Non-Pharmacological Interventions for Feeding and Eating Disorders in Persons with Dementia: Systematic Review and Evidence Summary

Abstract

Background:

Feeding and eating disorders related to cognitive and psycho-behavioral symptoms are strongly associated with health status in persons with dementia (PWD). Non-pharmacological interventions have been the priority selection to address this significant issue. However, the direct targets of non-pharmacological interventions are unclear and there is no consistent evidence of recommendations on the intervention of different dementia stages and the settings of intervention practice.

Objective:

To provide caregivers with a set of self-help non-pharmacological interventions for feeding and eating disorders in PWD.

Methods:

Based on the process of evidence summary, a systematic literature search was performed on dementia websites and seven databases. Two researchers screened the studies independently and appraise the quality. The evidence was graded by Joanna Briggs Institute Grades of Recommendation.

Results:

Twenty-eight articles were included. Twenty-three non-pharmacological intervention recommendations were categorized into six themes containing oral nutritional supplementation, assistance with eating and drinking, person-centered mealtime care, environmental modification, education or training, and multi-component intervention. These interventions corresponded to three direct targets including improving engagement, making up for loss ability, and increasing food intake directly. They were applied to different stages of dementia and most interventions were targeted at PWD in long-term care institutions.

Conclusion:

This article summarized the direct targets and the specific implementation of recommendations at different stages of dementia to provide caregivers with self-help non-pharmacological interventions. The practice of recommendations was more applicable to institutionalized PWD. When applied to PWD at home, caregivers need to identify the specific feeding and eating conditions at different stages and adopted the interventions in conjunction with the wishes of the PWD and professional advice.

Keywords

Alzheimer’s disease caregivers dementia feeding and eating disorders non-pharmacological intervention

INTRODUCTION

Dementia has become a global public health problem. According to the Alzheimer’s Disease International statistics in 2020, 50 million people worldwide suffered from dementia. The total number of persons with dementia (PWD) is projected to reach 152 million in 2050 [1]. As dementia progresses, feeding and eating disorders will occur, and about 50% of PWD will have such difficulties within 8 years of disease [2]. Feeding and eating disorders, including feeding, eating and meal behaviors, are caused by a variety of factors, such as loss of physical skills (e.g., impaired motor control, nerve damage, altered neck posture, and discoordination between oral and pharyngeal) that lead to dysphagia and inability to get food into the mouth, resulting in adverse outcomes such as aspiration pneumonia and pulmonary complications [3, 4]. Cognitive decline is a core symptom and behavioral and psychological symptoms of dementia (BPSD) are fundamental characteristics of disease [5]. Feeding and eating disorders due to cognition impairment and BPSD are unique to PWD and are more likely to appear early in the disease and worsen with disease progression. The definition of feeding and eating disorders associated with cognition and BPSD has not reached unity, and Edahiro et al. [6] attempted to summarize the eating behaviors associated with BPSD (E-BPSD), including difficulty in beginning a meal, difficulty in proper use of utensils, difficulty in recognizing the total amount of food provided, difficulty in maintaining proper attention and alertness while eating, etc. In addition, Takada et al. [7] summarized the signs and symptoms that accompany eating in PWD, including behaving violently to the caregiver, wandering about even though the meal is not finished and eating meals of other people, etc.

Feeding and eating disorders associated with cognition and BPSD also represent a significant momentous impairment affecting patients’ and caregivers’ lives. Eating disorders are recognized sources of ill health in PWD and are associated with dehydration, malnutrition, weight loss, obesity, and accidental poisoning [8, 9]. PWD who suffered from malnutrition and dehydration were 10 times more likely to be hospitalized than their peers [10, 11]. And the severity of these symptoms increases with the progression of the disease and may be a predictor of patients’ mortality [12, 13]. In addition, people who accept being fed instead of eating independently may develop a form of learned helplessness, leading to further deterioration of cognition [14]. For caregivers, the severity of patients’ aversive eating behaviors and their caregiving burden are closely related. PWD with long-term and repeated abnormal eating behaviors will make caregivers who assist them in eating feel hopeless and burned out [15]. A logistic regression analysis showed a positive association between aversive feeding behaviors (including refusal to eat and resistive) worsening and the initial caregiver’s burden after controlling factors [16, 17]. Moreover, various eating-related problems are commonly observed among PWD, and their caregivers must manage these on a daily basis, which burdens them a momentous caregiving task. Therefore, addressing feeding and eating disorders is identified as a significant issue for dementia caregivers.

Interventions for PWD feeding and eating disorders have been extensively researched, and the priority consideration for dealing with the problem is non-pharmacological interventions in most countries, as non-pharmacological interventions have fewer side effects and lower costs compared to drugs [18]. Previous systematic reviews (SRs) summarized that non-pharmacological interventions for feeding and eating disorders mainly consist of oral nutritional supplementation (ONS), eating or drink assistance, environmental modification, food type modification, and so on to directly or indirectly improve food intake or malnutrition status of PWD [19, 20]. However, problematic feeding and eating behaviors may occur in any type of dementia and are different during the phases of dementia development. For example, changes in “food preference” and “eating habits” usually occur in the mild or middle stage, and the patient gradually loses the ability to self-feeding in the final stage [21, 22]. This makes existing interventions inadequate to help caregivers, especially those at home, cope with feeding and eating at different stages of dementia. Although SRs and primary intervention research of non-pharmacological interventions to improve dementia feeding and eating disorders are increasing, an exploratory qualitative study still showed that caregivers often adopt negative attitudes and behaviors regarding the feeding and eating problems and needs of PWD in the community [23]. Therefore, it is essential to clarify eating intervention goals and provide targeted non-pharmacological intervention strategies to promote caregiver understanding and coping with feeding and eating problems. Whereas now, there are no detailed recommendations for caregiver management protocols regarding specific eating disorder problems at different stages of dementia. As well, evidence on the applicable setting for eating intervention practice remains scarce. Thus, this evidence summary aims to provide caregivers with a set of self-help non-pharmacological interventions for feeding and eating disorders in PWD from the perspectives of intervention targets, implementation stages and practice settings, through which to facilitate clinical evidence-based practice process and guide the implementation of interventions to improve eating problems in PWD.

METHODS

We conducted this review in accordance with the Shanghai Evidence based Nursing Centre of Joanna Briggs Institute Evidence Based Nursing Cooperation Centre reporting items for evidence summary [24]. The methodology is based on the summary of evidence produced by the Joanna Briggs Institute Centre for Evidence-Based Health Care in Australia. It adds recommendations for practice evidence diagrams to promote readability and dissemination of evidence, thereby helping dementia caregivers understand and use the evidence more easily. In addition, this methodology is a quick and concise decision support tool relative to the Cochrane methodology and is a relatively complete method to promote the transformation of evidence into clinical practice [24, 25]. We organized and extracted the evidence from each article and evaluated the evidence levels. We used the JBI Critical Appraisal Tools to evaluate the quality of the included systematic reviews, randomized controlled trials and quasi-experimental studies [26, 27]. The JBI evidence grading and evidence recommendation level system was used to evaluate and grade the included evidence [28]. Then grouped the evidence of the same category in all articles and selected the highest evidence level. All authors compared the evidence recommendation levels to obtain relatively consistent conclusions and formed a clearly relational map of the operational details of evidence-based practice.

Search strategy

PubMed, CINAHL Complete, the Cochrane Library, Embase, UpToDate, BMJ best practice, JBI, International websites about dementia (including Dementia Australia, Alzheimer’s Society, Alzheimers New Zealand, and Alzheimer’s Disease Association) were systematically searched. The publication date of the article was filtered ranging from January 1, 2016 to December 31, 2021 (specific date was based on the database setting). The search terms contained three constructs guided by PICOS principle, including the search terms of the population, outcomes, and study design. As the intervention is what we want to search for, the search term is not restricted. For the database where the advanced search is set, search strategies were developed using a combination of Medical Subject Headings (MeSH) Terms where applicable and text terms. Otherwise (e.g., UpToDate, BMJ best practice, JBI, and other international websites about dementia), only one text term related to the topic was used. The search strategy of each search resource is detailed in the Supplementary Material. Additionally, the references lists on related topics from identified articles were supplemented by hand search.

