Making the Most of Mealtimes (M3): protocol of a multi-centre cross-sectional study of food intake and its determinants in older adults living in long term care homes

Long term care (LTC), also referred to as care or nursing homes, is a residential option that provides for the instrumental and basic activities of daily living of their clients. Older adults??65 years of age) are the typical resident and dementia is a common condition that is managed in this setting. Poor food and fluid intake is the primary cause for long term care (LTC) malnutrition [1] resulting in falls, poor function, depression, mortality, poor wound care and decreased quality of life for residents [1–7]. Resident energy intake has been reported to be 1500 kcal/day or less [1, 8, 9] and up to 70% have lower than recommended intakes of protein, fibre, calcium, magnesium, zinc, and vitamins E, C, B6, thiamine, riboflavin, niacin and/or folate [1, 8, 9]. Improving food intake for residents in LTC has been identified as a priority by researchers, decision makers and clients (i.e., residents, families) to promote health and quality of life [10]. Yet, rigorous collection of dietary intake data across a diverse and large sample to truly understand prevalence of poor food intake in LTC is limited. For example, some researchers have used a single day of recalled (by resident or care staff) food intake to represent consumption patterns, resulting in potentially flawed conclusions on the adequacy of food intake [11–13]. Smaller studies with weighed food records, the gold standard, have demonstrated large inter-individual differences [12], but are commonly limited in measurement of covariates and the ability to model these due to the small sample size [12, 14]. Additionally, detailed descriptions of the food intake profile are often missing [15]. Understanding the extent of poor food intake and which nutrients are poorly consumed, is relevant in identifying strategies to support intake for residents in LTC. For example, if vitamin D is low, supplementation may be the best strategy, whereas if energy and a wide variety of micronutrients are low, interventions that increase food intake, such as quality eating assistance, may be required.

LTC malnutrition is both preventable and treatable; [1, 16] successful interventions can improve the health and function of residents [9, 17–19]. Innovative, multi-level (i.e., targeting residents, staff, dining environment) interventions are needed to address the problem of inadequate food intake and consequent malnutrition in LTC, as causes are likely to interact [20]. To develop interventions, a good understanding of the problem and its determinants is needed. At this point, we have a limited understanding of the problem of inadequate food intake in LTC worldwide, and analyses have often been focused on determinants that cannot be changed [21], such as dementia. However, persons with dementia often require some level of eating assistance [21] and this determinant can be modified; staff can be trained on eating assisting techniques [22], and sufficient time and accommodations can be provided to promote eating independently (e.g., finger foods) [23]. Thus investigations focused on describing and determining the relative importance of these amenable factors is needed to make advancements in improving food intake and thus malnutrition in LTC.

Research to date has failed to use a comprehensive conceptual framework to understand and intervene on the multi-level and inter-related determinants of food intake in LTC residents [7, 15, 20, 24]. The proposed study is built upon the M3 concept [25], which has its origins in the Social Ecological Model [26], the Five Aspects of Meal Model [27], and the Mealtimes as ‘Active Processes’ substantive theory [28]. The M3 conceptual model suggests that multi-level determinants influence food and fluid intake of residents. Specifically, it is hypothesized that regional government regulations and standards, LTC home policies and features, staffing levels, knowledge and practices, and resident characteristics will influence food intake. Within each of these levels, factors in three domains of Meal Quality (nutritious, appealing food); Meal Access (oral health, swallowing problems, eating ability); and Mealtime Experience (dining environment) are relevant to food intake. This model drives the design, data collection, analysis and interpretation of results in this project [25].

Meal quality is operationally defined as food and fluid offerings that are preferred and culturally appropriate, appealing (smell, taste, appearance), served at an acceptable temperature, and meeting the nutritional requirements (i.e., Dietary Reference Intake [DRI]) of residents. Quality food is favoured over oral nutritional supplements or meal replacements by residents and family for meeting nutritional needs and enhancing quality of life [29]. Government policies and home-level practices with respect to menu planning, choice of commercial or in-house food products, food variety, food budget, and the mandated role and time allocated to clinical dietitians and other allied health professionals (e.g., occupational therapists and speech-language pathologists) all have the potential to influence the types and quality of food provided in LTC, yet we know little about these determinants. Evidence to date suggests that increased funding for food improves energy intake in LTC residents [15]. A shorter menu cycle length is also associated with increased malnutrition [30]. It has been noted that LTC menus have been found to be low in protein and micronutrients [1, 31] and are likely insufficient to support adequate health and function [32]. Of specific concern are modified-texture foods (e.g., pureed) that are frequently provided to persons with dysphagia (i.e., swallowing impairment) to manage their swallowing difficulty. These foods often have poor sensory appeal and low nutrient density [33, 34].

