Table 3.

Mealtimes.

Study, country	Environmental intervention	Sample/setting	Research design	Main findings
Josephsson et al. (1995), Sweden	Tailored individual adaptations: ADL selected by individual (prepare drink/set table) supported by environmental solution, e.g. removing clutter, using labels, table with model process	Four people with dementia at psychogeriatric day care unit within hospital	Baseline, post-intervention video-recording assessment using assessment of motor/process skills and frequency of support noted	Three out of four of the participants demonstrated significant gains in the ability to perform the daily activities related to meal preparation
Starkhammar and Nygård (2008), Sweden	Assistive technology: Device attached to stove, automatically activated when turned on, switches stove off automatically, used as heat detector, alarm system	Nine people with dementia or memory impairment and five family carers Private home	Qualitative interviews following trials with stove timer device, observed in use (50 min–1 h 50 min)	Some irritated/frustrated, e.g. system shut off before finished, enabled cooking for some who’d given up, some problems with learning how to use it
Koss and Gilmore (1998), USA	Lighting and contrast: Increased light intensity and visual contrast for evening meal	Thirteen people with dementia High functioning dementia unit	Baseline, intervention and post-intervention measures of amount eaten and agitated behaviours over 21-day periods	Food intake highest in intervention condition Agitation decreased in high light/visual contrast
Dunne et al. (2004), USA	Lighting and contrast: Baseline white plates, white cups and stainless steel flatware Experimental high contrast red plates, red cups and red flatware	Nine men with advanced dementia, (MMSE M = 2.9) Care unit at ENRM Veteran Affairs Medical Centre	Food/liquid intake, ABA design. One year later with ‘blue’ high contrast condition	Significant increase for food and liquid in high contrast intervention (P < 0.001) Follow up showed significant for ‘blue’ suggesting high contrast has effect
Brush et al. (2002), USA	Lighting and contrast: Enhanced lighting in dining room and table setting contrast	Eleven nursing home. Fourteen assisted living facility people with dementia (MMSE score unknown) Two live in facilities	Three-day calorie count (all three meals) and changes in resident behaviour (meal assistance screening tool) and communication outcome measure pre/post changes	Calorie count: NH, +1000 calorie average increase (p < .16); AL, (p < .01) increase Communication: Significant increase at NH (p < .05), AL not significant (p < .115) Behaviour: No significant differences apart from for distractibility (p < .05)
McDaniel et al. (2001), USA	Quality of the environment: Extended care (EC) dining room: lower noise, higher lighting versus refurbished Alzheimer’s Unit (AU): higher noise, lower lighting, relaxing music, no television	Sixteen people with dementia One residential facility	Five-day nutritional analysis for breakfast and lunch in the two different dining environments	Mean total intake of calories and protein was higher in AU but not sig. Fluids at breakfast higher in AU over five days (p < 0.02). Significantly higher intake on day 3 (p < .5) and day 4 (p < .2)
Slaughter and Morgan (2012), Canada	Quality of the environment: Measured quality of dining environment: orientation, safety, privacy, stimulation, support, opportunities for control, familiar objects, facilitation of social contact	One hundred and twenty people with middle-stage dementia (assessed by global deterioration scale) Fifteen nursing homes	Observed residents abilities to walk to the dining room and to feed themselves	Environmental features that supported functional eating (e.g. finger foods) (p = .01), personal control (p = .033) and better regulation of stimulation (p = .027) reduced hazard of eating disability
Reed et al. (2005), USA	Quality of the environment: Compared dining environments in residential care, assisted living (less institutional facilities) and nursing home environments	Four hundred and seven people with mild–moderate–severe dementia Forty-five assisted living facilities, US	Structured mealtime observation – observing up to five residents during single meal, looked at amount of food/fluid consumed during single meal and alertness, utensil use, etc.	Lower food and fluid intake in nursing home environments (p < 0.05) RC/AL less likely to receive treatment for eating difficulty and less physical difficulties
Perivolaris et al. (2006), Canada	Quality of the environment: Renovated dining area to more homelike including fireplace, bright design, smaller room, aromas of food, menu board, removed tray service style, music, staff education	Eleven people with moderate–severe dementia (MMSE) Two memory support units in long-term care facility	Baseline, six weeks after environmental renovation and further measurement at 12 weeks with staff education. Measured calorific intake, focus groups with staff, satisfaction measure with residents	Significantly more food consumed after environmental intervention (p = 0.05) and further effects with staff training (p = 0.060) Staff feedback suggested residents more relaxed, sociable in new environment
Edwards and Beck (2013), USA	Environmental ambiance: Introduced a large aquarium into the dining area, each with light background and eight large fish	Seventy people mostly with severe dementia (MMSE, M = 5.57) Three specialist dementia units	Body weight and food/fluid intake were weighed at baseline for two weeks, two weeks when aquarium first introduced and then once a week for six weeks with aquarium	Significant increase in food intake with aquarium (P < 0.000), increasing trend for following six weeks, significant increase in resident body weight from start to end of study (P < 0.000)
Thomas and Smith (2009), USA	Environmental ambiance: Music during dining based on music preferences indicated by family members	Twelve people with middle stage dementia (global deterioration scale) Fourteen bed Alz unit	A–B–A design, for eight weeks, observed for 24 meals, visual monitoring by dietician for food intake, calorie intake measured	Twenty per cent more calories consumed when familiar music was played compared to no music. Anecdotal evidence of enjoyment of music: socially engaged, stayed in dining area longer
Desai et al. (2007), Canada	Choice: Traditional institutional setting with food delivered on tray versus newer, homelike environment with cafeteria style waitress service	Twenty-three (traditional facility), 26 (new facility), diagnosis of probable Alzheimer’s Academic Nursing Home	Twenty-one day energy and macronutrient intakes measured and behaviour measured using London psychogeriatric rating scale	Higher 24 h P < 0.001 and dinner P < 0.001 energy intakes in new facility due to greater carbohydrate intake More energy, carbohydrate and protein intake for residents with low BMI (p < 0.05) at new facility compared with higher No significant changes in behaviour
Altus et al. (2002), USA	Choice: Pre-prepared plates versus family style help yourself communal serving dishes, and further condition with nurse training to praise and prompt	Five people with moderate–severe dementia (MMSE) Dementia care unit	Observation/ratings by nurse measuring participation/communication and weight following intervention	Participation increase from 10 to 24% with family style. Communication increase from 5.5 to 10.6%. Further increase with family + training of 65% (participation) and 17.9% (communication). 3/5 gained weight
Melin and Götestam (1981), Sweden	Choice: Meal placed on table and patients able to serve themselves. Compared with providing meals on tray in chairs for patients in corridor	Twenty-one, mixed sample (19 dementia, two schizophrenia) Psychogeriatric ward	Communication: Observer recorded whether or not a given patient made contact with anyone else. Eating behaviour: Use of utensils, glass and napkins, each observed for 15 s, four times	Significant increase in communication in experimental group (p < 0.01). Improvement in eating behaviour (p < 0.01) in experimental
Namazi and Johnson (1992b), USA	Choice: Placing food and snacks in accessible area on kitchen surface with domestic style versus glass door refrigerators	Twenty-two people diagnosed with probable AD; 13 in early/mid, three severe, seven unknown (clinical dementia rating score) Dementia facility	Observation of opening fridge, taking snacks, requesting snacks, requesting assistance	Visible access to fridge didn’t affect independent snacking. But poor methodological design (e.g. red tape added to door handle in domestic fridge condition to facilitate opening)

ADL: Activities of daily living; MMSE: Mini Mental Status Exam.