Hospital accreditation systems and salience of organisational tensions

Abstract

Purpose

This study examines how an externally imposed management control system (MCS) – hospital accreditation – influences the salience of organisational tensions and consequently attitudes of management towards the system.

Design/methodology/approach

Data are collected using a case study of a large public hospital in Spain. In-depth interviews were conducted with 27 senior and middle managers across different functions. Relying on the organisational dualities classification in the literature, tensions are unpacked and analysed.

Findings

Evidence is presented of how hospital accreditation increases the salience of organisational tensions arising from exposition of the organisational dualities of learning, performing, organising and belonging. Salient tensions were evident in the ambivalent attitudes of management towards the hospital accreditation system.

Practical implications

The role of mandatory external control systems in exposing ambivalence and tensions will be of interest to organisational managers.

Originality/value

The study extends the management control literature by identifying an active role for an external MCS (accreditation) in increasing the salience of organisational tensions and triggering ambivalence. Contrary to the prior literature, the embedding of both poles of an organisational duality into the MCS is not a necessary precondition for increased tension salience. The range of attitudes towards MCSs beyond those specified in the previous literature (positive/negative/neutral) is extended to include ambivalence.

Citation

Robbins, G., Sweeney, B. and Vega, M. (2022), "Hospital accreditation systems and salience of organisational tensions", Accounting, Auditing & Accountability Journal, Vol. 35 No. 9, pp. 57-80. https://doi.org/10.1108/AAAJ-04-2020-4500

1. Introduction

I feel it is too intense, the process of accreditation itself. Those three days of the accreditation were difficult. There was lots of stress and tension. […] It was fourteen hours per day and was exhausting, particularly mentally. (Nursing Director)

Healthcare organisations face a myriad of tensions, such as shrinking resources and expanding targets (McCann et al., 2015), stability and change, and, as evident in the above quote, tension between time constraints and the additional work arising from a healthcare accreditation visit. While we know that the salience of organisational tensions can vary (Van der Kolk et al., 2020), previous research has questioned its drivers (Knight and Paroutis, 2017; Lewis, 2000). Tensions require continuous attention from managers (Van der Kolk et al., 2020), and to effectively manage them, there is a need to expose their drivers. Organisational tensions are socially constructed and made salient through environmental conditions (Lewis, 2000; Smith and Lewis, 2011) and adoption and implementation of complex internal processes, such as management control systems (MCSs). While the management control literature has long noted a role for internal MCSs in managing and balancing tensions (Curtis and Sweeney, 2017; Mundy, 2010; Simons, 1995; Taylor et al., 2019), further empirical work is needed on how MCSs influence the salience of tensions (Lewis et al., 2019). This study examines the implementation of healthcare accreditation systems and focuses specifically on the salience of organisational tensions for hospital managers, i.e. their experience of the relationship between alternate poles as both contradictory as well as interrelated (Smith and Lewis, 2011; see also Knight and Paroutis, 2017; Lewis, 2000).

Implementation of healthcare accreditation is an area which has been given inadequate consideration in studies to date (Brubakk et al., 2015) and merits additional scrutiny (Greenfield and Braithwaite, 2008; Wilson, 2018). There is recent evidence of ambivalent attitudes of hospital management towards ranking systems (Wallenburg et al., 2019). Tension salience can trigger ambivalent attitudes (Ashforth et al., 2014) through the experience of organisational dualities as both contradictory and interrelated (Smith and Lewis, 2011; Lewis, 2000). Ambivalent attitudes towards accreditation are not surprising, given that these systems do not always achieve intended outcomes (Brubakk et al., 2015; Wilson, 2018). We focus on attitudes towards the accreditation process of individual managers within a hospital management group who are responsible for accreditation implementation and address the following research question: How does the implementation of a hospital accreditation system influence the salience of organisational tensions experienced by hospital management and their attitudes towards accreditation? We address this question theoretically by mobilising the literature on tensions and ambivalence and empirically by conducting a case study of the implementation of an accreditation system in a large hospital in Spain.

We contribute to the literature by first, responding to calls to examine how organisational tensions become salient in organisations (Knight and Paroutis, 2017). We identify an active role for an external MCS (accreditation) in increasing the salience of organisational tensions through juxtapositioning underlying dualities of learning, performing, organising and belonging (Smith and Lewis, 2011). Second, we show that contrary to the prior literature (Lewis et al., 2019), the embedding of both poles of an organisational duality into the MCS is not a necessary precondition for increased tension salience. Third, our study adds to the small body of knowledge on MCSs and ambivalence by demonstrating a role for MCSs in triggering ambivalent attitudes through the increased salience of organisational tensions. It thereby addresses the need to better understand the “contextual” sources and triggers of ambivalence (Ashforth et al., 2014; Plambeck and Weber, 2010). Our study extends the range of outcomes of MCSs examined in the literature beyond positive, negative and neutral attitudes to control systems (Tessier and Otley, 2012) to include ambivalent attitudes. Finally, our findings highlight the complexity of accreditation implementation (Brubakk et al., 2015) and respond to calls to further examine implementation of health service accreditation (Greenfield and Braithwaite, 2008; Wilson, 2018).

2. Literature review

The review of the extant literature commences with a review of the literature on healthcare accreditation after which the broader management literature on organisational tensions is reviewed. The role of MCSs in increasing the salience of tensions and the importance of considering attitudes towards MCSs (particularly ambivalence) are then discussed. Finally, the research question addressed in the study is presented.

2.1 Accreditation systems

Systems which assess and certify organisations using technologies of quantification and verification have spread over the past 30 years – the audit explosion (Power, 2003). These systems are part of the expanding regulatory infrastructure and typically involve standard setting, certification and accreditation (Gustafsson and Tamm Hallström, 2018; Loconto et al., 2012). Those charged with evaluative responsibilities provide a surveillance role through audit and evaluation of government programmes frequently influencing the provision of future government funding (Donovan and O'Brien, 2016). When accreditation is granted to a public body, the public is entitled to trust that the entity in question meets minimum standards of operation (Baker et al., 2014). In fact, accreditation may be used by accredited organisations as a signalling mechanism of strong governance (Feng et al., 2016, 2019). Consequently, there may be reputational or financial consequences if accreditation is withheld, or not renewed, for an organisation (Andon et al., 2015).

In relation to healthcare, insufficient resources to meet healthcare needs in terms of quantity and quality have led governments globally to rely increasingly on external forms of certification in a search for improved performance from available healthcare resources (Bevan et al., 2019). Accreditation systems can be powerful forces for quality and change in complex systems (Greenfield et al., 2019), such as hospitals. In fact, healthcare accreditation systems are now pervasive reflecting a desire for external assurance of quality (Cooper et al., 2014). Hospital accreditation – an instrument of management control focused on efficiency and effectiveness – incorporates processes of self-assessment, evaluation, dialogue and change recommendations. Healthcare accreditation is a reconstitution of many of the elements that are used to define financial auditing, such as the generation of reports that become the subject of scrutiny, application of standards and identification of action required based on compliance and exception reports (Andon et al., 2015). “Making quality auditable” (original emphasis) involves a process of generating an “auditable front” for an organisation (Power, 1996). In the case of healthcare accreditation, this takes the form of a self-assessment report. This key element of the accreditation system's architecture specifies the domain of facts (Power, 1996) presented for external certification. The interplay between exerting management control and making visible the organisation's structures, systems and processes in this auditable front may be sizeable (Brivot and Gendron, 2011). The domain of facts to be audited is supported by the lexicon of audit and verification, which operates as a form of power through socialisation processes in different institutional and cultural contexts (MacLullich, 2003). However, it is acknowledged that there is a real risk that the audited system as revealed through its structures, systems and process can be detached from substantive organisational performance (Power, 1996). Yet, improved organisational performance is at the heart of accreditation efforts.

