From rational collective feeding to flexible individual choice: The governance of hospital foodservice in transition

Abstract

In this article, we critically discuss a Swedish hospital foodservice reform by putting foodservice in a historical and social context and analyzing a case, “the most flexible patient foodservice system in Sweden,” by using Bacchi’s what’s the problem represented to be (WPR) approach. We show how hospital foodservice governance in Sweden has become more focused on the individual patient as a consumer. Our analysis points out how in political protocols and related policy materials freedom of choice, flexibility, and efficiency were articulated as drivers of change. Freedom of choice was argued to enhance quality, increase food intake, diminish demand for special diets and lower costs. This can be seen as empowering patients, but also shifting part of the responsibility of nutrition care and health over to individuals, whether they want it or not. Flexibility was seen as beneficial for patient logistics regarding medical treatment and argued to improve allocation of staff resources. Inefficiency was a characteristic attributed to the old system, while the reform was an opportunity to modernize and be more cost-efficient. The new organizational structures may lead to tensions between foodservice and healthcare. The changes, system flexibility and patient freedom of choice, are salient within a broader neoliberal discourse.

Introduction

In this article, we analyze discourses pertaining a large-scale reform of the patient foodservice in a Swedish public hospital, which later came to be described as the most flexible patient food delivery system in Sweden. The initial process begun in the autumn of 2011 when a political decision was made to plan a reform of the hospital foodservice. The goal was to devise a high quality, flexible and economically sustainable solution to the problems perceived by the foodservice management and hospital administration. The main reason for this reconsideration of foodservice design was that meals produced on- site were considered cost in-efficient. Additionally, the existing hospital kitchen was old and worn down, and a transition to electricity as a power source was inevitable due to structural changes within the hospital’s infrastructure. This inevitability of renovation constituted a window of opportunity to argue budgetary allowances for a large-scale foodservice system change. In this article we examine the hospital foodservice reform in a broader historical and social context, which can be done from a discursive perspective. What is proposed as a solution in a reform also indicates what is considered to be problematic at present, which in turn reveals underlying assumptions and central discourses (Bacchi, 2009). Discourses in this case do not refer to language use as such, but to generally known ways of speaking, writing, and thinking that make it difficult to think in other ways (Bacchi, 2009). Therefore, studying the central discourses of a hospital foodservice reform makes visible meanings and values on a broader scale, as well as the governance. Despite their analytical potential to elucidate foodservice governance, which in this context, is understood as patterns of rule ¹ (Larner, 1997), analyses of discourses are at present scarce within hospital foodservice research. The aim of this paper is to critically discuss the premises of flexible hospital foodservice reform by analyzing underlying assumptions and the central discourses associated with the reform. We posed the following research questions:

• What is the problem represented to be and what are the likely effects of the proposed reform?

• What is left unproblematized by this way of representing the problem?

The case under study: The most flexible hospital foodservice in Sweden

The organization and responsibility of public health and medical care in Sweden, including hospitals, is governed by political regional councils that appoint regional executive boards. Major organizational changes therefore need to gain political endorsement. The full proposal for the new flexible foodservice system and its approximated costs were approved by the regional council in 2014. The reform was large, both conceptually and as a monetary investment. The foodservice system reform was a transition from what is often referred to in foodservice literature as a traditional system (Dunn and Williams, 1994; Edwards and Hartwell, 2006; Gregoire, 2013) to a cook-chill concept. The foodservice reform included the construction of a new central production line, new logistic patterns, and the building of new dining spaces in all hospital wards. These dining spaces were accompanied by a kitchen for reheating and a cold café-style display for snacks to be at the patients’ disposal. Additionally, a new labor structure was implemented. New routines followed, with service staff taking responsibility for foodservice on the wards from mornings to evenings and simultaneously relieving health care staff of most of these duties. Finally, the mealtimes and offered food choices were changed. Set mealtimes and a 6-week rotating menu with tray service were replaced with the offering of meals and snacks at any time from an à la carte menu, served from a kitchen space on the ward, which was kept replenished with menu items from the central production unit.

Background: History and development of Swedish hospital foodservice and current trends internationally

Hospital foodservice operations are in general complex systems. The main objective is to provide inpatients with all their daily meals, therefore being an important part of the nutritional treatment, patients’ quality of life, and providing a break from the medical environment (Johns et al., 2010; Jonsson et al., 2021; Kim et al., 2010). Hospital foodservice design is essential for how these objectives are achieved and choices include purchasing practices, menu design, cooking and serving methods, management and labor structures, hospitality, eating environment design, logistics, etc. (Gregoire, 2013). Reforms and interventions target one or more of these areas.

In the context of Sweden, public tax-funded foodservice meals have traditionally been seen as a welfare service guided by notions of equality and taking care of vulnerable groups (Mattsson Sydner and Fjellström, 2007; Persson Osowski and Fjellström, 2019). However, the logic governing hospital foodservice has developed over time and this can be illustrated by putting Swedish public foodservice in care settings into historical context and tracing how current arrangements have developed.

