Introduction
Numerous studies have analysed the importance of mealtimes at nursing homes from medical, social and cultural perspectives. Several studies focus on the nutritional aspects of meals, including the prevalence of malnutrition (Gaskill et al. Reference Gaskill, Black, Isenring, Hassall, Sanders and Bauer2008; Kenkmann et al. Reference Kenkmann, Price, Bolton and Hooper2010), how to improve older people's food intake (Desai et al. Reference Desai, Winter, Young and Greenwood2007; Simmons, Osterweil and Schnelle Reference Simmons, Osterweil and Schnelle2001) and how familiar food can affect recovery from illness (Evans, Crogan and Shultz Reference Evans, Crogan and Shultz2005). Other studies focus on the social and cultural aspects of mealtimes and have investigated a wide range of issues, such as care-giver behaviour during feeding (Schell and Kayser-Jones Reference Schell and Kayser-Jones1999) and perceptions of stress during meals (Hicks-Moore Reference Hicks-Moore2005; Sydner and Fjellström Reference Sydner and Fjellström2005). A number of studies have also examined the role played by the physical environment of dining rooms in long-term care facilities in terms of supporting the residents' abilities and habits (Chaudhury, Hung and Badger Reference Chaudhury, Hung and Badger2013; Kenkmann and Hooper Reference Kenkmann and Hooper2012; Lundgren Reference Lundgren2000). Research that more specifically focuses on mealtime interactions has, among other things, investigated how staff try to promote interactions during meals (Pearson, Fitzgerald and Nay Reference Pearson, Fitzgerald and Nay2003), how residents interact to avoid being regarded as trouble-makers (Sellerberg Reference Sellerberg1991), and how table allocation and table mates influence social interactions among the residents (Curle and Keller Reference Curle and Keller2010; Palacios-Ceña et al. Reference Palacios-Ceña, Losa-Iglesias, Gómez-Calero, Fernández-de-las-Peñas, Cachón-Pérez and Gómez-Pérez2013).
Much of the existing research has tended to view mealtimes as care situations, as illustrated by studies that analyse how staffing affects care at mealtimes (Kayser-Jones and Schell Reference Kayser-Jones and Schell1997) or how staff can increase the amount of calories older people consume during meals (Villarroel et al. Reference Villarroel, Sangra, Massaguer and Codina2012). Another tendency in previous research is that mealtime interaction generally has been regarded as something that takes place within the situation (e.g. Curle and Keller Reference Curle and Keller2010; Park Reference Park2009) not as something that shapes the situation per se.
A different approach, which we used in this study, focuses on how participants interactively shape mealtime situations using the available frames and scripts. Inspired by Fjær and Vabø (Reference Fjær and Vabø2013: 420), we viewed meals as situations that the actors could shape and ‘work on’. The use of this approach allowed us to investigate and pose questions about what kinds of mealtime situations are being accomplished at nursing homes.
Although policy makers and researchers stress that a good mealtime situation is important for nursing home residents (Kofod and Birkemose Reference Kofod and Birkemose2004; Mattsson Sydner and Fjellström Reference Mattsson Sydner and Fjellström2006; Philpin et al. Reference Philpin, Merrell, Warring, Hobby and Gregory2014; National Board of Health and Welfare (NBHW) 2014; Speroff et al. Reference Speroff, Davis, Dehr and Larkins2005), what constitutes the ideal mealtime remains unclear. This uncertainty mirrors the ambiguity surrounding the hybrid character of the nursing home and the use of different models in nursing home care. Are nursing homes medical settings, home settings, or a type of hotel or resort (Bland Reference Bland1999; Verbeek et al. Reference Verbeek, van Rossum, Zwakhalen, Kempen and Hamers2009)? In countries such as Sweden (Lundgren Reference Lundgren2000; Sidenvall Reference Sidenvall1999), Norway (Hauge and Kristin Reference Hauge and Kristin2008), the Netherlands (Verbeek et al. Reference Verbeek, van Rossum, Zwakhalen, Kempen and Hamers2009) and Australia (Tuckett Reference Tuckett2005), efforts to minimise the institutional character traditionally associated with nursing homes have resulted in the goal of making facilities more home-like. Some researchers argue that this can be difficult (Lundgren Reference Lundgren2000; Peace and Holland Reference Peace and Holland2001), and the medical and collective nature of mealtimes illustrates the difficulties inherent in making mealtimes more home-like. Even when nursing homes emphasise the personalised character of meals, and provide an environment that signals ‘private meal’ or ‘home meal’, the meals may still follow routines that are medical and institutional in nature. Identifying ways to make meals seem more private, personal or home-like may also be difficult and challenging if there is a lack of established ‘private scripts’ that can be used.
This study focuses on the framing of nursing home meals. Specifically, we investigate what kinds of meal situations are being accomplished and what kinds of scripts staff and residents use as guidance for their actions. The aim was to analyse how staff and residents shape meals by the way they initiate frames (Goffman Reference Goffman1974) and act according to established social scripts (St. Clair Reference St. Clair2008).
