Introduction

Feeding assistance is a commonly employed clinical technique. Cognitive, physical, and/or sensory impairments experienced by many patients often result in functional limitations that necessitate such assistance. Data suggest that at least 25%-55% of all nursing home residents require some level of feeding assistance (Dey, 1997; Harris-Kojetin et al., 2016; Steele, Greenwood, Ens, Robertson, & Seidman-Carlson, 1997). Among hospitalized older adults, up to 70% require assistance (Tsang, 2008), with one in three elderly acute stroke patients being dependent for eating (Unosson, Ek, Bjurulf, von Schenck, & Larsson, 1994). Over 50% of stroke rehabilitation patients have been found to be unable to eat without assistance, with nearly half being totally dependent for feeding (Westergren, Karlsson, Andersson, Ohlsson, & Hallberg, 2001). Finally, 17% of informal caregivers nationwide provide feeding assistance at home (Donelan et al., 2002).

Although often used to alleviate mealtime difficulties and prevent eating risks, feeding assistance is associated with a variety of potentially harmful consequences. Feeding dependency has been associated with malnutrition (Chavarro-Carvajal, Reyes-Ortiz, Samper-Ternent, Arciniegas, & Gutierrez, 2015), increased aspiration pneumonia risk (Langmore, Skarupski, Park, & Fries, 2002; Langmore et al., 1998), eating difficulties (Poels, Brinkman-Zijlker, Dijkstra, & Postema, 2006), loss of body cell mass (Unosson et al., 1994), and weight loss (Tamura, Bell, Masaki, & Amella, 2013). The loss of feeding-related proprioception can also alter typical eating-related mouth patterns (Shune, Moon, & Goodman, 2016). Dependence for activities of daily living, including eating, has also been associated with decreased subjective well-being, self-confidence, sense of control and value, and quality of life and increased feelings of conflict, struggle, and stigmatization (Iwarsson & Isacsson, 1998; Kayser-Jones & Schell, 1997; Strandberg, Norberg, & Jansson, 2003). Such physical and psychosocial risks may be partially attributable to the need to reconfigure a new "normal" pattern of eating that, for many, requires a renegotiation of roles based on co-constructed physical and social patterns between the patient and care provider (Martinsen, Harder, & Biering-Sorensen, 2008).

Eating and food-related activities are intricately tied to identity, reinforcing individuals' connections to their social networks and to themselves (Mintz & Du Bois, 2002; Plastow, Atwal, & Gilhooly, 2015). Furthermore, eating-related activity is an integral marker of "normality" in the presence of chronic illness, and eating difficulties can increase the impact of disease severity and influence functional outcomes (Klinke, Wilson, Hafsteinsdottir, & Jonsdottir, 2013; Perry & McLaren, 2003). Thus, it is not surprising that the transition toward feeding dependence is challenging. In addition to coping with a new medical diagnosis that prompted the change in functioning (e.g., an acute condition [stroke or spinal cord injury], a degenerative progression [dementia]), these individuals find themselves suddenly requiring help with tasks they have been completing independently since childhood. There is also a risk that such help does not correspond with an individual's own self-image, further exacerbating the situation and leading to decreased life satisfaction (Bogart, 2014; Martinsen et al., 2008).

It is unclear how to best support this transition in order to optimize the eating experience and alleviate the negative consequences associated with feeding assistance. It has been suggested that a model of interdependence, rather than dependence, is important to emphasize (Martinsen et al., 2008; Williams & Wood, 1988). As part of this model, individuals' sense of control is enhanced when care is provided in the context of a mutual or reciprocal relationship, supporting the reconciliation between themselves and their physical dependencies (Bogart, 2014; Cordingley & Webb, 1997; Roe, Whattam, Young, & Dimond, 2001; Sixsmith, 1986). In other words, the co-construction of a new feeding process that mutually relies on both individuals (interdependence), rather than a feeding process being driven by the "nondependent" care provider (dependence), appears to support this transition. However, much of this literature base has been with older adults, particularly older adults already receiving care. It is not known whether this model can be appropriately generalized to others. As much of this research is becoming dated, it is also unclear whether it continues to represent the current generation of older adults. Furthermore, younger adults make up a substantial proportion of patients who may suddenly require feeding assistance, and the average age of certain "age-related diseases" (e.g., stroke) continues to decrease (Chen, He, & DeVivo, 2016; Kissela et al., 2012; Mozaffarian et al., 2016). Perceptions regarding receiving care likely also change once individuals begin receiving care and thus do not offer insight into the premorbid, baseline perceptions that patients already possess. Relatedly, impairments in language and cognition that often co-occur with feeding/nutritional impairments and increased dependence may limit the ability for these individuals to express their opinions once there is a need for feeding assistance (e.g., Roque, Salva, & Vellas, 2013; Wotton, Crannitch, & Munt, 2008).

