Emerging and temporary connections in Quality work: An Ethnographic Study of Quality Coordinator Work in two Danish Hospital Departments

Madsen, Marie Henriette Doctoral Thesis Emerging and temporary connections in Quality work: An Ethnographic Study of Quality Coordinator Work in two Danish Hospital Departments PhD Series, No. 39.2015 Provided in Cooperation with: Copenhagen Business School (CBS) Suggested Citation: Madsen, Marie Henriette (2015) : Emerging and temporary connections in Quality work: An Ethnographic Study of Quality Coordinator Work in two Danish Hospital Departments, PhD Series, No. 39.2015, ISBN 9788793339613, Copenhagen Business School (CBS), Frederiksberg, https://hdl.handle.net/10398/9227 This Version is available at: https://hdl.handle.net/10419/208955 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc-nd/3.0/ PhD School in Organisation and Management Studies PhD Series 39.2015 PhD Series 39-2015 EMERGING AND TEMPORARY CONNECTIONS IN QUALITY WORK COPENHAGEN BUSINESS SCHOOL SOLBJERG PLADS 3 DK-2000 FREDERIKSBERG DANMARK WWW.CBS.DK ISSN 0906-6934 Print ISBN: 978-87-93339-60-6 Online ISBN: 978-87-93339-61-3 EMERGING AND TEMPORARY CONNECTIONS IN QUALITY WORK Marie Henriette Madsen EMERGING AND TEMPORARY CONNECTIONS IN QUALITY WORK AN ETHNOGRAPHIC STUDY OF QUALITY COORDINATOR WORK IN TWO DANISH HOSPITAL DEPARTMENTS Marie Henriette Madsen Supervisors: Morten Knudsen (Copenhagen Business School, Department of Organization) Lise Justesen (Copenhagen Business School, Department of Organization) Vibeke Normann Andersen (KORA) Doctoral School of Organisation and Management Studies Copenhagen Business Schoo Marie Henriette Madsen Emerging and temporary connections in Quality work 1st edition 2015 PhD Series 39-2015 © Marie Henriette Madsen ISSN 0906-6934 Print ISBN: 978-87-93339-60-6 Online ISBN: 978-87-93339-61-3 The Doctoral School of Organisation and Management Studies (OMS) is an interdisciplinary research environment at Copenhagen Business School for PhD students working on theoretical and empirical themes related to the organisation and management of private, public and voluntary organizations. All rights reserved. No parts of this book may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage or retrieval system, without permission in writing from the publisher. - 3 - ACKNOWLEDGEMENTS Thank you! First and foremost, I would like to thank managers and staff at ‘the hospital’. Without you, there would be no thesis at all. You welcomed me with open arms and allowed me to participate in and observe whatever I asked for. I hope this thesis does you justice, but also that you might be inspired to think differently about quality development in health care, what it takes and what it causes. Thanks to my supervisors Morten Knudsen and Lise Justesen for support during the study and for gentle supervision and guidance. I have not always embraced your comments with enthusiasm, but I have always stored them in my mind and most of them have found their way to this thesis. You were usually right! I also owe sincere thanks to my third supervisor Vibeke Normann-Andersen. You entered the project at a late stage, but you engaged with it on its own premises and provided me with new inspirations without imposing them on me. Thanks to Kristian Kreiner and Teun Zuiderent-Jerak who commented on my second WIP paper and encouraged me to push my analysis even further. Also thank you to Tine Murphy who infused me with new energy when I needed it the most, mostly by listening and letting me clarify my own thoughts, but also by convincing me of the relevance of my work. Thank you to Kirsten Rasmussen and Niels Coley for proofreading the thesis. To KORA (and the former DSI): thank you for giving me the opportunity to write this PhD thesis. Especially, I owe thanks to Jes Søgaard for persistent encouragement and help to pave the way to making this PhD possible. Thanks to Copenhagen Business School, Department for Organization (IOA), for providing inspiring work conditions, both socially and professionally, and especially to Center for Health Management for - 4 - excellent input and discussions of my own work and of health care in general. I look forward to return to new and old colleagues at KORA, but I will miss my IOA colleagues. I hope we will work together again at some point in the future. My daughter, Agnes, asked me one morning, “do you have a best friend at work?”’ I could truthfully say that I have more than one. Didde Maria Humle, Mette Brehm Johansen, Jane Bjørn Vedel and Amalie Martinus Hauge: you are the best office companions anyone could ask for. Our office has been my zone of comfort for more than three years, and the cheers, laughs, chats and loads of coffee (I might even owe thanks to Joe) have been significant and much appreciated. Maya Christiane Flensborg Jensen and Mie Plotnikopf: we definitely teamed up too late, but nevertheless it has been invaluable to share bits and pieces of this thesis with you, to receive your feedback on more or less unfinished work and to share the frustrations and insecurities that apparently come with this job. Thanks to you and to all the other IOA PhD fellows for being a fun, caring, clever, and inspiring network. I also owe thanks to family and friends for rock-solid support, but most of all for taking my mind off my work. Especially I owe thanks to my mom, dad and Elsa for helping out here and there, and for baby-sitting again and again, and to Stine who besides being the best of friends also proofread an early version of my paper. Last but not least: Thank you, Lars, Laurits and Agnes: for your love, patience and support, especially during the last months of this project. The summer holiday is just around the corner and I look forward to spending much more time with you again. This is by far the best thing about finishing this thesis! Marie Henriette Madsen Copenhagen, July 2015 - 5 - LIST OF CONTENTS PART I .............................................................................................................................. - 9 - Introduction .................................................................................................................... - 10 - Quality development in Danish health care ............................................................... - 13 - Research question ....................................................................................................... - 14 - The organizational set-up of the study ........................................................................ - 17 - The structure of the thesis .......................................................................................... - 18 - Quality development in a Danish context ..................................................................... - 21 - The many agendas of quality development ................................................................ - 21 - A Danish framework of quality development in health care ..................................... - 26 - The new institutions of quality development ............................................................. - 31 - The critique of the current organisation of quality development ............................. - 31 - Quality development beyond or after DDKM ........................................................... - 35 - Concluding remarks .................................................................................................... - 36 - Introducing the empirical case ....................................................................................... - 38 - The hospital ................................................................................................................. - 38 - The departments ......................................................................................................... - 41 - Concluding remarks .................................................................................................... - 46 - PART II ........................................................................................................................... - 49 - Framing the study and generating questions ................................................................. - 50 - Towards an ethnographic strategy .............................................................................. - 50 - Asking (new) questions: holding on to the empirical tensions .................................. - 52 - - 6 - Concluding remarks ................................................................................................... - 64 - Literature review ............................................................................................................. - 65 - Quality development – a study of medical practices .................................................. - 65 - From changed to emerging connections .................................................................... - 73 - Concluding remarks ................................................................................................... - 75 - Analytical framework ...................................................................................................... - 77 - Arcs of work and types of work .................................................................................. - 77 - A relational understanding of actors and orders ....................................................... - 80 - Articulation work ........................................................................................................ - 85 - Concluding remarks ................................................................................................... - 87 - 7. Methodology ............................................................................................................... - 88 - A study of quality coordinators at work ..................................................................... - 88 - Additional sites of observation ................................................................................... - 95 - Field notes ................................................................................................................... - 97 - Interviews .................................................................................................................... - 98 - From fragmented field notes to coherent analytical stories .................................... - 100 - Concluding remarks ................................................................................................. - 107 - PART III ....................................................................................................................... - 109 - ‘Operation joint future’ ................................................................................................ - 110 - ‘Operation joint future’ ............................................................................................ - 110 - Connecting doctors to the project: From rejection to critical participation .......... - 114 - Connections and de-connections: from unconditional to sceptical participation . - 121 - - 7 - Concluding remarks .................................................................................................. - 127 - Quality data: Constructions and their use ................................................................... - 130 - Data as achievements ................................................................................................ - 130 - Transforming clinical work into quality data ........................................................... - 133 - “Should we do something”: Moving quality data to arenas of reflection ................ - 145 - A repository of quality data ....................................................................................... - 153 - Concluding remarks .................................................................................................. - 159 - ‘A care pathway for medical infections’ ....................................................................... - 161 - Care pathway descriptions – a tool in quality development .................................... - 161 - Negotiating the “real” problem(s) ............................................................................. - 165 - Between order and disorder ...................................................................................... - 172 - Concluding remarks .................................................................................................. - 178 - PART IV ........................................................................................................................ - 181 - Discussion ..................................................................................................................... - 182 - The emergence of quality work ................................................................................. - 182 - Quality coordinator work .......................................................................................... - 186 - Reflections on the empirical and analytical strategy ................................................ - 194 - Conclusion .................................................................................................................... - 196 - Contribution ............................................................................................................. - 198 - References ..................................................................................................................... - 201 - Summary ....................................................................................................................... - 213 - Dansk resumé ................................................................................................................ - 217 - - 14 - hospitals and hospital departments (Knudsen, Fuglholm & Kjærgaard 2004). In this way, quality development has become a mandatory part of health care management and is consolidated as a distinct part of hospital life with its own agendas, assigned actors and tasks. As part of this development, the position of quality coordinators has emerged in some Danish hospital departments. The quality coordinators in these departments are given the responsibility for the implementation of the mandatory components of quality development, defined, for instance, by DDKM. Additionally, they are given the responsibility for the surveillance of the hospital departments’ quality level and for initiating processes of improvement when needed. Thus, quality coordinators are given a prominent role in this thesis as key actors in the organisation of quality development in the hospital departments, which constitutes the empirical case of this study. RESEARCH QUESTION In this thesis, I argue that we cannot fully understand how quality development affects the health care sector, if we are only investigating it as encounters between the various methodologies of quality development and the ordering of everyday clinical practices. Instead, I suggest, we have to investigate how quality development develops through practices where actors from the local health care settings are continuously connected and re-connected around different quality development initiatives. Hence, we have to study the practices related specifically to the construction of these connections. Accordingly, this thesis is empirically founded in a study of quality coordinators’ work in two Danish hospital departments. Hereby, I explore how particular purposes and tasks of quality development are constructed in the intricate relationship between local health care organisations and international and national ideas of quality development. In the analytical framing of the study, I argue that it can be fruitful to study quality work as sets of distinct but continuously emerging networks, and networks that need to coordinate - 15 - and cohere with the clinical context that this quality work is predestined to assess and improve. Thus, I consider quality work as a distinct arc of work in the hospitals, encompassing distinct actors and tasks {{97 Strauss,A. 1997; 88 Strauss,A. 1985} that develop through processes of translation (Callon 1986, Latour 1999b, Latour 2005). Hence, in this thesis, I study a particular kind of organizational work related to quality development and to the creation of the necessary intersections and coherence between quality development and clinical work. Accordingly I ask: How does quality work emerge in the hospital departments as local and specific processes in the intersection with standardised methods and requirements of quality development? What are the implications for the way in which quality work is organised and managed? These questions are rather broadly formulated and require some specification. First of all, the notion of work draws on the conceptualisations of Strauss and colleagues (Strauss 1985, Strauss et al. 1997) and resembles the tasks that needs to be done in order to carry out what they refer to as an arc of work. An arc of work relates to a particular object of work (for instance a patient), an end product (recovery or alleviation) and consists of tasks or groups of tasks carried out sequentially or simultaneously. Additionally, an arc of work entails the presence different actors or groups of actors to carry it out. Following this conceptualisation, I consider quality work as a particular set of tasks in the hospital departments related to various goals and requirements of quality development. I consider quality development as arcs of work that are different from, though deeply entangled with, clinical work, and hence work that can be studied in its own terms. It would be reasonable to argue that the quality of clinical work depends on much more than the initiatives specifically referred to as ‘quality development’. Resources in terms - 16 - of time and staff, budgetary constraints, the balance between clinical and administrative tasks, preand postgraduate education and training, and the health professionals’ experience and ability to react to unforeseen events related to the individual trajectories of treatment could all be expected to affect the quality of health care. Accordingly, quality work could refer to the work of health care staff when taking care of patients, to teaching and supervision etc. However, this thesis is primarily concerned with the work related to monitoring and assessing quality, and to the definition of quality problems and solutions, and hence a delimited part of the work in hospitals that contribute to quality. Evidently, it is still possible to include a vast array of practices in this definition of quality work, as the above could be the jobs of managers on different levels in the hospital or even the professional societies that develop clinical guidelines or standard operational procedures. However, the empirical scope of this thesis is further delineated by the focus on the tasks and responsibilities of the quality coordinators in the hospital departments. From this it follows that this thesis focuses primarily on a highly specific part of quality work related to the officially formulated requirements of quality development, such as monitoring and assessment of quality development, re-formulation of local best practices and so on. Until now, I have described quality work as a distinct type of work in hospitals. However, it is an important to note that quality work is not approached as a static entity, pre-defined through the many standardised methods of quality development. Inspired by the notion of translation (Callon 1986, Latour 2005, Latour 1986), I take an interest in how actors become connected and how particular definitions of quality problems and their solutions are made possible through these connections. Hence, quality work is not considered as a given, but as an achievement constructed through connections and reconnections of actors. - 17 - THE ORGANIZATIONAL SET-UP OF THE STUDY This study is funded and organised as an industrial PhD. The Danish industrial PhD setup is a scheme under the Danish Ministry of Science, Innovation and Higher Education involving collaboration between a company, a university and a PhD student. An industrial PhD is co-funded by the company and the Danish Ministry of Science, Innovation and Higher Education, and the university provides the PhD student with an academic environment and support in terms of academic supervisors, a PhD school and working space. The PhD student is employed by the company as well as being affiliated with a public research institution. In this case, I was employed by Danish Institute for Health Services Research (DSI) (now KORA, Danish Institute for Local and Regional Government Research 3) and affiliated with Department of Organisation (IOA), Copenhagen Business School. Typically, a PhD student in an industrial PhD setup provides research within the company, but in this case the research has been conducted in a Danish hospital, though within a subject of research of high relevance for KORA, whose mission statement is to contribute knowledge that can promote quality improvement, better use of resources and better management in the Danish public sector. The project proposal was initiated with support from DSI, and the initial research questions concerned the many new actors and institutions related to the organisation of quality work in the health care sector. This proposal was presented and discussed with 3 KORA was established on July 1st 2012 as the result of a merger of the three former Danish public research institutions DSI, AKF and KREVI. This project was initiated before the merger through my employment in DSI (Danish Institute of Health Services Research), and after the merger the project was continued through my employment in KORA. KORA is an independent institute under the Danish Ministry for Economic Affairs and the Interior. KORA carries out independent analysis and research for both public and private organisations. The institute advises public authorities and disseminates the results of its work to the relevant public and private stakeholders, and the public in general. - 18 - people from IOA, and this resulted in an application to the Danish Ministry of Science, Innovation and Higher Education in the summer of 2010. The project became accepted later the same year. However, the project was not initiated till 20124. Throughout the course of this PhD study, I spent about half of my time in IOA and participated as an integrated member in both academic and social events. Additionally, I have carried out work in KORA on projects related to the subject of the PhD study, spent time there while writing the thesis and participated in organizational events and meetings. THE STRUCTURE OF THE THESIS This thesis is structured into three parts. In the remainder of this the first section, I will introduce the empirical field and the broader context for quality development in Danish health care. Here, I introduce the current landscape of quality development and provide an account of the historical roots of this landscape, as well as the critique of the current national frameworks from stakeholders in the Danish health care sector. Then, I provide an introduction to the specific empirical case and describe the general organizational structure and how quality development is organised and embedded in both the hospital and the two departments. In this chapter, I also introduce the quality coordinator as an organizational figure that developed as a reaction to the many requirements related to quality development and changed organizational conditions within the departments. In part two, I turn to the methodological and theoretical framework of the thesis. I begin in Chapter 4 with a description of the original ethnographic strategy, and reasoning behind it, and of how I began formulating questions to the material that I constructed through the fieldwork, which consisted mainly of participant observation. Together with the literature review (Chapter 5), these questions became pivotal to the way I constructed the analytical framework, emphasising quality development as sets of emerging networks 4 Due to maternity leave in most of 2011. - 19 - of actors connected through processes of