Frames and scripts
In the study, we used frame analysis (Goffman Reference Goffman1974) to interpret meal situations, and the concept of social scripts (Abelson Reference Abelson1981; St. Clair Reference St. Clair2008) as an additional analytical tool. Goffman used the idea of frames to describe the outlines of contextualising interpretations that allow individuals ‘to locate, perceive, identify, label’ (1974: 21) events and, thus, to interpret meaning that will guide future actions. According to Goffman, frames are not static and do not come into use in a simple way. Rather, they can be continually reconstructed and maintained by individuals' actions. In this study, frame analysis was used to understand the process of framing and how interpretive frames can be reconstructed and maintained by individuals' actions. We also relied on the work of Gubrium and Holstein (Reference Gubrium and Holstein1997; Holstein and Gubrium Reference Holstein, Gubrium, Seale, Gobo, Gubrium and Silverman2004), who have studied the interplay between acts of crafting meaning and the available interpretive frameworks that are used in different institutional contexts. We use the concept of social scripts (Abelson Reference Abelson1981; St. Clair Reference St. Clair2008) to label an established type of frame with distinct episodic content and role expectations within meal situations. A ‘restaurant script’ may, for instance, include the following social roles and episodic functions: the customer enters the restaurant, the waiter approaches the customer and asks for an order, the customer places an order, and so on (St. Clair Reference St. Clair2008).
We do not regard social scripts as mental representations, which is a position used in some versions of script theory (St. Clair Reference St. Clair2008), but instead we consider scripts to be social phenomena that dictate episodic content, ‘action rules’ (Abelson Reference Abelson1981) and role expectations. Social scripts order and facilitate everyday interactions but also exert normative power in relation to the participants of the script. During data collection and analysis, we paid particular attention to the presence and absence of reoccurring and episodic events that could be defined as social scripts.
The context of the study
This study was part of a larger project entitled, ‘What Does it Mean to Live a Normal Life for People in Residential Care?’ That project focused on tensions inherent in efforts to provide personalised care in collective settings like residential care facilities. One aspect of the project was analysing how ‘normal’ meal situations unfolded. A range of empirical data that were collected at five nursing homes was used, including individual and focus group interviews with nursing home staff, managers, and residents and their relatives. In addition, participant observations were conducted in two of the five nursing homes. Access to participants was gained through contact with local authorities that selected the nursing homes that were included in the study. The present article was based primarily on field observations and semi-structured and unstructured field-based interviews with residents and staff at two nursing homes, Chestnut Park and Mill Lodge. The two facilities can be described as typical Swedish nursing homes in terms of standard and number of residents. Both provide housing and care to frail older people who are usually in their late eighties or in their nineties.
Chestnut Park is located in a medium-sized Swedish town (100,000 inhabitants) and contains four units, each with 15 small individual apartments. The observations and interviews took place on the first floor of one of these units. Each unit had a kitchen, dining area, communal television (TV) room, corridor, laundry room, a balcony and a small office for the staff. The dining room in the observed unit at the Chestnut Park nursing home had three tables that were used for meals, and each table had four to six assigned seats. On the bottom floor was a restaurant that was open to the public, and two residents of the unit usually chose to go down and eat their lunch there instead of in the common dining room. The residents' apartments were located along a long corridor, and each had private furniture and an en-suite bathroom.
Mill Lodge is located in a village that has become part of the outskirts of a medium-sized Swedish town. The unit that was observed at Mill Lodge had no communal TV room or living room. Instead, the dining room served as a combined living room and dining room, and several of the residents spent a lot of the time in their assigned seats either talking or dozing.
The meals at Chestnut Park and Mill Lodge were served at set times: lunch at 1 pm, afternoon coffee at 3 pm, supper at 5.30 pm and evening coffee around 7 pm. Breakfast times were more flexible, with breakfast served throughout the morning until about 10 am, to accommodate the individual morning habits and preferences of the residents. Similar to most Scandinavian nursing homes, table seating in Chestnut Park and Mill Lodge was predetermined, and residents had to sit at the same place at every meal (cf. Bergland and Kirkevold Reference Bergland and Kirkevold2008). The allocated seating arrangements were made by the staff, but the residents appeared to agree with their assignments in that we did not notice any complaints regarding this issue.
Data collection
This study was based on 21 interviews with residents, relatives, staff and managers at five nursing homes as well as on 19 sessions (approximately 100 hours total) of participant observations at two of these facilities. Fourteen interviews were conducted in the form of focus group interviews, and seven interviews were conducted with individual participants. A total of 45 people participated in the group or individual interviews. Since the aim of this paper was to analyse how staff and residents shape mealtimes, no data from relatives were included. A detailed description of sampling and methodology is described in Harnett and Jönson (Reference Harnett and Jönson2014).
Participant observations were conducted over a five-month period. To facilitate the identification of analytical themes and to make the study more accurate, we alternated observations at the two facilities so that both researchers conducted fieldwork in both settings. To capture a variety of situations, we conducted fieldwork at different times (between 7 am and 9 pm) and on different days of the week.
