Food Activities and Identity Maintenance Among Community-Living Older Adults: A Grounded Theory Study

Research on productive aging is one priority for meeting the goals of the Centennial Vision of the American Occupational Therapy Association (AOTA; 2007a). Nevertheless, a review of productive aging research published in the American Journal of Occupational Therapy between 2009 and 2013 (D’Amico, 2014) found only one article that addressed the link between occupational engagement and health (Wood, Womack, & Hooper, 2009). Being able to engage in food activities is essential for productive aging because diet and lifestyle have a considerable impact on health in old age (World Health Organization, 2002).

Older adults need to engage in a wide range of food activities, defined as food-related tasks necessary to meet one’s nutritional needs. These tasks include acquiring food, eating, drinking, preparing meals, and managing diet. The important role of occupational therapy in promoting this engagement has been emphasized in an AOTA (2007b) official document highlighting that occupational therapy practitioners have the skills and competence to modify the environment and improve social interactions to support and enhance feeding and eating performance. We believe clinical practice would be improved by focusing on not only the performance of specific food-related tasks but also the meaning of engaging in food activities.

Identity maintenance is a concept associated with meaning and productive aging in the occupational therapy literature. Identity maintenance is a person’s ability to preserve a consistent and stable view of who he or she is across the past, the present, and the future. Being able to maintain a sense of who one is in later life is associated with feelings of support and belonging (Kohinor, Stronks, Nicolaou, & Haafkens, 2011; Kong & Hsieh, 2012) and mastery and control over daily life, even when functional abilities decline (Daley, Newton, Slade, Murray, & Banerjee, 2013; Dollard, Barton, Newbury, & Turnbull, 2012). A balance between maintaining a composite sense of who one is while making changes to components of this identity is also associated with higher levels of self-esteem (Skultety & Whitbourne, 2004; Westerhof, Whitbourne, & Freeman, 2012) and affect (Troll & Skaff, 1997) in later life.

In his theoretical article, Christiansen (1999) proposed that occupational engagement contributes to identity maintenance. However, many previous studies with older adults that built on Christiansen’s theory have focused on the relationship between chosen activities and the maintenance of participants’ identities (Howie, Coulter, & Feldman, 2004; Phinney, Chaudhury, & O’Connor, 2007; Reynolds & Prior, 2011; Unruh, 2004). This focus has left gaps in the understanding of the relationship between food activities, which are essential, and identity maintenance in later life.

A recent systematic review found moderate evidence from 2 quantitative and 9 qualitative studies that identity is maintained in part through engagement in food activities (Plastow, Atwal, & Gilhooly, 2015). The review identified the following gaps in the literature:

• None of the identified studies intended to investigate whether and how food activities contribute to identity maintenance in later life.

• The unique experience of older adults was not made clear because much of the research included both younger and older adult participants in the same sample.

• Ten qualitative studies found moderate evidence that changes in health change food activities and, consequently, threaten identity.

• Little evidence was found for the impact of other life experiences and events, such as widowhood or retirement, on the relationship between food activities and identity maintenance.

In consideration of these gaps in the literature, the aim of this study was to explore the role of food activities in the maintenance of identity among British community-living older adults in West London, England’s capital city.

This article presents the results of the qualitative component of a larger concurrent mixed-methods study that also included quantitative and Q methodology components. Q methodology uses card sorting methods to explore participants’ subjective views on a topic (see Watts & Stenner, 2012, for further detail). In the qualitative component, we used Corbin and Strauss’s (2008) grounded theory methodology to explore the role of food activities in identity maintenance. Their interpretation of grounded theory draws on theories of pragmatism. For this study, a pragmatist approach to grounded theory involved the development of new and useful ideas and knowledge through listening to and observing how participants engaged in food activities and then using abductive reasoning to build and verify theories about identity maintenance (Strübing, 2007).

For the process of building and verifying theory, we followed the procedures described in Corbin and Strauss and used NVivo 8.0 (QSR International, Doncaster, Victoria, Australia) for data analysis. In keeping with this journal’s commitment to improving reporting standards for research (Chan, Heinemann, & Roberts, 2014), we used the Consolidated Criteria for Reporting Qualitative Research (Tong, Sainsbury, & Craig, 2007).