Organisations are increasingly pursuing change in work practices in an effort to improve performance (Eisenhardt, 2000). Yet, internal organisational forces – for example, established routines, inertia and staff familiarity with existing processes – promote stability in organisations (Ashworth et al., 2009; McNulty and Ferlie, 2004). Simultaneously, external institutional forces – government and the public (Schyve, 2000) – make increasing demands for public sector organisations to measure quality using performance evaluation strategies, such as accreditation (Ng et al., 2013). The result is a dynamic tension for individuals within these organisations as they deal with the push and pull of external forces, seeking, for instance, improved quality and reduced cost or short-term and long-term improvements. The next section reviews the literature on organisational tensions.

2.2 Organisational tensions

Organisations are facing increasing oppositions as new technologies are developed, new work practices emerge and as the boundaries of organisational life increasingly blur (Fairhurst and Putnam, 2019). The recent Covid-19 pandemic has dramatically increased the salience of some of these oppositions in organisations, including the short-term vs long-term tension (need to respond quickly to the crisis and also to take actions to ensure the long-term viability of an organisation); the use of information tension (the need for greater use of technology for knowledge sharing and also the need for stakeholder privacy) and the tension between cooperation and competition with other organisations (crisis revealed the need for coopetition between for example pharmaceutical companies to problem solve) (Carmine et al., 2021). The term oppositions used here refer to the concepts of “tensions (stress-inducing oppositions)” and “contradictions (interdependent oppositions that potentially negate each other)” (Fairhurst and Putnam, 2019, p. 917/8). These elements form a duality in that they are “oppositional to one another yet […] also synergistic” (Smith and Lewis, 2011, p. 386). For example, evidence of the imposition in hospitals of a challenging target for increased volumes of patient treatments without any additional resources is contradictory to the need for integrity of patient care (McCann et al., 2015), but is complementary to driving a search for innovations that increase efficiency of care and lead to better patient outcomes. The target in this case exposes the salience of the tension between treating more patients and protecting integrity of care.

Contradictory yet complementary “interrelated dualities are embedded in the process of organizing” (Smith and Lewis, 2011, p. 288). They are brought into opposition by environmental conditions, such as plurality, change and scarcity (Farjoun, 2010; Smith and Lewis, 2011, p. 288), and conditions reflective of implementation of accreditation in hospitals. Salient tensions reflect a practitioner's perspective on tensions and are built by making interpretive linkages over time (Knight and Paroutis, 2017). A focus on tensions is important as they require continuous attention from managers due to their dynamic, constantly changing and complex nature (Van der Kolk et al., 2020). However, attention can only follow their exposition as salient tensions, which is the focus of this study. It is through complex internal processes and environmental conditions that organisational tensions are socially constructed and made salient (Lewis, 2000; Smith and Lewis, 2011). A key component of the organisational architecture for management is the MCSs relied upon to manage organisational performance and deliver services. The next section reviews how MCSs can influence the salience of tensions.

2.3 MCSs and tension salience

While the MC literature has long noted a role for internal MCSs in managing and balancing tensions (Curtis and Sweeney, 2017; Mundy, 2010; Simons, 1995; Taylor et al., 2019), relatively little attention has been given to the role of either internal or external MCSs in increasing the salience of tensions. Lewis et al. (2019) focus specifically on the tension between control and empowerment and theorise how the design and use of an internal MCS can impact on the salience of the tension for an individual. They contend that the salience of a tension relates to whether an attempt is made to embed the competing elements of the tension in organisational processes. If “no attempt is made to embed both competing elements in organisational processes, the tension does not manifest in the design or use of MCS and is not obvious to organisational members” (Lewis et al., 2019, p. 489). Lewis et al. present the ideal type of rational bureaucracy as an example of a system where inherent tensions that exist between the empowerment of staff and tight managerial control remain latent as the control system does not engage these competing elements. They also provide an example of a situation where a system may create an illusion of engaging with competing elements, and in this instance, the tension also remains latent due this time to strong socio-ideological control working to disguise the source of the tension. Overall, while Lewis et al. (2019) provide some theoretical insight into how MCSs influence the salience of tensions, it is unclear how exactly tensions come about (Van der Kolk et al., 2015), and further empirical work is needed on MCSs and the salience of tensions (Lewis et al., 2019).

The MC literature has pointed to the importance of distinguishing attitudes to the system from the characteristics of the system and has identified positive, negative and neutral (but not ambivalent) attitudes to control systems (Tessier and Otley, 2012). Yet, a growing body of research elsewhere argues that generally in organisations, ambivalence may be the norm rather than the exception (Rothman et al., 2017). Ambivalence refers to the simultaneous presence of opposing attitudes or feelings towards a particular issue, object, etc. (in this instance accreditation) (Ashforth et al., 2014; Rothman et al., 2017). Ambivalence can be distinguished from neutral attitudes, which do not involve simultaneous opposite evaluations (Tessier and Otley, 2012), and from ambiguity, which relates to uncertainty or a lack of clarity (Ashforth et al., 2014). We focus on attitudinal ambivalence which encapsulates simultaneous positive and negative attitudes toward an object, event, etc. (Conner et al., 2002).

A greater understanding of what triggers ambivalence is important (Ashforth et al., 2014) given that expressing ambivalence can result in “collective responses” (Weigert and Franks, 1989, p. 223), and may offer an opportunity for motivating new action and a break from old routines (Piderit, 2000). Organisational events are one source of ambivalence (Rothman et al., 2017). Accreditation has become an important organisational event; indeed, as noted in our introduction, there is recent evidence of ambivalence of hospital management towards ranking systems (Wallenburg et al., 2019). The salience of persistent tensions – the individuals' view that the relationship between alternate poles is both contradictory and interrelated – is increased as organisations face conditions of change, scarce resources and plurality (Knight and Paroutis, 2017; Lewis, 2000; Smith and Lewis, 2011), conditions that characterise healthcare contexts. Salient tensions may present as a fundamental ambivalence about what is often called the search for “progress” which arises from the emergence of disquiet about accelerating managerial dominance in public sector organisations staffed by strong professional groups (Farjoun, 2010; Lapsley and Miller, 2019). While it has been established that ambivalence can assist the performativity of imperfect performance measures (Dambrin and Robson, 2011), it also has been associated with a range of negative outcomes (Conner et al., 2002; Guarana and Hernandez, 2015). Little attention has been given to the role of MCSs in generating ambivalence. An exception is Van der Kolk et al. (2015) who suggest that the design of internal MCSs needs to “allow” for human ambivalence between behaviour of employees as agents and stewards by incorporating constraining and facilitating elements.