Selection criteria

The inclusion criteria are as follows: 1) The study populations are the persons who are diagnosed with dementia according to related clinical criteria. 2) The interventions should be aimed to solve feeding and eating disorders of PWD due to cognitive impairment and neuropsychiatric symptoms, including but not limited to difficulty in psycho-behavioral symptoms while eating (e.g., behaving violently to the caregiver, wandering while eating, thinking someone poisoned the food), changes in appetite and preference (e.g., eating too much or too little, adhering to a single food), oral phase dysphagia specifically related to oral initiation of chewing and swallowing. 3) The interventions should be non-pharmacological and are administered by the caregiver. 4) According to requirements of study design in evidence summary formulation, evidence from guidelines, best practice sheets, SRs is prioritized. In the absence of SRs, however, high-quality and/or recent primary studies including randomized controlled trials (RCTs), and quasi-experimental studies (e.g., single-group pre-post design, PP; controlled clinical trials, CCTs) could be included. 5) The articles were published in English. 6) The publication year of studies were within 5 years. We excluded studies by the following criteria: 1) The study populations were not PWD. 2) The study populations were diagnosed with eating problems (e.g., dysphagia) due to physiological eating (e.g., upper limb neuromuscular injury, oral and dental abnormalities, cervical anterior displacement). 3) The interventions were pharmacological. 4) Review, protocol, interpretation of the guidelines, analytical cross-sectional studies, prevalence studies, qualitative studies. 5) The full text of the articles could not be obtained. 6) The studies were not peer-reviewed.

Article screening and data extraction

Two members of the team (Hongli Chen and Cheng Li) trained screened the title, abstract, and full text independently based on the selection criteria using Endnote X9. Data extraction for all studies included: first author, publication date, study design, study settings, interventions, and outcomes related to feeding and eating disorders. Sample size and age of study participants of RCTs and quasi-experimental studies were also extracted. The evidence recommendations were extracted and summarized according to the theme through full-text reading. Any disagreement was resolved through discussion and a third researcher would arbitrate.

Quality analysis

The risk of bias in each included SR was assessed using JBI Critical Appraisal Tools for systematic review and research syntheses, an 11-item tool. For assessing the quality of RCTs. JBI Critical Appraisal Tools for Randomized Controlled Trails, a 12-domain tool was utilized. JBI Critical Appraisal Checklist for Quasi-experimental Studies (Non-Randomized Experimental Studies) which is a 9-item tool, evaluated each included quasi-experimental study. Each item of all the tools mentioned above was rated as “Yes”, “No”, “Unclear”, or “Not applicable”. The evidence level was measured by JBI levels of evidence, and the grade of recommendation was measured by JBI Grades of Recommendation which is a FAME (Feasibility, Appropriateness, Meaningfulness and Effectiveness) scale may help inform the wording and strength of a recommendation. Two researchers Hongli Chen and Cheng Li discussed and reached a consensus after independently assessing. If two researchers encounter disagreement, corresponding author (Xianwen Li) will arbitrate.

RESULTS

General characteristics

The database search resulted in a total of 1,052 articles. Six related references to the identified article were supplemented by hand search. After de-duplication, we screened 870 studies by titles and abstracts, and 77 records were eligible for full-text review. Of these reviews, 17 studies were excluded because they were not SRs, RCTs, or quasi-experimental studies. Three studies were excluded because the populations were not mainly PWD. Four studies were excluded because the interventions were not non-pharmacological interventions. Three studies involved physiological dysphagia were also excluded. Additionally, 16 studies were excluded because the outcomes measured in the studies were not about feeding and eating disorders, the full text of 2 studies cannot be retrieved, and 1 study was not peer-reviewed. Finally, 28 articles fully met the inclusion criteria and were included in this evidence summary (see Fig. 1).

Fig. 1

Flow diagram of selection process.

The 28 included studies were published from the year 2016 to 2021. The studies were from Europe (14), North America (9), Australia (1), and East Asia (4). Of the 28 included studies, 14 were systematic reviews, 4 were RCTs, and 10 were quasi-experimental studies. No guidelines and best practice sheets were obtained and included in our evidence summary. The study participants lived at home, in the nursing home, in the hospital, in the day care center, etc. The sample size of PWD in each intervention group of RCTs and quasi-experimental studies ranged from 12 to 424 PWD. The age of the included sample was reported to be as young as 58 years, and the rest of the patients were concentrated in the age range of 65-90 years, thus indicating that the dementia patients with feeding and eating disorders were mainly older adults. The types of dementia in the included sample were concentrated in Alzheimer’s disease and vascular dementia. The interventions were ONS, food modification, environmental modification, education programs, assisting in eating and drinking, social support, empowerment, tailor-made dietary plans, and interval retrieval training, all have a positive impact on improving dietary problems in patients with dementia. Among such interventions, ONS, food modification, and assisting in eating and manual feeding can increase meals and reduce the incidence of weight loss, while spaced retrieval activities are conducted to enhance patients’ eating ability and nutritional intake. And the tailor-made dietary plans could improve the hyperphagic behaviors of patients with dementia. Reported outcomes of included studies can be categorized into direct outcomes related to feeding and eating disorders (e.g., hyperphagia, pica and eating ability, etc.) and consequences of feeding and eating disorders (e.g., nutrition status, body weight, body mass index (BMI), muscle mass, food intake, engagement in eating, etc.) (see Tables 1–3).

Table 1

Characteristics of included SRs

No.	First Author	Country	Year	Age of the participant (years)	Type of dementia	Setting	Interventions	Outcomes related to feeding and eating disorders
1	Keller [29]	Canada	2016	Unspecified	Unspecified	Long-term care institution	Multi-component intervention including training program (e.g., retraining of residents to self-feed), mealtime assistance, environmental modification	Eating performance, self-feeding ability
2	Abdelhamid [19]	UK	2016	Unspecified	Any type	Institutional setting (e.g., nursing home, long-term care facility, Alzheimer’s care center, etc.), hospital, day center, community	ONS, food or drink modification, swallowing problem management, eating assistance, supporting the social element of eating and drinking (family style meals)	Nutrition or hydration status, quantity/quality/ adequacy of or fluid intake, ability to eat independency, swallow without aspirating, enjoyment of food or drink or meaningful activity
3	Leah [30]	UK	2016	Unspecified	Unspecified	Long-term dementia specialist unit, home, ward	Education program, environmental or routine change, assistance with eating	Sitting time, food/energy/calorie intake, body weight or BMI, knowledge, positive attitude and behavior during mealtime, eating difficulty, MNA, self-feeding time and frequency, need for eating assistance
4	Bunn [20]	UK	2016	Unspecified	AD, other	Institutional setting (e.g., dementia unit, long-term care facility, nursing home, short stay assessment unit, etc.), hospital, day-care centers, home, community	Dining environment or food service modification, education or training for PWD or caregivers, behavioral intervention, exercise, multi-component intervention (>3 interventions, including at least one listed here)	Nutrition or hydration status, meaningful activity or enjoyment of food or drink, quantity/quality/adequacy of food or fluid intake
5	Sheppard [31]	Canada	2016	Unspecified	Any type	Nursing home, hospitals, assisted living facilities, day programs, home	Montessori-based activities (training)	Eating behaviors (e.g., self-feeding time, or frequency of verbal and physical feeding assistance)
6	Jensen [32]	USA	2017	Unspecified	AD	Long-term care facility, home	Environment based intervention (ambient music)	Engagement with eating, eating behavior
7	O’Neil-Pirozzi [33]	USA	2018	Unspecified	Any type	Any setting	ONS, food or drink modifications, dysphagia intervention, eating and drinking assistance, social meal environment manipulation	Nutrition or hydration status, quantity/quality/adequacy of food or drink, ability to eat independently, ability to swallow without aspiration, enjoyment of food/drink or meaningful food/drink related activity
8	Woodbridge [34]	UK	2018	Unspecified	Any type	Real-word setting (e.g., day care, home, care home)	Physical environment strategies (including changes to the global environment to architectural features, use of moveable environmental aids, and tailored individual approaches)	Everyday activity (e.g., eating and drinking, preparing food and drink, etc.)
9	Herke [35]	Germany	2018	75.8–86.9	AD, other	Home care, residential care, nursing home	Environmental (change of routine, context, or ambience) and behavioral (education and training programs for dementia or caregivers) modifications	Intake of food and fluid (calorie and protein intake), nutritional status (e.g., MNA, BMI, body weight, etc.), mealtime behavior
10	Fetherstonhaugh [36]	Australia	2019	>65	AD, VaD, Other	Any setting (e.g., long-term care setting, nursing home, dementia-specific care facility, assisted living facility)	Food presentation, interventions related to meal style with respect to seating/serving arrangements, environmental adaptations, music therapy, animal-assisted therapy, resident or caregiver training	Nutritional status, food intake, eating skills, communication and behavioral and psychological symptoms of dementia during mealtime
11	Yakimicki [37]	USA	2019	67–95	Unspecified	Any setting (e.g., memory care ward of extended care facility, memory clinic, adult day care center, assisted living community, geriatrics psych ward of hospital)	Animal-assisted intervention	Food intake
12	Borders [38]	UK	2020	Unspecified	AD, VaD, LBD, FTD, PDD, Other	Nursing home, hospital, day center, home	Patient, caregiver, and staff education, environmental modification, feeding intervention, oral supplementation	Oral intake, nutritional status (weight, BMI, MNA, etc.)
13	Backhouse [39]	UK	2020	58–100	Unspecified	Nursing home, specialized care unit, residential care facility, dementia care facilities, psycho-geriatric unit wards, continuing care center, long-term care facilities, cognitive support units, community dwelling	Intervention involving music, interaction and communication style	Refusal behaviors (agitation) in mealtime
14	Faraday [40]	UK	2021	>65	Any type	Care homes	Mealtime care (social connection, tailored care, empowering the resident, responding to food refusal)	Food intake, mealtime experience