During the fieldwork, we focused on how the staff and residents interacted before, during and after meals. While we were interested in the mealtime frames as such, we also wanted to explore how social roles could shift depending upon how a situation was framed. When writing field notes, we sought to follow the advice of Emerson, Fretz and Shaw (Reference Emerson, Fretz and Shaw1995) and strived to capture ‘scenes’. This meant that rather than trying to observe and document everything that was going on, we focused on shorter episodic events of everyday interactions. As participant ethnographers, we strived to make use of, rather than to minimise, our presence at Chestnut Park and Mill Lodge. One way of doing this was by using small talk to initiate field-based interviews during which staff and residents could clarify their views or elaborate on a theme that had just been observed.
Ethical considerations
All residents and staff were provided written and verbal information about the purpose of the project, and they were informed that participation was voluntary and that data would be coded for anonymity and stored in a secure manner. Written or verbal consent was obtained from all participants. The observation periods were preceded by group interviews, and as a result, most of the participants were acquainted with us when we initiated participant observations. The units observed at the facilities were not dementia units. However, since some residents might have cognitive impairments, their relatives were given the opportunity to provide their opinions before observations began. During observations, we did not regard the general consent among participants as an agreement to observe any situation, but paid close attention to signs that our presence was disturbing or upsetting for the residents. The project was approved by the regional ethics review board in Lund (Dnr 2013/349). Names, places and other information that could be used to identify the residents and staff have been omitted or changed.
Analysis
During our fieldwork, particular attention was paid to cues, or signals, and their responses; to the identification of social scripts within meal frames; and to staff and residents' explicit and implicit talk about how the meal situation should be understood. This focus helped us describe how staff and residents searched for frames and how they disputed the interpretations of meaning that could guide their actions.
Guided by the principles of analytic induction (Katz Reference Katz, Smelser and Baltes2001), we first identified the most fundamental example of meal frames, namely the institutional frame. Based on frequent statements in the interviews that residents at nursing homes should regard the setting as their homes and be able to eat according to personal preferences, we searched for a home frame. However, we only managed to identify fragments and sequences that we will refer to as private frames. This failure points to an important result of the study, i.e. the fundamental difficulty in enacting personal, home-like meals in a collective setting that is dominated by institutional routines. In the later phase of the study, we discovered a third frame, the restaurant frame. As our analysis gradually deepened, we identified more complex ways of initiating and maintaining meal frames.
Results
As noted above, we found that the staff and residents at Chestnut Park and Mill Lodge invoked three kinds of frames through which they achieved three very different versions of nursing home meals:
-
1. Institutional frame
-
• Meta-message: ‘This is an institution governed by medical routines, and staff should be in control’.
-
• Social roles: care recipients and care providers.
-
-
2. Private frame
-
• Meta-message: ‘This is a private situation’.
-
• Social roles: tenants, friends, acquaintances, family, colleagues.
-
-
3. Restaurant frame
-
• Meta-message: ‘This is a restaurant and customers should be in control’.
-
• Social roles: waiters and customers.
-
Institutional frames
During the interviews, the staff and managers stated that meals should be personalised and home-like and not be governed by the routines of the facility. This position is in line with official Swedish policy, which states that nursing homes are not institutions per se but are the homes of the residents (NBHW 2011). Yet, despite efforts to dissociate meals from institutional characteristics, the meals at both Chestnut Park and Mill Lodge were in many aspects framed as medical and institutional situations. The logic and the social roles of the institutional frames had similarities to the ‘social care approach’ (Bland Reference Bland1999) in that residents acted as care recipients and staff as care providers. Staff members were in control as representatives of the institution, and both residents and staff usually adjusted smoothly to their roles and to a shared understanding of the situation. Open defence of institutional priorities was most visible in cases in which the staff found it necessary to clarify the premises of the meals for residents, for instance in statements like ‘after all this is a ward’. One example of this occurred during a morning staff meeting that involved resident Bea, who frequently complained about the fact that the facility often served food with rice. Bea had never liked rice.
At the meeting, Angela informs her colleagues that the resident Bea raised a question about more diversity during meals at a meeting between staff and residents, but her request was declined:
Angela: We had to explain to her that you need to have a doctor's recommendation about a special diet, like they have in schools. You can't just say that you want this or that.
Sandy: No, not in a place like this.
Angela: No, then we would have to serve ten different kinds…
The staff members argue that the attempts of Bea to get special food are annoying and jointly confirm that such requests are only accepted for medical reasons. (observer HJ)
Staff members Angela and Sandy asserted that an increase in food diversity based on preferences was impractical in ‘a place like this’, but diversity that was medically motivated was acceptable. The comparison to school lunches, in which large numbers of pupils are processed through a canteen during a limited time-frame, was used to strengthen the position of the facility. A second framing technique was to juxtapose the logic used in an individual home and the logic used in the nursing home as marking differences between then and now. Staff member Dragan explained to the observer (TH) that some residents found it hard to accept the order of the nursing home: ‘Some care recipients are old fashioned, you know, chicken on Friday. It can't be on a Saturday. You explain, yes but that was the way you had it at home. We can't do it like that here’. In his account, Dragan framed nursing home meals as being fundamentally different from home meals. This prevented the normality of home from being extended into the care setting (Jönson and Harnett Reference Jönson and Harnett2015).