Building on insights from the literature studies reviewed on MCSs, tensions and ambivalence, our study addresses the following research question: How does the implementation of a hospital accreditation system influence the salience of organisational tensions experienced by hospital management and their attitudes towards accreditation?

We adopt Smith and Lewis's (2011) categories of learning, performing, organising and belonging as analytical tools to examine the organisational tensions experienced by hospital management. Smith and Lewis view learning tensions as emanating from the “efforts to adjust, renew, change, and innovate” (p. 383). Learning tensions arise when organisational members are challenged to focus on both today and tomorrow and to look forward and backward simultaneously. Learning tensions arise from “efforts which involve ‘building upon, as well as destroying, the past to create the future’” (O'Reilly and Tushman, 2008; Smith and Lewis, 2011, p. 383). Performing tensions stem from the multitude or plurality of stakeholders and can sometimes result in competing strategies. Highlighting goals and strategies, such as the relative importance of local vs global, brings to light what is and is not emphasised and creates performing tensions (Smith and Lewis, 2011). Organising tensions occur in relation to collaboration and competition, empowerment and direction, and control and flexibility (e.g. Adler et al., 1999; Siggelkow and Levinthal, 2003). As Smith and Lewis point out, defining how an organisation operates also defines how it will not operate and hence creates organising tensions. Conflicting identities, roles and values create belonging tensions for those within organisations when they are asked to respond to changing organisational requirements (Smith and Lewis, 2011). Identifying as an individual and as part of a collective fosters belonging tension (e.g. Pratt and Foreman, 2000), a tension that individuals are exposed to in organisational life. These four tensions are used to analyse our findings and provide insight into the drivers of tension salience.

3. Research method

We adopt a case study approach given its suitability to understand the complexities and multifaceted aspects of social phenomena in complex settings (Ahrens and Chapman, 2006; Yin, 2009). Furthermore, it is useful in examining social aspects of real-life contexts where individuals and organisational processes have strong and intricate relationships (Yin, 2009). Data are collected on a large high-technology government hospital in Catalonia, Spain (referred to as Hospenda) with 600 beds serving almost half a million population. Hospenda underwent an external accreditation process for the second time in 2013.

The unit of analysis is the collective of the management group of Hospenda involved in the adoption and implementation of accreditation, comprising individuals with responsibility and discretion for changing policies and practices. This is the group referred to as “management” in this organisational entity for which we aim to interpret the findings on implementation of an external control system using the analytical tools of Smith and Lewis' (2011) organisational dualities of learning, performing, organising and belonging. We draw on the concepts of organisational tensions, salience and ambivalence from the literature as these have emerged as having meaning and relevance from detailed analysis of the data. It is through a synthesis of individual views that a collective view emerges on findings related to the research question. To enable a comprehensive understanding of the accreditation system, we also collect data from external assessors and regional health department advisors.

Data collection consisted of semi-structured interviews, archival records, review of internal and external reports, and observational information gathered during site visits. A semi-structured interview approach was used. Interview questions based on a critical review of the literature studies in the previous section of this paper and focusing on structures, processes and outcomes of accreditation were designed to be “open-ended, neutral, singular, and clear” (Patton, 2002, p. 295). Data were collected in two phases over a 12-month period with 27 face-to-face interviews conducted in total, averaging 45 min each (8 in the first phase and 19 in the second) (refer Appendix 1 for list of interviewees from Hospenda, the regional Catalan health department and the independent accreditation assessors).

A purposive sampling strategy (Patton, 2002) was adopted for identifying hospital staff and external connected parties – surveyors, assessors, advisors and regional health department staff who had responsibility for accreditation. During the first set of interviews, the key role of the accreditation system in securing contractual agreements with the main purchaser of public healthcare services emerged. Furthermore, at that time the hospital was in the process of being re-accredited and hospital management were vocal in articulating their attitudes towards this external control system. Following this, the research question was refined, and the researchers requested to interview candidates with extensive experience of accreditation during the second phase of data collection. Individuals from a range of hospital departments were identified, including, nursing, medical, economic and finance, human resources, general services and projects, and organisation systems (see Appendix 2 for summary hospital organisational management chart). Where possible, interviews were recorded (participants agreed to be recorded in 18 of the 27 interviews). Hand-written comments and notes were used to support recorded and non-recorded interviews.

We reviewed documents including strategic plans, annual reports, regional management reports, management awards, newsletters and annual, monthly and weekly financial and non-financial performance measurement system information. In addition, we visited the hospital facilities (i.e. logistics and purchasing, facilities management and information technology departments) on three occasions to observe in situ how individuals carried out their work activities and work practices. This proved valuable in understanding everyday practices; for example, during the 2.5 h interview with the Logistics and Purchasing Manager, one of the researchers had the opportunity to observe the Logistics Assistant, the Purchasing Assistant and a Pharmacist adjusting the hospital coding process for stock items to avoid duplicates and enhance traceability. This non-participant observation period was useful in highlighting how the accreditation implementation process required detailed consideration and refinement of the purchasing software used to manage large volumes of stock.

It is acknowledged that qualitative data are messy and transforming them is challenging (O'Dwyer, 2004), and this was the case with our data. It demanded “continuous back and forth questioning of interpretations and discussion of recorded field data” (Ahrens and Chapman, 2006, p. 833). We followed Marshall and Rossman's (2011) data analysis steps: organisation of data, data immersion, identification of initial codes, initial interpretation, discussion of patterns, further coding, search for negative evidence and presentation of findings. First, we transcribed and translated the interviews, notes and other documents. We then organised the data files and immersed ourselves in the data by reading and re-reading transcripts and other documents several times to ensure thorough comprehension. Based on this, we generated “thick” (Geertz, 1973, p. 6) comprehensive descriptions in word documents of issues surrounding the implementation of the accreditation system, such as the features of the accreditation system (e.g. mandatory form, implementation process, types of metrics used, etc.).

This process of familiarisation with data allowed aggregation of data into an initial set of themes or “higher order headings” (Graneheim and Lundman, 2004, p. 106) linked closely to the research question and literature review. For example, attitudes towards the accreditation system were divided into (1) positive, (2) negative and (3) simultaneously positive and negative. We discussed patterns in these interpretations and evidence to support them. The ease/hesitation with which interviewees moved between expressing negative and positive attitudes and their recognition of interrelationships and contradictions between different aspects of accreditation were not always immediately apparent from transcripts but came to light from the researcher's notes and during discussions between the researcher who carried out the interviews and the other two members of the research team. For example, tensions were observed during interviews in the body language and hesitation and apparent challenge experienced by interviewees in responding to questions about their experience of implementation of this external control system. These discussions between the researchers resulted in a second round of coding to refine themes emerging (e.g. contradictory relationships, different types of tensions, etc.). To aid our analysis and the questioning of field data, we developed templates and tables to search for patterns in different attitudes towards accreditation and assess the evidence supporting our interpretations. We sifted through emerging themes and then accepted, rejected or relegated based on their prevalence. This iterative analysis facilitated a constant dialogue between the researchers, the data and theory (Bazeley, 2013) and served to validate or invalidate themes through the emergence of patterns that supported a coherent story. In drawing conclusions, we re-examined data to search for inconsistencies and other evidence as suggested by Miles and Huberman (1994). Finally, we selected quotations which expressed consensus views so as to allow us present the “thick description” (Denzin, 1994) reported in the next section.