AD, Alzheimer’s disease; BMI, body mass index; FTD, frontotemporal dementia; LBD, Lewy body dementia; MNA, Mini Nutrition Assessment; ONS, oral nutrition supplementation; PDD, Parkinson’s disease dementia; SR, systematic review; VaD, vascular dementia.

Table 2

Characteristics of included RCTs

No.	First Author	Country	Year	Group	Sample size (n)	Age of the participant (years)	Type of dementia	Setting	Interventions	Period	Outcomes related to feeding and eating disorders
1	Kao [8]	Taiwan, China	2016	I1	46	83.10±5.00	Any type of exclusion of VaD	Dementia special care unit	Spaced retrieval training	40 min/session, 30 sessions, 6 weeks	Hyperphagic behavior (intention to eat, rapid eating, increased eating, inappropriate dining etiquette), pica behavior, change in eating habit, short meal frequency, distress of the caregiver to hyperphagic behaviors
				I2	49	82.69±6.81			Spaced retrieval and Montessori-based activities training
				C	45	81.82±5.89			Routine activities
2	Hsu [41]	Taiwan, China	2017	I	50	81.55±7.68	Any type of exclusion of VaD	Care center for patients with dementia	Spaced retrieval training	40 min/session, 30 sessions, 6 weeks	Hyperphagic behavior, food intake, BMI
				C	47	81.93±6.52			Daily care
3	Ohmana [42]	Finland	2017	I1	57	77.9±5.2	AD	Community dwelling	Group exercise	60mim/one time, twice a week, 12 months	NPS (including appetite and eating abnormalities)
				I2	63	77.4±5.3			Home exercise
				C	59	78.1±5.3			Usual community care
4	Reverté-Villarroya [43]	Spain	2019	I	15 (PWD and caregivers)	78.69±6.58	AD, VaD	Cognitive and behavioral disorders unit	Cognitive-behavioral education with periodic telephone contacts	Education: 90–120 min/one time, telephone visit: 30 min/one time, 3 months	NPS (including eating disorders)
				C	13 (PWD and caregivers)	79.20±6.91			Routine clinical practice

AD, Alzheimer’s disease; BMI, body mass index; C, control group; I, intervention group; NPS, neuropsychiatric symptoms; PWD, persons with dementia; RCT, randomized controlled trail; VaD, vascular dementia

Table 3

Characteristics of included quasi-experimental studies

No.	First Author	Country	Year	Study design	Sample size (n)	Age of the participant (years)	Type of dementia	Setting	Interventions	Period	Outcomes related to feeding and eating disorders
1	Krolak-Salmon [44]	France	2016	PP	424	84.0±7.2	Unspecified	Home, nursing home	Mobile team intervention	30 days	NPS (including appetite and eating abnormalities)
2	Batchelor-Murphy [45]	USA	2017	CCT (cross over study)	30	88.5 (68–99)	Unspecified	Nursing home	Supportive handfeeding techniques (DH, OH, UH or OH, UH, DH or UH, DH, OH)	6 consecutive meals, changing technique every 2 days, 18 meals	Time required for feeding assistance, meal intake, feeding behaviors
3	Hansen [46]	Norway	2018	PP	12	65–85	Unspecified	PWD unit, nursing home	Use of colored and porcelain plate	3 weeks	Food intake, appetite, mealtime dignity
4	Simmons [47]	USA	2018a	PP	53	83.8±10.7	AD, Other	Nonprofit assisted living facility	Feeding assistance with quality improvement system managing	About 5 min/meal, 12 months	Meal intake, body weight
5	Simmons [48]	USA	2018b	PP	57	85.1±9.1	AD, Other	Nonprofit assisted living facility	Feeding assistance with quality improvement system managing	3 meals and 2 snack offering between-meal/day, 13months	Meal time nutritional care quality, meal consumption, amount of staff assistance provided to encourage intake, offers of alternatives to the served meal, weight
6	Sulmont-Rosse [49]	France	2018	PP	32	86.8 (75–98)	AD	Special care unit	Olfactory priming	15 min/lunch, 1 lunch/2 weeks, 4 lunches (2 control, 2 primed)	Food intake, eating behavior (e.g., willingness to eat, staying sitting at table)
7	Gaspar [50]	USA	2019	PP	85	86.08±7.37 (70–98)	Unspecified	Assisted living memory care	Montessori-inspired lifestyle	7-12 months	Engagement during meals
8	Visscher [51]	Netherlands	2020	PP	15	84.6±9.8	AD, VaD, Other	Nursing home	Finger foods	Once a day around 4 pm, 6 weeks	Consumption of the finger foods, fruit and vegetable consumption, food intake during dinner
9	Jung [52]	South Korea	2020	PP	23 dyads (PWD and caregivers)	PWD: 86.09±5.98 Caregiver: 60.83±6.93	Unspecified	Long-term care facility	Mobile application-based meal assistant training program	Twice a week, 4 weeks	Eating behavior, eating time, caregiver attitude toward feeding, caregiver behavior regarding feeding
10	Lee [53]	South Korea	2021	PP	30	85.63±6.67	Unspecified	Long-term care facility	Caregiving approaches during mealtime (person-centered behavior, task-centered behavior, on interaction)	6 months	Behavior symptoms during mealtime

AD, Alzheimer’s disease; CCT, controlled clinical trial; DH, direct hand, the caregiver holds the object (e.g., fork, spoon, cup) intended to provide food or fluids to the resident without any active involvement on the part of the resident; NPS, neuropsychiatric symptoms; OH, over hand, The caregiver puts his/her hand over the resident’s hand in an effort to guide/support/assist the resident with the activity; PP, single-group pre-post design; PWD, persons with dementia; UH, under Hand, the caregiver holds the object (e.g., fork, spoon, and/or cup) and places his/her hand under the resident’s hand. This technique theoretically allows the resident to feel as though he/she initiated the movement, and is in control; VaD, vascular dementia.

Quality assessment

The quality assessment results of included SRs are shown in Table 4. There was a clear statement for study aims and recommendations for policy or practice in all included SRs as questions 1 and 10 were all assessed as “Yes”. Eight criteria (questions 2–8, and 11) of the JBI quality check were satisfied by most of the SRs. A high bias of risk was found in publication bias and only 3 out of 14 SRs were measured as “Yes”. Table 5 presents the critical appraisal results of the RCTs. There was an unclear bias risk for allocation concealment, and the participants were not analyzed in the groups to which they were randomized in all the RCTs. Few RCTs used blinded participants, delivering treatment, and outcomes assessors. Six questions (questions 1, 2, and 10–13) of JBI critical appraisal tools were satisfied by all the RCTs. The results of the critical appraisal of quasi-experimental studies were reported in Table 6. In all the quasi-experimental studies, there was a clear statement for what is the ‘cause’ and ‘effect’, and the outcomes of participants included in any comparison were measured in the same and reliable way. Questions 2–6 and 9 were judged as unclear or high risk of bias.