When regarded as social scripts (St. Clair Reference St. Clair2008), the episodic functions differed slightly depending on whether the meal was breakfast, lunch, coffee break or supper, but at both Chestnut Park and Mill Lodge, the institutional lunch script can be summed up as follows. Residents entered the dining room a couple of minutes before lunch time. Residents sat down at their assigned chairs. Staff members asked residents what they would like to drink. Staff members transferred food into serving bowls and placed on each table. Staff members helped residents serve the food. Staff prepared trays with food for residents who had their food in their private rooms. Some residents received help with eating, and often the staff sat down with the residents and had some lunch themselves (in Sweden it is a common practice that staff have some food together with residents, often considered as ‘pedagogical eating’). When the residents had eaten their main course, the staff removed their plates and immediately served dessert. Thus, some people ate dessert while others were still working on their main course. When the dessert was finished, the residents left the dining room, either on their own or accompanied by a staff member. When residents had finished eating, the staff cleared the dishes and wiped the tables, and by doing so, they marked the end of the institutional lunch script.
The discussion and performance of medical and institutional routines were essential parts of the institutional meal, as in a case in which staff members who were sitting at a table discussed the need to check the blood pressure of a resident:
‘We need to check Geraldine's blood pressure, but we can do that tomorrow morning’, one staff member says to another.
‘On me?’ asks Geraldine, who sits next to them. (observer TH)
The relationship between staff and residents that unfolded within the institutional frames was similar to the ‘pragmatic relationship’ described by Wilson, Davies and Nolan (Reference Wilson, Davies and Nolan2009). Although comments about and the performance of care routines had little to do with the actual meals, care routines appeared to be integrated into the institutional meal script, particularly in the way that they signalled the end of meals. Sometimes the staff at Chestnut Park and Mill Lodge even performed care routines during meals, such as during one breakfast when they rolled in a portable scale and started to weigh the residents who were sitting around the tables. The staff asked Dean, a resident, to stand up at the breakfast table and sit on a portable scale-chair that they had placed behind him. After being weighed, Dean sat down and continued to eat while the frustrated staff members retreated after realising that the battery of the device had discharged. Through their comments and actions, the staff directed attention to ongoing and upcoming care-giving routines and sent the meta-message that ‘this is an institution, and the people eating here are care recipients’. On several occasions, staff could only provide general information about the food being served, for instance, checking and then stating that it was ‘fish’. The reluctance to describe the meal as something that was served in order to taste good sent a meta-message about the meaning of the meal – to regulate the health of residents. This meaning became evident in interactions during mealtimes, when staff were clearly in charge, as illustrated by this excerpt of a conversation during lunch at Mill Lodge:
Inga, a staff member, is sitting down on Kiara's [a resident's] walker. Everyone is silent.
‘Have some more, Dean! There are many hours until supper’, Inga says.
‘Do you want some more, Geraldine?’, Inga asks.
‘Oh, I am stuffed’, Geraldine answers.
Inga turns to Kiara and tells her that she has to eat potatoes and vegetables, too, not only fish. (observer TH)
The institutional lunch script provided normative expectations both on what residents ate (e.g. not only meat and potatoes but also vegetables) as well as how much they ate (e.g. ‘enough’ so that they will not be hungry before the next meal is served). The lunch was the most substantial meal of the day and staff often talked about the food as a way to regulate bodily needs and also as a way to maintain the sequential order of the lunch script. This was done through reminders such as, ‘Are you not going to finish your greens, Geraldine?’ or ‘Eat slowly, John’ or ‘Should you not have dessert, Kelly?’ In the excerpt above, Inga told Dean how much he should eat (i.e. ‘more’) and oriented him (and other residents) about the timing of the next meal. Dean did not object and continued to eat a little more. When the residents were not present, the staff told the observers that the quality of food was sometimes not defendable, but during meals they acted jointly to mitigate complaints.
Although institutional frames and certain established scripts shaped meals at Chestnut Park and Mill Lodge, it would be wrong to suggest that the residents acted as ‘interpretive dopes’ (Holstein and Gubrium Reference Holstein, Gubrium, Seale, Gobo, Gubrium and Silverman2004: 269), trapped within a frame's roles and episodic patterns. Some managed to introduce competing frames (see below), and some were never urged to ‘eat up’. There were also cases in which residents protested against procedures that the staff employed to facilitate the distribution of food. An example of how dissatisfaction could unfold was displayed at the beginning of a lunch at Chestnut Park when nursing assistant Mary was helping residents cut their meat using a special pair of scissors. When Ida, who had recently moved into the facility, was asked if she wanted her food to be cut, Ida vigorously protested: ‘Nothing is to be cut around here’. Then, perhaps in an attempt to defuse the tension of the situation with humour, she added: ‘Except perhaps some hair’. While a pair of scissors is a practical tool for quickly dividing meat into smaller pieces, the comment by Ida suggested that the scissors appeared to be out of their proper context.