4. Findings and discussion

This study focuses on a hospital in Catalonia where mandatory external accreditation for acute care hospitals was introduced in the early 1980s to provide assurance on quality standards. The introduction of hospital accreditation in Catalonia is consistent with worldwide regulation trends on hospital accreditation (Touati and Pomey, 2009). Accreditation audits are performed by independent for-profit audit firms selected by the Catalan Health Ministry and involve an assessment of hospital structure, processes and results. Hospitals can choose their auditor from a number of approved auditing firms.

Findings are reported on how the hospital accreditation system influences the salience of organisational tensions and the attitudes of management. Findings are structured around the different features of the accreditation system: standardisation, comprehensiveness, multiple objectives and mandatory external form. While it is known that tensions can exist at the individual, group or organisational level or indeed across these levels (Smith and Lewis, 2011, p. 384), our focus is on tensions at the organisational level expressed by individual managers who have either clinical, administrative or quality management roles. Discussion of findings at the organisational level relies on a synthesis of the attitudes of individual managers involved in accreditation and their reflections of tensions and the interpretive linkages they have drawn (Knight and Paroutis, 2017). Our reporting and discussion of findings provides this synthesised analysis of evidence gathered from documents, interviews with individual staff and non-participant observation. The interview process was used to capture a situated account of the attitudes of management to implementation of this external control system. Engeström and Sannino (2011) point to the importance of focusing on participants' language, especially their words and phrases to reveal oppositions. In each of the following sections describing the attitudes of managers towards the main features of accreditation, we draw out the oppositions that underlie mixed positive and negative views of interviewees towards the implementation of the accreditation system. Tensions result from the plurality of organisational dualities. We rely on the organisational dualities as classified by Smith and Lewis (2011) of learning, performing, organising and belonging to analyse our data. Using these analytical tools, we unpack how the mandatory quest for relevant standardised performance indicators that are intended to be comprehensive brings these organisational dualities into juxtaposition. We show how the mixed positive and negative attitudes of Hospenda management are rooted in these dualities, which reveal and reflect inherent and persistent tensions.

4.1 Standardisation

The essence of accreditation is to apply a common set of standards to evaluate organisations and award accreditation. The accreditation system was consistently described as a “one size fits all” approach (Sax and Marx, 2014), which did not allow any type of customisation of essential standards by hospitals. It was viewed as “restrictive” (Medical Director) and “strongly regulated” (Human Resources Assistant Director). There was evidence of tensions related to the use of a standardised list of performance indicators and a mix of positive and negative views in relation to the desirability of standardisation. For example, interviewees recognised that standardised processes resulted in improvements and that this was very useful in managing change:

We reviewed the whole process related to the Emergency Room (ER) services because this department was going to be relocated in a new building by the end of the year […] this was a good exercise for reflection. The accreditation process helped us to develop the standards and processes and to prepare the functional plans for this new ER department (Medical Director).

However, many interviewees juxtapositioned the benefits and drawbacks of accreditation as evident in the following quote:

My first thought when someone talks about the accreditation is “Ugh, it gives me the creeps!”, but honestly when you use it and you see good results, then you realise it is worth the effort again (Administrative Contracts Manager).

Interviewees expressed concern that there was little scope for creativity and flexibility in this second round of hospital accreditation in Catalonia. In recognising the contradictory interrelationship between the improvements that result from increased structure and the constraints to innovation from reduced flexibility, there was evidence that the Assistant Director of Human Resources experienced tension salience between the two poles (standardised processes and flexibility):

I am in favour of having these mechanisms [accreditation system] because they help you in some way to improve processes and indicators, but sometimes they are so structured that they unfortunately restrict people's capacity to use their own initiative […] I believe that sometimes having too much structure limits things related to innovation, creativity, looking at different approaches with an open mind to say: what will happen if we do things differently? (Assistant Director of Human Resources).

There was also evidence of a salient tension between the reduced usefulness of standards that were not tailored to individual hospitals and the advantages of increased comparability of accreditation results between hospitals: “Because it is the same tool for everyone” (Deputy General Director). While the Medical Director enthusiastically recalled the usefulness of having developed standards and processes for accreditation in managing the move of the ER department to a new location in the earlier quote, he believed that accreditation standards should reflect the nature and complexity of services provided by various types of hospitals (i.e. community hospital, high-technology hospital etc.) and that

The same accreditation model should not be used for all hospitals. Performance should be evaluated across different hospital types. They should use at least two or three hospital types based on similar complexity to be able to compare hospital results (Medical Director).

The Quality Manager pointed to the absence of standards around critical Hospenda priorities as a high technology hospital, such as the adoption of technology and teaching and research activities. However, the use of the same tool for everybody increased the scope for “external collaboration and knowledge sharing” (Nursing Director). The ambivalence of the quality manager is evident in relation to the common standards and knowledge sharing. She commented on the negative side of standardisation in terms of the absence of standards on key areas and here she refers positively to the benefits in relation to the ability to share several documents created for accreditation (such as the Code of Ethics). In addition, she refers to the opportunities created for more widespread sharing of information between hospitals and considered the accreditation process a “constructive experience”:

As a result of the common accreditation, we offered help to a nearby hospital. We created a group of quality coordinators integrated by the hospitals of the Catalan Health Institute to share information […]. We have worked together for the past year having meetings every two months. We do not just talk about the accreditation, but also about other problems that we share. The accreditation has been a great opportunity in this sense (Quality Manager).

The benefit of a standardised system in promoting closer working relationships was also confirmed by the external auditor:

I can say that more or less they have achieved similar scores in all the hospital centres […] because they have been working very closely together. Sometimes if a hospital did not understand something, it asked others for support. This helped them with [the development of] many indicators for accreditation (Auditor).

However, there was evidence of some envy of hospitals who sometimes benefited from the experience of other hospitals. For example, the Quality Manager referred to other hospitals that had achieved the same or even higher scores than Hospenda by just making a final sprint:

There are hospitals that scored the same or even higher scores than us […] I could say that the accreditation is an administrative formality where you only need to make a final sprint during [the final] three or four months to prepare documentation, submit the supporting documentation, and indoctrinate staff for the surveyors' [auditors] visits. The actual accreditation per se is not used as an improvement tool (Quality Manager).

Overall, the use of a standardised system brought a number of contradictory and interrelated aspects of accreditation to the fore (structure vs flexibility and uniformity vs tailoring of standards), and this resulted in a mix of positive and negative views of hospital managers towards accreditation. Organising tensions relate to the tensions resulting from structure and leadership, including cooperation vs competition, empowerment and direction, and control and flexibility (Smith and Lewis, 2011). These tensions were evident in our findings in relation to structure vs agility where there was a desire for both standardised metrics that are the same for every hospital and a desire for flexibility to tailor metrics to areas central to Hospenda's mission, such as technology, teaching and research. Belonging tensions, which relate to tensions between the individual and group and between competing values, roles and memberships (Smith and Lewis, 2011), were also evident in the desire for both standardisation and tailoring. There was evidence that hospital management wanted accreditation to accommodate the needs of their individual hospital but also wanted their hospital to belong to a region of hospitals (Catalonia) that are evaluated in the same way. Further evidence of organising tensions emerged in relation to cooperation vs competition. There was a desire for learning from shared problem-solving and policy development (e.g. Code of Ethics), but this was coupled with an envy for those competitor hospitals who benefited from shared processes and documents and who accelerated through the accreditation preparation process relatively fast as a result of this shared learning.

