University of Canberra Research Publication Collection Peer-Reviewed Journal Article for Peter Williams (Adjunct, University of Canberra)

Transcription

1 - University of Canberra Research Publication Collection Faculty of Health Peer-Reviewed Journal Article for Peter Williams (Adjunct, University of Canberra) This is the Accepted version of the following article: Citation: Walton, K., Williams, P. Tapsell, L. (2012). Improving food services for elderly, long-stay patients in Australian hospitals: adding food fortification, assistance with packaging and feeding assistance. Nutrition and Dietetics, 69 (2), doi: /j x Find this item in the UC Research Repository: Copyright: 2012 Walton et al. Nutrition & Dietetics 2012 Dietitians Association of Australia This article has been published in final form at: Version: This is the authors final version of a work that was accepted for publication in Nutrition and Dietetics. Changes resulting from the publishing process, such as editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. University of Canberra, ACT 2601 Australia, Switchboard The University of Canberra is located on Ngunnawal Country. CRICOS number: University of Canberra / University of Canberra College #00212K.

2 Improving food services for elderly, long-stay patients in Australian hospitals: adding food fortification, assistance with packaging and feeding assistance Authors: Karen Walton Smart Foods Centre, School of Health Sciences, University of Wollongong Wollongong, New South Wales 2522, Australia Ph: kwalton@uow.edu.au Associate Professor Peter Williams Smart Foods Centre, School of Health Sciences, University of Wollongong Wollongong, New South Wales 2522, Australia Ph: peter_williams@uow.edu.au Professor Linda Tapsell National Centre of Excellence in Functional Foods, University of Wollongong Wollongong, New South Wales 2522, Australia Ph: linda_tapsell@uow.edu.au Short title: Survey of barriers and priorities to improve nutrition of long stay patients. Keywords: food service, hospital, menu, feeding assistance, food fortification, packaging Word count:??? Address for correspondence: School of Health Sciences University of Wollongong Wollongong, New South Wales 2522, Australia Ph: Fax: kwalton@uow.edu.au

3 Conference presentations: The key findings have been peer reviewed and presented by KW at DAA and ADA conferences, with abstracts being published in the following journals: 1. Walton KL, Williams P & Tapsell LC (2006). Nutr Diet; 63 (Suppl. 1): Walton KL, Williams PG, Tapsell LT (2006). J Am Diet Assoc 106 (Suppl2):A-11

4 ABSTRACT Aim: To highlight barriers and feasible opportunities to enhance nutrition support of long-stay patients in Australian hospitals. Methods: A total of 218 dietitians, nurse unit managers and food service managers from medical and rehabilitation wards of 184 hospitals completed a web based survey about current practices, perceived barriers and priority opportunities to enhance nutrition support. Results: Cook-fresh was the most commonly reported food-service system (50%). Eighty-eight percent still used paper menus and one- or two-week cycles were the most common menu cycle lengths. Lack of choice due to special diet, boredom arising from the length of stay, a lack of feeding assistance, limited variety and inadequate flexibility of food service were the key barriers identified. Food fortification, assistance with packaging, additional feeding assistance by nurses, non nursing feeding assistance and further nutrition assessment were key priorities for improvement. Conclusions: A toolbox of strategies is needed as no one intervention will improve nutrition support of all patients. Further practice-based outcomes and cost-benefit studies are needed to enhance support and advocacy for feasible food service interventions in the future.

5 INTRODUCTION The prevalence of malnutrition in Australian hospitals has been reported to be up to 49% 1-7. There are many possible causes 8-10 and long stay and elderly patients are particularly vulnerable 7, 11. Malnutrition in the hospitalised elderly is preventable and treatable Providing nourishing snack options, food fortification and nutritional supplements can help improve dietary intakes While factors that influence intakes and some interventions have been investigated in other countries, an Australian survey of key stakeholders about barriers and feasible interventions is lacking. This study builds on an earlier focus group study 23 with patients, nurses, food service managers, food service assistants, dietitians and nutrition assistants in NSW hospitals, which asked their views about all aspects of food service for long-stay patients. This revealed five key themes of concern: the food service system; menu variety; preparation to eat/feeding assistance; packaging and serve size. Those findings were used to plan a comprehensive national survey of key stakeholder views, the results of which are presented in this paper. The aims of this study were to: 1. Explore current practices of food service provision in Australian hospitals 2. Determine the key barriers to adequate dietary intakes 3. Prioritise the most practical interventions for ongoing improvements to food service provision.