Historically, public foodservice in Swedish care settings used to be strict and collectively regulated. Preserved instructions on how to manage the foodservice at a care facility for the mentally ill from 1853 illustrate some of the conformity considered effective at the time (Sondén, 1853). Breakfast consisted of 13.3 g of butter and one piece of bread each day (p. 48). Sunday’s, Tuesday’s and Friday’s dinner ought to be meat cooked with turnips, grain, and flour. The following excerpt further illustrates what was considered appropriate:

Meals, however monotonous, are in general fairly good and sufficient for the poorest class, who haven’t been accustomed to better, and does not compare to the more well-off classes’ habits and needs: and any specific diet for the sick, so-called sick diet, is not possible to arrange here (Sondén, 1853: 47–48)

Emphasis was put on the benefits of conformity, and both the foods to be included in patients’ diets and their preparation were strictly regulated. In general, simplicity was preferable in general, but a beautiful table and cutlery were believed to contribute to order and comfort and also to “a small increase in feeling human for those who are mad” (p. 49). This text allows a glimpse of the early public foodservice thinking, the so-called “collective feeding.”

In Sweden, guidelines and reports regarding hospital foodservice and diets have been published periodically by national authorities. In 1965, The Central Board of Hospital Planning and Equipment, responsible for standardization and rationalization of hospital operations at the time, published an investigation of 53 hospitals as a basis for decision making for future rationalization measures (Centrala Sjukvårdsberedningen, 1965). In the investigation, work spent on tasks other than cooking were problematized, and measures to rationalize transport were suggested (p. 51). Conventional meal distribution was criticized for its deficiencies (p. 67), and the advantages of technological advances like centralized tray service were highlighted, although additional administration in the form of individual labels for each tray were seen as negative because of the labor cost. Single portion microwaveable trays served through an automated supply system, using single-use plastic cutlery and plates, were promoted by being framed as revolutionary (p. 72). The investigation was concluded by a commentary on two topics – freedom to choose what to eat in a hospital setting and the hospital kitchen of the future. It was stated that while examples from the United States suggested a menu-based system could be introduced without additional staff, this would probably not be the case in Sweden (p. 74). It was seen as an organizational possibility in relation to the modern tray services presented earlier, but not without additional costs. Two apprehensions were presented; that patients would always choose the most expensive menu items, and that there would be a lack of control over patient nutrition. Implementing freedom of choice was thought to “be questionable from a medical perspective” (p. 74). In terms of hospital kitchens of the future, the investigation concluded that centralization reduces costs related to the building and staff but requires good communication between production units and receiving units (p. 75).

In 1971 (second edition 1976), the National Food Agency published recommendations for two general hospital diets, one for adults and one for children, and 17 therapeutic diets for example, fat-reduced diet, diabetic diet, etc. (Statens Livsmedelsverk, 1976). The recommendations included detailed calculations and examples of daily meal plans. In 2003, revised recommendations focused on three hospital diets; general diet, sick diet, and energy-dense diet, and the number of therapeutic diets had been reduced (Nilsson Balknäs and Andersen, 2003). Greater focus was put on individualized care, where nutritional care was to be ordained like medical care and individual care plans established for patients in need of dietetic intervention. The previous year, the National Food Agency had also published a report concerning hospital foodservice management where they concluded that staff on hospital wards lacked adequate food safety education, highlighting an increased focus on the professionalization of foodservice work and differentiating it from nursing (Lantz and Svensson, 2002). More recently, in 2017, the three regulated hospital diets were removed with the goal of achieving more flexibility in menu planning, more individually tailored meal plans, and less food waste (Livsmedelsverket, 2022: 26).

Moreover, and separately, in 1985, the National Board of Health and Welfare had published general guidelines for nutrition in hospitals including some criticism of the earlier rationalizations, stating that big central units and reductions in working hours could make management more difficult (Socialstyrelsen, 1985). In these guidelines, malnutrition was highlighted as an issue, and the regulation of mealtimes was promoted. Malnutrition was also highlighted in a publication by the Nordic Council of Ministers in 1995 (Nordiska Ministerrådet, 1995), but emphasis here was put on the individual needs of the patient who had “lost the freedom to choose what one wants to eat and when” (p. 53), as well as the collaboration between the healthcare and foodservice organizations that patients depend on.

This development toward individualization can also be seen in the international hospital foodservice research literature which shows a gradual systemic shift to more personalized forms of foodservice. The terms used to describe the changes in foodservice organizations have varied, including for example individualized meal service (Holst et al., 2017) and patient-centered foodservice (Sathiaraj et al., 2019).

Sathiaraj et al. (2019) called the move toward patient-centeredness a paradigm shift. However, the idea was not novel as more than 20 years earlier, Miller and Kinsel (1998) had discussed the implications of patient-focused care specifically for nutrition practice.

The change from a collective to an individual focus in public foodservice in Sweden occurred during the 1980s and 1990s (Mattsson Sydner, 2002). Medical science and nutritional value for money had thus far been guiding the planning of foodservice operations. The emergent customer orientation in foodservice was characterized as one of the most significant ideological shifts for the welfare state to date. These changes in public foodservice in Sweden were also influenced by New Public Management (NPM), a concept originally described by Hood and encompassing both ideological and administrative trends (Hood, 1995). Although introduced during the 1980s and 1990s, NPM continues to be influential to this day. For example, cost-effective cook-chill production methods and offering meal choice to promote individualism are foodservice reforms that have been discussed in relation to NPM (Skinnars Josefsson et al., 2018).

Miller and Kinsel (1998) noted that in patient focused care “the individual patient’s needs are the determining factors for providing services, unlike the traditional hospital structure in which institutional or departmental efficiency considerations are the driving factors of service delivery” (p. 179). However, they also described how costs were functioning as a driving force behind change. Patient-focused care initiatives were initiated with intended outcome goals such as improved customer and staff satisfaction, but also with operational and financial efficiency (p. 179). Contemporary interventions build upon concepts that are focused on changing the menus or changing service-related features like room service, electronic bedside meal ordering, and à la carte menus (Dijxhoorn et al., 2018; Doorduijn et al., 2016; MacKenzie-Shalders et al., 2020; McCray et al., 2018a, 2018b; Ottrey and Porter, 2016; Prgomet et al., 2019). In other words, they have some common elements, including trying to offer more possibilities to increase the patients’ freedom of choice.