Previous research has theorised how the design of internal MCSs can result in tensions being latent or salient for an individual (Lewis et al., 2019). Lewis et al. maintain that if no attempt is made to embed a tension (in their study, empowerment vs control) into organisational processes, then the tension will remain latent and imperceptible for organisational members. They present an example of a highly bureaucratic form of control (iron cage) and maintain that because there is no desire for this form of control to be empowering (i.e. empowerment is not a feature of the system), the tension between empowerment and control is latent. Our findings on the tension between flexibility and standardisation, however, contrast with this and suggest that the salience of the tension is not a reflection of whether the tension is embedded into the design or use of the control system. We find that the tension between flexibility (in allowing for creative responses to problems in the context of individual hospitals) and standardisation (in allowing comparability of accreditation results across the hospital sector) is a salient tension for managers in Hospenda. Yet, flexibility was not embedded into the system as a feature of accreditation (i.e. healthcare authorities did not claim to offer hospital staff flexibility in how the accreditation system was implemented). Hospital management's recognition of the contradictory and interrelated aspects of both sides of the organisational duality – flexibility and standardisation – reflects the salience of the tension and suggests that it is not necessary for a tension to be embedded into the design or use of an external MCS for it to be salient.

4.2 Comprehensiveness

Comprehensiveness was a feature of the accreditation system, and this also elicited mixed positive and negative views on managing the enormous volume of standards and the value of a holistic view of the organisation. The accreditation model is based on the European Foundation Quality Management model (EFQM) divided into different dimensions, including structures, processes and results, and includes 1,302 standards (696 essential and 606 non-essential), each of equal value. Hospenda achieved a final score of almost 95% in 2013–2014 failing only 37 of the 696 standards evaluated.

In relation to the volume of standards, the Economic Management Manager commented: “When I first saw the ‘brick’ [manual] in my hands I thought this is huge!”. Yet, this manager appreciated the comprehensive insight provided by the “brick”:

Apart from trying to understand the overall vision, it [accreditation] made me rethink that maybe now it is a good idea to arrange some of the processes that we do not have systemised […] We should spend a little more of the time on processes because very often we skip them and we go directly to results, which is the reason why sometimes we are not able to achieve the desired results (Economic Management Manager).

The benefit of the “comprehensive and integrated view” was perceived to be increased understanding of internal linkages with the work of others, and this was viewed positively:

For example, we rely on the Logistics Department to identify our planning needs and then to go ahead with the necessary contracts. In the past, we did not know how long it would take to make certain types of contracts […] but now we know more about each other's work which has helped us in our daily work (Accounting Manager).

The Nursing Director described the multifaceted and integrated approach provided by the three components of accreditation:

All three [structures, processes, results] are complementary. The structure is the basis of the accreditation and processes help to identify needs […] Results are closely related to processes. So, if you have a process rigorously controlled, then it will lead to achieving better results (Nursing Director).

The comprehensiveness of the system was viewed positively in terms of its beneficial impact on stimulating top management to analyse strategic issues from varied angles and unconventional perspectives (Plambeck and Weber, 2010).

However, despite the benefits of a comprehensive set of standards, interviewees were very conscious of the difficulties in managing the vast number of standards:

You get overwhelmed when you see the number of standards. We also have our day-to-day activities. We need more time to revise and create new protocols, etc. […] The main problem is that you have to read them all [standards] and confirm that they are still operational. If you do not already have staff doing this, it is complicated (Nursing Director).

The juxtapositioning of these two alternate poles (need for comprehensiveness and challenge of managing the volume of data) was evident in the comment from the Billing Manager:

Sometimes you only realise how good things are when you write them down. But it is a time-consuming process, it requires lots of resources and the evaluation of indicators takes so much time, while you still have to fit in your day-to-day activities (Billing Manager).

The need for the volume of standards and what were perceived as very intense accreditation visits were questioned:

My only criticism is: do we need 700 [confirmed as 696] essential standards? Perhaps not? (Deputy General Director).

I feel it is too intense, the process of accreditation itself. Those three days of the accreditation were difficult. There was lots of stress and tension. […] It was fourteen hours per day and was exhausting, particularly mentally (Nursing Director).

For some, the salience of this tension reduced following the first cycle of accreditation in 2007/2008 due to efficiencies and learning which reduced the volume of work for the second accreditation:

The first time everything was created in a hurry and we had to start from scratch. It generated lots of work because we had much less knowledge then, than we have now. […] Overall, I view it as a very positive thing, especially regarding “results”. […] In the end, when you do some analysis like you have to do for accreditation, you always see new things to be done, things that need to be improved, things that are no longer needed or perhaps things that we need to do in a different way. I appreciate it now more positively than the first time (Medical Director).

However, while the medical director was of the view that the process was more manageable the second time, there was a tension around changes between the accreditation cycles and the timelines of changes. He referred to the incorporation of 25% new essential standards (178 out of 696 essential standards) only “three or four months” before the accreditation audit, which caused significant frustration among hospital staff:

I think that the [Health] Department should have prepared these things in advance. If they planned to begin the audits in the second half of the year, everything should have been prepared by [the previous] December (Medical Director).

The Quality Manager also referred to these last-minute changes and the late delivery of the self-assessment tool, which limited the hospital's capacity to achieve better results.

To try and manage the tension between a comprehensive system and the potential for information overload, the hospital restricted access to information in accordance with work profile, needs and responsibilities of employees:

I only have access to my indicators, not to others. We have access to the ones at the Finance and Economic department, but only for individuals like me [managers] who have some kind of responsibility (Accounting Manager).

Transparency was limited to top and middle management, and there was a view that individuals at lower levels would only be aware of accreditation from their involvement in very specific activities related to it and from the announcement posted on the hospital's intranet showing the final score achieved.

If I have to explain the concept of accreditation I can say that it is a requirement from the Catalan Government, etc., etc., but I find it hard to believe that they [employees at lower levels] will understand its benefit. Sometimes it is even difficult for us to understand it, so, just imagine for someone who is not involved […] almost impossible […] Besides, even if they are interested in knowing more about it, they will find that it is very difficult to get access to information […] Maybe having access to all the information is too much, but I would certainly allow access to a larger number of things (Economic Management Manager).

Supporting this view, the Quality Manager believed that the accreditation process failed to permeate to lower organisational levels and that individuals at lower levels delivering services to patients did not understand the full accreditation process because they only interacted with accreditation requirements related to their particular area of expertise.

While the standards were numerous and were viewed by hospital staff as comprehensive, the outcome of the accreditation process reported by the accrediting body was a single score. There were views that this single score did not reflect or capture differences in levels of effort expended in implementation of the accreditation system and that the comprehensiveness of the system was lost.