translation. Accordingly, in Chapter 6 I explicate my analytical framework inspired by the conceptualisations of work (Strauss 1985, Strauss et al. 1997) and Actor Network-Theory. In Chapter 7, I outline the methods used in this study. In the third part of this thesis, I offer the empirical explorations of the continuously emerging and temporal connections of quality work in relation to three empirical cases. In the first analysis (Chapter 8), I explore the mutual process of formulating (and reformulating) the purpose of a quality development project and engaging significant actors in this project. The purpose defines what should be improved, but implicitly also who becomes significant participants and accordingly obligatory and/or unavoidable to engage in the process. Towards the end of this analysis, I discuss the quality coordinators’ attempts to foresee what motivates the different groups of staff to be engaged in the project, and how this is challenged by unforeseen shifts in motivation. The next analysis (Chapter 9) explores the efforts of constructing, storing and utilising quality data. This analysis shows how quality data is a result of a process of translation that relies on specific criteria for the judgment of the departments’ quality levels. Thus, the quality data enables the quality coordinators to prompt reflection among staff and managers upon specific parts of the clinical work. Still, the utility of these data – as a point of connection in a further process where distinct quality problems are defined – relies on a greater effort than the generation of quality data. Quality data are simplified representations of a far more complex network of actors and tasks, and accordingly of possible reasons for the indicated (low) level of quality. Thus, in order to delineate a distinct focus of attention in the further process of quality improvement, the reintroduction of the details of the clinical work that quality data represents is needed. The third and final analysis (Chapter 10) investigates how the quality coordinators contribute to the processes of quality development from an organizational position - 20 - outside both management and clinical work. A characteristic of the process outlined in this chapter, as well as the other analyses, that it is a short-term project aimed at changing a delimited section of work. Each new quality development project is a new situation in which purposes and problems need to be defined and negotiated anew, and where the involved actors are not aligned beforehand. Hence, an important task for the quality coordinators is to frame these alignments in a way that is both strategic and adaptive to the specific, local and temporal contingencies. In the final part, I present the main findings of the thesis and discuss the implications of these for practical conduct of quality development and for future research. - 21 - QUALITY DEVELOPMENT IN A DANISH CONTEXT In this chapter, I describe quality development in a Danish health care context with the aim of providing the reader with an understanding of the historical, technological/methodological and organizational context of the analyses in this thesis. First, I will describe quality development as being of a fluid and changeable nature, after which I will counter this by describing it as consisting of firm rules of method, dedicated technologies and organizational structures in Danish health care. Finally, I will turn to the prevailing critique of the current national framework for quality development and position this study in the midst of a field of multiple agendas, concerns and opinions. THE MANY AGENDAS OF QUALITY DEVELOPMENT According to WHO’s definition (World Health Organization 2006), quality in health care is related to effective and evidence-based care, efficiency, patient preferences, accessibility and equitability (see Figure 1). - 22 - Figure 1: WHO's definition of health care quality (World Health Organization 2006) The last two bullets in this definition call for more structural and political actions, whereas the first three are closer to the aspect of quality development studied in this thesis. This definition was referred to during the fieldwork of this study and is also explicitly mentioned in the majority of Danish textbooks on quality development, as well as in many policy documents (Mainz et al. 2011, Kjærgaard et al. 2001, Kjær et al. 2004, Det nationale råd for kvalitetsudvikling i sundhedsvæsenet 2002). Starting at the top of the WHO definition; evidence and evidence-based medicine (EBM) play a predominant role in health care quality development. EBM emerged as a reaction to the findings that great variation exists in the treatment provided to patients with the same symptoms and diagnoses (see (Wennberg, Gittelsohn 1973) for an often cited reference), and is now considered a cornerstone in medicine (Vallgårda 1992, x Effective: delivering health care that is adherent to an evidence base and results in improved health outcomes for individuals and communities, based on need x Efficient : delivering health care in a manner which maximises resource use and avoids waste x Acceptable/patient-centred: delivering health care which takes into account the preferences and aspirations of individual service users and the cultures of their communities x Accessible: delivering health care that is timely, geographically reasonable, and provided in a setting where skills and resources are appropriate to medical need x Equitable: del ivering health care which does not vary in quality because of personal characteristics such as gender, race, ethnicity, geographical location, or socioeconomic status x Safe: delivering health care which minimizes risks and harm to service users. - 23 - Timmermans, Berg 2003). In quality development, the importance of EBM is also emphasized as an active ingredient, because it provides the evidence-based basis of the clinical practice (Mainz, Påske Johnsen & Bartels 2010); in other words, it sets a standard for the delivery of health care services. In several Danish textbooks on quality development in health care, quality is described as occurring in a dynamic relation between medical research, health technology assessments (an assessment of the conditions for and the consequences of a form of health technology) and quality development. Where medical research lays the ground for definitions of ‘best practice’, health technology assessments judge the feasibility of these supposed best practices and functions as a tool of prioritisation, and quality development is practices related to the implementation of quality standards and priorities depicted by the former two components (Mainz et al. 2011, Kjærgaard et al. 2001, Mainz, Påske Johnsen & Bartels 2010). Thus, quality development encompasses the following list of activities: (…) problem identification, definition of quality goals, quality measurement, quality assessment, feedback to the involved health care staff, analysis of underlying causes of quality break downs, implementation of changes aimed at improving the quality and renewed quality assessment in a continuous cycle (Kjærgaard et al. 2001, author's translation and emphasis) Additionally, EBM can also be seen as related to quality development by being part of a standardisation movement. Here, the purpose of standardisation is to ensure that all patients receive the same effective and efficient treatment (Timmermans 2010, Light 2010, Light 2000), when they are treated for the same conditions. However, objections have also been raised against standardisation and what is referred to as ‘cook book’ medicine, which, it is held, erodes the individualised adjustment of medical care and devalues the worth of professional expertise and autonomy (Timmermans 2010). Nevertheless, standards constitute a considerable part of quality development - 30 - revised (Institut for Kvalitet og Akkreditering i Sundhedsvæsenet 2013), and currently Danish hospitals are facing their second round of accreditation. The components of DDKM In the preface to first version of DDKM, the model is described as follows: Over time, DDKM will become a comprehensive, integrated and joint system for quality and assessment of important services and activities in health care. Overall, DDKM is to support and promote systematic, continuous quality development in the [health care] sectors (Institut for Kvalitet og Akkreditering i Sundhedsvæsenet, author's translation) The programme follows a cyclical template structured by the PDSA cycle, which takes care of the continuous monitoring of compliance with predefined goals (e.g. predefined quality standards), so practices of clinical work can be redesigned or changed accordingly. Additionally, the programme is a system of accreditation and follows a fouryear cycle of external assessment, where the hospitals are evaluated by an external survey team. The second version of the programme (Institut for Kvalitet og Akkreditering i Sundhedsvæsenet 2013) comprises 80 standards with at least four indicators, following the steps of the PDSA cycle. Hence, indicators are formulated on four levels. The first level is concerned with the presence of a guideline describing the services related to the given standard. The second level of indicators is concerned with the staff and managers awareness of the presence of this guideline, whereas the third level of indicators is concerned with monitoring of compliance with the standard. Finally, the fourth indicators are concerned with implementation of changes with a view to increase the compliance with the standard. The methods of assessment within this regime can take many forms. However, audit stands out as a distinct and preferred method in the programme. - 31 - THE NEW INSTITUTIONS OF QUALITY DEVELOPMENT In 2004 (i.e. during the developmental phase of DDKM), the Ministry of Health, the National Board of Health and the counties agreed to establish a national accreditation institute (IKAS) (Knudsen, Christiansen & Hansen 2008). This institute was to be responsible for the establishment and further development of DDKM, for the provision of the standards that are to serve as the basis of the quality evaluation of the health care institutions and for conducting the accreditations. The Ministry of Health, the National Board of Health and the counties were represented in this process as members of IKAS’s board, but the responsibility for the institute was given to a director and a secretariat. The establishment of IKAS is indicative of a general development according to which quality development has become an increasingly consolidated and permanent part of health care with its own institutions, organisations and assigned actors. In a presentation6 from the annual meeting of the Danish Society for Quality in Health Care in 2014 the chairman, DMSc, Knut Borch-Johnsen presented the development of quality work from 1990 to 2013. Here he characterized this development as a shift from work carried out by dedicated individuals to work embedded in a ‘matured organisation’ that encompasses quality organisations in both regions and local health care institutions (including hospitals), in national societies for both quality and patient safety, in a national institution for accreditation (IKAS) and a central secretariat for the national databases of clinical quality. In the next chapter, I will return to this theme of ‘quality organisations’ in relation to the specific hospital departments that constitute the empirical cases of this study. 