However, reducing costs is often one of the explicitly stated driving forces behind the hospital foodservice interventions (Dijxhoorn et al., 2019; McCray et al., 2018a, 2018b). Increased competition within the market and patient demands for improved quality of care are also recurrent factors mentioned as incentives for interventions (Dall’Oglio et al., 2015). Most of these interventions have reported positive patient and foodservice outcomes, for example, increased patient satisfaction, increased energy and/or protein intake, and decreased food waste and cost (Cheung et al., 2013; McCray et al., 2018a, 2018c; Sathiaraj et al., 2019). In sum, hospital foodservice governance in Sweden has over time become more focused on the individual and detailed regulations have been dissolved. Internationally, although some hospitals are private and others public, recurring norms of foodservice governance include competition and consumer demand, as well as patients acting as active decision makers in their own care. Currently, person-centeredness and person-centered care are emphasized in both guidelines and research concerning hospital foodservice in Sweden (Jonsson, 2021; Livsmedelsverket, 2022). In the latest Swedish guidelines for hospital foodservice, adapting to individual needs is encouraged, as well as flexibility, responsiveness, and patient participation (p. 7) The guidelines also state that a well thought out menu for different patient needs should be available around the clock. Further, hospitality as a way to enhance the meal experience for patients has emerged as an area of research (Jonsson, 2021).

Methods

Data collection

This article builds on policy analysis with data collected from one of Sweden’s 21 regions wherein the hospital of concern was located. Three types of material were included in the analysis: (1) official records, (2) internal and external communications material, and (3) additional material about the reform gained through collaboration with the region. Public authorities and official entities in Sweden are legally obliged to follow the principle of public access to official records, which means that regional councils keep their meeting protocols and similar records publicly available. In this case, most data was accessed through an online archive from the year 2015 onward. All regional executive board protocols and regional council protocols and their enclosed attachments between January 2015 and March 2020 were initially included. Both the regional executive board and the regional council met on average once monthly with breaks during the summer months (June, July, August), and the number of issues raised during each meeting varied. However, foodservice was not a frequently appearing topic. ²

Initially 112 meeting protocols and attachments from between 2015 and 2020 were read, totaling to over 20,000 pages, looking for foodservice related discussions. Additionally, when documents of interest referred to earlier ones, access to material from before 2015 was gained through collaboration with archivists employed at the region. The archivists also did a screening of material related to the foodservice reform from before 2015 and forwarded two records of decisions from the regional executive board concerning the reform which were added to the analysis.

All documents that mentioned food, foodservice, meals, service, nutrition, kitchen, dietitians, or food production were considered for analysis and included if they described the foodservice reform in any way. This resulted in approximately 200 pages from 12 meeting protocols including attachments, which formed the main body of analysis. This material depicted the decision process to accept the construction of the new foodservice system, including the final proposal of the new foodservice system presented for approval, a feasibility study with detailed arguments and analysis, and all included attachments, such as statements from concerned parties, clarifications, and budgets. Additionally, information material was included to elucidate the core ideas and values of the foodservice and healthcare organizations, and how those ideas and values were communicated within the region. Internal and external communications materials, for example patient information brochures and presentation material, were gathered through the region’s web page for this purpose. Lastly, the new foodservice system underwent a pilot phase in 2016, and a pilot project evaluation was included as was presentation materials used to promote the pilot and an approved grant application. Access to the pilot project evaluation was gained through collaboration with the region.

Analytical approach: What’s the problem represented to be?

The analytical framework in this paper is based on the “What’s the problem represented to be” (WPR) approach (Bacchi, 2009). The focus of the WPR approach is on practical texts, texts that aim to produce certain effects in their audiences (p. 34), as materials which allow insight into governing practices through “problem representations.” The core idea of the WPR approach is that a proposal for change implies that there is a specific situation that needs changing, that there is a “problem,” and the proposed “solutions” reveal what is considered problematic. It is argued that it is important to draw attention to various meanings ascribed to problems as concepts (Bacchi, 2015, 2016). Bacchi positions the WPR approach as an analysis of discourses, not discourse analysis (Bacchi, 2005). With this distinction she wants to bring attention to the two different foci of analysis in the two traditions, where the latter usually focuses on language use and the former on systems of thought. She writes; “the goal is to identify, within a text, institutionally supported and culturally influenced interpretive and conceptual schemas (discourses) that produce particular understandings of issues and events” (p. 199). Discourses are social knowledges or ways of thinking that establish frames of references that are difficult for people to think, and act, outside of. Political practices are affected by discourses (Bacchi and Bonham, 2014), and as the aim of this paper is to critically discuss the premises of hospital foodservice reforms, the WPR approach is suitable because by analyzing underlying assumptions and central discourses it refuses to take problem representations for granted. Engaging in an analysis of how different positions on change are argued for can elucidate what is considered important in a community at a specific time. However, the purpose of the WPR approach is not to point out advantages or disadvantages with certain problem representations (Bacchi, 2012). That is to say, the purpose is not to point out wrong-doings. Problems in the WPR approach are neither seen as naturally existing and uncontested, nor socially negotiated to be solved. A focus on “problematization” refers to the underlying meaning ascribed to an explicit or implicit verbalization of a problem. The contribution of the WPR approach thereby is its ability to illustrate underlying conceptual meanings and logics, rather than present solutions to the problems under scrutiny.