6 METHODS The 2005 Australian Hospitals Directory listed 748 public and 549 private hospitals, of which 670 were initially deemed eligible to participate in the survey. As longer stay, elderly patients were the focus of this research, we excluded day only, maternity, paediatrics, psychiatric, palliative care, dental, eye and endoscopy hospitals, those with fewer than 20 beds, and those with no medical ward. A letter of invitation was mailed to the food service manager and chief dietitian at each hospital during August Information for nurse unit managers (NUMs) was sent to the director of nursing, along with a letter outlining the study and asking that the information be forwarded to the most appropriate NUM. Three versions of a web-based questionnaire were developed. Each had the same 35 core questions, plus others specific to each professional; eg: food service system (food service managers), nutrition assessment (dietitians) and foods brought in from outside (nurses). Initial versions of each questionnaire were pilot tested with four dietitians, two food service managers and two NUMs, which resulted in some changes to the order of questions and response options. Copies of the questionnaires are available from KW. The key questions related to barriers to dietary intakes and priority interventions. Participants were asked to choose and rank their top 10 barriers from a possible list of 20 (identified in the earlier study 23 ), where number one was the most important barrier and 10 was the least important. Participants then chose their top 10 intervention priorities (in no particular order) from a list of 20, and provided a feasibility rating for each (where 1 = very easy, 2 = somewhat easy, 3 = possible, 4 = somewhat difficult and 5 = very difficult). Fifty-five of the original 670 hospitals were unable to take part in the survey as they had either amalgamated, closed or advised that they didn t routinely see long stay, elderly patients, leaving 615 eligible hospitals. Survey responses were received from 92 dietitians, 58 food service managers (FSMs) and 68 nurse unit managers (NUMs) from 184 hospitals (Table 1). Public hospitals represented 68% of the invited sample and were the workplace of 77% of the participants. Responses represented 42% of hospital beds across Australia. Stakeholder responses to each question were determined and a combined mean feasibility rating was calculated for each priority intervention. The number of responses for each option was recorded and a factor applied to account for the importance of order. For example, the number of times that an option such as limited variety was rated as 1 (most important) was multiplied by 10; the number of times it was rated as 10 (least important) was multiplied by

7 one. Simple addition gave a cumulative total for each barrier for each stakeholder group to be determined, which was the resultant raw, unweighted data. To calculate a combined summary of the 10 most important barriers and priority interventions identified by all the three stakeholder groups, the number of responses from each stakeholder (92 dietitians, 58 FSMs and 68 NUMs) was multiplied by an appropriate weighting factor: 615/92 for dietitians, 615/58 for food service managers and 615/68 for NUMs. These weighted results are reported in Tables 4 and 5. Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS Version 15 for Windows, 2006, SPSS Inc., Chicago, IL). Chi Square analyses or Fisher s Exact Test were used to determine any statistically significant differences between categorical data. The p-value was set at <0.05 and all tests were two-tailed. Ethics approval was obtained from the [relevant] Committee. Information sheets were sent to each participant, and completion of the survey was viewed as consent given.