Table 4

Quality assessment of included SRs

Study	Q1	Q2	Q3	Q4	Q5	Q6	Q7	Q8	Q9	Q10	Q11
Keller [29]	Yes	Unclear	Unclear	Unclear	Yes	Yes	Unclear	Unclear	Unclear	Yes	Yes
Abdelhamid [19]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Unclear	Yes	Yes
Leah [30]	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Unclear	Yes	No
Bunn [20]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Unclear	Yes	Yes
Sheppard [31]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Unclear	Yes	Yes
Jensen [32]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
O’Neil-Pirozzi [33]	Yes	Yes	Unclear	Yes	Yes	Yes	Yes	Yes	Unclear	Yes	Yes
Woodbrid [34]	Yes	Yes	Yes	Yes	Unclear	Unclear	Yes	Yes	Unclear	Yes	Yes
Herke [35]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Fetherstonhaugh [36]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Unclear	Yes	Yes
Yakimicki [37]	Yes	Unclear	Yes	Yes	Unclear	Unclear	Unclear	Yes	Unclear	Yes	Yes
Borders [38]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Unclear	Yes	Yes
Backhouse [39]	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes
Faraday [40]	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Unclear	Yes	Yes

Q1. Is the review question clearly and explicitly stated? Q2. Were the inclusion criteria appropriate for the review question? Q3. Was the search strategy appropriate? Q4. Were the sources and resources used to search for studies adequate? Q5. Were the criteria for appraising studies appropriate? Q6. Was critical appraisal conducted by two or more reviewers independently? Q7. Were there methods to minimize errors in data extraction? Q8. Were the methods used to combine studies appropriate? Q9. Was the likelihood of publication bias assessed? Q10.Were recommendations for policy and/or practice supported by the reported data? Q11.Were the specific directives for new research appropriate?

Table 5

Quality assessment of including RCTs

Study	Q1	Q2	Q3	Q4	Q5	Q6	Q7	Q8	Q9	Q10	Q11	Q12	Q13
Kao [8]	Yes	Unclear	Yes	Unclear	No	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes
Hsu [41]	Yes	Unclear	Yes	Unclear	No	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes
Ohmana [42]	Yes	Unclear	Yes	No	Unclear	Unclear	Unclear	No	No	Yes	Yes	Yes	Yes
Revert $\overset{´}{e}$ -Villarroya [43]	Yes	Unclear	Yes	Yes	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes

Q1. Was true randomization used for assignment of participants to treatment groups? Q2. Was allocation to treatment groups concealed? Q3. Were treatment groups similar at the baseline? Q4. Were participants blind to treatment assignment? Q5. Were those delivering treatment blind to treatment assignment? Q6. Were outcomes assessors blind to treatment assignment? Q7. Were treatment groups treated identically other than the intervention of interest? Q8. Was follow up complete and if not, were differences between groups in terms of their follow up adequately described and analyzed? Q9. Were participants analyzed in the groups to which they were randomized? Q10. Were outcomes measured in the same way for treatment groups? Q11. Were outcomes measured in a reliable way? Q12. Was appropriate statistical analysis used? Q13. Was the trial design appropriate, and any deviations from the standard RCT design (individual randomization, parallel groups) accounted for in the conduct and analysis of the trial?

Table 6

Quality assessment of quasi-experimental studies

Study	Q1	Q2	Q3	Q4	Q5	Q6	Q7	Q8	Q9
Krolak-Salmon [44]	Yes	Yes	Yes	No	Yes	No	Yes	Yes	Yes
Batchelor-Murphy [45]	Yes	Unclear	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Hansen [46]	Yes	Yes	Yes	No	No	Yes	Yes	Yes	Not appliable
Simmons [47]	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes
Simmons [48]	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	No
Sulmont-Rosse [49]	Yes	Yes	Yes	Yes	Yes	Unclear	Yes	Yes	Yes
Gaspar [50]	Yes	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes
Visscher [51]	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes
Jung [52]	Yes	Yes	Yes	No	Yes	Yes	Yes	Yes	Yes
Lee [53]	Yes	Unclear	Unclear	Yes	Yes	Yes	Yes	Yes	Yes

1. Is it clear in the study what is the ‘cause’ and what is the ‘effect’ (i.e., there is no confusion about which variable comes first)? 2. Were the participants included in any comparisons similar? 3. Were the participants included in any comparisons receiving similar treatment/care, other than the exposure or intervention of interest? 4. Was there a control group? 5. Were there multiple measurements of the outcome both pre and post the intervention/exposure? 6. Was follow up complete and if not, were differences between groups in terms of their follow up adequately described and analyzed? 7. Were the outcomes of participants included in any comparisons measured in the same way? 8. Were outcomes measured in a reliable way? 9. Was appropriate statistical analysis used?

Summarized evidence

Twenty-three non-pharmacological interventions recommendations were categorized into 6 themes containing ONS, assistance with eating and drinking, social engagement during mealtime, environmental modification, education or training for PWD and caregivers, and multi-component intervention. Most interventions were targeted at moderate and severer stages of PWD who lived in institutional settings with diverse purposes (see Table 7 and Fig. 2). The levels of evidence are almost level 1 or level 2, as they were extracted from SRs of RCTs, quasi-experimental or mixed methods studies, and single RCT or quasi-experimental studies (see Table 7).

Table 7

Recommendations of feeding and eating disorders for PWD extracted from included studies

Category	Recommendation	Evidence level	Recommendation grade	Stage of dementia^*	Setting^*
ONS	1 ONS is recommended to increase energy intake, BMI, and weight in PWD at any stage living in any place with a small positive in a short-term	Level 1a	A	Mild, moderate, severe	Institutional, home
	2 Between-meal snack delivery can be considered	Level 2d	B	Severe	Institutional
	3 Providing rich finger foods to PWD seems feasible for both PWD and caregivers	Level 1c	A	Severe	Institutional
Assistance with eating and drinking	4 Assistance with eating and drinking (e.g., under hand^#, direct hand^#, and assistance in support meal preparation) who required is recommended and sufficient time should be given to finish their meals	Level 1b	A	Moderate and severe	Institutional
	5 Under hand and direct hand feeding are viable options to reduce feeding behaviors relative to over hand^# feeding	Level 2c	B	Unclear	Institutional
	6 High-quality improvement systems could be recommended to apply during mealtime feeding assistance	Level 2d	B	Severe	Institutional
	7 Responding carefully and skillfully to food refusal is a significant element of mealtime care for caregivers which may encourage PWD to eat more	Level 2	B	Moderate and severe	Institutional
	8 The use of an assistive device or technologies supporting meal preparation for PWD seem to enable people to begin or carry-on cooking	Level 3	B	Unclear	Home
Person-centered mealtime care	9 When caregivers encounter behavioral symptoms during mealtime (e.g., agitation), it is recommended to promote person-centered behaviors and social interaction with PWD, and avoid no response and task-centered behaviors, especially verbal controlling and inappropriate touch	Level 1b	A	Any (most are severe)	Institutional
	10 Mealtime care should be tailored to the PWD by knowing their needs and preferences, providing choices for PWD at mealtimes (e.g., giving their choice of what to eat, where to eat, and when to eat), and being receptive to their changes	Level 2b	A	Moderate and severe	Institutional
Environmental modification	11 It is recommended that pleasant environments (e.g., playing the most preferred music, fish tank with fish placed in meal setting, flowers in dining settings, etc.) for PWD	Level 1b	A	Mild, moderate, severe	Institutional
	12 Home-like environments (e.g., family-style meals, renovating dining area in the institutional setting) could be recommended	Level 2a	B	Mild, moderate, severe	Institutional
	13 Tailoring individual adaptation for PWD such as removing clutter, using labels, and providing a table with a model example is recommended to improve their ability to prepare drinks and set the table	Level 2d	B	Moderate and severe	Institutional
	14 It is recommended that modifying light intensity (higher light level), and visual contrast (higher visual contrast) using in cups or plates	Level 2b	A	Moderate and severe	Institution
	15 Olfactory priming is recommended for PWD to increase food consumption and interest in the meal during a short period	Level 2c	B	Unclear	Institutional
Education or training for the PWD and caregivers	16 Intensive and long-term education, counselling, or training on nutrition, feeding skill, and preventing weight loss are recommended to deliver to caregivers or with PWD through face-to-face sessions or web-based program	Level 1b	A	Mild, moderate, severe	Institutional, home
	17 The mobile application-based meal assistance training program may be useful for caregivers who need meal assistant skills education	Level 2d	B	Unclear	Institutional
	18 The non-pharmacological therapy of individual therapeutic-cognitive intervention (cognitive-behavioral education) directed to family caregivers carried out by nursing personnel expert in its management is recommended	Level 1c	B	Unclear	Institutional
	19 Spaced retrieval training or/and Montessori is recommended for PWD	Level 1b	A	Mild, moderate, severe	Institutional, home
	20 Spaced retrieval training with the use of a visual aid was functionally related to improvements in 2-3 compensatory swallowing behaviors	Level 2d	B	Mild-moderate	Institutional home
Multi-component intervention	21 Individual geriatric assessment is recommended to be involved in multidisciplinary interventions if it is available	Level 2c	B	Severe (most)	Institutional
	22 Multifactorial rather than single-component interventions (including ONS, food or drink modification and assistance, environmental modification, psychomotor rehabilitation program, etc.) are more recommended	Level 1b	B	Mild, moderate, severe	Institutional
	23 The mobile team for Alzheimer’s disease which is composed of a geriatrician, a psychiatrist, a nurse, a psychologist, and a social worker is recommended to propose some recommendations for PWD (e.g., non-pharmacological interventions, pharmacological therapies, and medico-social interventions)	Level 2d	B	Mild, moderate, severe	Institutional, home