To proactively support the transition to feeding assistance, it is necessary to characterize current attitudes toward receiving such assistance across the lifespan prior to the necessity of care. The purpose of this exploratory qualitative study was to characterize these attitudes among healthy younger and older adults following an assisted feeding experience.

Methods

Participants

Participants (N = 36) were self-rated healthy, community-dwelling adults: 17 younger adults (M_age = 24.2 years, SD = 3.8; 10 male) and 19 older adults (M_age = 76.0 years, SD = 4.8; 11 male). Demographic data are presented in Table 1. All participants had normal or corrected vision and hearing; normal oral motor and upper extremity function; normal oral, facial, and upper extremity sensation; a Mini-Mental State Examination score of >=26; and a negative history of confounding medical, neurological, or musculoskeletal disease and medication use that could influence neurological and/or motor function. Purposive sampling techniques were used to recruit participants, and all participants were recruited from the local community via posted flyers and university e-mail listservs. All participants signed written informed consent prior to participation.

TABLE 1 Participant demographics

Task Procedures and Data Set

The data for the current study were collected as part of a larger mixed-methods study examining the impact of age and preoral (proprioceptive, exteroceptive) sensation on eating behaviors that is described in detail elsewhere (Shune & Moon, 2016; Shune et al., 2016). The institutional review board at the participating institution approved all procedures. Briefly, participants consumed food and drink under various sensory loss conditions. These included self-feeding and assisted feeding conditions given the presence/absence of vision and audition. After eating under the multiple conditions, a subset of the participants from the larger study participated in semistructured interviews that lasted between 10 and 30 minutes regarding their opinions on and feelings toward the experience. All task procedures, including the interviews, were conducted in a research laboratory within a university building and were video-recorded for later analysis. All interviews were conducted by the author (S. S.), a certified speech-language pathologist and doctoral candidate at the time of data collection, who was not known to the participants prior to their participation.

The interviews were guided by a list of questions to be covered (see Table 2); however, conversations were allowed to stray from the guide if appropriate and relevant. Participants also frequently commented throughout the study procedures (i.e., between tasks). Although no specific attempt was made to elicit feedback during the simulated eating experience itself, these informal comments were included in subsequent analyses as further described below. Data collection continued until the topic was saturated; in other words, data saturation was reached when no new themes were revealed in the interviews.

TABLE 2 Semistructured interview guide

Discourse was transcribed and analyzed across the entire session for both the participant and the researcher. Transcription was completed using a three-stage process (Duff, Hengst, Tranel, & Cohen, 2008). First, audio portions of the videotaped sessions, including utterances and audible sounds, were transcribed with unintelligible utterances marked. Next, the original transcriber added gestures that contributed unique content (e.g., facial expressions indicating displeasure) from the video and made any corrections to the audio content of the transcript. Lastly, a second transcriber watched the video, with the original transcriber as needed, to generate a final transcript that reflected any corrections or additions made according to discussion between the two.

Data Analysis

Interview data were analyzed following an inductive thematic analysis procedure (Braun & Clarke, 2006). The qualitative portion of this study was designed to understand participants' experiences with eating under different sensory conditions, particularly when being fed. In other words, the objective was to provide a participant-informed description of this experience or description of the phenomena itself. Therefore, more generic qualitative research methods ("qualitative description") were used in which an explicit set of theoretical assumptions in the form of the known qualitative methodologies did not guide the analysis (e.g., Sandelowski, 2010). Qualitative description is not without a guiding philosophy and draws from the general tenets of naturalistic inquiry (Sandelowski, 2000).

All analysis was performed by a team of two research assistants and the author. Following initial readings of the transcripts, a subset of data-derived thematic codes relevant to the purpose of the current investigation was identified and described. Manual coding of the entire data set then proceeded using a three-phase consensus coding procedure. First, a primary coder identified all eating condition-related comments, including the beginning and ending boundaries of each comment, and characterized the sentiment (positive/negative/neutral) and sensation (physical/emotional) of each comment. Physical sensations were defined as those tied to the body or the state of the body (e.g., moving the mouth differently; swallowing more times per bite). Emotional sensations were defined as those tied to a mental state or feeling (e.g., feeling like a child when being fed; sympathizing with nursing home patients). Individual comments could be coded with more than one sensation, but each comment received only one sentiment code. It was also documented which eating condition(s) each comment was referring to (e.g., being fed in general, being fed in the absence of vision and hearing). Next, a secondary coder reviewed the transcripts, noting agreements and disagreements with the first-phase boundaries and codes. The two initial coding passes resulted in 88.9% agreement between the two coders. Coding differences were primarily attributable to episode boundaries and the addition of new episodes coded in the second pass, with few disagreements on sentiment and sensation codes. Disagreements between the first and second coding passes were resolved through discussion and consensus. The boundaries and codes were finalized by the primary and secondary coders with a consensus coder. In order to further characterize participants' attitudes and perceptions related to feeding assistance, all comments were then analyzed for common themes across responses.