Thus, the key element of WPR as an analytical approach is that a proposed solution reveals what is considered problematic and what therefore needs to change (Bacchi, 2009). In practice it consists of questions building on each other that are meant to facilitate critical reflection on premises of thought, adjacent genealogies, and effects of these specific problem representations. The first step of analysis was to identify solutions proposed in the material by applying the question; “What’s the problem represented to be in a specific policy or policies?” (Bacchi, 2009). Through multiple readings of the material, solutions were identified, thematized, and analyzed using question 1 to describe represented problems, resulting in three problem representations. The second step was to apply the question; “What deep-seated presuppositions or assumptions underlie this representation of the “problem”?.” Thereafter the analysis focused on the questions; “What is left unproblematic in this problem representation (Where are the silences)?” and “What effects are produced by this representation of the “problem”?”. Finally, reflexivity was applied to the analysis. We considered the empirical material in relation to the historical developments that over time normalized individualism. The authors each represent different perspectives and research backgrounds, but jointly agreed that the material reflected a continuum of neoliberal thought. The first author conducted coding according to the WPR approach, which was continuously discussed with all authors. All authors contributed to the analytical interpretation and agreed on the main discourses identified in the analysis.

Neoliberalism as governmentality

In addition to using Bacchi’s “What’s the problem represented to be” approach (Bacchi, 2009) to deconstruct problem representations in the material we have used an understanding of neoliberalism as governmentality (Larner, 2000) to critically discuss the discourses we identified as central and meaning-bearing. Neoliberalism as governmentality can be understood as the ways in which neoliberal ideas and practices embedded in society shape the way that a society is governed (Larner, 2000). Further, when promoting individual choice and limiting the scope of the state, governance is not decreased, it is shifted from state actors to the market (Larner, 2000). This means that both individuals such as patients, and institutions such as hospitals and regional boards, are encouraged to conform to the norms of the market, so called market governance, which emphasizes self-acting subjects, deregulation, promoting competition, and corporatization (Larner, 1997). The ways through which this encouragement happens are in turn discursive, as discourses in this case are understood to be generally known ways of speaking, writing, and thinking that make it difficult to think in other ways (Bacchi, 2009).

Although critical discussions of neoliberal discourse often have a focus on the expansion of markets and the minimization of government intervention, the influence of neoliberal discourse within the public healthcare structure is just as prominent (Veitch, 2010). As Ayo writes “neoliberalism is by no means just an economic or political term – it is inherently social and moral in its philosophy” (Ayo, 2012). The broader culture of consumerism makes individual choice and control appealing in the hospital context (Veitch, 2010). From a patient empowerment perspective this might seem as an unequivocally positive development, as freedom of choice and control resonate with us on a fundamental level (Harvey, 2005). However, an uncritical focus on empowernment risks overlooking the broader effects of meaning-making in neoliberal governance.

Analysis

Three recurrent, meaning-bearing problem-representations were identified in the material: lacking freedom of choice, inflexibility, and inefficiency. Their counterparts, freedom of choice, flexibility, and efficiency were thus articulated as the drivers of change, or the goal of the policies. In line with the WPR approach these key terms are not defined beforehand, but instead they are seen as open to meaning-making processes, and thus the terms can carry various meanings. In the following analysis we will discuss how the concepts are represented and the various connotations that they carry. Each discourse is examined below, ending with a discussion of silences.

(Lacking) freedom of choice

Emphasis was placed on how the proposal for the new foodservice system called for patient freedom to choose, thus representing the problem as a lacking freedom of choice. The importance of patients’ freedom to make choices was emphasized in relation to benefits for the patients’ well-being.

Implementing the system . . . . . . will entail a quality enhancement for the patients. Flexibility is a theme that pertains the whole service system. There should be a wide selection of menu items and meals to choose from, possibility to eat at a suitable time and when one is hungry.

A wide selection of choices and service was also stated to guarantee future patient needs optimally, however, what was meant by future patients’ needs was not clear, as the meaning of the word needs can be ambiguous and it was not further elaborated on in the material. It was also specified that freedom of choice could be associated with increased food intake despite medical conditions making it more difficult to eat. Freedom of choice was also argued to be associated with lowering costs of meal production, building upon the assertion that providing patients with the freedom to choose their meals from a menu would lower the demand for special diets. The food was described as customized, and the menu described as follows,

The food is customized. The menu includes approximately 20 main courses, a bunch of snacks, appetizers, soups, sweets, and desserts. The patient gets to choose sides to their meal by themselves, e.g., pasta, cooked potatoes, rice, or mashed potatoes.

The possibility to choose when and what to eat seems to follow the underlying logic that individual choice is inherently favorable because it is not further argued how the benefits mentioned will develop or be realized. Treating choice as inherently good aligns with a neo-liberal discourse (Larner, 2000), and can also be related to the historical developments described, where individual centeredness has ideologically emerged. However, some perspectives were left unproblematized, which are discussed in the following section.