^*The description of stages of dementia and living settings in the table are pragmatically stated in included studies of this evidence summary. It doesn’t mean the recommendation is not suitable for PWD at other stages living in other settings. This requires more research to be conducted to support the feasibility. In addition, institutional settings in Table 7 contain nursing homes, long-term care centers, day care centers, hospitals, etc., which are the organizations or establishments founded for PWD or old adults. ^#Direct hand, the caregiver holds the object (e.g., fork, spoon, cup) intended to provide food or fluids to the resident without any active involvement on the part of the resident; Over hand, The caregiver puts his/her hand over the resident’s hand in an effort to guide/support/assist the resident with the activity; Under Hand, the caregiver holds the object (e.g., fork, spoon, and/or cup) and places his/her hand under the resident’s hand. This technique theoretically allows the resident to feel as though he/she initiated the movement, and is in control; ONS, oral nutrition supplementation; PWD, patient with dementia.

Fig. 2

Self-help practice guidance of targeted non-pharmacological interventions for PWD with feeding and eating disorders at different stages.

Oral nutritional supplementation

Three SRs study assessed the effect of ONS on nutritional status and food/energy intake (SR 2, 7, 12). All the results of included studies support the recommendation that ONS is a feasible measure to increase energy intake, BMI, and weight in PWD, and to improve their malnutrition status resultant from feeding and eating disorders such as poor appetite, forgetting to eat and drink, chewing and swallowing problems, etc., which could happen in any stage of dementia and any setting. Two among the three SRs (SR 2, 7) conducted the meta-analysis and the other performed narrative descriptive synthesis on the effect of ONS. Two meta-analysis results suggested small, statistically significant effects of ONS on weight (2.02 kg, 95% CI: 1.53–2.50, 344 participants, I² = 0%) and BMI (0.91 kg/m², 95% CI: 0.56–1.25, 221 participants, I² = 0%), when the span of intervention is over 3 to 12 weeks. However, if the duration of the intervention is over 12 weeks, the meta-analysis suggested no statistically significant effect on weight (0.72 kg, 95% CI: 1.02–2.45, 382 participants) but with high heterogeneity (I² = 89%). These effects were too small to be considered as the long-term sustainability of these outcomes. Additionally, between-meal snacks may contribute to the low prevalence of weight loss and increase food consumption among PWD which could be considered as a form of ONS (SR 2; PP 4). The results of the PP study found between-meal snack delivery with a quality improvement system had a positive effect on the prevalence of unintentional weight loss of PWD residing in assisted living facilities averaged 1.3% across 12 months, while the SR suggested no significant effect on weight or waist circumference. Thus, the evidence was mixed and low-quality in the SR and PP study.

The recommendation of providing rich finger foods for PWD was summarized from one SR in which two CCTs were narratively analyzed, and one single pre-post-test design study (SR 2; PP 8). The intervention seems to be feasible for both PWD and caregivers and could provide a pragmatic approach to enhance consumption and total food intake. The settings where study participants lived in of included studies were long-term care institutional settings containing nursing homes and day care centers. Only the PP study mentioned that the study participants were advanced dementia, while the stage of PWD who participated in trials of SR was vague. Food consumption was found to increase (p < 0.05) during the provision of finger foods in a CCT of included SR and PP study, while eating independence or ability in self-feeding improved with limited evidence, and the effect on body weight was unclear.

Assistance with eating and drinking

Four SRs and a CCT study (SR 1, 2, 3, 7; CCT 2) evaluated the effect of assistance with eating and drinking (e.g., under hand, direct hand, assistance in support meal preparation, etc.). PWD at the middle and late stages may require more assistance during mealtime and should be given sufficient time to finish their meals. A little effect of assistance on the nutritional outcome of PWD increasing their energy intake and reducing feeding behaviors were found through the narrative summarize of SRs and CCT study. The study settings of included studies were nursing homes or wards. Among hand feeding assistance techniques, under hand (UH) and direct hand (DH) feeding were viable options to reduce feeding behaviors relative to over hand (OH) feeding according to the results of a CCT study (CCT 2). It revealed that meal intake in UH (65±15.0%) and DH (67±15.2%) groups were both significantly greater than OH group (60±15.1%). Feeding behaviors (e.g., refusal to open mouth, refusal to eat, refusal to swallow, spitting food out, etc.) were more frequent in OH group (8.3±1.8%), while the results are 7.7±1.8% in UH and 8.0±1.8% in DH group.

Additionally, high quality improvement system could be recommended to apply during mealtime feeding assistance. The results from 2 PP studies (PP 4, 5) found that it could increase the consistency of staff during meals and reduce the unintentional prevalence of weight loss compared with the national average (0% –3% versus 7%) among PWD who were severely cognitively-impaired living in the assisted living facilities. However, the statistical significance was not measured in the studies. As well, responding carefully and skillfully to food refusal using techniques is an important element of mealtime feeding assistance for caregivers which may encourage PWD to eat more based on the results of an SR (SR 14). The study participants in the trials of this SR were moderate or severe dementia living in nursing homes or long-term care facilities. For PWD living at home who may have a certain ability to prepare meals by themselves, the results of an SR (SR 8) suggested that the use of an assistive device or technologies supporting meal preparation seem to enable them to begin or carry-on cooking.

Person-centered mealtime care

The narrative synthesis of 3 SR studies (SR 7, 13, 14) and a PP study (PP 10) demonstrated that when caregivers encounter behavioral symptoms during mealtime (e.g., agitation), it is recommended to promote person-centered behaviors and social interaction with PWD, which are possible to improve important aspects of quality of life including autonomy, communication, mood, involvement and participation in meaningful activity in the context of food and drink. No response and task-centered behaviors (especially verbal controlling and inappropriate touch) are avoided. The results of the PP study showed that the total duration of the Cohen-Mansfield Agitation Inventory was associated with no verbal response (β=9.09) and task-centered verbal behavior (β=8.43), specifically verbal controlling (β=7.87) during mealtime. The mean score of Mini-Mental State Examination and Clinical Dementia Rating in the PP study and primary studies of included SRs indicated that most of the participants from long-term care facilities or nursing homes had severe cognitive impairment.

In addition, to better provide person-centered care, mealtime care also should be tailored to the PWD by learning about their needs and preferences, providing choices for PWD at mealtimes (e.g., giving their choice of what to eat, where to eat, and when to eat) and being receptive to their changes, which may increase their food consumption, positive mealtime behaviors, and communication with fellow. The recommendation was summarized from the narrative synthesis of two SRs (SR 8, 14). The study settings of included studies were nursing homes or wards, and most of the participants are in the middle and late stages of dementia.