Furthermore, the Quality Manager pointed out that there was no recognition for achieving better accreditation scores than the minimum required, which limited the motivational value of the system:

From time to time, it is nice when others [the Health Department] give you a pat on the back and appreciate your work […] Some kind of recognition is needed because it seems that scoring 81 or 95 is exactly the same thing (Quality Manager).

Despite a lack of recognition for achieving a high accreditation score, the quality manager appreciated the opportunity presented by standardisation and accreditation for discussing non-accreditation problems with other hospitals as noted earlier in this section.

Overall, salient tensions were revealed in the contradictions and interrelationships between the comprehensiveness of this multifaceted and integrated system, the increased volume of work resulting from it, the potential for information overload and restrictions on access to information. Furthermore, tensions were evident in the reduction of the comprehensiveness of the system to a single score. These tensions resulted in the simultaneous presence of positive and negative views on the comprehensiveness of the system as hospital staff struggled to deal with performing tensions – those that relate to the difficulty of attending to multiple and competing goals (Smith and Lewis, 2011). The intensity of the accreditation preparation process was exacerbated by outside forces – the Department of Health – delaying accreditation system design and circulation to Hospenda and stipulating additional standards in a tight time frame for hospital staff who were also responsible for operational performance and the delivery of frontline clinical services. This increased the salience of performing tensions.

Learning tensions result from efforts to adjust, renew, change and innovate and relate to building upon and destroying the past to create the future (Smith and Lewis, 2011). This tension is evident in the interlinkages between the positive and negative views on the comprehensiveness of the system. Organisational learning was enabled for those involved in accreditation in better understanding colleagues' job demands, while the comprehensive set of accreditation metrics offered opportunities for learning and reflective space. However, negative views stemmed from the lack of time to engage with the extensive volume of metrics, a failure to benefit from additional staff learning, restricted access to information resulting in lost opportunities for greater staff engagement at lower levels and a concern that the accreditation process was simply “an administrative formality”, detached from substantive organisational performance.

4.3 Multiple objectives

While healthcare accreditation is a system focused on continuous quality improvement (Touati and Pomey, 2009), it was also used by the government as an obligatory mechanism to manage economic and cost objectives. Interviewees expressed positive views on the ability of accreditation to combine cost and quality objectives. A close examination of the accreditation system revealed a number of indicators combining cost and quality objectives classified under broad categories dealing with issues related to efficiency, effectiveness and quality. These included readmissions, re-interventions, complications, avoidable or preventable hospitalisations and average length of stay. Improvements (decreases) in these metrics could address both cost and quality objectives. For example, striving to reduce the readmission rate is closely connected with delivering improved quality of care, which if successful may result in overall lower costs. These metrics addressed the contradictory relationships that can occur between cost and quality. For example, the readmission rate could potentially be reduced by delaying discharge, but this would increase the average-length-of-stay and cost metrics. Hence, an integrated approach was needed for the management of the various metrics and accreditation was seen to offer this. While in the short term, the goals of cost and quality could clash, many standards drove the hospital to take a longer-term outlook on cost. The Medical Director provided an example of this in relation to replacement of catheters every 72 h:

In order to reduce catheter-related infections, it is most likely that certain financial investments comprising antiseptics, dressings and other items will have to be made to guarantee that the catheter is replaced every 72 h. This obviously means an increase in costs. However, fewer catheter infections represent improved outcomes on hospitalisations – earlier discharge, decreased use of antibiotics and [an improvement in the] mobility of patients (Medical Director).

While many of the indicators seemed to focus on efficiency and processes rather than outcomes, interviewees were positive in terms of their ability to ultimately lead to improved quality of service. However, they expressed concerns around the ultimate value of accreditation as the volume of patients treated was not captured by the accreditation system. While accepting the positive features of the accreditation system, staff questioned why hospitals should strive to improve processes. Hospitals that were more efficient were not rewarded in any way and were not able to increase the number of patients treated (i.e. provide a higher number of clinical and medical procedures), which was seen by staff as the ultimate goal of hospitals. In fact, due to the financial cutbacks between 2009 and 2013, Hospenda had lower levels of clinical activities than heretofore, despite achieving a better score in the 2013/2014 accreditation compared to 2007/2008:

The problem with the cutbacks is “what” [the surgeries/operations] you do not do anymore, not what you actually do. […] The real problem is all we cannot do anymore because we have less budget […] What really worries me is all the patients we cannot treat, all the hip replacements that we cannot perform, all the spinal surgeries we performed in the past and we are now not doing and so on (Medical Director).

The hospital performed a similar number of procedures but with a lower budget, thus achieving an increase in efficiency based on a reduced bed capacity. This reduction in capacity, due to a reduction in funding, was not within the control of the hospital and was outside the scope of the accreditation system. It served to limit the perceived benefits of greater efficiency as an outcome of accreditation as additional throughput of patients was not permitted within the terms of the agreed service contract with the regional health authority. Between 2009 and 2013, Hospenda suffered a significant decline of approximately 25% in bed capacity. The Human Resources Assistant Director recognised that some of the main priorities in the hospital were directed at reducing the patient length of stay (measured by the accreditation process) with a reduced number of beds (capacity) so as to maintain the same level of activity. However, he affirmed that people normally perceived that “a decrease in the number of beds meant a reduction in our capacity to perform”. This was demotivating for staff and led to underlying concerns about the ultimate value of accreditation in this context of reduced resources.

The complex interrelationship between levels of activity, revenue and costs was recognised by the Economic Management Manager who pointed out that “sometimes working more efficiently can be counterproductive” for Hospenda. He explained how concurrent efficiency gains, such as reduced average hospital length of stay for the patient (a key focus of quality improvements), free up bed capacity to take additional patients. However, the annual service contract price agreed with the regional health authority is for a maximum number of procedures/treatments, and treating additional patients above this number does not result in additional revenue. However, it will result in additional costs, e.g. drug therapies for medical patients or implants for surgical patients, which vary directly with patient intake and treatment and lead to budget overruns. He further explained that bed reductions resulted in negative perceptions externally. Hospenda's inability to treat additional patients as a consequence of reduced budget, notwithstanding the increased efficiency resulting from the accreditation system, was at the root of negative views for many interviewees.

As Smith and Lewis (2011) point out, different stakeholders have different views of success, and this performing tension is evident in Hospenda through the multiple and competing goals (e.g. improved quality of care, reduced cost and increased volume of patients treated). Interviewees recognised the interrelated and contradictory aspects of the different objectives. Performing tensions emerge when identification and goals clash (Smith and Lewis, 2011), such as when a short-term view is challenged and a longer-term view is juxtaposed on clinical treatment protocols, such as Hospenda's revised catheter management clinical protocols. Interviewees generally perceived that the accreditation system achieved a balance between quality of care and longer-term efficiency and led to a reconciliation of these objectives. However, the absence of any accreditation focus on the volume of patients treated and the inability of the hospital to turn efficiency gains into increased patient throughput exacerbated the contradictory nature of the multiple objectives. The inability of the hospital to treat more patients was seen as particularly regressive in light of the high accreditation score achieved.