We used convenience sampling and snowballing to recruit community-living older adults at two locations: a community interest group meeting and a community research fair in a shopping mall. Inclusion criteria included being age >60, being retired from full-time employment, and living in the community in West London. We excluded older adults who were living in nursing or residential care or unable to give informed consent to participate. All potential participants received an information leaflet from the first author (Plastow) or another participant, a contact sheet, and an addressed stamped envelope. The first author telephoned all potential participants who returned the contact sheet to answer questions and arrange the interview. We excluded 4 of the 43 participants whom we contacted; 1 did not live within West London, 1 lived in a nursing home, and 2 were unable to provide a convenient interview time.

The age at interview of the 39 participants ranged from 61 to 89 yr (mean age = 74, standard deviation = 7.29). All but 1 of the participants described themselves as White. The overwhelming majority of participants were women (n = 31). More than two-thirds were unmarried, including those who had never married (n = 5), were divorced (n = 9), and had been widowed (n = 14). However, the proportion of participants living alone (n = 20) versus living as a couple (n = 13) or with family (n = 6) was similar. Most participants were classified within the highest socioeconomic classification group (n = 26) on the basis of their preretirement occupation. No significant differences were found in gender, marital status, living arrangements, and socioeconomic status between participants at the two recruitment sites (p > .05). However, the community interest group participants (mean age = 76.7 yr) were significantly older than those recruited at the community research fair (mean age = 70.4 yr; p = .006).

We expected our sample size of 39 participants to be sufficient to achieve theoretical saturation of core categories because it fell within the parameters of the 4 to 40 participants most often included in qualitative studies (Holloway & Wheeler, 2010). This sample size also fell within the midrange of 13 to 92 participants included in other qualitative studies of food and identity (Plastow et al., 2015).

A single semistructured interview was conducted with each participant in the participant’s own home (n = 36), a participant’s sister’s home (n = 1), a local library (n = 1), or the researchers’ university (n = 1). The interview schedule was piloted with 4 other older adults from the first recruitment site and 1 participant who also took part in the main study. The first two interviews in the main study were conducted jointly by the first and second authors (Atwal). The remaining 37 interviews were conducted by the first author. Most interviews were conducted with a single participant (n = 31); four interviews were completed in pairs (n = 8), including 3 couples (n = 6) and 2 widowed sisters who were neighbors. Semistructured interviews allowed participants and interviewers to take the conversation beyond the scheduled questions.

The semistructured interviews included four data collection components. Part 1 included responses to the open question, “Tell me about your everyday food activities.” In Part 2, participants completed two possible-selves Q sorts. Possible selves are future-oriented aspects of identity (Christiansen, 1999). In this card-sorting activity, participants rank-ordered 35 statements each printed on a card. Half of the cards included hoped-for self statements such as “I will be loved” and “I will be able to cook.” The other half included feared self statements such as “I will be lonely” and “I will be useless to others around me.” In the first sort, participants ranked the cards from the selves they hoped for most to those they feared most, and in the second, they ranked the cards from those they expected most to those they expected least. Participants commented on the content of their cards.

In Part 3, participants completed the Occupational Performance Measure of Food Activities (OPMF; Plastow, Spiliotopoulou, Atwal, & Gilhooly, 2014), in which they rated their food activities on a 15-item scale across three subscales: Importance (α = .68), Performance (α = .62), and Satisfaction (α = .60). The participants were shown a prompt sheet with each scale but reported their ratings verbally. Most participants spontaneously explained why they had given themselves a specific score. Participants were also asked, “Is there anything that stops you doing your [food activity] in the way you would like to?” Part 4 collected demographic data. At the end of each interview, participants were thanked and given a £10 (∼$15) voucher as a token of gratitude.

Interviews were audiorecorded and transcribed verbatim for analysis. Recorded interview time was between 34 min and 1 hr 51 min. Mean recorded interview time was 60 min. Postinterview field notes included observations of the home environment, comments participants made before or after the tape-recorded interview, the interviewer’s reflections on the content of the interviews, and any concepts the interviewer thought interesting enough to follow up on in data analysis.

The textual data for analysis included everything participants said to the interviewers and the interviewers’ field notes. Numerical data (e.g., ranked card position, the numerical value of OPMF scores, and date of birth) were excluded from the analysis in the qualitative component reported in this article. The first author coded all data and met with the second author at multiple points to discuss emerging categories and concepts.

The process of analysis is illustrated in Figure 1. The first part of the analysis focused on identities associated with food activities. Open coding was used to identify concepts and categories evident in the content of participants’ interviews. Axial coding was used to describe the relationships among codes, concepts, and categories (Corbin & Strauss, 2008). This process of data analysis was repeated until each category and concept was fully described in relation to its properties (characteristics and attributes), dimensions (variation of properties along a continuum), and inconsistencies using the data available.