Firstly, in the excerpt above the food is called “customized,” and the selection of meals is arguably quite wide, but designed to cater to the average patient in this context. This is inclusive to a certain point, unless a patient’s dietary preferences fall outside the assumed norm. This not only concerns the menu items, but also seasonings and cooking methods. As Larner points out, the definition of who the “consumer” is can be broad to include many people or narrow to exclude some, which affects who the system takes into account and serves (1997). That being said, sometimes it is of greater value to be able to dismiss an option rather than choose a personal favorite. It should not be ignored, that part of the value of including menu options is the possibility to avoid disliked foods or foods with low compatibility with the patients’ cultural or ethical beliefs.

Secondly, the freedom to choose does not replace the necessity for quality control in terms of the sensory quality of the meals. If all menu options are perceived as equally poor, the argued value of the freedom to make choices becomes redundant. The topic of sensory quality of hospital meals was non-existent in the original proposal, it was never mentioned. However, at the time of implementation in 2020, sensory quality was brought up in the information material presenting the new foodservice system,

The meal is more than just food, especially at hospitals. For us the meal should be the highlight of the day. That is why our new foodservice system is based on the patient’s needs and is focused on the meal experience. To offer meal joy cooked on site is, for us, more than just food. There should always be tasty dishes, cooked with the best quality, to choose from. Food should contribute to faster recovery and recuperation.

Here sensory quality is mentioned, as dishes should be “tasty” and “cooked with the best quality.”

The freedom to choose also shifts responsibility of choice onto the patient. In Larner’s terms, choice works as a mechanism of responsibilization (2000). In theory, if patients are presented with the freedom to make choices and subsequently feel unhappy about the quality of the meals they are served, they indirectly become unhappy with their own choices because they have been presented with the opportunity to choose something else. Nutrition-wise, menu options and the freedom of choice also means that the responsibility of health is more actively shifted onto the patient. However, this is a complicated matter. In relation to nutritional guidelines concerning the proposed reform, the hospital had an explicit focus on food intake. A regulatory document for nutritional content stated that 85% of the main courses in the menu are required to have a minimum of 430 calories and 75% are required to have at least 20 g of protein per portion. Soups, appetizers, snacks, and desserts are exempt from this. The goal stated in the regulatory document is for the food to be appetizing and taste good to ensure it gets eaten, rather than following strict nutritional requirements. In regard to the old system with tray service at certain set times with a 6-week rotating menu, there was no way to control what patients chose to eat from their plates. The freedom to choose raises some questions in relation to nutrition care. How little may a patient choose to eat? How much may a patient choose to eat? How monotonous may they choose their diet to be? Also, the effects of these questions are tied to the length of stay of patients. In the end, these are issues that need to be solved in practice among the staff responsible for the patients.

Further, how “healthy eating” is understood in this context also affects how governing takes place, and how much of the responsibility to make the right choices is shifted onto the patient. If hospital malnutrition, and risk of developing malnutrition, is seen as the problem, and food intake in a general sense is seen as positive, then all menu choices can be seen as equally right. To exemplify, this would mean that a patient may choose based solely on hedonistic reasons, for example to eat only desserts from the menu and still be considered to make the right decisions. In turn, patients will make bad decisions if they choose not to eat or to eat very little. If “healthy eating” is understood as making the nutritionally most optimal decisions, then making the right decisions becomes more complicated. Choosing meals from the menu in this case shifts more responsibility to the patient. However, in both cases, for a patient-centered foodservice approach to truly be patient-centered, support from staff in decision-making is essential (Munthe et al., 2012). For patients at risk for malnutrition, monitoring food and drink intake is also essential, and whether this would be a task done by meal attendants or healthcare staff is important to discuss in the practical implementation.

There are some complexities associated to “freedom of choice” and the menu that are interestingly not discussed in the material. There are some cases where the a la carte menu design may pose a problem for certain groups of patients, such as patients with a long length of stay at for example, psychiatric wards, palliative care, or oncology wards, since the menu might become repetitive for them. In general, the proposal does not address practical issues like how the role of dietitians change when patients are free to choose what to eat, or how dietary restrictions caused by medical issues might limit the range of appropriate choices from the menu, thus reducing the emphasized “wide selection” of meals. What effect special diets may have on menu options in foodservice has also been recognized as a silence in other settings (Skinnars Josefsson, 2018: 19–20).

In sum, patients generally perceive choice as positive (see e.g. Dall’Oglio et al., 2015; Dijxhoorn et al., 2019; McCray et al., 2018a), and introducing freedom of choice facilitates the possibility to choose favorites and avoid disliked foods with low compatibility with the patients’ cultural and ethical beliefs. However, support in decision-making is essential for a patient-centered approach. Introducing freedom to choose shifts more responsibility onto the patient, and assuming a broad or narrow definition of the “customer” affects who’s needs are the focus of the menu.

(In)flexibility

Flexibility is a distinct characteristic of the new foodservice system, and emphasis is placed on providing flexibility, thus representing the problem as inflexibility. In the material, flexibility takes on double meanings and the first, “freedom of choice” as a guiding logic concerning the menu, has been discussed in the previous section.

However, flexibility was also presented as an organizational solution.

There is presently a desire within healthcare that meals are more flexible, that there are more dishes to choose from and that it is possible to serve patients at different times during the day. To facilitate ensuring that patients get all the meals needed and that someone takes responsibility for the entirety surrounding the meal it would be positive if somebody else rather than healthcare was responsible for food at the wards.