Multiple steps were completed to ensure data fidelity. To promote dependability and trustworthiness of the findings, weekly peer debriefing and the creation of an audit trail or a step-by-step process of data collection and analysis procedures (Creswell, 2008) were employed. All research team members completed code cross-checking, which allowed for cross-validation of coders' data interpretations. Furthermore, the majority of coding was completed by the two research assistants who were considered to be unbiased coders as they did not conduct any of the interviews and they were not involved in study design. Although member checking was not completed, reflective listening and summarizing techniques were used throughout the semistructured interviews themselves to further ensure data accuracy.

Results

Overall, participants made 112 unique comments about the eating conditions (53 and 59 comments by younger and older adults, respectively). Of these, 70 addressed being fed and/or self-feeding (30 and 40 comments by younger and older adults, respectively) and were selected for further analysis.

Feeding assistance comments produced by all participants were predominately physical (82.8% and 79.5% of the total comments for younger and older adults, respectively). For the purposes of the current analyses, sentiments were classified based on assisted feeding (i.e., a positive comment about self-feeding was coded as a negative comment about assisted feeding). Although similar patterns were observed in the distribution of sensations between the groups, differences emerged with regard to sentiments. Older adults were more likely to make neutral comments (65.0% and 36.7% of the total comments for older and younger adults, respectively), whereas younger adults were more likely to make negative comments (60.0% and 32.5% of the total comments for younger and older adults, respectively). One participant in each group made a positive comment about assisted feeding (3.3% and 2.5% of the total comments for younger and older adults, respectively).

Five specific themes were identified within the three overarching sentiments associated with feeding assistance (see Table 3). Negative comments subdivided into two themes, one related to independence and the other related to difficulties associated with being fed. Overall, both younger and older adults described many difficulties encountered while being fed as compared to self-feeding that frequently related to decreased naturalness and coordination (e.g., "I just, like, know my own speed of how to angle my glass[horizontal ellipsis]it wasn't as natural"; "I didn't know if I was supposed to drink all of it or when to stop[horizontal ellipsis]"; "It made it more difficult[horizontal ellipsis]because you had to kinda move [your mouth] differently than you do normally"). These often stemmed from differences between how they would normally feed themselves or "want to be fed" and how they were actually fed (e.g., "You had to do something different than how you prefer"). However, younger adults were more likely than older adults to discuss independence, commenting on how dependency was difficult (e.g., "I don't think I can ever be dependent on somebody"; "I wasn't really in control") and how it evoked negative emotions (e.g., "I felt like an infant").

TABLE 3 Emerging themes related to feeding assistance based on 70 comments stated by younger and older adult participants

Neutral comments also subdivided into two themes, one encompassing named differences related to feeding status and one encompassing a general expression of "feeling fine." Participants frequently commented that being fed felt fine, providing no further description. They also described differences between being fed as compared to self-feeding that did not appear to have any underlying positive or negative sentiment (e.g., "I felt like I swallowed less [number of] times when I fed myself[horizontal ellipsis]just different"). Overall, older adults were more likely to make neutral comments of both types as compared to younger adults. Of interest, although younger adults indicated strongly negative attitudes toward being fed, including when thinking of their future selves as previously described, many older adults more neutrally indicated that feeding assistance was likely to eventually occur (e.g., "I was just thinking that, I'm 85, people are going to be feeding me one day"; "I mean that is a long time, seven or some years before I will have to have that happen").

Two participants, one younger adult and one older adult, indicated some level of enjoyment with being fed ("It was nice to be fed"; "It was fun"). Both of these individuals also made additional comments that fell into the other sentiment categories.

Discussion

This study aimed to investigate attitudes toward feeding assistance in healthy younger and older adults, particularly after an assisted feeding experience. Overall, the majority of comments made by both groups related to physical aspects of feeding and eating. Furthermore, although older adults were more likely to make neutral comments about assisted feeding, younger adults were more likely to make negative comments, especially as related to independence.