8 RESULTS Table 2 shows cook fresh was the most common food service system. Fifty percent of respondents were in hospitals with a cook fresh system, 31% had cook chill and 17% used a combination of systems. Findings from a 2001 NSW food service survey are included for comparison 24. Approximately 93% of all hospitals plated meals centrally, while 7% used decentralised plating; there was no difference based on hospital size. Patient meals were mostly delivered by a food service assistant (88%), followed by ward assistants (5%) and nurses (2%), with no significant association between the staff member who distributed meals and the hospital size. Paper menus were used in 88% of hospitals. A combination of paper menus and palm pilots were used in 10.7% of hospitals with >100 beds, while only 3.3% of hospitals with <100 beds used a palm pilot. There was no significant association with hospital size (p=0.402). One- and two-week menu cycles were used in 53.7% of responding hospitals and were commoner in the larger hospitals (65% of those >100beds), while a four-week cycle was more common for the hospitals with 100 beds (40%), although overall there was no significant association with hospital size (p=0.094). Ninety percent of FSMs felt that special dietary, religious and cultural needs were adequately met in their hospitals, significantly higher than the proportions of NUMs (58%), and dietitians (48%) who were satisfied with the adequacy of choices available for patients on special diets (p=0.001). There was no significant difference between the stakeholder views that adequate serving sizes were offered: 87.6% of dietitians, 96% of FSMs and 92.5% of NUMs agreed. There was also agreement that small serves were available when required: 94% of dietitians, 88% of NUMs and 95% of FSMs. Seventy-six percent of dietitians, 67% of NUMs and 78% of FSMs agreed that additional main meal options to those on the standard menu could be offered to long stay patients. Patients on high-protein, high-energy (HPHE) diets appeared to have regular options of HP milk, HP commercial supplements, yoghurt or dairy dessert, or cheese and biscuits. Most nurses (86.5%) reported that additional items were also available on the ward for patients when required. Approximately 73% reported that long stay patients often had food and beverage items brought in for them by friends or relatives, the most popular items being chocolates and lollies, fruit, soft drinks, main meals and desserts.

9 There was agreement that the setting up of patients to access their meals and assisting those unable to feed themselves is primarily the responsibility of nurses. It was estimated that 42% of patients required some form of assistance with food and/or beverage packaging. The mean reported time available for each main meal was 40 minutes. Almost all nurses (98.5%) felt that they had adequate time to assist and feed patients who required it, although they did report the need to divide their time between several patients when there were numerous patients on the ward requiring assistance. Fifty-five percent of dietitians and 59.5% of FSMs reported that some non-nursing feeding assistance was provided, mostly by food service assistants and visitors. Few dietitians (14.5%) or FSMs (21.5%) indicated that trained, nonnursing staff was available to assist with feeding and only one site mentioned a volunteer feeding assistance program. Only sixty percent of dietitians thought patients nutritional needs were adequately assessed, while a significantly greater proportion (87.5%) of nurse unit managers agreed (p=0.001). Recording the dietary intakes of patients not eating well was reported by 94.5% of dietitians, but the usefulness of these records varied: 26.7% rated them very useful, 46.7% useful. The reasons for limitations were that they were not always filled in at the time of the meal so they were not always accurate, or up to date. More than 80% of hospitals offered nutritionally fortified versions of some foods and beverages, particularly soup, mashed potato, milk and juice. The most common fortificants were protein powder, skim milk powder, glucose polymers and cream, with no difference between availability in small or larger hospitals (p=0.726). Table 3 summarises the top 10 barriers to dietary intakes identified by each stakeholder group and the combined totals using the raw and the weighted data. The combined list included the same top 10 barriers, whether the data was unweighted or weighted, with only the order of barriers 2-4 being changed by the weighting. Several other barriers were identified by at least two stakeholder groups. Dietitians and FSMs both noted limited nutritional assessment and communication between staff as barriers, while dietitians and NUMs identified lack of flexibility of food service and taste of food. Table 4 summarises the top 10 priority interventions. The combined list included the same top 10 interventions, with only the order of priorities 2-5 changed by the weighting. In addition to the top 10, both dietitians and FSMs identified additional feeding assistance by nurses, non-nursing feeding assistants available at meals, and additional assistance to set up for meals, while dietitians and NUMs both identified more nourishing between mid-meal snacks, and improved variety of menu options as priorities for change.

10 Table 5 provides the mean feasibility ratings of the interventions. The top ten priorities from Table 4 are indicated in bold. Food fortification, the highest rated intervention, was also rated favourably regarding ease of implementation, closely followed by packaging assistance and provision of more nourishing between meal snacks. However interventions such as additional feeding assistance, improved menu variety, nutrition assessment and greater menu flexibility were perceived as harder to implement, although on average none were rated as very difficult.

