5); and offering dietary advice (Excerpt 6). This variation in response, to when a person's wishes and choices conflict with the provision of safe and adequate nutrition and hydration,indicates there is no predetermined, or prescribed, response when managing tensions between protecting a person from harm and respecting their autonomy. As such, DSS are free to exercise a degree of creativity or imagination in how they understand and react to these situations. This is evidenced by the "compromise," which allows Mr Davies to eat his favourite food despite the risks this entails (Excerpt 1); the Mr Bruce: Yeah, I would.

Box 3 Unhealthy Choices
Daniel: He likes it as a treat, as something to look forward to, but fortunately I think he's got the capacity to appreciate that that's not ideal every single day, and he likes too many other foods.
Inter: So, it's not a real problem.
Daniel: No. I don't … I think we can safely say that would never be an issue, hey Jono.

| CON CLUS ION
SLTs, managers and supervisors in residential services, along with the bodies responsible for setting standards and regulating care practices, need to look beyond simple ideas of choice and safety.
Mealtime support is a complex responsibility. DSS are engaged in finding workable to solutions to everyday dilemmas, a fact that needs to be recognised in the development of models of care and, in the training received by DSS. Ideally, this training should address how best to enable the people being supported to develop their capabilities (Gawande, 2014), and how responsibility for any risk might be shared (Mol, 2006). For then, and perhaps only then, can services begin to address the incidence of diet-related ill-health among men and women (Tyrer & McGrother, 2009;Glover & Evison, 2013) while also furnishing men and woman with intellectual disability with opportunities for self-expression.