In this case, the problem was represented to be the old system’s inflexibility in relation to medical treatment. The underlying assumption was that the new foodservice system’s flexibility is beneficial for healthcare and improves allocation of staff resources. The problem was represented to be not having meal attendants, in other words having care staff serving the hospital meals, and the solution were the meal attendants. In the reformed system, service staff relieve health care staff from the responsibility of meal related tasks, so that everyone can focus on the areas in which they have professional training. The argument presented was that care staff do not have the time to focus on meal related tasks. Healthcare professionals are hard to recruit, staffing will get even more challenging in the future, and streamlining the job description of nurses is favorable. The time of healthcare staff was thus portrayed as something valuable.

Service staff “unburden” care staff so that they may have more time for patient care which was presented as their main function in the organization. Simultaneously the service staff, whose professional title at this point in time was meal attendant, were portrayed to be competent professionals in their own field who would take responsibility for the meals and meal related tasks.

Every ward should have meal attendants. They take responsibility for the process of meals throughout the whole day and are the ones who meet the patients during meal service. It takes time and care to convince patients the importance of eating, and healthcare staff does not have that time and care at present.

Interestingly, as the flexibility of the new system allows patients to eat whenever it suits them, the proposal itself also creates a need for service professionals such as meal attendants. Shifting roles and responsibilities to accommodate consumer needs is typical for neoliberal governance (Larner, 1997, 2000). The added focus on meals and service, presenting menus as well as manning the dining room and café counter showcase makes the system more labor intensive and time consuming compared to the mere distribution of meal trays at set meal times. Thus, the proposed changes expand the position of the foodservice within the hospital, by claiming that meal related tasks at the wards should be done by professionals knowledgeable in meals. In the material it was further proposed that there should be a clear distinction between the meal attendants and the care staff, as can be seen in the following:

The meal attendant should be a service-minded person with nutrition training, and their uniform should be clearly distinguishable from healthcare. This increases the intelligibility of them being a person working with food and not with healthcare. The meal attendant should also work towards a nice eating environment.

Thus, it was suggested that service staff should wear different colors to clarify they do not work with care. However, this distinction can be problematized.

While service staff might be organizationally differentiated from healthcare through the use of different colored uniforms, research has argued that it is not always easy to distinguish between work related to meals and work related to care. Hospitality in the hospital context illustrates this, as hospitality in this context can be viewed as a therapeutic effort to enhance patient well-being (Patten, 1994). Building on this view on hospitality, it has been shown to be an important part of the hospital mealtime experience (Jonsson, 2021). Further, while the relationship between hospital meal experience and food intake is not simple because of the challenges hospitalized patients may face with eating because of for example illness, staff at hospital wards have expressed time with patients as a way to increase their food intake (Jonsson and Nyberg, 2022).

Meanwhile the relationship between food intake and health outcomes is well known (Agarwal et al., 2013). In line with this, the statement “food is medicine” is used in the material to highlight the importance of meals in the hospital setting. However, there seem to be some tensions because while service staff is said to “unburden” healthcare staff and relieve them from work concerning meals it is simultaneously stated that meals have to be treated with the same respect as medical care. Multiple studies have shown this tension between medical care and mealtimes when it comes to prioritization at the ward (Eide et al., 2015; Heaven et al., 2013; Ottrey et al., 2019; Ross et al., 2011; Sathiaraj et al., 2019). In the material it was also stated that the reality of healthcare is that many patients are at medical appointments during mealtimes, which is why flexibility is needed.

While the new foodservice system is more labor intensive because of the flexibility it aims to provide in relation to medical treatments, the expansion is also necessary for the new system to realize its goal of being patient-centered. Patient- centeredness necessitates a partnership with a professional who helps patients make conscious and well-informed decisions, and Munthe and colleagues argue from an ethical point of view, that shared decision making and patient-centered care is resource and time-consuming task (Munthe et al., 2012). Time, in turn, is associated to labor costs which will be further elaborated in the next section.

In sum, flexibility was presented as an organizational solution, a strategic separation between healthcare and service staff aiming to delineate responsibilities and streamline operations. Notably, food was positioned as a therapeutic agent “food as medicine,” emphasizing its important role in patient care and recovery.

(In)efficiency

The proposed new foodservice system is described as efficient and modern, representing the old system as inefficient, thus representing the problem as inefficiency which is primarily articulated as “being costly.”

This proposal will entail a modern production and service system for us. . . . . . with regards to dietary management. Introducing this concept implies a substantially higher flexibility towards the patient, a more cost-efficient food production, and lower amounts of food waste.

The opening statement in the proposal for change was that the current foodservice had “an inefficient production leading to expensive meals.” The financial argument presented was that the remodeling from steam to electricity would be costly no matter what, and therefore it should be viewed as an opportunity to modernize the foodservice and reap additional benefits, described to be “efficiency, flexibility, and lower amounts of food waste”. Financial resources will be the focus of this section, making costs part of the problem representation.

First, some ambiguity exists in relation to effectiveness and efficiency (Zidane and Olsson, 2017), but often effectiveness is defined as the level of achieving objectives, while efficiency is defined as the relationship between the amount of resources used and the level of objectives achieved. In the term used in Swedish (effektivitet), both these factors are accounted for which should be noted as to not be lost in translation. Efficiency is presented as a neutral and desirable quality, but in line with Larner, cost metrics also act as political tools that shape governance (Larner, 1997, 2000). If efficiency means using resources to meet set goals, cost numbers are used as tools to claim the reform has met those goals.

High cost per meal in the existing foodservice system was stated as an explicit problem, and the new foodservice system was stated to be cost-effective and produce lower amounts of food waste. Food waste was described as an environmental burden, but also promptly associated to cost. In the material, an example was given of a month during which 383 meal trays were “sent back to the central kitchen untouched,” described as a huge environmental burden and associated to a cost of 35,833 SEK, a cost for health care, which was also explicitly highlighted. In similar fashion, the adoption of menus and introduction of freedom of choice ³ was also associated to lower total costs because it was reasoned it would lower the demand for special diets.