6 http://www.dsks.dk/filer/aarsmode%202014/fre_dsks_borch_johnsen.pdf - 32 - Despite the increased institutionalisation of quality development in health care, the current framework for quality development in Danish health care has not been implemented without resistance. It has been criticised on several points, but the lack of ‘evidence’ stands out as one of the most prominent. In the health professions, and especially among doctors, there is a strong quest for evidence-based quality development, and one of the critiques of DDKM is that the effects of accreditation remain unproven. In the Journal of the Danish Medical Association, there is an ongoing debate about the lack of clarification of the evidence behind DDKM. In 2011, an international study published by the Cochrane Collaboration revealed that only two studies were performed to investigate the effects of accreditation, and none of these studies were able to show consistent and convincing effects. The authors concluded that: “No firm conclusions could therefore be drawn about the effectiveness of external inspection on compliance with standards” (Flodgren et al. 2011: 2). The response from the director of IKAS, Jesper Gad Christensen, was that the result was unsurprising. Accreditation is a multifaceted intervention, and it would be impossible to design an evaluation that would accommodate the research requirements of the Cochrane Institute and allow it to become part of their review, he argued (Rasmussen 2011). To accommodate the critique that followed, he and other contributors to this debate (Steenberger 2011a) referred to another newly published study from Australia, which underlined the possibility that accreditation had influenced the organizational culture and management in 20 Australian hospitals (Braithwaite et al. 2010). However, this study has not succeeded in silencing the critique of the evidence-based assets of accreditation and hence DDKM. In order to accommodate this critique, IKAS developed a research strategy that focuses on the provision of evidence-based knowledge about the effects of the model7. 7 http://www.ikas.dk/IKAS/Virksomhedsgrundlag/Forskningsstrategi.aspx THE CRITIQUE OF THE CURRENT ORGANISATION OF QUALITY DEVELOPMENT - 33 - However, the lack of evidence-based knowledge about the methods used in quality development has not only led to criticism of the methods themselves, but also criticism of the actors who advocate or are directly involved in the implementation of the methods. When arguments cannot be based on facts, the criticism states they must be based on beliefs. Connotations such as ‘polemic’ and ‘argumentation based on subjective accounts’ are used to describe of the parties in the debate over who favours the present methods of quality development (Kristensen 2011). In opposition to the apparent religiosity of actors dedicated to quality work, doctors are accused of being absent in quality development. In relation to this absence, the former president of the Danish Medical Association (Hansen 2013) has also implied that professional irrelevance and uncoordinated initiatives are a major reason for the doctor’s disinterest in the quality development processes. Some even go so far as to speak of a ‘busy quality and safety industry’ preoccupied with commissions, initiatives and programmes, and totally separated from the world of doctors, “… that are actually delivering the health care with little buy-in to the quality and safety agenda” (Buist, Middleton 2013). Others have stated that the quality work, when it is placed in the hands of assigned quality workers, has moved too far away from the clinical realities in hospitals to be truly relevant for the clinical staff (Nørrelund 2012). In contrast, the doctors’ engagement in the quality agenda is supposed to be a premise for the design of initiatives programmes etc. that are perceived as relevant. Accordingly, a need of not requiring the people engaged in quality development to leave their offices and face the clinical realities is formulated, as well as a need for getting the doctors engaged in the quality agenda, (Buist, Middleton 2013, Gerdes 2013, Steenberger 2012, Hansen 2011). As a continuation of the reluctancy of doctors debate, a debate on the implications of the model when used in practice is put forward. The main concern in this debate is the workload related to documentation. For instance, in a chronicle written for Politiken, - 34 - one of the major Danish newspapers, a young doctor told of how he spent half of his shift on documentation. In an interview to the Journal of the Danish Medical Association he elaborated on his opinions: My point is that what we do in order to improve quality and safety takes up so much time that it ends up harming exactly the things that it is supposed to benefit. If you look at the overall treatment efforts, you have to ask whether the quality is improved because of the documentation. The answer is no. Maybe it has improved on some parameters – but the time it has taken to achieve a higher score here has been taken from other areas, in which the quality has decreased. Overall, there is no improvement (Steenberger 2011c, author's translation) This quote also reflects disbelief in the basic components of DDKM. Thus, the practices of documentation are not merely considered irrelevant because of the time it takes from other types of work, but also because the prime reason for documentation is related to ‘achieving a higher score’. This quote is of course related to a pending debate and makes use of a strong rhetoric, but it reflects a more general criticism that considers documentation and accreditation as related to ‘showing off’ and as separated from the clinical work (see for instance (Wiener 2000)) A fourth theme in the debate about DDKM is the focus on processes8 instead of the core tasks in health care (performance standards). Standards related to direct patient care are ruled out, and this is pointed out as a weak spot in the Danish as well as the international accreditation models. This was an argument for moving away from the international models that were already in use and instead developing a Danish model (Larsen 2003). However, a remaining point of criticism that the model – even in its Danish form – fails to evaluate the most crucial aspects of the health care service, and it has been called a ‘a ridiculous model’ that uses ‘surrogate measures of quality’ (Bjerre 2010, 8 For instance, there are standards for the presence of a quality organisation, for guidelines on everything from screening of pressure ulcers to refrigerator temperatures and for correct administration of guidelines. - 35 - Holm-Petersen, Wadmann & Vejen Andersen 2015). Instead, it is argued, attention should be paid to the quality of the core activities in health care and a national model of quality development should include clinical guidelines specifically (Steenberger 2011b, Heinskou 2011). QUALITY DEVELOPMENT BEYOND OR AFTER DDKM DDKM is given considerable space in this chapter. Redraw DDKM is a clear indication of the considered importance of decision makers and stakeholders in making quality development a systematic and integrated part of health care. Still, a very recent development in the field of quality development in Danish health care is the announcement of the Danish Ministry of Health, in April 2015, that DDKM would be Withdrawn and replaced by a simpler model with fewer nationally formulated quality standards. Instead, there should be more emphasis on the local hospital department, right to defining own goals of quality and the provision of relevant real-time data on performance through investments in the national clinical quality databases. This announcement was part of the ministry’s strategy paper ‘National quality programme for the health care sector 2015-2018’, which also emphasised the manager’s role and what they call ‘a learning culture’, which is defined as a situation where: “(…) you [the health professionals] go to work every day with the ambition of doing your job a little better than you did yesterday” (Ministeriet for Sundhed og Forebyggelse 2015: 6). Interestingly, this strategy paper also criticises DDKM for being too bureaucratic and placing too much emphasis on control and documentation, and calls for a greater space for health care professionals to work with quality development initiatives that are meaningful to them in their specific clinical situation (Ibid). Hereby, it voices the criticism of DDKM put forward by the health care professionals for several years, but still emphasise the need for a national model that retains a systematic focus on quality development and that this work should be data driven – though with slightly different goals and means. - 36 - The ministry’s announcement sent a shock wave through the system and led to reflections on what would be gained and lost by ‘sacking’ DDKM. In a meeting which I attended in the beginning of May 20159, some Danish hospital directors expressed a concern for losing an important tool of motivation for change (or a means of putting pressure on the clinical departments) with the removal of the sense of importance provided by accreditation. Others expressed a hope that quality development using this new approach would become much more closely attached to the clinical managers and clinical staff instead of being a concern mostly for employees in the hospitals’ quality organisations. Clearly, this newest development is not reflected in the empirical material of this study. But it is interesting to mention here as a telling case of quality development as a dynamic field, in which new frameworks of how to conduct quality development seem to emerge, live side by side and replace each other at a considerable pace. However, even the most minimalistic frameworks for quality development apply for a systematic approach that encompasses methods of quality assessment, problem identification and analysis. CONCLUDING REMARKS In this chapter, I have provided an introduction to the landscape of quality development as being embedded in a set of general ideas, and I have described how these ideas have settled in a Danish context. Quality is a broadly defined concept that allows many ideas, activities and technologies to be placed under the heading of quality development. Many of the concrete initiatives include predefined goals for quality development, as well as firm framings for how quality development should be performed. These framings are not always compatible with each other, and critics have called this landscape incoherent. In a Danish context, the development of DDKM stands out as a manifestation of many years 9 The annual meeting of the Society for Hospital Directors. - 37 - of work towards a national framework for quality development that encompasses most of these ideas and practices. Even though DDKM was developed as a framework that was supposed to ensure quality development as a uniform and integrated part of health, it is interesting to note that the debate in the health care sector points to the possibility of disintegration of the efforts of quality development. This lack of integration is not only related to the plural nature of quality development, but to the challenges of making quality development activities cohere with clinical work and engaging the staff in them. However, quality development has solidified as an explicit part of health care through separate organizational structures, and the question is whether this has brought quality development closer to the health care organisations or led to a gap that needs to be breached. This study is positioned in the middle of this ‘blurred picture’ of quality development and embedded in or performed next to everyday clinical practices. I wish to look into those practices that enable the coexistence between different frameworks and between quality development efforts and clinical work. How are the tensions –increased by the heated debate about the current frameworks for quality development – acted upon and resolved? And last but not least: who carries out this work? These are some of the questions that initiated this study, which is also reflected in the choices behind this study’s empirical strategy and analytical framing. I will return to these choices in Chapters 4, 6 and 7, and in the following chapter the concrete empirical case of this study will be introduced. - 38 - INTRODUCING THE EMPIRICAL CASE In the following, after the introduction to the broader empirical context of quality development in Denmark, I will introduce the hospital and the two hospital departments in which this study took place. Additionally, I will provide a description of the pivotal empirical figure in this thesis; the quality coordinators. The reasons behind the specific choices of these empirical fix-points are given in Chapter 4. THE HOSPITAL The study was performed in a hospital10 in the Copenhagen area. This hospital had the status of a community hospital, with a medical department and emergency ward, and was a hospital with specialised functions in areas such as neurology, ophthalmology, rheumatology and severe back diseases11. As such, the hospital served the citizens of the five surrounding municipalities (133,000 citizens), and in the specialised areas also the citizens from the Capital Region12 and the rest of the country. The hospital was managed by a hospital director (administrative director) and two deputy directors (a nurse and a doctor), and besides the clinical departments the hospital organisation included administrative and servicing departments (see Figure 2). 