Environmental modification

The effect of environmental modification on improving feeding and eating disorders of PWD was evident across narrative synthesis in 10 SRs (SR 1, 3, 4, 6, 8, 9, 10, 11, 12, 13). It is recommended that a pleasant environment (playing the most preferred music, fish tank with fish placed in meal setting, etc.) for PWD which may increase food intake, body weight, BMI, reduce incidences of eating behaviors (e.g., agitation and refusal behavior), and improve engagement with eating at mealtimes. This recommendation may fit well in both long-term care institutional settings and home settings for PWD at any stage. The variety of interventions and outcomes of primary studies in SRs made meta-analysis unfeasible. Seven SRs (SR 3, 4, 6, 8, 10, 12, 13) demonstrated an overall positive influence of music involved in mealtimes. Quiet, relaxing, soothing, familiar, pop and PWD’s preferred music were often utilized in the intervention. The positive effect of animal-assisted especially aquariums with fish placed during mealtime was evident based on the findings of 2 quasi-experimental studies analyzed in 4 included SRs (SR 8, 10, 11, 12).

As well, for mild to severe PWD living in nursing homes or dementia units, 4 SRs (SR 2, 4, 8, 10) found that home-like environments were associated with positive outcomes related to meal tasks, which suggested that home-like environments (e.g., family-style meal, renovating dining area of long-term care institutional settings) could be recommended to increase food intake, improve levels of autonomy (eating independently, helping with serving and clearing up), and promote participation and interaction during mealtimes. Also, body weight was reported significantly increased in a CCT study (+5.64 kg, p < 0.024) and an RCT study (+3.4 kg versus –2.2 kg, p = 0.02) of included SRs. Individuation is also an important element in the environmental modifications. Tailoring environmental changes based on individual adaptation for PWD such as removing clutter, using labels, and providing a table with a model example is recommended to improve their ability to prepare drinks and set the table, which was only summarized from a PP study in SR (SR 8). The populations who participated in the included study were PWD at moderate to severe stage from the geriatric hospital.

External environment stimulation in the perception was proved to be beneficial for increasing food intake of PWD in 4 SRs and 2 PP studies (SR 4, 8, 10, 12; PP 3, 6). It is recommended that modifying light intensity (higher light level) and visual contrast (higher visual contrast) may increase food intake of PWD (SR 4, 8, 10, 12; PP 3). The color visual contrast of tableware such as cups, plates, and cutlery can be manipulated. In most research, recruited study participants were at moderate to severe stage of dementia living in the long-term care institutional settings. One trial in included SRs focusing solely on the effects of visual contrast of dining ware found significant increases in food (>25%) and fluid (>84%) consumption for high visual compared with low visual contrast (p < 0.001) regardless of color. While the positive effect of modifying light intensity was proved with limited evidence as most of the trials explored the effect of light intensity combining with visual contrast. In addition, olfactory priming is also recommended for PWD to increase food consumption and interest in the meal (SR 14; PP 6), which can target nonconscious memory processes to stimulate appetite. However, this effect was no longer observed when the intervention was replicated two weekslater.

Education or training for the PWD and caregivers

Five SRs (SR 1, 3, 4, 9, 10) investigated the effect of education or training programs. Intensive and long-term education, counselling, or training on nutrition, feeding skills, and preventing weight loss could be recommended to deliver to caregivers or PWD which may contribute to improving body weight and food intake slightly through face-to-face sessions. The intervention may be feasible for PWD at any stage from any setting. However, only 2 CCTs analyzed in included SR showed significant improvement in weight (+4.6 kg, p < 0.01;+1.4 kg, p < 0.05), and an RCT showed significant decrease in a percentage of weight loss, while other studies presented no difference in weight, so did the same in BMI and eating behaviors. Nutritional training and counselling programs may increase the intake of food and liquids slightly (MD: 0.11 g/kg, 95% CI: –0.01 to 0.23, n = 78) as well. When regarding how to improve the feeding skill of caregivers, the findings of an SR and a PP study (SR 3; PP 6) suggested the mobile application-based or web-based meal assistance training programs may be useful for caregivers who needed meal assistant skills education according to the experience of caregivers in qualitative research, although most quantitative outcomes (e.g., eating assistance, food consumption, etc.) were not significant.

The non-pharmacological therapy of individual therapeutic-cognitive intervention (cognitive-behavioral education) directed to family caregivers carried out by nursing personnel experts in its management is recommended, as it has a favorable impact on the control and reduction of the eating disorders of PWD. Evidence showed that eating disorders were improved (p = 0.019) in the intervention group at 3 months in an RCT (RCT 4). As well, according to the findings of 5 SRs (SR 3, 4, 5, 9, 10), 2 RCTs (RCT 1, 2), and a PP study (PP 7), Spaced retrieval training or/and Montessori is recommended for PWD as long-term effectiveness of improving feeding and eating disorders (e.g., hyperphagic, rapid eating, and pica behaviors), self-eating ability, and increasing meal intake. However, it is uncertain whether nutritional status such as the score of MNA, BMI, and body weight was improved as the evidence was mixed. Spaced retrieval training with the use of a visual aid was functionally related to improvements in 2-3 compensatory swallowing behaviors from the result of a PP study (PP 2).

Multi-component intervention

The SR study (SR 3) demonstrated that the use of individual geriatric assessment of eating difficulties is crucial to address the individual needs of dementia diets. Due to the factors of feeding and eating disorders being multiple, 3 SRs (SR 1, 2, 4) revealed that multifactorial rather than single interventions (including ONS, food or drink modification and assistance, environmental modification, psychomotor rehabilitation program, etc.) were more likely to improve eating performance and increase weight, BMI, and food intake of PWD. However, due to the components of interventions in different studies being variable, the effectiveness of multi-component intervention should be carefully considered. Regarding the implementor of multi-component intervention, a PP study (PP 1) suggested the mobile team for Alzheimer’s disease which is composed of geriatricians, psychiatrists, nurses, psychologists, and social workers could be considered, who can propose some recommendations for PWD (e.g., non-pharmacological interventions, pharmacological therapies, and medico-social interventions) reducing a partial proportion of hospitalizations related to appetite and eating abnormalities and may help to reduce eating behavioral disorders. The Neuropsychiatric Inventory score in the study declined after mobile team intervention (45.8–29.9, p < 0.001). Sleep and appetite disorders were more associated with a higher risk of hospitalizations.

DISCUSSION

To our knowledge, this is the first evidence summary concerning targeted non-pharmacological interventions for caregiver to cope with feeding and eating disorders in PWD. Despite the guideline and best practice sheet were not obtained through systematic searching, higher level evidence containing SRs, RCTs, and quasi-experimental studies were preferentially considered according to the principle of the evidence pyramid. Six types of feeding and eating recommendations were extracted and graded, then we formed the evidence into practice recommendations and help caregivers understand the application of the evidence by drawing simple practice recommendation diagrams. The feeding and eating management strategies consist of ONS, assistance with eating and drinking, person-centered mealtime care, environmental modifications, education or training for the PWD and caregivers, and multi-component intervention. These 6 types of interventions corresponded to 3 direct eating targets and were applied to different stages of dementia. Environmental modification and education or training were recommended for the mild to severe stages of dementia, with the aim of altering feeding and eating disorders and improving engagement with eating. Person-centered mealtime care and assistance with eating and drinking were recommended in the moderate to severe stages of dementia, with the former intervention targeting to improve engagement with eating and the latter to make up for some loss ability in eating. ONS was recommended in the severe stage of dementia to increase food intake directly. Multi-component intervention was applied throughout dementia and intent to solve eating problems in a better and targeted way. Additionally, some studies (SR 4; RCT 3) evaluated the effect of exercise on PWD. However, limited evidence revealed the improvement of nutritional status in any study. Thus, exercise intervention was not included in the recommendation list.