4.4 Mandatory external form

The accreditation system is a mandatory external control system, and interviewees accepted this and the resulting low level of discretion they had over how they implemented the system. The Assistant Director of Human Resources viewed accreditation as an essential process:

Next week we have the accreditation […]. There is a protocol for everything […] healthcare and non-healthcare activities. Here, at the Human Resources department, we have many protocols. Why? Well, because it is essential […] it is clear that it is the only way to be organised and to do things properly (Assistant Director of Human Resources).

The low discretion was regarded as inevitable in the context of a mandatory externally imposed system. While there was some discretion in relation to the adoption of non-essential standards, this was viewed cynically as “non-essential standards in one period tended to become essential standards in the next and could not be ignored” (Quality Manager).

There was also evidence of perceived resistance towards change that resulted from externally mandated systems at lower organisational levels due to the prior experience of staff in relation to changes in employment legislation. The Economic Management Manager spoke about the resistance of nurses and a perceived association between flexibility and worsening employment conditions. Similarly, the Nursing Director, while expressing positive views on how the accreditation system helped “analyse our limitations, our errors, our needs and also our advantages”, spoke about the difficulty of motivating nurses as she felt a high score on accreditation did not compensate for the additional constant pressures experienced by nurses. She was of the view that in the current climate, employees were not very open to managerial ideas and proposals because of the difficult economic situation experienced by hospitals in the region. Hospitals had to re-negotiate a new contract with the main purchaser of public healthcare services, which resulted in a considerable reduction in budget that negatively affected staff salaries and their working conditions. The increased volume of work resulting from accreditation was difficult to accept in this context:

After the accreditation process there is a process of relaxation, and we are now in a phase where we are trying to maintain everything we have achieved. […] we have a real worrying factor which is the demotivation of our [nursing] teams due to external economic problems. The past four years have been really difficult. We put too much pressure on them to achieve good results and we have achieved almost 95% which it is very close to the excellence rate. We are now trying to motivate them. Let's see […] Perhaps in the past you could motivate staff by offering special training, involvement in large projects, etc. […] Nowadays, these things are not motivating factors (Nursing Director).

Staff at lower levels were perceived to have made a linkage between hospital management requests for flexibility, increased work and worsening employment conditions. While the low level of discretion associated with a mandatory external system was seen as inevitable and accepted by hospital management and while they recognised the improvements that had resulted from the accreditation system, they were concerned about the contradictions of increased work, requests for flexibility and deteriorating employment conditions at lower levels. Recognising these interrelated and contradictory aspects of a mandatory external system made this a salient tension. Belonging tensions (Smith and Lewis, 2011) were evident in the perceived demotivation of nursing staff whose reach permeates throughout the hospital and who find their role at odds with accreditation intentions in the context of constrained budgets and reduced bed capacity. We also see evidence of performing tensions (Smith and Lewis, 2011), and the need to manage additional objectives relating to enhanced working conditions and ensuring staff motivation by facilitating vocational goal achievement. The nursing group were disadvantaged (as were other groups), by the reduction in pay and deterioration in working conditions (nurses to bed ratio) that occurred in the three-year period leading up to accreditation. As the dominant (in number) staff group in this and all hospitals, their unique vocational identity was challenged by the overwhelming demands on their ability to do their day-to-day job while concurrently engaging with examination and review of the structures, processes and results that underpinned the majority of the 1,302 accreditation standards.

As organisations comply with requirements to implement external control systems, internal organisational processes become more complex and the salience of contradictory tensions increases (Lewis, 2000). Accreditation implementation involves complexity (Brubakk et al., 2015). This complexity was evidenced in the attitudes of hospital staff who evaluated 1,302 standards relating to hospital structures, processes and results. When individuals evaluate more complex systems that contain a large number of aspects or features, they are less likely to perceive the system as simply good or bad (Meyerson and Scully, 1995; Plambeck and Weber, 2010). Ambivalence relates to the expression of the two sides of a dualism (Meyerson and Scully, 1995) and the increased salience of organising, performing, learning and belonging tensions led to the generation of ambivalent views among Hospenda management staff responsible for its implementation. Hospital management involved in a mandatory review of structures, processes and results during accreditation developed a “holistic” and comprehensive view, recognising the benefits of accreditation while acknowledging both the costs in terms of managerial time and the deficiencies of both the process and the results of the process. In this era of ambivalence (Weigert and Franks, 1989), simple depictions of positive and negative views (Pratt and Doucet, 2000) fail to capture the complexity of attitudes of management towards hospital accreditation. Ambivalence can create dialogue, reflection and un-learning that may stimulate change (Eisenhardt, 2000; Piderit, 2000). While responses to ambivalence were not the focus of the study, there is clear evidence of a questioning of practices and processes as hospital management recognised the contradictory and interrelated aspects of different tensions. The next section sets out the contributions of our study, implications for practice and areas for future research.

5. Conclusion

This study examined how an external mandatory hospital accreditation system influences the salience of organisational tensions and attitudes of hospital management. It focused on capturing the attitudes of senior and middle management towards the features of the system. By requiring managers to reflect on a wide variety of issues related to structures, systems and processes of the accreditation system, we show how its implementation provokes tensions. Interpretive contexts can reveal the contradiction and interdependence between opposing poles (Knight and Paroutis, 2017; Weber and Glynn, 2006). Accreditation implementation is a temporal interpretive context occurring in this hospital every six years. The second accreditation implementation process began with the creation of an “auditable front” (Power, 1996) in the form of a self-assessment report, which required creation or review of standards covering hospital structures, systems and processes. This report together with a three-day on-site visit represented the domain of facts (Power, 1996) presented for external certification. The previous section presented evidence on how the organisational dualities of learning, performing, organising and belonging (Smith and Lewis, 2011) are made salient through the accreditation system. Furthermore, it presented evidence of the coexistence of favourable and unfavourable attitudes – an “ambivalent” orientation (Merton, 1976) to the imposition of this external mandatory control system. The contributions of our study are four-fold and are set out below.

First, we respond to calls to examine how organisational tensions become salient in organisations (Knight and Paroutis, 2017). We identify an active role for an external MCS (accreditation) in increasing the salience of organisational tensions. We find that an external MCS increases the salience of tensions by bringing the organisational dualities of learning, performing, organising and belonging (Smith and Lewis, 2011) into juxtaposition. Increased tension salience revealed by exposition of organisational dualities during accreditation implementation is of benefit given that a duality-sensitive mind-set recognises the merits of both sides of the duality continuum, as opposed to choosing one extreme over the other (Graetz and Smith, 2008).

Second, we extend the existing MC literature which has focused on creation and balancing of tensions through internal MCSs (Curtis and Sweeney, 2017; Mundy, 2010; Taylor et al., 2019; Van der Kolk et al., 2020). In particular, we find that contrary to the prior literature (Lewis et al., 2019), the embedding of both poles of a duality into a MCS is not a necessary precondition for increased tension salience. While standardisation was embedded in the external control system in this study, flexibility was not. Yet, we find a salient tension between flexibility and standardisation arising from the accreditation system. Lewis et al. (2019) maintain that internal MCSs can increase the salience of a tension by embedding both poles of the duality into the design and use of the control system. They also suggest that tensions arising from one MCS can be disguised through the operation of other MCSs, such as socio-ideological controls. An explanation for our finding on the salient tension between flexibility and standardisation may relate to fewer opportunities to disguise the tension in the context of a mandatory external MCS compared to an internal MCS.