10 I have made a deliberate choice of referring to the hospital instead of calling it by its actual name. To my knowledge, I have not observed or presented anything that could compromise the hospital, the departments or the staff in this study, and none of my informants requested anonymity. However, in order to avoid any unintentional disrepute I have chosen to blur the identity of both the hospital and the persons in the fieldwork. When referring to the departments, I refer to the surgical or medical department, for instance, and hence by their medical specialty instead of their name. The same goes for the department wards. All persons in this text are given other names (see Appendix 1), though I refer to their actual positions. 11 After I had finalised the fieldwork, the Capital Region implemented a new hospital structure, and the presented organizational traits have already been radically changed. 12 One of the five Danish regions – a politically led authority managing the hospitals. -39 - Figure 2: The organizational chart of the hospital (adapted by the author from the hospital’s webpage) The quality organisation of the hospital As previously mentioned, the quality and safety movement did not only materialise into a number of methods, programmes and technologies of quality development. Quality had also become part of the hospital’s organisation through quality departments, quality policies and assigned quality staff. The establishment of quality organisations in the local health care institutions such as hospitals can at least to some extent be ascribed to - 46 - nurses of either a ward or a department. They were familiar with the daily routines of a hospital department, including the planning horizon and the typical workflows within and across departments. Often they referred to this as a particular important skill, as many of their activities were about timing – or more precisely making the right people meet at the right place at the right time. Besides their background as nurses and former managers, the quality coordinators had supplemented their skills through various additional courses and educations. One quality coordinator held a master’s degree in IT management and the other a master’s degree in Public Governance. The insight into the more technical aspects of ‘quality development’, such as the detailed requirements of audits, and the patient and safety programme was gained through courses offered by, for instance, IKAS, introducing methods of audits and surveys. Finally, project management courses and courses in LEAN facilitation and other process-related courses could be seen in their CV’s. CONCLUDING REMARKS In this chapter, I have provided a description of the empirical case, i.e. the hospital, the departments and the quality coordinators studied in this thesis. Hereby, I have first and foremost provided an overview of the specific, local context, of which the empirical material is a part. Secondly, I have provided a description of the way quality work is embedded in a specific organizational context that is closely connected to the many political goals and elaborate national programmes of quality development, and not least how this organisation, including the quality coordinator position, is a result of an expansion of the requirements to quality development, as well as changed conditions for managing and coordinating hospital work in general. As a continuation of the previous chapter, this chapter also points to the significant position of quality development, this time in the hospital. Here, quality development is - 47 - organised in independent units containing both actors and tasks, but is still deeply interrelated with the organisation of clinical work through a) the hopes and assumptions about the way that quality work will have e beneficial impact on clinical work and b) the way that quality work claims resources (time and staff) already in use in the clinical work. This raises questions about how this structure of simultaneous independence and interrelation condition the development of quality work in the hospital departments, and the way quality work is organized and managed. Questions I have also carried into the development of the analytical framework (Cchapter 6) and the literature review (Chapter 5), as a point of reference in the reading of existing research literature on quality development in health care. Now that the empirical field of quality development in health care and the specific empirical case consisting of two hospital departments in a Danish hospital have been introduced, I will move on to a description and discussion in Part II of how I developed the analytical and methodological framework used to explore the development and organisation of quality work. - 48 - - 49 - PART II - 50 - FRAMING THE STUDY AND GENERATING QUESTIONS In the previous chapter, I framed quality development as a field of many ideologies, goals, methods and frameworks, which, despite its continuous developments and additions, has become a consolidated part of the health care system. This was the starting point of the present started: an overall interest in quality work as a broadly appealing, expanding and yet amorphous and elusive part of the health care sector, as well as a curiosity about the way quality work became organised as part of, or in relation to, clinical work and with what consequences. An intriguing starting point, but one that is not sufficiently specified. In the following, I provide a further specification of how this particular study progressed into its form as presented in this thesis. The purpose here is not to provide a detailed account of my own personal journey from the day I initiated this study and till it was completed. Rather, I wish to emphasise some of the turns this study took, because I believe that it is here that I am able to show and explain some of the crucial choices made in order to narrow down, specify and develop the questions raised in the study. Hence, this chapter is neither a methodological nor a theoretical outline (see Chapters 6 and 7), but a chapter about how I framed the study through different encounters with the empirical field. TOWARDS AN ETHNOGRAPHIC STRATEGY As I have already described, this study was initiated through an empirical interest in the expansion of quality work and how it was about to settle as an independent field within the hospitals. Hence, I set out to study quality work as something that was both an - 51 - organisation in itself and a type of work that had to be organised and embedded in an existing hospital organisation and hospital work. My aim was exploratory, and I basically wanted to know what these organising activities were about. As Neyland (Neyland 2008: 32-35) notes, this is a completely valid starting point, but the need to define a focal point – a place, a group of people, a certain activity – for the ethnographic strategy is inevitable. However, I was undecided on how to construct and delineate an ethnographic field to study. Following Madden (Madden 2010), an ethnographic field is not the same as a geographical field, but an attempt “(…) to marry the interrogative and investigative inclination of the ethnographer to the place that has been made by a group of people. (…) It is the synthesis between concrete space and investigative space that defines the ethnographic field and gives it its reason for being – it exists to describe, to interrogate, to question, to problematize, to theorise (…)” (Ibid: 39). The question became, however, where this concrete space of quality work was, and where I would be able to study it. Where should I perform my study, and who or what should I study? I ended up performing the fieldwork in two hospital departments in a Danish hospital mentioned above. The choice of both hospital and departments was pragmatic rather than strategic, as I went along with an existing contact I had from my employment in the Danish Institute of Health Services Research14. This hospital was about to develop a programme of organizational development based on a new concept, the concept of ‘Inter Professional Learning’ (IPL), aiming at improving quality in terms of increased patient satisfaction and more coherent care (Vyt 2008, Oandason, Reeves 2005). This programme included an educational course in which clinical managers or health professionals from the hospital departments would be educated as facilitators of increased and improved communication between the various groups of health 14 Now KORA, Det Nationale Institut for Kommuners og Regioners Analyse og Forskning [Danish Institute for Local and Regional Government Research]). See the organizational set-up of the study in the introduction (Chapter 1). - 52 - professionals and patients involved in the individual illness trajectories 15 . More specifically, these facilitators were to become initiators of development processes aiming at IPL, and the hospital was interested in an evaluation of this programme. I was not interested in IPL as such, but I decided to meet with the hospital’s quality manager and those consultants from the hospital’s HR & quality department who were engaged in developing the concept and the education programme, in order to search for possible ‘windows’ that would allow me to study the quality work of interest. During this meeting, it occurred to me that these future facilitators of IPL were not the only group of staff in the hospital assigned to play a key role in quality development processes. In this hospital, the quality coordinators’ position (see Chapter 3) was institutionalised, as every department had assigned someone to assist the department managers in the implementation of national quality frameworks and national, regional and local quality development initiatives. At that moment, this position revealed itself as a possible way to study quality work, whatever and wherever that was, and hence a solution to my selfimposed challenge of engaging with a field that could potentially include almost anything that looked like improvement, development and change. This position presented itself as an empirical demarcation, and at some point also as the unit of analysis (Yin 1989:31). ASKING (NEW) QUESTIONS: HOLDING ON TO THE EMPIRICAL TENSIONS Although I was approaching a firmer empirical framing, I still had to develop a research question that precisely defined my analysis of what would become a rather comprehensive body of empirical material. In this section, I will provide a description of the strategies used to revisit, explore and problematise some observed tensions in the 15 I followed one of these IPL courses in the fall of 2012, because I initially had an idea of performing the study as an observation of both IPL facilitators and quality coordinators. However, as the study progressed I decided to abandon this idea, and hence the insights I gained through the observation of this course have not been used in the analysis of this thesis. - 53 - empirical material and how this informed the questions asked in this thesis and the positioning of the thesis analyses in relation to the existing research literature on quality development. Paying attention to the mysteries According to Ybema & Kamsteeg (2009), the interesting insights from an organizational ethnographic study can only be achieved