Special diets are in general costly for foodservice because they are prepared in a labor-intensive diet kitchen, a separate part of the kitchen, and mass production advantages are lost in terms of cost control. Altogether, it was suggested that the new foodservice system could be more cost-efficient, stating possible savings of up to 30%. The new design was described as much more efficient in terms of labor;

If described foodservice design is accepted the production unit could be much more efficient and performance in the kitchen would be much higher. The kitchen area could be decreased and during weekends run with only minimal staff who would take care of any special diets and doing dishes.

The new proposed foodservice system was a cook chill production model, which is a less labor intense way of producing meals compared to the cook serve system which it would replace (Burns and Gregory, 2007). Since labor is one of the largest expenses in foodservice, cook chill is usually the less expensive alternative although the initial investment can be costly. However, the new foodservice solution also includes the addition of meal attendants, staff that are proposed to take over the task of serving meals to patients at the hospital wards. While changing the form of meal production from cook serve to cook chill may lower staff requirements in the kitchen, the concept of the new foodservice increases staff needs at the wards. The salary of meal attendants is not calculated into the cost per meal since it is not a foodservice expense at the point of production. Interestingly, the cost of this new category of staff is argued for in the long term, relating the financial burden, to shorter patient length of stay and increased quality of life. Although financial facts are used as leverage to argue for the proposal, it is unclear if there is any immediate financial net gain from adopting the new foodservice system since the meal attendants will be an “additional cost.” It is, however, also argued that care staff spend a lot of their working hours dealing with meal related tasks. In the analyzed material it is stated that based on interviews from 2012, care staff spent on average 8.8 hours per ward on meal related tasks while working hours for the meal attendants were calculated to be 7 hours per day and ward. A restructuring of labor and cost between the healthcare and foodservice budgets, which are separate, is never explicitly stated in the material as a part of the proposal, but it is implied through presentations of estimates of working hours such as described above. It seems like discussions of how costs should be divided has been a concern for those representing the healthcare in discussions concerning the proposal.

Meal attendant with nutrition training who is organised by foodservice is, according to the health and medical care administration, an internal allocation of staff, based on described reduction in staffing needs in the new kitchen. Reduction of healthcare staff according to previous description is not deemed feasible.

While positive to the concept of the new foodservice as a whole, and especially to the flexibility the system allows healthcare, lowering the amount of care staff at the wards in favor of recruiting meal staff is out of the question. Lowering the amount of care staff was never explicitly stated in the proposal, but evidently it was a feared effect among healthcare management.

In 2014 the proposal for the new foodservice was accepted by the regional council and the investment was calculated to be 26.8 million Swedish Crowns. In 2018 it became clear that the remodeling needed was much more extensive, and the investment became thrice the amount at 82 million. This raised questions about the profitability of the foodservice system, but at this point it was too late to change the concept. Two of the political parties of the regional council signed resignation to the decision of accepting these increased costs, stating that the decision making took place before election of the current regional regime. The following excerpt shows the reasoning of one of the two:

The remodelling of the hospital kitchen was decided before the elections of 2014 and the investment was calculated to be 26,8 million Swedish Crowns. The intention was to adjust food production into a new flexible system so that inpatients would be given more freedom to choose menu items and time of their meals. This is a good thought that reduces the risk of malnutrition. However, the benefit for patients declines in tandem with increasingly shorter lengths of stay and the financial investment which has now been calculated to be tripled to 82 million Swedish Crowns has to be considered in relation to this. [Party name]’s judgement is that the new total cost by far exceeds the patient benefit of the new flexible foodservice. . . (shortened)

The perceived benefits of the foodservice, lowering risk for malnutrition, were put in relation to the cost and deemed inadequate. These perceived benefits relate to how hospital meals are understood, and what the value of meals are perceived to be.

In sum, efficiency was emphasized as one of the main arguments for replacing an outdated system with a desirable cost-effective and contemporary alternative. The less labor intense cook chill production kitchen came hand in hand with a service concept with meal attendants at the wards. In the end, costs can be perceived both as a problem representation but also as something left unproblematic, as it is never really clarified how the costs shifted from the kitchen to the wards will be divided between foodservice and healthcare or what the net gains or losses of the reform will be. The proposal focused on the overall cost benefits of stream-lining production and reducing food waste, but left the cost of service-staff unallocated and politically invisible – although representatives of healthcare raised concerns about their budgets.

Silences left unproblematized

In relation to the presented problem representations, we suggest at least two “silences,” according to the analytical approach (Bacchi, 2009); firstly the practical implementation of the new organizational structure between foodservice and health care which includes negotiations of staff responsibilities and budget allocations but also the working situation of the production kitchen staff, and secondly addressing a framework for the menu which in turn forms the basis of the foodservice output. Our analysis suggests that these are taken for granted, which can be seen as either trust in the implementation at the practical level, or indifference at the political level.

Firstly, a lot of focus is put on how the meal attendants can “unburden” care staff, and while some attention is given to their role as service professionals the effects of this are left largely undiscussed in the political forums. Hospitality has gained a lot of attention in hospital foodservice and meal science literature the past years for its importance for the meal experience (Jonsson et al., 2021; Justesen et al., 2014; Sporre et al., 2017). This was hardly given attention in the 2014 proposal presented before the regional board, however, in a research & development grant application to Vinnova, Sweden’s innovation agency, “meal experience,” and “the last 30 cm to the patient” were explicitly stated as the meal attendants’ responsibility, who were also stated to be key figures.