Third, our study adds to the small body of knowledge on MCSs and ambivalence by demonstrating a role for MCSs in triggering ambivalent attitudes through the increased salience of organisational tensions. It thereby addresses the need to better understand the “contextual” sources and triggers of ambivalence (Ashforth et al., 2014; Plambeck and Weber, 2010). Our study extends the range of outcomes of MCSs examined in the MCS literature beyond positive, negative and neutral attitudes to control systems (Tessier and Otley, 2012) to include ambivalent attitudes. While previous research points to the need for MCSs to “allow” for ambivalence (Van der Kolk et al., 2015), we find evidence of a more active role for MCSs in increasing, revealing and exposing the salience of tensions and in doing so generating ambivalence. The complexity of attitudes towards a mandatory externally imposed system is revealed where singular attitudes of positive and negative are insufficient to describe how management viewed the system (Pratt and Doucet, 2000). Given the range of positive and negative outcomes (outcomes are not the focus of this paper) associated with ambivalence (Conner et al., 2002; Guarana and Hernandez, 2015), the existence of ambivalent attitudes is likely to be important in explaining the consequences of different MCSs. Previous research points to the need to consider cognitive processes in MCS studies (Hall, 2016), and our study demonstrates the potential for research on ambivalent attitudes as a response to MCSs to provide insight into cognitive processes.

Finally, we respond to calls to further examine implementation of health service accreditation (Greenfield and Braithwaite, 2008; Wilson, 2018). Our findings highlight the complexity of accreditation implementation (Brubakk et al., 2015) where multiple organisational dualities are brought into juxtaposition through the various features of an external control system. A key concern with quality control systems is that there may sometimes be a decoupling of organisational performance from that represented by the audited system (Power, 1996). We find evidence of reflection on best practice in reviews of hospital structures, systems and processes undertaken as part of the audit of current practice in preparing both the self-assessment report and for the on-site accreditation visit. We provide further evidence that the lexicon of audit operates as a form of power through socialisation processes (MacLullich, 2003) in the self-reflections and discussions triggered by the imposition of an external control system and articulated by hospital staff relating to learning, performing, organising and belonging organisational dualities.

The widespread ambivalent attitude of hospital management found in this study has implications for healthcare management and healthcare authorities as it provides insights into the sources and triggers of ambivalence resulting from the plurality of learning, performing, organising and belonging organisational dualities. Accreditation is a low discretion context where there is a significant staff investment in seeking accreditation but few opportunities for feedback from those involved in the accreditation process. To gain a more realistic understanding of responses to organisational change, the triggers and existence of ambivalence must be understood (Oreg and Sverdlik, 2011). Understanding hospital management's attitude towards accreditation and the underlying organisational tensions that become salient can help healthcare authorities in managing the implementation process for accreditation systems and help hospital management understand how to realise benefits from the accreditation process.

The study points to a number of areas for future research, including an examination of how the features of accreditation impact on attitudes of those without management responsibilities and at lower organisational levels. While the accreditation system is designed as a management tool to improve structures, processes and outcomes, our study is limited by its sole focus on perceptions of management albeit with medical and non-medical backgrounds. In addition, future research could examine responses of management to ambivalence and implications for organisational outcomes. This would be particularly useful for contributing additional insights of practical relevance for hospital management and healthcare authorities. A comparison of high and low discretion contexts would be useful here. Sverdlik and Oreg (2009) maintain that there is likely to be greater ambivalence associated with imposed as opposed to voluntary change, providing support for our findings on the widespread existence of ambivalent attitudes towards an imposed control system. The mandated external form resulted in low discretion for hospital management in how the accreditation system was implemented, and there was evidence of an attitude of reluctant acceptance of the features of the system. However, contradicting this, Plambeck and Weber (2010) maintain that complex, as opposed to prepacked, issues are more likely to elicit ambivalent responses, and accreditation could be considered a prepacked issue. Lastly, future research could usefully examine how packages of control (Malmi and Brown, 2008) increase the salience of tensions. The various MCSs in use in an organisation interact and may affect each other's outcomes (Van der Kolk et al., 2020). Further research is needed on the interplay between external and internal control systems in increasing the salience of tensions. Interactions between different MCSs may explain the presence of a salient tension between standardisation and flexibility even though flexibility was not embedded in the design or use of the accreditation system as Lewis et al. (2019) suggested was necessary.

Table A1

Details of interviewees

1st phase of data collection
Hospenda staff
Job description and hospital department	Recorded
Assistant Director Human Resources/Human Resources	Yes
Maintenance Manager/General Services and Projects	Yes
Information Systems Director/Organisation Systems	Yes
Nursing Head of Quality and Safety + 2 nurses/Nursing	Yes
Technical Office Manager + Technical Office Assistant/Organisation Systems	Yes
Medical Director/Medical	Yes
Assistant Director IT Systems/Organisation Systems	Yes
Economic and Finance Director/Economic and Finance	No
2nd phase of data collection
Hospenda staff
Billing Manager/Economic and Finance	Yes
Accounting Manager/Economic and Finance	Yes
Economic Management Manager/Economic and Finance	Yes
Logistics and Purchasing Manager/Economic and Finance	Yes
Administrative Contracts Manager/Economic and Finance	Yes
Quality Manager/Medical	Yes
Hospital Deputy Director/Hospital Deputy Director	Yes
Medical Director/Medical	Yes
Nursing Director/Nursing	Yes
Economic and Finance Director/Economic and Finance	No
Health department and independent assessors
Quality and Accreditation Director	No
Institute Director (“Health” Institute)	No
Quality and Accreditation Director + 2 committee members	No
Foundation Director (“Health” Foundation)	No
Auditing firm surveyor	Yes
Quality and Accreditation Director + 2 committee members	No
Quality and Accreditation Director	No
Quality and Accreditation Director + 1 committee member	No
Other
Applications Systems Manager (different hospital)	No

Appendix 1

Appendix 2 Organisational chart Hospenda

Departments and sub-departments:

1. Nursing → Two sub-departments: Outpatients and Support (five services) and Medical and Surgery (ten services)

2. Medical → Four sub-departments: Outpatients (12 services, e.g. pharmacy, emergencies, psychiatry), Medical (seven services, e.g. neurology, infectious diseases), Surgery (ten services, e.g. plastic surgery, urology) and Medical Coordination (six services, e.g. research, teaching, quality)

3. Economic and Finance (known as Accounting and Finance in other countries) → Five sub-departments: Billing, Economic Management (Cost Management), Accounting, Administrative Contracts, and Logistics and Purchases

4. Human Resources → Five sub-departments: Professional Development, Payroll and Social Security, Personnel Administration, Staff Welfare and Ongoing Training

5. General Services and Projects → Five sub-departments: Development, Maintenance, Hospitality, Security and Clinical Engineering

6. Organisation Systems → Three sub-departments: Technical Services, Information Technology and Ancillary and Facilities Management