The core targets of the recommended intervention strategies are the same, all addressing feeding and eating disorders or related nutritional outcomes. However, direct targets among them are different, which can be broadly classified into three categories. ONS is mainly aimed at directly increasing food intake and improving the nutritional outcomes (e.g., body weight, BMI) of PWD resulting from eating and feeding disorders. Assistance with eating and drinking can be regarded as a compensatory measure to make up for some loss of the ability of PWD in eating and drinking, such as the ability to food preparation, eat dependently, correctly use tableware, etc. The direct purposes of the recommendations including person-centered mealtime care, environmental modification, and education or training for the PWD and caregivers are to alter eating habit and behavior (e.g., agitation behavior, refusing to eat, failing to recognize food, pica), and improve engagement with eating. Sometimes, PWD or caregivers often encounter multiple problems in eating rather than single, thus, multi-component intervention should be utilized in practice. In addition, in order to better and targeted solve feeding and eating disorders, assessment of the problems and needs of PWD and their caregivers is a vital process before intervention is implemented. Last but not least, geriatricians, psychiatrists, nurses, psychologists, and social workers should be involved in the implementation of some professional interventions such as education and training programs to guarantee effective solving of eating problems.

In addition to considering the target condition of intervention implementation, it is also worth considering which stage of PWD and where the recommendations mentioned above can be used. Although most PWD who participated in included studies were from long-term care institutional settings and at moderate to severe stages, the applicability of recommendations should not be restricted to these settings and these stages of dementia. In real life, to better, more efficiently and effectively improve feeding and eating disorders, we should take full account of the needs or state of PWD and caregivers, existing resources combined with these recommendations. In the mild stage, cognitive deficits cause them to eat very frequently, having forgotten they had just eaten. Cognitive-behavior training or Spaced retrieval training is necessary for cognitive improvement. The application of internet technology-based interventions is convenient for caregivers and residential dementia to access professional resources. At the moderate and severe stages, the metabolic rate gets down-regulated as the disease cause increasingly severe atrophy and physical activities reducing. As the disease progresses, PWD may completely lose the ability to eat independently, then the caregiver’s assisted feeding methods are more worthy of attention. In addition, they have difficulties in using cutlery and recognizing food, and are easily distracted by changes in the environment. Apparently, apart from measures to solve feeding and eating disorders, external nutrition supplementation is needed for frailty improvement.

The limitations of our study include the high statistical heterogeneity as well as complex categories of interventions among the RCTs of several included SRs making it difficult for researchers to perform a meta-analysis, and only narratively summarized the effectiveness of interventions in these included SRs. Not all of the included studies describe the specific period and content of the intervention, and these may lead to decreased reliability of recommendations on feeding and eating disorders in PWD. Most of the available recommendations came from long-term care institutional settings, and more original research should pay attention to residential PWD to provide more high quality and targeted evidence in home care, as most PWD are preferred to live at home according to the report of the Alzheimer’s Association.

Conclusion

This article illustrated the direct targets of recommended interventions on different themes, including directly increasing food intake, making up for some loss of ability in eating, and improving engagement with eating. Directly increasing food intake was mainly a specific target in the severe stage of dementia disease, while the other two eating targets are throughout the disease process. In addition, we summarized the types of interventions and specific recommendations applicable to different stages of dementia to manage feeding and eating problems. For the dementia with mild stage, measures such as environmental modifications and education or training for the PWD and caregivers were mostly recommended. For the dementia with moderate to severe stage, all six interventions summarized in this study were applicable, with the ONS theme of between-meal snack supplementation and finger food being more specific to the severe stage of dementia. In addition, multi-component intervention can be implemented at any stage of the disease, but the combination and sequence of interventions remained to be further explored. Most of the evidence for the recommended interventions came from the studies on the population living in the long-term care institutions, while less evidence included the home dementia population. Therefore, the practice of recommendations was more applicable to institutionalized PWD, and when applied to PWD at home, patient wishes, caregiver capacity, and available resources should be adequately considered. A large number of high-quality studies are still needed in the future to determine the appropriate time, place, and specific interventions for feeding and eating disorders.

Footnotes

ACKNOWLEDGMENTS

The authors have no acknowledgments to report.

FUNDING

This study was supported by the Ministry of Education of Humanities and Social Science Project (Grant No: 22YJCZH089), the Jiangsu Commission of Health Foundation (Grant No: Z2020028), the Research Project of Philosophy and Social Sciences in Jiangsu Universities (Grant No: 2022SJYB0298), and the Connotation Construction Project of Nanjing Medical University for Priority Academic of Nursing Science (Grant No: 2022-12).

CONFLICT OF INTEREST

The authors have no conflict of interest to report.

DATA AVAILABILITY

The data supporting the findings of this study are available within the article and/or its supplementary material.

The supplementary material is available in the electronic version of this article: .

References

Alzheimer’s Disease International. Dementia statistics. https://www.alzint.org/about/dementia-facts-figures/dementia-statistics, Accessed on July 26, 2021.

Chang

, Lin

, Chiu

, Liao

, Ho

, Lin

, Chou

, Liu

(2017) Prevalence and factors associated with food intake difficulties among residents with dementia. PloS One 12, e0171770.

Aselage

(2010) Measuring mealtime difficulties: Eating, feeding and meal behaviours in older adults with dementia. J Clin Nurs 19, 621–631.

Flynn

, Smith

, Walsh

, Walshe

(2018) Modifying the consistency of food and fluids for swallowing difficulties in dementia. Cochrane Database Syst Rev 9, CD011077.

Cerejeira

, Lagarto

, Mukaetova-Ladinska

(2012) Behavioral and psychological symptoms of dementia. Front Neurol 3, 73.

Edahiro

, Hirano

, Yamada

, Chiba

, Watanabe

, Tonogi

, Yamane

(2012) Factors affecting independence in eating among elderly with Alzheimer’s disease. Geriatr Gerontol Int 12, 481–490.

Takada

, Tanaka

, Hasegawa

, Sugiyama

, Yoshiike

(2017) Grouped factors of the ‘SSADE: Signs and symptoms accompanying dementia while eating’ and nutritional status-An analysis of older people receiving nutritional care in long-term care facilities in Japan. Int J Older People Nurs 12, 10.1111/opn.12149.

Kao

, Lin

, Wu

, Lin

, Liu

(2016) Effectiveness of different memory training programs on improving hyperphagic behaviors of residents with dementia: A longitudinal single-blind study. Clin Interv Aging 11, 707–720.

Ching Lim

(2018) Eating disorders and dementia. J Alzheimers Dis Parkinsonism 8, 22.

10.

Natalwala

, Potluri

, Uppal

, Heun

(2008) Reasons for hospital admissions in dementia patients in Birmingham, UK, during 2002– 2007. Dement Geriatr Cogn Disord 26, 499–505.

11.

Meijers

, Schols

, Halfens

(2014) Malnutrition in care home residents with dementia. J Nutr Health Aging 18, 595–600.

12.

Gillette-Guyonnet

, Lauque

, Ousset

(2005) Nutrition et maladied’ Alzheimer [Nutrition and Alzheimer’s disease]. Psychol Neuropsychiatr Vieil 3, S35–S41.

13.

Holm

, Söderhamn

(2003) Factors associated with nutritional status in a group of people in an early stage of dementia. Clin Nutr 22, 385–389.

14.

Majer

, Simon

, Csiba

, Kardos

, Frecska

, Hortobágyi

(2019) Behavioural and psychological symptoms in neurocognitive disorders: Specific patterns in dementia subtypes. Open Med 14, 307–316.

15.

Manthorpe

, Watson

(2003) Poorly served? Eating and dementia. J Adv Nurs 41, 162–169.

16.

Shinagawa

, Honda

, Kashibayashi

, Shigenobu

, Nakayama

, Ikeda

(2016) Classifying eating-related problems among institutionalized people with dementia. Psychiatry Clin Neurosci 70, 175–181.

17.

Rivière

, Gillette-Guyonnet

, Andrieu

, Nourhashemi

, Lauque

, Cantet

, Salva

, Frisoni

, Vellas

(2002) Cognitive function and caregiver burden: Predictive factors for eating behaviour disorders in Alzheimer’s disease. Int J Geriatr Psychiatry 17, 950–955.

18.

Sopina

, Sørensen

(2018) Decision modelling of non-pharmacological interventions for individuals with dementia: A systematic review of methodologies. Health Econ Rev 8, 8.

19.