The findings that younger adults viewed feeding assistance more negatively may be reflective of broader negative views toward aging and disability. According to aging stereotypes, older adulthood is viewed as a period of frailty, disability, illness, and dependence. In Western culture, the independence and vitality of early and middle adulthood is cherished (e.g., "the young grow up") whereas older adulthood is framed within the metaphor of childhood and viewed as a regression (e.g., "the old go downhill"). Research has indicated that 80% of younger adults fear old age, with two-thirds citing loss of independence and poor health as primary concerns (Hockey & James, 1993). Older adults and aging simulations can induce anxiety in younger adults, potentially due to the reminder of what may or will happen to everyone eventually (Greenberg, Schimel, & Martens, 2002; Rittenour & Cohen, 2016). Overall, younger adults demonstrate explicit preferences for younger as compared to older adults (Chopik & Giasson, 2017). Given this perceived link between aging, disability, and ultimate mortality, it is not surprising that younger adults would be more resistant to increasing dependency. It is plausible to suggest that this link strongly contributes to younger adults' increased desire for independence, reflected here as an increase in independence-related negative comments.

Although older adults view independence as a priority, their more neutral sentiments toward receiving feeding assistance observed here may be attributable to changes in life views and social behavior. Older adults admit that some degree of dependence is inevitable with age, and their desire for independence is mostly an "illusion" (Sixsmith, 1986). Such a state of inevitability was reflected in the older adults' comments; even when participants did not view themselves as needing feeding assistance in the near future, many indicated that it would (definitively) happen one day. Furthermore, it has been proposed that perceived limitations on time with advancing age (mortality) result in a shift toward seeking out feeling-related goals linked to the obtainment of more immediate emotional gratification (socioemotional selectivity theory; Carstensen, 1992; Carstensen, Fung, & Charles, 2003). Receiving assistance provides increased opportunities for social interaction and physical contact, two activities that the socioemotional selectivity theory posits become more meaningful in older adulthood given their ability to provide satisfaction in the "here and now." Thus, it is possible that beyond just accepting the inevitability of feeding assistance to meet a physical need, older adults may be more tolerant of such assistance given that it can serve an emotionally meaningful purpose. Although both younger and older adults value independence, older adults may emotionally benefit when accepting some level of dependence. In other words, receiving feeding assistance may actually be beneficial for older adults beyond just meeting their immediate physical needs and opportunity for increased social engagement during mealtimes should thus be emphasized.

It is essential to understand how independence is defined. Two concepts that emerged in the comments of the younger adults here and that are reflective of previous findings in older adults are (1) the ability to perform a task alone and (2) autonomy and decision-making capacity (Bell & Menec, 2015; Sixsmith, 1986). In the current study, all participants experienced loss of independence for the transport of food to the mouth and the physical feeding process. However, the younger adults in particular expressed negative sentiments regarding loss of freedom and autonomy or their decision-making capacity (e.g., it was difficult to not be in control; it felt like being an infant). The ability to retain control through choice and reciprocity in the relationship between a care provider and care recipient is part of the model of interdependence thought to ease the transition to receiving assistance (Roe et al., 2001; Sixsmith, 1986). The results here suggest that, particularly for younger adults, addressing feelings of lost autonomy is critical for successful interdependence. This is especially important as younger adults have the potential for greater lifetime burden of disability.

Eating is intricately tied to an individual's identity. It is therefore necessary that the process of assisted feeding led to an individualized, co-constructed, and coordinated pattern between the patient and the care provider. This meets the needs of patients wanting their own values around meals to be realized; they want to recognize themselves in mealtime routines and rituals (Martinsen et al., 2008; Milte et al., 2017). Such a theme was reflected among both the younger and older adults (e.g., "I wasn't really in control of amount or[horizontal ellipsis]speed at which I wanted to eat my food"; "You had to do something different than how you prefer"). This desire further ties into the maintenance of aspects of independence despite requiring feeding assistance: By maintaining freedom of choice (e.g., what and when to eat, how food is presented), the consequence of feeding assistance does not become dependence, but rather a new pattern of interdependence. Here, the patient is fully recognizable as a mutually important member of the feeding process. Higher disability self-efficacy is associated with increased life satisfaction (Bogart, 2014). Despite needing extensive assistance, individuals can still have high disability self-efficacy if they believe in their abilities to participate in and manage desired tasks by, for example, effectively directing a caregiver and making their own mealtime decisions. However, although caregivers frequently want to individualize feeding-related care and often do make these adjustments, there frequently is tension between the importance of these tasks and the low priority they are assigned as compared to other care-related responsibilities (Martinsen & Norlyk, 2012). Thus, the integration of individuals' own eating-related patterns and preferences into assisted feeding is not only essential but must also be fundamentally valued.