The second part of the analysis focused on the processes associated with identity maintenance evident in the data. The potential for each previously identified concept, category, subcategory, or code to be part of an identity process was considered, and memos were recorded (Corbin & Strauss, 2008). The goal was to identify patterns of responses or patterns of behavior within the coded data. In the final stage of analysis, we used selective coding to raise the analysis to a theoretical level by telling the story of the dynamic relationship between food activities and identity maintenance, grounded in the data. Extensive use of diagrams to graphically represent the emerging process facilitated the data analysis and interpretation.

This study was approved by the Brunel University research ethics committee. Pseudonyms are used throughout this article to protect the anonymity of participants.

The data analysis identified three processes by which food activities were associated with identity maintenance: (1) participation and maintenance, (2) threat and compensation, and (3) changes in meaning and identity. Participation and maintenance explained how participating in food activities helped participants maintain their food identity as either “food lover” or “nonfoodie” across the life course. Threat and compensation indicated how participation in food activities was threatened and changed by a variety of life experiences that required compensation. Finally, changes in meaning and identity explicated how an accumulation of changes led some older adults to develop the food identity of “not bothered” about food (see Figure 2).

For this study, we defined food identity as a composite sense of who one is derived from one’s experience of participating in food activities. Within the participation and maintenance process, participation in food activities in a way that was consistent with meaning and identity led to the maintenance of a lifelong food identity as either a food lover (n = 18) or a nonfoodie (n = 10; see Table 1).

Food lovers talked about food and food activities in a positive way. Most used the word love to describe the pleasure food gave them. Sharon described the pleasure and enjoyment she experienced from food and food activities: “I love my food. . . . I enjoy cooking, I enjoy food, and it’s the way I was brought up. . . . No, I’m afraid I’m a bit of a foodie.” Maisy similarly described herself as “a foodie, really; I love food.” Daphne and Gloria used the word enjoy to describe their feelings about food and food activities. Identity-consistent participation for these participants included “spending time and money” on food activities, “cooking together” with and for others, and “resisting temptation” through “choosing well” and making “moral choices” during food activities.

The process of identity maintenance was somewhat different for the nonfoodies. They were uninterested in their food activities. Linda described herself as “disinterested”: “I wouldn’t say I’m a foodie at all.” Instead of “loving” food, these participants talked in neutral terms about their food activities. Food activities were important to these participants only because they had to do them. Otherwise, food and food activities did not play an important role in their lives. The nonfoodies maintained this identity by spending “as little time as possible” on their food activities by “minimizing” cooking and eating out. Rupert most clearly described how the nonfoodies spent as little time as possible on their food activities: “Eating keeps you going, but . . . I’d be quite happy not to eat but enjoy life in other ways. . . . If I had an astronaut-type squeezer tube, I’d be quite happy to do that [laughs].”

Regardless of food identity, almost all participants experienced changes in the way they participated in food activities. “Being alone at the table” was the greatest threat to food activities and identity, especially for the food lovers who gained pleasure from the social aspect of their food activities. Edna, Bettie, Saul, Barbara, Carol, Dorothy, and Martha talked about widowhood. Sarah and Daphne talked about divorce. Sarah, Anne, Judith, and Helen talked about their children moving away from home. Judith and Bettie had moved from different parts of the country to be nearer their children. Marjory and Sally talked about retirement and age changing their experience of mealtimes.

Being alone at the table was “difficult and boring” and meant that food activities became “less routine.” For example, Barbara talked about how her cooking had become precise and controlled after her husband died: “It is quite difficult to cook for one person, getting it reduced down in quantity. . . . In the end, you actually get down to the stage you actually know precisely what you’re going to eat.” As a result, she viewed her meals as “mundane.” Adapting to cooking smaller portions was a challenge for those who were widowed or divorced and living alone. It was also a challenge for younger participants whose children had left home. The way being alone at the table affected Anne’s routines also reflects the difficulty other participants had in keeping their routines without somebody else to cook for:

Changes in food activity participation were also caused by deteriorating health, which led to reduced mobility, a change in food choices, and routines becoming controlled by dietary requirements or medication regimens. The cost of food, expressed in the concept “Cost is one thing,” also led to restrictions in the food items participants bought, less eating out, and difficulty making savings, especially through bulk-buy special offers.