11 DISCUSSION A number of the trends in food service practices identified in this survey may be barriers to the provision of adequate patient nutrition. Firstly, while cook fresh was the most common food service system, it seems that the overall usage has decreased (from 53.8% in 2001 to 38% in 2005 in NSW), while cook chill and combination approaches have increased. Cook-chill may limit the range of foods that can be provided on the menu and affect patient meal acceptance 25. Secondly, meal plating was still predominantly centralised, although some sites used bulk delivery carts in some wards. Kelly (1999) reported the benefits of decentralised plating on the intakes of patients in medical wards 26, and several recent studies have supported this delivery option with various patient populations Thirdly more hospitals are offering menus with cycle lengths of one week or less (particularly the larger hospitals). This may be a result of changes to streamline production, and limit human resources and consumables in a bid to save costs 4, 6. This is likely to increase the problems of discontent with menu variety. Lastly, while almost all nurses felt they had sufficient time to assist the feeding of patients, the other two stakeholders saw inadequate assistance as a barrier and felt further feeding support was required. The reported average time available (40 minutes) should be adequate for patients to eat in an unhurried manner. All stakeholders agreed on six key barriers: lack of choice due to a special diet, boredom due to length of stay, limited menu variety, lack of feeding assistance, packaging difficult to open and lack of meal set up assistance. The first three issues relate to a lack of customisation in hospital food services while the last three barriers relate to a lack of time by food service and nursing staff. All three stakeholders agreed on five feasible priority interventions: food fortification, assistance with packaging, nutrition assessment, adequate monitoring of intakes and flexibility of menu choices. Of note were the additional priorities identified by the dietitians and the FSMs (additional feeding assistance by nurses, non-nursing feeding assistants available at meals, and assistance to set up for meals) and by both the dietitians and the NUMs (more nourishing between meal snacks and an improved menu variety). The first three items concern the ward environment, while the last two concern the food service system. This indicates that stakeholders tended to focus on solutions outside their own areas of control.

12 Food fortification was ranked the most feasible intervention (mean 2.3), and more than 83% of hospitals were already using this in some form. This intervention has been evaluated favourably in the literature 15, 16, 30 and is valuable because it can increase the intakes of small eaters. However it is reliant on other strategies such as assistance with feeding and opening packaging to be successful. Improved packaging had a feasibility of 2.5, suggesting it might be relatively easy to address. Forty nine percent of patients surveyed in two Queensland hospitals had difficulty opening packaging 7 and this problem was identified also in our previous study 23. The use of packaged items appears to be increasing, which can adversely affect the visual appeal of the meal, the environment and negatively impact food consumption. Further research is required to investigate what types of packaging patients find easiest to open. The priority of more nourishing snacks (feasibility rating 2.7) reflects literature that indicates that many patients find meals too large and less appealing 7, 31. Items such as chocolate, cheese and biscuits, high energy cakes, and sandwiches could be incorporated as mid-meal snack offerings, and are recommended in the new nutrition standards for NSW hospitals 32. Trials are needed to examine the cost-effectiveness of this strategy compared to the use of more traditional commercial supplements. Nutrition assessment of all patients (mean feasibility 3.5) and monitoring of intakes (3.0) were rated harder to implement, although regarded as a high priorities by dietitians. Three of the identified barriers (lack of choice due to a special diet, boredom due to length of stay and limited variety) could be addressed by improved menu variety (mean feasibility 3.3.). Other research suggests that customisation is an important determinant of overall patient satisfaction with hospital food services 33. Additional feeding assistance by nurses or trained volunteers and additional meal set up assistance were ranked more highly as priorities by the dietitians and FSMs, than the NUMs. Nurses in the survey reported they could assist their patients when required in most instances. This is in contrast to the findings of other researchers who suggest that busy nurses have little time to encourage and assist those who need it 34, 35. These three potential strategies (nutritional assessment, improved menu variety, and feeding assistance) were viewed as more difficult to implement. Implementation of a protected meal times could help in addressing these three issues because meal times would be prioritised and nursing staff would be available to assist with setting up, feeding and monitoring 36. The possible role for trained volunteers in assisting with feeding assistance, opening of packages

13 and monitoring also requires further consideration. Trained volunteers have assumed this role in one aged care ward of a Sydney hospital, with apparent improvements in patients intakes 37. Although the survey response rate was low, it included responses from hospitals with almost half the hospital beds in Australia. There were more responses from dietitians than from FSMs or NUMs, but the weighted results take account of the differing numbers of responses from each stakeholder group. Hospitals smaller than 20 beds were not included, however the responses from 29 hospitals with less than 100 beds provides information on issues likely to affect the smallest sites.