Private frames
Private rooms at Swedish nursing homes are formally regarded as apartments, and the residents are free to eat whatever they like. Staff are also educated to treat semi-public areas like corridors and dining rooms as private areas if the care users are engaged in private activities like talking to a relative (Jönson and Watanabe Reference Jönson and Watanabe2013). In our interviews, the staff and managers stated that meals in all areas should be personalised and defended arrangements for private food or items like smoked salmon for sandwiches or wine to be brought out to the common dining room.
During our observations, however, private frames were elusive and did not appear as developed scripts but rather as members' temporary efforts to create glimpses of non-institutional meals. One exception was the evening coffee script at Chestnut Park. According to this script, the residents acted as if they were in charge of the situation. Every night, about one hour after supper, the staff rolled out a serving trolley with coffee, biscuits and some snacks to a TV room at the end of the corridor. Usually only a few of the residents turned up for this event, while the majority stayed in their rooms watching TV or getting ready for bed:
I am in the TV room, and it is 7 pm . . . Esther [resident] enters the room and grabs a coffee cup from a serving trolley that the staff has prepared with coffee, biscuits and butter. I ask if they have assigned seats, and Ida [resident] tells me that they don't but that Esther usually sits on the sofa and she herself often sits in an armchair. Esther asks if I want some coffee and says, ‘Usually there are not very many of us here in the evenings. It is us and maybe a few more, so you can definitely have some coffee’. (observer HJ)
The evening coffee scripts were shaped as private meals following episodic functions that looked very different from the other meals: a staff member rolls in a trolley with coffee, biscuits and some snacks to the TV room. A resident or a staff member turns on the TV. Residents help themselves to coffee and food. They sit down on the sofa or in the armchairs, and they can sit wherever they feel like. If staff members are present, they also sit down and have some coffee and snacks. Staff and residents discuss TV programmes, news, gossip or personal matters. They act like acquaintances or colleagues that are making small talk. Residents leave the room when they wish, and by doing so they mark the end of the script. The evening coffee times at Chestnut Park illustrate that, although it may be difficult, it is possible to invent and establish private meal scripts in a collective setting. The residents at Chestnut Park probably had very different previous experiences in terms of evening coffee routines. Nevertheless, the evening coffee script appeared to be well-practised with well-learned ‘action rules’ (Abelson Reference Abelson1981). While other meals were shaped by the action rule that ‘all residents have to eat a nutritious meal’ (preferably together in the dining room), the action rule for the evening coffee script was not about nutritional intake, and each person decided whether he or she was in the mood for joining the script.
The evening coffee script bears similarities to the way Holstein and Gubrium (Reference Holstein, Gubrium, Seale, Gobo, Gubrium and Silverman2004) interpret institutions at rest. In a study of a juvenile correction facility, they showed that non-institutional routines may be invoked at times when the regular activities of an institution are at rest, such as during the weekends. There was a similar division of time and space at Chestnut Park. What made the evening coffee script different from the other meal scripts at Chestnut Park was that the residents acted as if they were hosts and were therefore free to help themselves and to offer snacks and drinks to others without asking the staff. The shift from an institutional frame to a private frame also meant a shift in social roles. The staff and residents acted like friends who had coffee together and talked about private matters or discussed the TV show that was playing. Staff re-entered the role of ‘staff member’ when they were called away by alarms on the portable phone and subsequently left the room. The TV room, which is where staff had meetings in the morning, was transformed into a private space during evenings. During the evening coffee script, it would have been inappropriate for a member of the staff to urge residents to eat more slowly or to finish their snacks.
The cues for private meal frames included both what the actors talked about, consisting of comments about the world outside of the nursing home context, as well as how and when the actors talked about these matters (cf. Gubrium and Holstein Reference Gubrium and Holstein1997). When we systematically searched for attempts to frame meals as private, we found that fragments of such frames were sometimes performed when the institutional demands had been fulfilled, typically during the extended phase of meal activities. The frames thus blended into the institutional frame, as is illustrated by this account of nursing assistant Andrew at Mill Lodge:
It is supper, and Andrew has served soup to the residents around the two tables. Andrew prepares a tray with food and walks out into the corridor to a resident's room. The residents in the dining room sit silently and eat. When Andrew returns, he mixes some powder and milk in a plastic cup. He explains that he gained too much weight when he went on vacation. He says that the powder is called Herbalife. It is 6 pm, and Andrew takes a seat next to two women at one of the tables. Beatrice [resident] asks Andrew if that powder really makes him full. They talk about different diets. Geraldine [resident] says that her daughter lost a lot of weight when she joined Weight Watchers. (observer TH)
A crucial point in the excerpt above was when Andrew deviated from the institutional supper script and sat down rather than clearing the dishes away. Usually the staff removed the residents' plates and put them in the dishwasher shortly after they finished eating. This signalled the end of the institutional supper script, and residents usually left the dining room. When Andrew sat down and talked with the residents, he did not act as a staff member but as a friend, similar to how he may have acted with his colleagues during a break. After Andrew told residents about his weight-loss problems, he listened to their stories on the same subject. The relationship between Andrew and the residents had similarities to Wilson, Davies and Nolan's (Reference Wilson, Davies and Nolan2009) concept of ‘reciprocal relationships’. However, while reciprocal relationships mean that staff members recognise the needs of residents (and vice versa), private frames are characterised by a role shift in which staff and residents act like friends or acquaintances.