In general, arguments in the proposal were positioned to benefit the healthcare, and this may at least in part explain some of the problem representations and silences. Kitchen staff are invisible in the proposal, although a remodeling of the kitchen affects their workplace. They are mentioned only once in a comment to the proposal by the foodservice manager who brings up concerns of their working environment, especially the dishwashing station which is 35 years old. The silence concerning kitchen staff is especially interesting in relation to discussions concerning costs, where it is mentioned that the new foodservice system is more cost-effective in terms of labor. Lay-offs are never mentioned or even implied, but it is unclear where staff will be relocated if less kitchen staff is needed in the new foodservice system.

In general, the arguments are positioned to be perceived as favorable by healthcare, but also to protect it, which can be seen following the reasoning, of unburdening healthcare staff, protecting their time, and making meals flexible around medical treatments. However, many practicalities of the new organization concerning meal related tasks are left unaddressed. Allowing patients to order meals at any time demands meal hosts manning the wards and serving the meals, but care staff are never free of all responsibility concerning the meal situation. Only care staff may for instance touch bed-ridden patients to ensure patient safety, and thus need to be available for assistance when these patients wish to eat. The practicalities of these situations are left unaddressed, as is the division of costs between healthcare and foodservice in terms of labor related to meals at the ward since it was unclear at the point of decision-making who would bear the cost of the meal hosts at the ward. Additionally, food first approaches (Cheung et al., 2013), like protected meal-times, in which the hospital meal would be prioritized were not brought up, and neither was additional nutritional training for health care staff.

Secondly, the menu was not addressed in-depth. The menu in foodservice can be seen as a management control tool (Gregoire, 2013) since it is the deciding factor on what will be produced, and in the proposal focus was put on freedom of choice and not the menu options. However, they could be seen as taken for granted qualities of the foodservice, as prerequisites for any hospital foodservice and therefore unnecessary to address. This suggest either a high level of trust or indifference at the political level. Nevertheless, since freedom of choice is a strong point of focus, the menu provides a framework within which these choices will be ultimately made, affecting what ends up on the patients’ plate.

In the proposal for the new foodservice, allusions to private sector restaurants are made both explicitly and implicitly. Part of the proposed concept is a dining room at each hospital ward with a chilled counter showcasing snacks and desserts, described to look like a café. Patients are often referenced using the word customer, end-customer, or guest. Distancing the proposed hospital foodservice system from earlier known foodservice systems is made to seem preferable. The new foodservice system is portrayed as something new and different, however, the technologies of the foodservice system itself are not novel. Additionally, there are some interesting choices of words in the description of the new solution, like “cooked on site,” and “plated on china” that would not be necessary if they were thought to bear no meaning. These words target some of the problems that have been associated with public foodservices in the past, like unsightly plating of food, and aim to signal that this new foodservice system is different. In the end however, the actual quality of a meal depends on the operational adaption of the foodservice system and the professionals trained to work in it. Hence, the implication of these silences will be visible in the practical implementation of the reform, which we will address in upcoming work.

Conclusions

The foodservice reform analyzed and discussed in this paper, which reflects many similar reforms in Sweden and elsewhere globally, illustrates the double-edged sword of neoliberal governance. These insights can inform reforms beyond the Swedish context by elucidating how problems are represented in neoliberal discourses and what effects these discourses might have. For example, the introduction and general approval of choice in public services is salient. The new foodservice system allows patients more freedom of choice, thus empowering patients but also shifting part of the responsibility of nutrition care and health over to them making them active agents in their own healthcare whether they want to or not. The meal host plays a mitigating factor concerning individual responsibility, as the role implies helping patients make menu choices by providing hospitality and information, inspiring ill patients to eat, and trying to make sure patients with many medical appointments have the time for all their meals. Notably, a focus on individual responsibility of health and nutrition is not limited to the hospital context, it is widespread in public health promotion and dietary advise discourses as well (cf. Bergman et al., 2018, 2019, 2020; Crawford, 1980; Harwood, 2009; Michailakis and Schirmer, 2010; Qvarsell, 2005).

In this paper, choice, flexibility, and cost-effectiveness (efficiency) are not simply concepts that reflect policy goals or operational priorities. Rather, they are discourses that function as political instruments that help shape realities, construct subject positions, and legitimize particular technologies of governance. Choice is a tool of responsibilization and consumer framing that shapes the position of the patient as a customer expected to make individual decisions. Choice is not necessarily by itself a neoliberal logic, however in this case it is framed within a broader culture of consumerism. The question remains what effects this will have on patients who are unable or unwilling to make decisions. If there is a lack of support in decision making, due to for example time constraints of staff, it might have a negative health impact on certain groups of patients. Flexibility and cost-effectiveness (efficiency) produce professional roles that are streamlined and centered around cost metrics rather than in-patient care practices. What effects this will have for the professions working in these conditions is uncertain.

It was clearly stated in the material that the initial catalyst for change was to decrease the cost per meal, and this can be achieved by decreasing kitchen labor, but by introducing the additional staff category the meal host entails it is not evident that a decreased cost is achieved overall. In turn, budgets have to be renegotiated between healthcare and foodservice. Which party should bear the cost of the meal hosts was not fully solved at the time and may lead to tension between foodservice and healthcare in the practical implementations of the reform.