14 CONCLUSION There was significant agreement between stakeholders regarding many key barriers and priority interventions to improve dietary intakes of long stay patients. Limited menu variety, boredom due to length of stay, food packaging and a lack of feeding and set up assistance are particular barriers that warrant further consideration. Priority interventions include the application of food fortification, additional assistance with packaging, meal set up and feeding when required, improved menu variety, more nourishing between meal snacks and an increased use of nutritional assessment. Clearly there are numerous barriers to adequate intakes by hospital patients, and a toolbox of interventions is required as no one approach will fix all the problems. Further cost-benefit and outcomes studies will enhance support and advocacy for feasible food service interventions in the future. ACKNOWLEDGEMENTS Funding was provided by the Smart Foods Centre, University of Wollongong. Our thanks to Greg Abernethy (web designer), Prof David Steele (statistical advice) and three dietetics research students (Christine Wirtz, Natasha Ainsworth and Lauren Lynch) who assisted with the 29 data analysis.

15 REFERENCES 1. Beck, E; Patch, C; Milosavljevic, M, et al. Implementation of malnutrition screening and assessment of dietitians: malnutrition exists in acute and rehabilitation settings. Aust J Nutr Diet. 2001; 58: Middleton, M; Nazarenko, G; Nivison-Smith, I; Smerdley, P. Prevalence of malnutrition and 12-month incidence of mortality in two Sydney teaching hospitals. Int Med J. 2001; 31: Visvanathan, R; Penhall, R; Chapman, I. Nutritional screening of older people in a subacute facility in Australia and its relation to discharge outcomes. Age & Ageing. 2004; 33: Lazarus, C; Hamlyn, J. Prevalence and documentation of malnutrition in hospitals: a case study in a large private hospital setting. Nutr Diet. 2005; 62: Neumann, S; Miller, M; Daniels, L; Crotty, M. Nutritional status and clinical outcomes of older patients in rehabilitation. J Hum Nutr Diet. 2005; 18: Vivanti, A; Banks, M. Length of stay patterns for patients of an acute care hospital: implications for nutrition and food services. Aust Health Rev. 2007; 31: Vivanti, A; Banks, M; Aliakbari, J, et al. Meal and food preferences of nutritionally at-risk patients admitted to two Australian teaching hospitals. Nutr & Diet. 2008; 65: Garrow, J. Starvation in hospitals. BMJ. 1994; 308: Kowanko, E; Simon, S; Wood, J. Energy and nutrient intake of patients in acute care. J Clin Nurs. 2001; 10: Lipski, P. Improving delivery of food services for acute geriatric inpatients: a quality assurance project (letter to the editor). Austral J Ageing. 2003; 22: Walton, K; Williams, P; Tapsell, L; Batterham, M. Rehabilitation inpatients are not meeting their energy and protein needs:. e-spen Eur e-j Clin Nutr Metab. 2007; 2: e120-e Dickinson, A; Welch, C; Ager, L; Costar, A. Hospital mealtimes: action research for change? Proc Nutr Soc. 2005; 64: Gazzotti, C; Arnaud-Battendier, F; Parello, M, et al. Prevention of malnutrition in older people during and after hospitalisation: results from a randomised controlled clinical trial. Age & Ageing. 2003; 32: O'Flynn, J; Peake, H; Hickson, M; Foster, D; Frost, G. The prevalence of malnutrition in hospitals can be reduced: results from three consecutive cross-sectional studies. Clin Nutr. 2005; 24: Gall, M; Grimble, G; Reeve, N; Thomas, S. Effect of providing fortified meals and between-meal snacks on energy and protein intake of hospital patients. Clin Nutr. 1998; 17: Barton, A; Beigg, C; Macdonald, I; Allison, S. A recipe for improving food intakes in elderly hospitalised patients. Clin Nutr. 2000; 19:

16 17. Kondrup, J; Johansen, N; Plum, L, et al. Incidence of nutritional risk and causes of inadequate nutritional care in hospitals. Clin Nutr. 2002; 21: Larsson, J; Unosson, M; Ek, A, et al. Effect of dietary supplement on nutritional status and clinical outcome in 501 geriatric patients - a randomised study. Clin Nutr. 1990; 9: Potter, J; Landhorne, P; Roberts, M. Routine protein energy supplementation in adults: systematic review. BMJ. 1998; 317: Nolan, A. Audit of supplement use on care of the elderly and rehabilitation wards. J Hum Nutr Diet. 1999; 12: Roberts, M; Potter, J; McColl, J; Reilly, J. Can prescription of sip-feed supplements increase energy intake in hospitalised older people with medical problems? Br J Nutr. 2003; 90: Schenker, S. Briefing Paper. Undernutrition in the UK. Nutr Bull. 2003; 28: Walton, K; Williams, P; Tapsell, L. What do stakeholders consider the key issues affecting the quality of food service provision for long stay patients? J Foodserv. 2006; 17: Mibey, R; Williams, P. Food services trends in New South Wales hospitals, Food Serv Technol. 2002; 2: McClelland, A; Williams, P. Trend to better nutrition on Australian hospital menus and the impact of cook-chill food service systems. J Hum Nutr Diet. 2003; 16: Kelly, L. Audit of food wastage: differences between a plated and bulk system of meal provision. J Hum Nutr Diet. 1999; 12: Marson, H; McErlain, L; Ainsworth, P. The implications of food wastage on a renal ward. Br Food J. 2003; 105: Carr, E; Mitchell, J. An assessment of factors affecting consumption of entree items by hospital patients. J Nurs Stud. 1991; 28: Pietersma, P; Follett-Bick, S; Wilkinson, B, et al. A bedside food cart as an alternative food service for acute and palliative oncology patients. Support Care Cancer. 2003; 11: Kennewell, S; Kokkinakos, M. Thick, cheap and easy: fortifying texture-modified meals with infant cereal. Nutr & Diet. 2007; 64: Barton, A; Beigg, C; Macdonald, I; Allison, S. High food wastage and low nutritional intakes in hospital patients. Clin Nutr. 2000; 19: NSW Health Greater Metropolitan Clinical Taskforce. Nutrition Standards for Adult Inpatients in NSW Hospitals Available online: [Accessed on 14 March 2011] 33. Dube, L; Trudeau, E; Belanger, M. Determining the complexity of patient satisfaction with foodservices. J Am Diet Assoc. 1994; 94: Xia, C; McCutcheon, H. Mealtime in hospital - who does what? J Clin Nurs. 2006; 15:

17 35. Kowanko, I; Simon, S; Wood, J. Nutritional care of the patients: nurses' knowledge and attitudes in acute care settings. J Clin Nurs. 1999; 8: British Dietetic Association. Delivery nutritional care through food and beverage services Available online: ge_services.pdf. [Accessed on 13 March 2011] 37. Walton, K; Williams, P; Bracks, J, et al. A volunteer feeding program can improve dietary intakes of elderly patients - a pilot study. Appetite. 2008; 51:

18 Table 1: Stakeholder types and hospital size Bed numbers Dietitian Food Service Manager Nurse Unit Manager Total 100 beds % 52% 41% 38% Diet 32% FSM 30% NUM >100 beds % 48% 59% 45% Diet 23% FSM 32% NUM Total