Meal access is operationalized as those factors that influence food/fluid access and specifically: the availability of food (meal timing, between meal snacks); the ability to taste and smell; dysphagia; dentition and oral health status; capacity with self feeding, time it takes to eat, and any eating challenges; and the time provided for and quality of eating assistance. Taste and smell are commonly impaired in older adults and specifically in those with dementia [35, 36], potentially impairing food intake and nutritional status [36]. Dysphagia is a significant comorbidity that influences food intake and malnutrition [21], although prevalence in LTC is elusive. Additionally, dental factors such as loose teeth, poorly fitting dentures, and/or poor oral health may contribute to limited food intake, preference for foods low in micronutrients but easier to eat (e.g., ice cream, commercial puddings, mashed potatoes) or prescription of modified-texture foods (e.g., minced meat) or diets [37]. Oral health status of LTC populations is rarely investigated, but problems are prevalent with estimates of 37% of residents reporting a dry mouth and 51% having untreated cavities [38]. The resulting pain and distress of poor oral health has been shown to affect food intake and malnutrition [21, 37]. In addition, difficulty accessing food due to packaging, lids, and dishes causes stress at mealtimes and is associated with poor food intake [30, 39]. Agitation and decreased ability to eat independently also result in decreased consumption [40]. Requiring assistance commonly results in inadequate food intake [7, 21, 39]. Assistance with food intake ranges from setting up the meal and opening packages, to encouragement, to partial assistance with some foods or total assistance for those unable to eat independently. For those requiring assistance, food access is influenced by the number of qualified, trained care staff, the number of residents requiring assistance, the type of assistance needed, and the presence of family, volunteers or paid meal helpers who can provide one-on-one assistance [22]. For instance, a staffing ratio of 3:1 (residents:staff) as compared to 5:1 significantly improves energy intake for residents in LTC settings [15]. It is anticipated that as many as 50% of LTC residents requiring total assistance could consume at least some of their food independently if changes were made to the environment and supports were put in place to promote resident autonomy [23, 41], such as sufficient time to eat [22].

Mealtime experience is operationalized as the physical and psychosocial mealtime environment and mealtime processes that can influence food intake [42]. Mealtimes can be the highlight of the day for residents, providing opportunities for social interaction as well as development of relationships with care providers and tablemates [43]. Apathy and depression have been found to be independently associated with weight loss in LTC residents [44] and may be influenced by a negative mealtime experience [42]. Two theories specific to persons with dementia (Mealtimes as Active Processes in LTC and the Life Nourishment Theory), demonstrate the importance and potential influence of the psychosocial environment on food intake [28, 45]. It is hypothesized that positive social connections and honouring individual identities (e.g., food preferences) at mealtimes will promote food intake and quality of life. Consistent with these theories, family style dining provides greater opportunity for social interaction and choice of food offerings and has been shown to improve energy intake in LTC [46–48]. The Eden Alternative®, which is a ‘household’ model providing care to a small number of residents in a resident-centred and homelike setting has some benefit with respect to maintaining body weight [49]. This association may be linked to tailored individualization of care including participation in food preparation, meal choices and honoring food preferences. Social models of care that recognize and encourage staff-resident relationships and resident-centered care are seen as the preferred approach to promote residents quality of life in LTC [49, 50]. To date, the relational and resident-centeredness of mealtimes has been qualitatively explored [43, 47, 51, 52], but has yet to be quantitatively assessed to determine its association with food intake. The physical environment can also influence eating [41, 53] and ‘homelike’ environments with music, decorations, and table dressings have been shown to improve residents’ food intake [46, 54]. Food being plated on the unit, rather than in the kitchen, enhances consumption [54]; residents see and smell the food and can interact with staff about choosing their meal and desired portions at the point of service. On the other hand, bulk delivery systems can create noise and distraction [39].

In sum, we lack rigorously measured dietary consumption data to understand the prevalence of inadequate food intake in LTC residents. Multi-level determinants that address meal quality, meal access and mealtime experience have yet to be collectively assessed to understand their relevance to food intake. Without this knowledge, we cannot undertake the development of effective interventions and policy directives to address the high prevalence of LTC malnutrition and its costly human and system consequences. Specifically, determinants amenable to an intervention, which could have strong potential to improve food intake, have rarely been explored. The objectives of this cross-sectional, multi-site study were to:

1. 1.

   establish the prevalence of inadequate energy, protein, micronutrient and fluid intake of residents in LTC, across and within four Canadian provinces, and;

2. 2.

   identify the independent and inter-related associations between multi-level (i.e., resident, unit, LTC home, province) determinants of energy and protein intake of residents in Canadian LTC homes.

The purpose of this paper is to fully describe the protocol used to rigorously measure food intake and to examine determinants of food and fluid intake among older adults participating in the Making the Most of Mealtimes (M3) study.