Both food lovers and nonfoodies maintained their food identity by successfully compensating for these changes. Successful compensation meant that the meaning of food activities stayed the same and participation remained consistent with each of the two lifelong identities. Food lovers often compensated for being alone at the table by cooking for others, including their friends, neighbors, and extended families and at volunteer organizations. They also learned to cook well for one person and enjoyed being “selfish” in the food choices they made. Nonfoodies, in contrast, made their food activities even more efficient—for example, by buying microwaveable meals instead of ingredients to prepare.

Both food lovers and nonfoodies compensated for deteriorating health by making food activities quicker and easier. Food lovers were more likely to describe themselves as “being reasonable” in the food choices they made and the timing of their meals. Sally, who had diabetes, noted, “I think a bit slower if I have a bit more sugar one day, and I’ve noticed it’s up, and . . . it’s slower to go down, but it does go down.” “Being reasonable” meant they could continue to gain pleasure from their food activities, but in moderation.

The cost of food was an issue faced by food lovers and nonfoodies alike. Some not-bothered participants talked about “spending less” by “eating out less” and “restricting food choices” by buying food on special offer or not buying food that was too expensive. Only the food lovers talked about growing their own food to compensate for cost.

An accumulation of threats and increasing difficulty adapting led to changes in the experience of food activities. These changes went further than a change in habits and context. For not-bothered participants (n = 11), the meaning of food activities had changed. Food activities became “less important,” “less pleasurable,” and “too much effort.” All of these participants talked about food activities being less important than they used to be: As Edna expressed it, “There was a time when cooking was very important; now it’s just necessary. . . . You get that you can’t really be bothered.”

Food activities became less important because these participants no longer felt socially connected when they shopped, cooked, and ate. When Sally retired, her eating became less important. Edna, Marjory, Ruth, and Howard talked about how their food activities were less pleasurable because they did not enjoy cooking anymore; for example, Marjory stated, “I used to quite enjoy cooking, but I just can’t be bothered these days [laughs]. And that’s the story of my life at the moment: can’t be bothered.” Many participants also became not bothered because their food activities had become too much effort. Edna described herself as merely “coping” with her shopping.

Some not-bothered participants also found it was too much effort to eat healthily or improve their diet to manage their diabetes and osteoarthritis. It was evident that these health conditions led to an increasing importance in being a healthy eater in later life, but participants remained unmotivated to change their eating behavior. For example, Howard noted,

These participants did not describe themselves as always loving food, like the food lovers. Nor did they talk about a lifelong disinterest in food, like the nonfoodies. Instead, it was consistently evident across all of their interviews that a change in meaning had led to a new identity as not bothered about food activities.

Three different identities were evident in participants’ descriptions of themselves: food lover, nonfoodie, and not bothered. Food lover and nonfoodie were lifelong identities. Not bothered was a new identity that emerged through a cumulative change in food activities in old age. The findings of our study suggest that food identities vary in their importance to the individual. The food lover identity was an important identity that contributed to participants’ mental well-being. In contrast, both the nonfoodie and not-bothered identities were unimportant and did not contribute to mental well-being. This finding adds to Christiansen’s (1999) theory that activities contribute to the development and maintenance of identities by showing that only important activities and associated identities contribute to mental well-being in later life.

Engaging in food activities was an important part of productive aging for the food lovers. Both men and women who were food lovers felt socially connected when they cooked for and ate with their families. Similar findings were evident in studies with women across different cultural groups (Plastow et al., 2015). However, our findings contrast with those of other studies that described cooking as being in conflict with men’s masculine identity (Moss, Moss, Kilbride, & Rubinstein, 2007) and role expectations (Atta-Konadu, Keller, & Daly, 2011). Food lovers also gained pleasure from being “healthy eaters,” “naughty,” and “moral” during their food activities.

Negotiating between health-seeking and pleasure-seeking behaviors did not seem to be difficult for these food lovers, in contrast to Jallinoja, Pajari, and Absetz’s (2010) focus group study of health-seeking lifestyles among participants in a health promotion intervention. In this study, food lovers gained pleasure from a sense of achievement when they resisted temptation, indicating that following healthy eating guidelines did not pose the same threats to identity as was found in other studies (Plastow et al., 2015). In those studies, loss of control was a source of negative self-evaluation. Instead, many of the food lovers in this study positively evaluated themselves because they resisted temptation. Indeed, Christiansen (1999) argued that a sense of self-efficacy gained from reaching goals and having control over one’s environment contributes to the sense of meaning and identity derived from engagement in occupations.