During our observations, private frames were more likely to be initiated in the evening during supper than during lunch. The rhythm of the institution appeared to have a daily pattern in which meals and other routines usually followed clear institutional framing until the afternoon coffee at 3 pm. In the afternoons, however, the pace often slowed down. Even if the ‘bed-and-body-work’ (Gubrium Reference Gubrium1975) could be stressful for the staff, it was as if the institution switched to a less strict rhythm during afternoons and evenings.
In the excerpt above, it is clear that Andrew diverged from the institutional supper script at Mill Lodge. Did he also display a lack of professionalism by making his personal problems the centre of attention? When reviewing our field notes for similar events, it was possible to discern a pattern in which staff at Mill Lodge managed to initiate a conversation that involved several residents by introducing subjects of personal or common interest. Previous research about mealtimes (Curle and Keller Reference Curle and Keller2010) has shown that comments from staff about general matters can be a useful way for staff to engage residents in interactions with each other. Andrew's small talk can also be interpreted in the context of what Ryvicker (Reference Ryvicker2009: 20) described as ‘conscious efforts to preserve and acknowledge residents’ adult identities’, as the talk about body weight invited residents to reflect on their life experiences.
Such efforts were often linked to the personal lives of the staff, as when Maggie, a staff member, sat down on the back of a walker during the extended phase of a lunch and engaged two residents in a discussion about the local village where the facility was located and where she and several of the residents had grown up. She opened up the conversation by asking if one of the residents remembered her as a girl and if the resident remembered Maggie's grandfather, who was known locally for his ever-present worries. The conversation evolved into memories of people and places in a way that is common in conversations between younger and older relatives. Maggie and one of the residents agreed that they had both failed to live up to expectations because they had married men from less posh areas. Finally, a third resident at the table objected that she did not know what the other three people at the table were talking about. She seemed to feel excluded from a conversation that had become private.
Personalisation as a concealed use of institutional frames?
The private frames that we identified were mostly fragmented, and some cues that appeared as personalising could also be regarded as part of the institutional frame. This is illustrated by an excerpt from a conversation that took place at Mill Lodge. Inga, a staff member, used a personalising approach when serving lunch:
‘What do you want to drink, boys and girls?’ Inga [staff] asks the residents sitting around the two tables in the room.
‘Something tasty’, Kiara [resident] replies.
‘Pear juice’, says John [resident].
‘Berry juice will do’, says Inga.
Kiara asks Linda, another staff member, what kind of fish is being served. Linda says that she does not know. (observer TH)
Inga did not ask the residents individually what they would like to drink. Instead, she addressed them collectively as a group of ‘boys and girls’. Her question served as a contextualisation cue, sending a meta-message that ‘meals are collective situations’. In this case, it had recently been decided by the manager that the staff was no longer allowed to buy pear juice when making food orders. When one resident asked for pear juice, Inga simply replied, ‘Berry juice will do’. Through her response, Inga clearly communicated that it was the staff members, not the residents, who were supposed to be in control and that the residents could be treated as a group. The risk highlighted in the case above is that cues that make a meal seem personal and private in fact belong in the institution frame. This can also be understood to be in line with Rockwell's (Reference Rockwell2012) findings about how nursing homes operate according to structures that prioritise the organisation at the cost of residents' individual requests.
A similar duality between private and institutional frames concerns assigned seats. When resident Ida moved into Chestnut Park, staff decided which table and seat would be best for her; the fact that she knew some of the women at another table was secondary to the need to balance the personalities in the room. At one of the facilities where we conducted interviews, the staff complained that it was actually the group of residents that upheld the system of assigned seats: ‘But they are very fond of their seats and God help you if you sit on someone's seat . . . Actually, there shouldn't be any assigned seats, but that is the way it has developed’. What remains open for interpretation is if this use of assigned seats by residents should be interpreted as an attempt to mark out personally owned, private space in a common room, or as a case where residents are upholding the institutional order at the facility. During our observations there were several times that acts by residents that appeared to create personal space or community worked to enforce the basic pattern of the institutional frame. This duality was present when Dean, a resident at Mill Lodge, arrived for lunch at 1.15 pm and his tablemate Richard remarked: ‘So, you have your own meal times?’ The two men usually talked about private matters as friends during meals, and the comment by Richard could be regarded as a sign of frustration by a person who had to wait for a friend at lunch. But Richard did not refer to their relationship, and his remark about Dean having his own ‘meal times’ sent a meta-message about normative expectations of timeliness at lunch.
Restaurant frames
Comparisons to restaurant meals were mentioned during our interviews. Specifically, they were used to back claims that residents should be allowed to bring private food to common meals and that this should be accepted by others at the table. A nursing assistant explained: ‘It's similar to when people have a today's special at a restaurant. Some things are included and others are not. If you want to add a Danish pastry, you have to pay extra. It's a bit like that here too’. Similar comparisons were rare during our observations, but when used, they had the capacity to challenge the institutional frame.