Abdelhamid

, Bunn

, Copley

, Cowap

, Dickinson

, Gray

, Howe

, Killett

, Lee

, Li

, Poland

, Potter

, Richardson

, Smithard

, Fox

, Hooper

(2016) Effectiveness of interventions to directly support food and drink intake in people with dementia: Systematic review and meta-analysis. BMC Geriatr 16, 26.

20.

Bunn

, Abdelhamid

, Copley

, Cowap

, Dickinson

, Howe

, Killett

, Poland

, Potter

, Richardson

, Smithard

, Fox

, Hooper

(2016) Effectiveness of interventions to indirectly support food and drink intake in people with dementia: Eating and Drinking Well IN dementiA (EDWINA) systematic review. BMC Geriatr 16, 89.

21.

Kai

, Hashimoto

, Amano

, Tanaka

, Fukuhara

, Ikeda

(2015) Relationship between eating disturbance and dementia severity in patients with Alzheimer’s disease. PloS One 10, e0133666.

22.

Amella

, Grant

, Mulloy

(2008) Eating behavior in persons with moderate to late-stage dementia: Assessment and interventions. J Am Psychiatr Nurses Assoc 13, 360–367.

23.

Hsiao

, Chao

, Wang

(2013) Features of problematic eating behaviors among community-dwelling older adults with dementia: Family caregivers’ experience. Geriatr Nurs 34,, 361–365.

24.

Xing

, Hu

, Zhou

, Zhu

, Gu

, Zhang

, Chen

(2020) Promoting the transformation of evidence into clinical practice: Making and writing evidence summary. J Nurses Train 35, 1129–1132.

25.

Munn

, Lockwood

, Moola

(2015) The development and use of evidence summaries for point of care information systems: A streamlined rapid review approach. Worldviews Evid Based Nurs 12,, 131–138.

26.

Aromataris

, Fernandez

, Godfrey

, Holly

, Khalil

, Tungpunkom

(2015) Summarizing systematic reviews: Methodological development, conduct and reporting of an umbrella review approach. Int J Evid Based Healthc 13,, 132–140.

27.

Joanna Briggs Institute. JBI Manual for Evidence Synthesis. https://synthesismanual.jbi.global, Accessed on July 30, 2021.

28.

JBI Levels of Evidence and Grades of Recommendation Working Party. Supporting Document for the JBI Levels of Evidence and Grades of Recommendation. https://jbi.global/sites/default/files/2020-07/Supporting_Doc_JBI_Levels_of_Evidence_Grades_of_Recommendation.pdf, Accessed on July 30, 2021.

29.

Keller

, Slaughter

(2016) Training programmes and mealtime assistance may improve eating performance for elderly long-term care residents with dementia. Evid Based Nurs 19, 32.

30.

Leah

(2016) Supporting people with dementia to eat. Nurs Older People 28, 33–39.

31.

Sheppard

, McArthur

, Hitzig

(2016) A systematic review of Montessori-based activities for persons with dementia. J Am Med Dir Assoc 17, 117–122.

32.

Jensen

, Padilla

(2017) Effectiveness of environment-based interventions that address behavior, perception, and falls in people with Alzheimer’s disease and related major neurocognitive disorders: A systematic review. Am J Occup Ther 71, 7105180030p1–7105180030p10.

33.

O’Neil-Pirozzi

, Hilton

(2018) Limited evidence to support effectiveness of interventions directly targeting food and drink intake in individuals with dementia. Evid Based Commun Assess Interv 12, 77–80.

34.

Woodbridge

, Sullivan

, Harding

, Crutch

, Gilhooly

, McIntyre

, Wilson

(2018) Use of the physical environment to support everyday activities for people with dementia: A systematic review. Dementia (London) 17, 533–572.

35.

Herke

, Fink

, Langer

, Wustmann

, Watzke

, Hanff

, Burckhardt

(2018) Environmental and behavioural modifications for improving food and fluid intake in people with dementia. Cochrane Database Syst Rev 7, CD011542.

36.

Fetherstonhaugh

, Haesler

, Bauer

(2019) Promoting mealtime function in people with dementia: A systematic review of studies undertaken in residential aged care. Int J Nurs Stud 96, 99–118.

37.

Yakimicki

, Edwards

, Richards

, Beck

(2019) Animal-assisted intervention and dementia: A systematic review. Clin Nurs Res 28, 9–29.

38.

Borders

, Blanke

, Johnson

, Gilmore-Bykovskyi

, Rogus-Pulia

(2020) Efficacy of mealtime interventions for malnutrition and oral intake in persons with dementia: A systematic review. Alzheimer Dis Assoc Disord 34, 366–379.

39.

Backhouse

, Dudzinski

, Killett

, Mioshi

(2020) Strategies and interventions to reduce or manage refusals in personal care in dementia: A systematic review. Int J Nurs Stud 109, 103640.

40.

Faraday

, Abley

, Beyer

, Exley

, Moynihan

, Patterson

(2021) How do we provide good mealtime care for people with dementia living in care homes? A systematic review of carer-resident interactions. Dementia (London) 20, 3006–3031.

41.

Hsu

, Lin

, Wu

(2017) The effects of spaced retrieval training in improving hyperphagia of people living with dementia in residential settings. J Clin Nurs 26, 19–20.

42.

Ohmana

, Savikko

NRN

, Strandberg

, Kautiainen

, Raivio

, Laakkonen

, Tilvis

, Pitkala

(2017) Effects of frequent and long-term exercise on neuropsychiatric symptoms in patients with Alzheimer’s disease – Secondary analyses of a randomized, controlled trial (FINALEX). Eur Geriatr Med 8, 153–157.

43.

Reverté-Villarroya

, Zaragoza-Brunet

, Matamoros-Obiol

, Inglada-García

, Escalante-Arroyo

, Forcadell-Ferreres

, Esteve-Belloch

(2020) Individual therapeutic-cognitive intervention of the psychological and behavioural symptoms of patients with dementia (PRESTA study). Rev Cient Soc Esp Enferm Neurol 52, 7–18.

44.

Krolak-Salmon

, Roubaud

, Finne-Soveri

, Riolacci-Dhoyen

, Richard

, Rouch

, Leperre-Desplanques

, Dauphinot

(2016) Evaluation of a mobile team dedicated to behavioural disorders as recommended by the Alzheimer Cooperative Valuation in Europe joint action: Observational cohort study. Eur J Neurol 23, 979–988.

45.

Batchelor-Murphy

, McConnell

, Amella

, Anderson

, Bales

, Silva

, Barnes

, Beck

, Colon-Emeric

(2017) Experimental comparison of efficacy for three handfeeding techniques in dementia. J Am Geriatr Soc 65, e89–e94.

46.

Hansen

, Frøiland

, Testd

(2018) Porcelain for All - a nursing home study. Int J Health Care Qual Assur 31, 662–675.

47.

Simmons

, Coelho

, Sandler

, Schnelle

(2018) A quality improvement system to manage feeding assistance care in assisted-living. J Am Med Dir Assoc 19, 262–269.

48.

Simmons

, Coelho

, Sandler

, Schnelle

(2018) A system for managing staff and quality of dementia care in assisted living facilities. J Am Geriatr Soc 66, 1632–1637.

49.

Sulmont-Rossé

, Gaillet

, Raclot

, Duclos

, Servelle

, Chambaron

(2018) Impact of olfactory priming on food intake in an Alzheimer’s disease unit. J Alzheimers Dis 66, 1497–1506.

50.

Gaspar

, Westberg

(2020) Evaluation of the Montessori-Inspired Lifestyle^® as the foundation of care in assisted living memory care. J Gerontol Nurs 46, 40–46.

51.

Visscher

, Battjes-Fries

, van de Rest

, Patijn

, van der Lee

, Wijma-Idsinga , Pot

, Voshol

(2020) Fingerfoods: A feasibility study to enhance fruit and vegetable consumption in Dutch patients with dementia in a nursing home. BMC Geriatr 20, 423.

52.

Jung

, De Gagne

, Lee

, Choi

, Chung

(2020) Feasibility of a mobile meal assistance program for direct care workers in long-term care facilities in South Korea. Clin Interv Aging 15, 2019–2029.

53.

Lee

, Lee

, McConnell

(2021) Mealtime caregiving approaches and behavioral symptoms in persons living with dementia: A longitudinal, observational study. BMC Nurs 20, 104.