Within the identity process of threat and compensation, we found that food activities were temporarily disrupted by life events and experiences. In our study, the most significant threat to participants’ identity was a change in the social meaning of food activities. When participants were alone at the table, food activities became more difficult and boring and less routine. Two other qualitative studies similarly found that eating becomes a necessity rather than a pleasure for older adults who eat alone (Gustafsson & Sidenvall, 2002; Kullberg, Björklund, Sidenvall, & Åberg, 2011). Our finding is important because previous research has focused on the effect of changes in health on identity (Plastow et al., 2015). In addition, our finding that four of the five participants who were living with relatives also talked about being alone at the table suggests that occupational therapy professionals should not assume that living with others automatically brings a social dimension to food activities.

Deteriorating health was also a threat to participants’ identity. However, contrary to our expectations, deteriorating health did not always lead to a change in identity. In nine previous studies, the links were clear among changing health, changes in food activities, and threat to identity (Plastow et al., 2015). We found that deteriorating health made food activities more difficult and less pleasurable because participants were less mobile and found their food activities more restricted and controlled.

More food lovers reported a change in their food activities than participants with other identities. We draw two conclusions from this finding. First, deteriorating health may not lead to a change in identity if older adults are able to compensate successfully. Second, the identity of food lovers may be threatened more by deteriorating health than that of nonfoodies, because engagement in food activities is likely to change more significantly for this group of seniors.

Devine (2005) suggested that specific turning points signal a drastic and permanent change in food activities. In our study, no such turning points were identified. Instead, participants talked about a cumulative change in their food activities. For example, when Bettie told us how widowhood had changed her food activities, she started by talking about cooking for a large household, then her children leaving home, and then her husband’s death. We found no consistency among participants between specific life experiences and particular changes in meaning. For example, eating became less important after retirement for participants like Sally and Marjory because they ate alone more, whereas food activities became more important to participants like Jill and Peter because they had more time for such activities. Nevertheless, our findings also suggest that as older adults experience one life event after another, they become less able to compensate for the accompanying changes in their food activities. This accumulation of threats leads to changes in identity.

We found no evidence of any differences between the food lover and not-bothered groups’ expectations of achieving their hoped-for possible selves or of having a good quality of life. Both findings suggest that a change in food identity did not affect these participants’ mental well-being. This finding is at odds with the sense of loss that was also evident in their interviews. Christiansen (1999) suggested that a loss of meaning (similar to that experienced by the not-bothered participants) can lead to depression, self-harm, and alcohol abuse. He described these conditions as “diseases of meaning” (Christiansen, 1999, p. 556). In support of this theory, studies by Whitbourne and colleagues found a negative relationship between identity change (accommodation) and self-esteem (Skultety & Whitbourne, 2004; Sneed & Whitbourne, 2003; Westerhof et al., 2012). However, these studies also found that adults and older adults who achieve identity balance by maintaining a consistent and positive identity while changing some self-descriptions are more likely to have higher self-esteem. The findings of our study indicate that maintaining a food lover identity also maintained those participants’ mental well-being. However, it is not yet clear precisely whether and how mental well-being is negatively affected by a change in food identity.

Although we believe we reached theoretical saturation in explaining the role food activities play in the maintenance of identity, we believe we do not fully understand the experience of participants whose identity changed. Most important, it was not always possible to determine which lifelong identity participants had held before becoming not bothered. A life history or narrative approach to future interviews would address this limitation. The data provide insufficient evidence to reach a fully grounded explanation of how not-bothered participants’ mental well-being was affected by the loss of meaning of their food activities. We will be exploring this issue in future research. Another important limitation is the homogeneity of our participant sample. Nevertheless, this study addresses an important gap in the literature by exploring the relationship between food and identity within a dominant ethnic group.

Although this study was largely theoretical, some implications for occupational therapy practice can be drawn:

• Assessment of feeding and eating performance should include questions about the importance and meaning of food activities to clients.

• Participation in meaningful activities and compensation for changes in these activities maintain lifelong identities. A conversation with older adults about their most important daily activities, why they love or enjoy them, and what they do to make sure participation is consistent with the way they view themselves is a good starting point for an assessment of occupational identities.

• Older adults’ food identity may have an effect on the success of occupational therapy interventions. Finding ways to compensate for changes in food lovers’ food activities while still maintaining the meaning of food activities should be the focus of occupational therapy intervention for these older adults. For nonfoodies, finding ways to spend as little time as possible on food activities while still achieving optimum nutritional health should be the focus of occupational therapy intervention. Interventions with not-bothered older adults may focus on either restoring meaningful participation in food activities or finding new meaningful activities that maintain important identities.

This research was generously supported by the British Geriatrics Society/Dunhill Medical Trust Research Fellowship held by the first author.