through a deliberate effort to distance oneself from the empirical material and hereby foster a capacity for being surprised or discovering the unexpected, the mysterious, the strange or the irrational. The advantages of an ethnographic fieldwork are the intimate relationship between the researcher and the studied field, and hence the ability to approach an insight from within the field being studied (de Jong, Kamsteeg & Ybema 2013). This holds particularly true for the classic studies of anthropology, where researchers left their homes to study foreign cultures and approached the field as outsiders. Being an outsider heightened the researcher’s sense of sensitivity and awareness of the nuances of what she was studying. Organizational ethnographers often perform studies in settings that are somehow familiar to them and close to them socially and culturally. Ybema & Kamsteed (2009) argue that the ability to ‘open a field’ by engaging with and being immersed in the field through ethnographic methods is counteracted by the ease of becoming blindfolded by normality. This is not an argument against organizational ethnography, but a reminder that the strange and the unforeseen is not a given, but something the organizational ethnographer has to pay particular attention to in order to develop an understanding of the studied field that goes beyond “what [is] conventionally ‘there’ to be seen” (Ibid: 8). The fieldwork of this thesis consisted of days that were very much the same; the quality coordinators worked on their computers, they participated in meetings, they small-talked with someone who dropped by their offices or someone they met in the corridor and so - 54 - on. To be honest, they were sometimes tedious days that pretty much resembled my own working days (when I was not performing fieldwork), and many of the situations that I observed were very familiar. Additionally, I was well acquainted with the health care sector in general and the hospital organisations in particular from studies in my previous job. Furthermore, I was warmly welcomed by both the quality coordinators and their closest colleagues, and I soon began to feel at home. Hence, the odd feeling of being an intruder or trespasser, and the period in which everything is new (Neyland 2008: 100101) did not last for long. In this process, the mysteries of what I studied became less obvious, and I had to deliberately detach myself from the material in order to reflect upon what I studied. Ybema & Kamsteeg encourage organizational ethnographers to think from ‘within’, through an open and empathetic approach, and from ‘without’ by constantly problematising the observed, cultivating strangeness and preserving an intellectual distance to the empirical material (Ybema, Kamsteeg 2009, 8-9). Accordingly, they suggest several strategies and I was particularly inspired by one of these: the strategy of holding on to those mysteries that arise from the researchers own naïve wonders or the surprise of those who are researched (Ibid: 12-14)16. A strategy that I did not utilise deliberately (though sometimes unintentionally) throughout the field study, but a strategy I attempted to apply when I revisited my field notes and some of the very early analytical work and conference presentations, and some of the first attempts to describe my empirical material. In this empirical and early analytical material, a set of tensions or paradoxes kept disturbing my otherwise ‘un-problematized’ view on what I observed and became essential for the questions I ended up pursuing in the thesis. 16 Additional strategies suggested by Ybema and Kamsteeg (2009) are ‘Looking for the irrational’, ‘Making it look strange’, ‘Breaking the friendship bond’, Distancing by immersion’ and ‘This fellow is wise enough to play the fool.’ - 55 - Overall, these tensions were concerned with the elusiveness and negotiability of quality and quality as an object of work (Casper 1998), the emerging and dissolving connections among actors in quality development, and the quality coordinator’s position as being closely related to decisions of change but without managerial authority. In the following, I will present these tensions by drawing on some of the empirical encounters and my reflections upon them. What is quality? First and foremost, I was struck by the diverse nature of what was referred to as quality work. During the observations, several definitions of ‘quality’ were emphasised and ‘quality’ was rarely the same, neither in its definition or execution. In one of the outpatient clinics, a doctor described quality as ‘giving time’, ‘paying attention to the interaction with patients’ or ‘conferring with colleagues’ when cases were difficult to understand. During a surgery, I observed another doctor mentioning quality as related to ‘a treatment that fits the patient’s conditions’, when he explained to me the surgical procedure he was about to perform, and a surgical nurse emphasised ‘protecting patients from infections’ as being part of performing health care services of a high quality. In some of the accounts of quality it was possible to track them back to national or even international definitions of quality and guidelines describing practices of high quality. In other accounts, ‘quality’ seemed to be defined in the situation by the actors I observed or linked to the specific professional backgrounds of the people I happened to be talking to – not least the quality coordinators. I had the impression that quality, and following from this also quality problems, could apparently be almost anything. Although I usually understood why each notion of ‘quality’ presented to me was important, it also led to the question: how does something become defined as a quality problem, and – following from this – how does something become a subject of quality - 62 - methodology they were themselves conducting an additional amount of work that was not described in the official manuals. Searching for the presence of health professionals in the booking systems, preparing material and presentations for a workshop process and having the head nurse make a workshop invitation are examples work performed by the quality coordinators that is not otherwise described as part of the methodology for quality development. Additionally, the quality coordinators were not only paying attention to the work captured in the various quality assessments. On the contrary, they were deliberately seeking and using alternative insights into the various work practices in the department, in order to understand the entire sequence of work. In that respect, they were operating on the edge of what was standardised and made visible on the one hand, and the situated, complex and invisible work (Star, Strauss 1999) on the other. I have now described some of the ‘mysteries’ that caught my interest during and after the fieldwork. In the following section, I will describe how this made me reconsider the original emphasis on the quality coordinators as the primary unit of analysis. A shift in attention: From ‘quality coordinator work’ to emerging connections The meeting with the quality manager and the consultant in the hospital department of HR & Quality mentioned earlier and the subsequent decision to go along with the idea of studying quality coordinators’ work took place before I even started to formulate the initial PhD application. As I implied above, this choice also ended up influencing the inquiries I made at that point in time. Hence, this study was initiated as an inquiry into the actors performing quality work, and I formulated a research application and research question that focused on the emergence of this new group of actors in hospital settings, working in the intersection between clinical work and national regulation. As such, this preoccupation with a particular group of actors in the hospital and the expectations of their potential achievements that I met in my first encounters with the empirical field - 63 - was crucial for how I initially framed the study. This was reflected in the way I designed the collection of empirical data, I decided to pay particular attention to the work of quality coordinators, and the preoccupation with the quality coordinator function was also highly explicit in the earliest analysis (Madsen 2014, Madsen Forthcoming). However, as my analysis developed I became more and more aware of the limitations of this perspective, because it favoured the quality coordinators in a way that did not resemble what I observed empirically. One point of criticism that I was often faced with was that my analyses gave the quality coordinators a status as more strategic than other actors and as unaffected by the actions and the other actors to whom they and their work was related. Central to this challenge was my emphasis on one particular actor (the quality coordinators), which was reflected in both the imposed strategy for my fieldwork and the research questions that I formulated. Other actors were too easily allocated marginal positions, where they appeared as ‘context’ or ‘tools’, and hence a premise rather than someone or something that interacted actively with the quality coordinators. At some point, this problem made me rethink my preoccupation with the quality coordinators as representatives of quality work. What I saw and what I found interesting in the empirical material was that quality work was apparently constructed through emerging and existing group formations. These group formations had caught my attention from the very beginning of the fieldwork, but I had primarily considered them as a result of the quality coordinators work. At a certain point, my attention shifted, and I started to pay attention to these group formations as ongoing processes in which actors where connected and disconnected, and where certain aims and actions became more or less important. The quality coordinators were still the only actors present in every encounter that I came to study, and it was inevitable that they would play a pivotal role in the empirical material as well as in the analysis, even with this reframing, but the attention shifted from their individual achievements towards their performances in - 64 - unison with actors in and beyond the local departments. Thus, this shift in attention was not made to prevent myself from discussing the actual work of quality coordinators, and it was not made to have to perform a completely new fieldwork. It was a shift in the analytical unit from ‘quality coordinator work’ to the shifting connections among actors engaged in quality development processes in the departments and how the specific details of quality work were constituted through these connections. CONCLUDING REMARKS In this chapter, I have described how the present study came into being and how my engagement with the empirical field led to the emergence of new questions. In the above, I have narrowed these questions down to a list of three, and the tensions inherent in these issues are in different ways related to the fuzziness, elusiveness and instability of quality work on the one hand and the observation of emerging assemblages and interrelations on the other. These questions did not merely develop through empirical encounters, but also as an iterative process in which different analytical concepts were tested and discarded or used along the way. In these iterations, the empirical observations of a dynamic fluctuation between emerging and fading coherences in quality development processes became even more intriguing, because it became possible to contrast and supplement the existing theoretical and analytical perspectives used in existing research on quality development and perhaps contribute with new insights. Below, I will explain how I developed the analytical framework in dialogue with existing literature and the empirical encounters mentioned above. - 65 - LITERATURE REVIEW In the previous chapters, I have described the context of quality development in Danish health care as a superordinate of technologies and initiatives that attempt to reorganise clinical work for the better. I have demonstrated how quality