When residents at the two nursing homes complained about the food during observations (which they frequently did), they usually invoked the quality of food at other nursing homes or during a previous time at the facility as a comparison. In contrast, Richard based his complaints on a comparison to meals served at hotel restaurants:
I sit and talk to Richard about breakfast. He talks about the eggs and says that the staff often over-boils them.
‘One should have two kinds of eggs. That is how they do it in hotel restaurants', says Richard. He explains his argument by saying that ‘at hotels, they usually serve runny eggs that are boiled for seven minutes and hard-boiled eggs that are boiled for 12 minutes or so’. (observer TH)
By stating, ‘That is how they do it in hotel restaurants’, Richard implied that breakfast at hotel restaurants were for him a logical comparison. This also made it reasonable for him to demand a wider choice of eggs.
Richard usually had his breakfast later in the morning, at around 9.30 am. He wanted the same breakfast every day, except on Wednesdays, when he also wanted a boiled egg with his meal. He was a rather frail man in his nineties, but intellectually he was sharp, and as a well-educated man, he was informed about his rights. Richard's breakfast consisted of a cup of tea, a bowl of sour milk with cornflakes and sugar, a slice of soft rye bread and a slice of thin, crisp bread. On the side, he had butter, two slices of cheese, liver pate and pickled gherkins. Richard expected the staff to serve his breakfast exactly the way he wanted it, which sometimes caused irritation among the staff members:
Richard grabs his spoon to have some sour milk and tells Sarah, a staff member, that they have given him a dessert spoon.
‘All right’, says Sarah, and takes out another spoon.
‘No, it does not matter, as there is so little food [sour milk] in the bowl’, Richard says.
‘I will put it here if you change your mind’, says Sarah, and puts the spoon on the table.
There is an egg in an eggcup placed by Richard's place.
‘I only have eggs on Wednesdays’, he says.
Sarah removes the egg from the table.
‘It is only on Wednesdays that I have an enhanced breakfast’, Richard explains.
Sarah asks why, and Richard explains that he misses lunch on Wednesdays as he goes away and plays bridge. (observer HJ)
In the excerpt above, Richard skilfully used language to transform his breakfast into a situation in which he was the one to make demands and Sarah the one to serve. Even when he told Sarah that it did not matter that his spoon was too small, he managed to convey tacit criticism by implying that his small spoon was proportional to his small serving of sour milk. During this mini-drama, Sarah's position was similar to that of a new waitress who fails to recognise the established habits of a loyal customer. The logic and the social roles seen in restaurant meals have some similarities to Bland's (Reference Bland1999) notion of a social service approach in which residential homes are run more like a hotel than like a care home. Sarah later explained to the observer that she usually worked at a dementia unit in the same building and that it was very different from working at this unit:
Sarah talks to the observer while she is wiping tables in the dining room.
‘One feels like a servant at this unit’, she says and shrugs her shoulders.
‘He who sat here [pointing to the place where Richard sat] … he could very well have taken a spoon himself. But I don't want to be rude’. (observer HJ)
Richard's actions during breakfast were an example of how residents could use comments to signal ‘this is a service facility’. Several staff members at Mill Lodge perceived the role of waiter to be patronising but still performed it in accordance with the frame that was invoked. From Richard's point of view, this arrangement can be seen both as a way to gain control over his morning meal and as a way to present himself as a person. It could thus be argued that the reference to the restaurant helped him enact his meals in the form of a private script. All of the regular staff appeared to be well informed that Richard had an egg for breakfast on Wednesdays and that he was going to be picked up to go to bridge at 11.30. Sarah's ambiguous attitude towards the role of a servant illustrates an uncertainty about what kind of frame to act according to and what kind of meal situation she wants to achieve. She acknowledged that non-compliance on her behalf would be rude and did not argue that Richard should be persuaded to change his habits. She did not use the common idea of ‘help to self-help’ in care for older people in Sweden, which suggests that care users should do as much themselves as they can in order to stay active (Damberg Reference Damberg2010). The possibility of Richard performing outside the institutional frame was probably increased by the fact that he was very old, well-educated and intellectually capable, and his self-presentation included the habit of socialising in restaurants. Sarah and her colleagues complied with the demands of Richard but did not use the restaurant frame when interacting with other residents.
Discussion
The use of frame analysis (Goffman Reference Goffman1974), and in particular our efforts to identify established social scripts (Abelson Reference Abelson1981; St. Clair Reference St. Clair2008) at Chestnut Park and Mill Lodge, helped identify three important findings: (a) an institutional frame was dominant; (b) there were substantial difficulties in introducing private frames and established private scripts for meals; and (c) the successful creation of private or home-like meal situations illustrates an often-overlooked skill in care work.