19 Table 2: The food service system operated by hospital size compared with the results from a 2001 NSW survey Food service system This national survey conducted in 2005 NSW survey in Percentage of hospitals with 100 beds Percentage of hospitals with >100 beds Total NSW total % of hospitals with <100 beds % of hospitals with 100 beds Total (n=29) (n=29) (n=58) (n=21) (n=47) (n=46) (n=93) Cook fresh * Cook chill Frozen N/A N/A N/A Combination Total Legend: * A Chi-square analysis between hospital size ( 100 or >100 beds) and cook-fresh vs all other systems (cook chill, frozen and combination) had a statistically significant P value of

20 Table 3: The top 10 barriers (raw, unweighted and weighted data) for each stakeholder group and combined stakeholder totals Barrier Number Dietitians 1 Lack of feeding assistance Raw Wt. Food Service Managers Boredom due to length of stay Raw Wt. Nurse Unit Managers Raw Wt. Combined Stakeholder Totals Limited variety Lack of choice due to special diet Raw Wt Lack of flexibility of food service 3 Lack of choice due to special diet Lack of feeding assistance Lack of meal set up assistance Lack of choice due to special diet Packaging difficult to open Boredom due to length of stay Lack of feeding assistance Boredom due to length of stay Communication between staff & patients 5 Limited variety Limited nutritional assessment 6 Lack of meal set Lack of choice due up assistance to special diet 7 Packaging difficult Limited monitoring to open of intakes 8 Limited nutritional assessment Communication between staff 9 Taste of food Packaging difficult to open 10 Communication between staff Lack of flexibility of food service Limited variety Boredom due to Lack of flexibility of length of stay food service Taste of food Packaging difficult Lack of culturally appropriate food Lack of meal set up assistance Lack of feeding assistance Limited variety Temperature of food to open Lack of meal set up assistance Limited nutritional assessment Taste of food Communication between staff & patients Legend: Raw are the unweighted totals for each barrier, while Wt. refers to the values after the weightings are applied. A complete description of this is available in the Methods.

21 Table 4: The top 10 priorities (raw unweighted and weighted data) for each stakeholder group and combined stakeholder totals Priority Number Dietitians Raw Wt. Food Service Managers 1 Food fortification Additional feeding assistance by nurses 2 Additional feeding assistance by nurses 3 Non nursing feeding assistant available at meals 4 Assistance with packaging 5 Nutrition assessment of all patients 6 Adequate flexibility of menu choices 7 More nourishing between meal snacks 8 Adequate monitoring intakes 9 Improved variety of menu options 10 Additional assistance to set up for meals Adequate monitoring intakes Nutrition assessment of all patients Non nursing feeding assistant Raw Wt. Nurse Unit Managers Assistance with packaging Improved variety of menu options Adequate flexibility of menu choices available at meal Food fortification More nourishing between meal snacks Assistance with packaging Additional assistance to set up for meals Adequate flexibility of menu choices Raw Wt. Combined Raw Wt Food fortification Assistance with packaging Additional feeding assistance by nurses Food fortification Non nursing feeding assistant available at meal Improved taste of meals Nutrition assessment of all patients Additional culturally appropriate meals Serve size options Adequate monitoring intakes More information Additional foods about food from home choices Nutrition assessment of all patients Adequate monitoring intakes Adequate flexibility of menu choices More nourishing between meal snacks Improved variety of menu options Additional assistance to set up for meals Legend: Raw are the unweighted totals for each priority, while Wt. refers to the values after the weightings are applied. A complete description of this is available in the Methods.

22 Table 5: Feasibility rating for each priority intervention and stakeholder group Priority intervention Serve size options (small offered) Mean feasibility rating Dietitian rating FSM rating NUM rating Food fortification More information on menu choices Packaging assistance Adequate time allowed Improved communication staff & patients More nourishing between meal snacks Additional foods brought from home Improved communication between staff Improved layout and appearance of meal tray Adequate monitoring of intakes Additional assistance to set up for meals Additional feeding assistance by nurses Improved taste Additional culturally appropriate dishes Dining room Adequate flexibility of menu choices Improved variety of menu options Nutrition assessment of all patients Non nursing assistant available at meals Legend: 1=very easy, 2=somewhat easy, 3=possible, 4=somewhat difficult, 5=very difficult The bold text indicates the top ten cumulative priorities for the three stakeholders 21