development in health care encompasses many types of ideas, technologies and practices, and also how it has succeeded as an overall project in becoming a strong and independent component of Danish health care. Additionally, I have explained how this intrigued my curiosity, and how I began asking questions regarding this as an empirical phenomenon. In this chapter, I will turn to the literature on quality development in health care, in order to place this study in the broader academic context. At the end of this chapter, I will provide a discussion of how these studies relate to the questions raised in the previous chapters and clarify the contribution of this thesis. QUALITY DEVELOPMENT – A STUDY OF MEDICAL PRACTICES The immense interest in ways to improve quality has not only attracted the interest of policymakers, managers and professionals in health care, but also researchers from various scientific and theoretical fields who take an interest in the ways in which quality development affects the service delivery of health care institutions. According to Zuiderent-Jerak & Berg (2010), this interest is an extension of an interest in medical sociology: The quality and safety improvement movement that is emerging in response to the problems in the delivery of health care is highly involved in a practice that medical sociology has for many decades been exploring: analysing and problematizing medical practice in substantial and increasingly influential ways (Zuiderent-Jerak, Berg 2010: 325). - 66 - Furthermore, they argue that the involvement from the quality and safety movement in medical practice has raised special interest among medical sociologists because of the quality and safety movement’s close relationship with influential national institutions with strong agendas for change. This has, according to Zuiderent-Jerak & Berg, made medical sociologists specify this interest in medical practices and inquire into the effects of the many subfields of quality development (Ibid). Accordingly, analyses of the underlying ideas of quality development and how they manifest themselves in concrete practices have been performed, as part of a critical discussion of the ambivalences and ambiguities emerging from the many attempts to improve or control quality. In general, these studies point to resistance, tensions and unforeseen effects in the meeting between these methods and technologies of quality development and the health care organisations and concrete medical practices. In this way, the studies pinpoint particular drawbacks of the technologies and methods used and illuminate the efforts required from both staff and managers in relation to their implementation. As I will return to later, this study contributes to the latter in particular by taking departure in the efforts related to the constructions of emerging and temporary connections of actors around particular quality development processes. In the following, I will present the overall findings of the research literature through three broad categories of research. Inquiries into the underlying assumptions and basic components The first group of research approaches quality development from an interest in the underlying assumptions about how quality is improved. One such example is Knudsen’s analysis of how the Danish accreditation programme (DDKM) was developed and how this contrasts with the inherent assumption and related rhetoric, stating that accreditation is a method of all-encompassing evaluation. Rather, he holds, the - 67 - predefined quality standards on which accreditation relies, must be considered as a selected and limited insight and based on the attentiveness of the programme developers. Hence, the exact activities that are subject to evaluation are based on these generalised formulations of indicators, which must be considered as limited in scope (Knudsen 2011a, Knudsen 2011b). In a similar way, the underlying assumptions of the patient safety movement have been questioned in relation to the implicit suppositions of the concrete methods of patient safety. These suppositions, it is argued, neglect already existing structures for safety work, and assume that there is a dominant discourse of blame among health care workers and that it is possible to erase uncertainties and errors (Zinck Pedersen 2013, Mesman 2008, Waring 2007b, Waring 2005, Jerak-Zuiderent 2012). For instance, the existence of ‘a culture of blame’ that is believed to make health professionals reluctant to share their mistakes and learn from them is a principal assumption inscribed into many patient safety models. This assumption is, however, questioned by Zinck Pedersen (2013), who suggests that errors are rather handled in informal structures of collegial regulation, which indicates that there is no ‘culture of blame’ to be replaced by a ‘culture of safety’. A similar argument is brought forward by Waring (Waring 2007b), who challenges the assumption that doctors are biased towards ‘person-centred’ explanations of errors rather than explanations referring to ‘the system’. Additionally, in another paper he points to a rather different set of barriers to doctors’ incident reporting than those put forward by the patient safety movement. A set of barriers that is rooted in ‘a culture of medicine’, where medical errors are regarded as inevitable and unmanageable, and where incident reporting is also resisted, because it potentially opens up for the possibility of interference and control from managers and others outside the medical profession (Waring 2005). - 68 - Also, within the specific subfield of patient safety researchers have paid particular attention to the way ‘safety’ as a practice is already achieved in the health care. Such studies have pointed to different types of errors in the performance of health care and to how some of these are already addressed by resilient practices of care. In this way, these studies invite to a discussion about how and whether attempts to standardise practices in the name of ‘patient safety’ support the elimination of errors or produce new forms of risk (Mesman 2008, Jerak-Zuiderent 2012). Health professions in transition A second category of research pays particular attention to what may be referred to as ‘health professions in transition’; old figures change and new figures emerge in the attempt to respond to changing public demands and increasing interest in regulation of health care. As Allen & Pilnick phrase it, “(h)ealth care has emerged as a popular case for the study of the world of work and occupations” (Allen, Pilnick 2005: 683) for decades. This especially holds true in light of the continuous efforts to regulate the health care system through ideologies of efficiency, transparency and patient involvement that have challenged and changed existing work practices, divisions of labour and professional autonomy (Ibid). A major concern in this category of research is the impact of regulatory reforms on professional autonomy. The increase in such reforms and the associated requirements of making professional work available for control and scrutiny are said to challenge the professional autonomy (Kuhlman 2006a, Light 2010, Light 2000). Accordingly, the medical professions’ strategies to either adapt, subvert or avoid the managerial control mechanisms have been a subject of study, in which colonisation or decoupling has been stated to be an end in a continuum of possible reactions to these demands (Power 1997, Numerato, Salvatore,D., Fattore,G. 2012). Others, however, have provided insights into - 69 - other kinds of reactions, where the professionals adapts technologies and methods in order to take control of them, and hence also take control of the degree of transparency of their work (Waring 2007a, Levay, Waks 2009). Another theme in this category of research is the discussion of changed professional identities and jurisdictions (Kuhlman 2006b). Kirkpatrick et al. (2011) stress that management has become a contested terrain as a result of new opportunities for engaging in health care management: “(…) this system [referring to Abott (1988)] is in constant flux, with external changes in regulation and technology generating opportunities for established and aspiring professionals to colonize new areas of work” (Ibid, 490). In other words, the work and power relations of existing professions and occupational groups tend to be displaced. Scarce resources and patient concerns are other examples of factors that lead to changing frameworks of management, and distinctive administrative elites are emerging in the clinical professions (Kirkpatrick,I.,Kragh Jespersen,P., Dent,M.,Neogy,I. 2009). The notions of hybrid professions (Kragh Jespersen 2005, Kurunmäki 2004, Numerato, Salvatore,D., Fattore,G. 2012) and hybrid managers (Kragh Jespersen 2005) have emerged as a way of understanding the necessity of merging and balancing dimensions of professional work and more general dimensions of management. This emergence of new categories in the professions can be understood as a strategy of maintaining control through direct involvement in management or regulation technologies (Waring 2007a, Waring, Currie 2009). Additionally, research has revealed how less powerful professions are able to reposition themselves through strategic involvement in the construction of the regulatory frameworks in a manner that promotes their own interests (Levay, Waks 2009, Martin, Waring 2012, Evetts 2011). In this way, management has become a contested terrain, in which doctors are concerned with maintaining their autonomy, and nurses increasingly engage in management and - 70 - strengthening their position in the hierarchy (Kirkpatrick, Dent & Kragh Jespersen 2011). Quality development technologies in practice Previously, I have described how there is a debate in the health care sector about what constitutes ‘evidence based quality development’. There is a general interest in the effects of different approaches to quality development and the preconditions for these effects (for instance staff motivation and methods of implementation). Consequently, journals such as BMJ Quality and Safety and International Journal of Quality in Health Care, as well as the Cochrane Institute, are rich sources of this kind of study, which approaches questions of effect in a variety of medical specialties and health care sectors. Studies such as these can be defined as subscribing to a functional (Wadmann et al. 2013) or rational (Albæk 2003) perspective, where quality development methods and initiatives are evaluated in relation to their original design and purpose as tools of quality development, assessment and control. As such, studies have paid particular attention to the exploration of the effects of, for instance, accreditation/external assessment (Braithwaite et al. 2010, Falstie-Jensen et al. 2015, Øvretveit, Gustafson 2002, Flodgren et al. 2011) and the effects of methods for implementing clinical guidelines (Øvretveit, Gustafson 2002, Grimshaw,J.M.,Thomas,R.E.,Mclennan,G., Fraser & Ramsay,C.R.,Vale,L. 2004)}. This approach, however, has been critiqued by social scientists, who argue that it pays undue attention to the complexities of making the many methods and technologies of quality development work as intended (Zuiderent-Jerak, Berg 2010, Vikkelsø, Vinge 2004). Most of research in this category of literature takes as its point of departs the tradition of Science and Technology Studies, according to which a well-functioning technology is considered an achievement based on a particular local practice (Latour - 71 - 2005). Referring to the field of quality development, it is argued that it is impossible to understand the success or failures of these attempts to improve quality, without paying attention to the complexities of the various practices related to their execution and the investments made in order to make them work. Additionally, without these tracings, the assessment of whether and to whom the quality development initiatives have become a success is impossible to make (Zuiderent-Jerak, Berg 2010, Timmermans, Berg 2003, Vikkelsø, Vinge 2004). This technology-in-practice