The dominance of an institutional frame
Nursing homes have a tradition of being medical institutions where people are treated as a group and where attempts to personalise care and provide a home-like environment constitute attempts to change this situation (Kitwood Reference Kitwood1993; Nolan et al. Reference Nolan, Davies, Brown, Keady and Nolan2004; Rahman and Schnelle Reference Rahman and Schnelle2008). One finding of this study was that although official policy in Sweden states that ‘special housing for older people’ should be as similar to ordinary homes as possible, nursing home meals are still often performed according to the logic of the institution (NBHW 2011; Persson and Wästerfors Reference Persson and Wästerfors2009). In many ways, the physical characteristics of the two dining room settings in this study were in line with what researchers have described as ‘family-style dining environments’ (Nijs et al. Reference Nijs, de Graaf, Kok and van Staveren2006; Philpin et al. Reference Philpin, Merrell, Warring, Gergory and Hobby2011): meals were served from glass dishes on tables rather than on trays, the tables had small flower arrangements, and there were between three and six people sitting around each table. Still, it was evident that meals usually followed an institutional script with established roles and a sequential order of action. We found that nursing home meals were not only shaped by food, furniture and seating arrangements, but also by the members' interactive efforts to create an appropriate version of a meal situation.
The difficulty of introducing non-institutional frames
Given the goal of providing a non-institutional personalised environment to people living in nursing homes in Sweden, it is of particular interest that it was so difficult to identify home-like and private meal scripts during our observations. One reason for this could be that such scripts are personal and difficult to generalise and import into a collective setting, in particular for persons who have had to abandon their homes for reasons of health and frailty. Differences between here (a nursing home) and there (a private home) tend to take the form of differences between now (being frail and dependent) and then (before frailty and dependency) (Jönson and Harnett Reference Jönson and Harnett2015). This study highlights the need for research that specifically focuses on what, when and how frames and scripts that are external to the institutional setting can be introduced as alternatives to the institutional meal. Comments that invoked the logic of a restaurant appeared during interviews but during our observations we found only one resident who acted as if this logic should apply to his meals. However, the example has theoretical implications, and several participants commented on other types of meals that were external to the context of care that have known and established scripts. These examples were, however, mentioned as being extraordinary events: a Nobel Prize dinner, a Christmas lunch and traditional Swedish goose or crayfish parties. While the meaning of a private, personal or home-like meal at the dining room of a nursing home is inherently unclear, restaurant scripts are established and known outside the context of care. They can thus be made visible and called into action as a way to challenge institutional or paternalistic arrangements. Such scripts could, for example, be successfully used to reframe the case of residents not eating their food. Instead of framing this as a problem related to the resident's nutritional intake, it could be reframed as a problem related to the taste and quality of the food and the context of the meal. One challenge is that the restaurant script includes the waiter/servant role that echoes traditional subordination of the mainly female or ethnic minority service staff. This aspect of social roles needs to be acknowledged when the restaurant script is discussed and evaluated. For instance, is the type of behaviour that Richard displayed at Mill Lodge related to class, gender or ethnicity, and should it be regarded as an unacceptable relic of an unequal society? Or should it be hailed as a case in which arrangements at a facility allowed a resident to live according to his established lifestyle?
Success in framing meals as personal and private situations
A third finding of our study was related to how to best apply and expand the good mealtime practices that we identified. First, we were struck by the way that staff initiated discussions among residents using self-disclosing comments, personal anecdotes and the introduction of topics of common interest. Individual staff members varied greatly in terms of having this skill. Such discussions involved staff acting outside institutional meal scripts, and, to some extent, outside their expected work roles. The effect was the creation of a situation that was both personal and private, which was clearly appreciated by residents and staff alike. Creating a non-institutional mealtime highlighted an often-overlooked skill in care work, namely how staff, by how they interact with residents, can shape situations that are favourable or unfavourable for the people who live in nursing home settings. Second, it was clear that the dominant institutional script was enacted and at rest according to a rhythm, running from the strictly scripted lunch to the less formalised supper, the rather personalised breakfast and the private evening coffee that the residents were (allowed to be) in charge of. The implication here is that practices that have already been established for evening coffee and breakfasts may be integrated into the lunchtime routines as well. The costs of such a shift are that the staff give up some control; in addition, there may be more obstacles to running the facility smoothly and a reduction in the opportunities to exchange information about upcoming care routines during meals. During our observations, a staff member who had participated in an advanced course in dementia care told us that her attempts to stop colleagues from talking about care routines during meals had made her unpopular. Still, making meals as ‘care-free’ as possible can be regarded as a way of operationalising and ‘testing’ the seriousness of the goal to provide a non-institutional environment, a goal that was embraced by both staff and residents during our interviews at the facilities.
Acknowledgements
The study was carried out as part of the research programme ‘Individual Care and Universal Welfare: Dilemmas in the Age of Marketisation’ and the research project ‘Improving Everyday Conditions by Reconceptualising Elder Care Through the Lens of Disability Policies’, funded by the Swedish Research Council for Health, Working Life, and Welfare (FORTE grants 2013-2296 and 2014-4565). This study was also carried out within the context of the interdisciplinary Centre for Ageing and Supportive Environments (CASE) at Lund University, funded by the Swedish Council for Working Life and Social Research. The authors would like to thank Mette Raunkiær PhD for valuable comments on an earlier draft of this paper. The authors also wish to thank the two anonymous reviewers for useful comments on earlier versions of this article.