A few limitations of the current study are worth noting. First, participants were interviewed specifically about eating conditions that were presented in the context of a research study, which did not replicate a completely natural mealtime setting. Furthermore, all participants were fed by a research assistant who was not trained in providing feeding assistance. Thus, it is possible that the experience did not fully capture a dependent feeding situation and that other feelings related to feeding assistance were not evoked (e.g., most comments were physical in nature). To build on this, future studies should use a more naturalistic setting and/or focus on additional attitudes toward feeding assistance (e.g., hone in on the emotional experience).

It is also necessary to discuss the limitations involved with using a simulated experience as a proxy for the real-world experiences of individuals who have aged into or acutely sustained disability. The results of the current study may not accurately represent the range of attitudes that individuals who require feeding assistance have toward the process, particularly when feeding assistance becomes a necessity for obtaining adequate nutrition/hydration. However, the intention of the current investigation was to characterize the premorbid attitudes that can contribute to shaping these individuals' later experiences and that may (partially) explain the previously identified negative consequences of feeding assistance. As recognized in the International Classification of Functioning, Disability and Health, personal factors that are not a part of a health condition or health state play a role in disability and can ultimately impact intervention outcomes (World Health Organization, 2001). These factors, including premorbid attitudes, likely play an influential role in how an individual grieves, copes, and adapts given an acquired disability. Thus, this exploratory study provides a first step in broadly outlining those factors that may interact with assisted feeding, providing a foundation for future studies in this crucial area.

Conclusion and Implications for Nursing Practice

In summary, this study suggests that healthy younger and older adults differ in their baseline attitudes toward feeding assistance; whereas older adults view the process more neutrally, younger adults have a more negative view of such assistance. Such findings indicate the need to further investigate the potentially differing clinical needs between younger and older adults who require feeding assistance. It may be especially important for younger adults to disentangle the perceived link between aging, mortality, and disability in order to emphasize continued autonomy and empowerment in the presence of feeding dependence. Defining the steps necessary for this process and identifying facilitators that can be used in clinical practice should be a focus of future research. Furthermore, in order to optimize the eating experience given feeding assistance and foster feelings of patient autonomy, it is necessary to individualize the assisted feeding process for all adults, promoting interdependence between patients and care providers.

This emphasis on individualized care is well aligned with current models for professional rehabilitation nursing that include competencies such as delivering patient-centered care, implementing interventions based on best evidence to manage disability and/or chronic illness, promoting health and preventing disability, and fostering self-management (Association of Rehabilitation Nurses, 2014). Central to these competencies is utilizing a collaborative approach to planning and delivering care that takes into account a patient's values, beliefs, and culture. Through such a patient-centered approach, patient self-efficacy is also incorporated, yielding increased capabilities for achieving the highest quality of life in the presence of chronic illness and/or disability. Eating is fundamental not only to an individual's daily life but also to an individual's own identity. Thus, targeting patient-centered feeding and eating practices needs to become prioritized in rehabilitation nursing practice. In order to optimize the eating experience given feeding assistance and foster feelings of patient autonomy, it is also necessary to individualize the assisted feeding process for all adults. Such an approach necessitates a better understanding of relevant personal factors, including premorbid attitudes and individual preferences, which appear to vary across the lifespan. Identifying ways to support the transition from being fully independent and operationalizing the steps needed to move toward a model of interdependence during eating and feeding will ultimately improve quality of care, patient outcomes, and patient well-being.

Key Practice Points

* The results of this study suggest that healthy younger and older adults differ in their baseline attitudes toward feeding assistance.

* On the basis of a simulated feeding experience, healthy younger adults viewed receiving feeding assistance more negatively than older adults and viewed eating-related dependency as difficult.

* The findings that younger adults viewed feeding assistance more negatively may be reflective of broader negative views toward aging and disability.

* Both younger and older adults want their preferences recognized during assisted feeding. In order to optimize the eating experience given feeding assistance and foster feelings of patient autonomy, it is necessary to individualize the assisted feeding process for all adults.

Acknowledgments

This work was supported by grants from the National Institute on Aging at the National Institutes of Health (F31AG042255) and the University of Iowa Executive Council of Graduate and Professional Students. The author wishes to thank Hsin-Wei Huang and Susie Nam for their assistance in data coding. The author declares no conflicts of interest.

References