approach (Timmermans & Berg 2003) considers technologies as actors with the ability to change workplaces, but this is neither because of the attributes of the technologies nor the attributes of its users. Transformation and change happen in the interrelationship between technologies and users, and hence the technologies’ accomplishments can only be studied empirically: Research in this field [technology-in-practice] has demonstrated that tools do not slip into some predefined space in an existing workplace; getting a tool to work requires negotiations with all the actors involved and sensitivity to the local work organisation. It is through a process of convergence by the use of such technologies in practice that tools and workplaces transform each other (Allen 2009: 356). Accordingly, it is difficult to single out and isolate the functions of technologies, as they are embedded in heterogeneous networks consisting of other tools, practices, groups, professionals and patients (Timmermans, Berg 2003). This approach has inspired many empirically sensitive studies of medical technologies initiated with the aim of improving practices of care. Such studies have paid particularly attention to the specific processes related to, for instance, the implementation of quality indicators (Jerak-Zuiderent, Bal 2010), accreditation systems (Wiener 2000, Hatting 2007), electronic patient records (Vikkelsø 2005), care pathways (Allen 2009, Pinder et al. 2005, Allen 2013) and quality improvement collaboratives (Zuiderent-Jerak et al. 2009). Among those are studies that pay particular attention to the transformations of specific quality development tools from general and vaguely defined technologies to specific local versions of the - 78 - patient). More generally speaking, it refers to any project of organised actions towards a given product such as: “ (…) inventing a new model of computer, building a house, getting a voluntary organisation off the ground etc.” (Strauss 1985, 4). An arc is constituted by a variety of tasks performed in a sequence or in different phases of a project: “An arc for any given project consists of the totality of tasks arrayed both sequentially and simultaneously along the course of a trajectory or a project.” (Strauss 1985, 4) These tasks can be bundled into types of work, and in the study of illness trajectories Strauss and colleagues identified five types of work (Figure 4). These types of work refer to the illness trajectory studied by Strauss and colleagues and hence to a specific project within a hospital. The types and their combinations will vary by different arcs, as well as by the sequential ordering of the tasks or bundles of tasks. Strauss et al. refer to a trajectory sequence point as any point at which it is decided to do certain tasks, and hence they suggest that tasks are allocated in time. Furthermore, they suggest that tasks have an organizational base such as: “(…) proper skills, a sufficient workforce, appropriate equipment, necessary drugs, enough time and so on” (Strauss et al. 1997, 30-31), which allows them to be carried out. Specifically, Strauss and colleagues emphasise that for every task there is an actor (or several) assigned to that task. In other words: the distinction between different tasks is followed by the assignment of these tasks to different actors. These different projects or trajectories with their implicated arcs of work entail different divisions of workers (persons or classes of persons or units of organisations) in order to get the constituent tasks done. (Strauss 1985, 4) As revealed in this citation, an actor can be a person, a department or an organisation etc., and hence actors can also vary in their attributes, for instance experience, skill or knowledge. Actors may share tasks or perform them individually and, as was the case with the tasks included in the arc of work, it is an empirical question how the division of labour should be. - 79 - Figure 4: Types of work identified by Strauss and colleagues in relation to an illness trajectory (Strauss et al. 1997) I have already described how this thesis grew out of an interest in the quality movement and particularly in how it became a formal part of health care, introducing distinct quality organisations, distinct demands to the way quality development is performed, Machine work: Work related to the monitoring of the equipment for instance by safety engineers from a service department done in order to avoid hazardous situations where machines are a danger to patients or personnel or are providing false information about the patients’ condition. Service and provision of supplies are also included in this category of work and may be performed by both service departments, but also from personnel within the departments. Finally, there is the machine work related directly to a patient or a patient’s body; connection of machines with the patient’s body, the use of diagnostic equipment in the laboratories, but also to the transportation to and from examinations. Safety work: Work related to the managing and shaping of courses of illness so they are the least hazardous and protect the patients from the contingencies of their disease. The content of this work includes anticipation of the potential hazards of a given condition and the implied potential risk of the medical intervention (risk of infection, adverse effects of medication etc.) as well as the hazards related to the use of equipment (and hence related to the above mentioned machine work), and organisation of the available resources accordingly. Comfort work: Work related to the management and relief of discomfort caused by the illness itself, the medical interventions or the hospital environment. Several kinds of tasks can be mentioned here like preparing a patient for the discomfort, minimising or relieving of discomfort, or assessing and reporting discomfort. Sentimental work: Work related to the patients’ psychological discomfort of being ill, going through various treatment and diagnostic procedures. This encompasses empathic gestures in direct relation to a current discomfort but can also be work related to the preparation of patients to life with the illness after hospitalisation. Articulation work: Work related to the management and coordination of the individual types of work in order to make them cohere and merge into a totality. This type of work, I will return to in the following. - 80 - distinct technologies and designated actors to support and manage this work. If arcs of work consist of tasks, clusters of tasks and various actors to perform those tasks, I argue that quality development can be considered as an arc of work that, despite the obvious interrelatedness, is different from the clinical work related to the care of patients. Or more specifically, quality development consists of several arcs of work, related to different quality development processes that each constitutes what Strauss and colleagues refer to as a trajectory sequence point. However, where Strauss and colleagues describe these arcs of work as relatively stable entities, I wish to approach the arcs of quality work (in the forthcoming just quality work) as emerging, and for that purpose I now turn to some of the analytical principles developed in Actor Network-Theory (ANT). A RELATIONAL UNDERSTANDING OF ACTORS AND ORDERS ANT was originally developed with a minimum of theoretical concepts and introduced as a set of analytical principles that could be used to explore the construction of science, technology, society etc. (Latour 2005, Latour 1999a). One of these key principles is that any scientific fact, any societal norm or structure or any actor17 should not be considered as acting according to some inherent or basic properties or essential truth, but as effects of an actor-network. Latour (2005) also refers to ANT as a sociology of associations and hereby emphasises that every stabilised entity or structure is the result of negotiations and compromises that align actors in a stable network. When the entity is stabilised or 17 In this thesis, I consistently use the term actors rather than actants, as otherwise suggested by Latour (Latour 2005: 54-55). Because the concept ‘actor’ may suggest the intentional actions of humans, the term actant was suggested to reflect the belief that people and things should be studied on similar terms. The argument, and a second analytical principle of ANT, is that agency – regardless of whether it is the agency of humans or nonhumans (organisation, ideologies, technologies, animals, microbes etc.) – should be considered as an achievement given by its relations to other actants. In this thesis, I relate to this understanding of agency and symmetrical explanations of human and non-human agency (Latour 1987). Nevertheless, I insist on using the term actor instead of actant, but solely due to a linguistic preference, and not to a preference for human agency or asymmetrical explanations of human and non-human agency. - 81 - black-boxed (Latour 1987), no one questions these underlying efforts. However, in order to understand how and why something becomes stable enough to be considered an unquestionable truth, whether it is a scientific fact or a social structure, the researcher must trace these associations that precede its emergence (Latour 2005). Translation Accordingly, ANT agitates for empirical studies of the way the stability of a scientific truth, an organizational or societal order, emerges in processes of translation, in which negotiations and controversies have been dealt with and actors have been aligned in an actor-network. Hence, translation refers to actors’ abilities to influence other actors according to their own strategies or interests (Latour 1999b, Latour 1987, Latour 1991). Latour (Latour 1991) explains this as the meeting between a programme and an antiprogramme: one actor engages in a relation with other actors with a particular ambition, as well as with expectations to their actions and specific capabilities. In the beginning, this programme may be fragile and contested by other actors, who also engage in the relation with a programme – or, from the perspective of the first actor, an antiprogramme. The programme is always in the hands of the other actors, who can react with ‘resistance, carelessness or savagery’ (Latour 1991: 105). Hence, the first actor must respond to these anti-programmes and try to convince the other actors of his/hers programme. However, this controversy or contest between programmes can only be overcome through compromises in which the actors modify their original programmes. This is done through the attachment of other actors, who support the programme, and in this manner the chain of associations expands. However, the mobilisation of this programme has a price; it has to change in order to make other actors attach to the programme. Through this process, the identity of the involved actors and their possible forms of interaction and action are also negotiated and changed (Ibid). - 82 - Contested and interrelated networks The relational or non-reductionist (Law 1994) principle of ANT is not only used to explain scientific facts or larger societal structures, but also to understand how organisations are working and to challenge existing ideas about how organisations are managed. For instance, Law has challenged the idea of ‘order’ and calls it a modernist dream of purity; something that signifies ‘the best possible’ and something that is possible to reach (Law 1994). Instead, he suggests ordering is neither a stable nor unified entity: (…) first, the notion of order goes. Perhaps there is ordering, but there is certainly no order. This is because (…) orders are never complete. Instead, they are more or less precarious and partial accomplishments that may be overturned. They are, in short, better seen as verbs rather than nouns. Second, the idea that there is a single order (‘the’ social order) goes. This is the dream, or the nightmare, of modernity. But there never was a root order, so we have to replace this aspiration by a concern with plural and incomplete processes of order. (Ibid 1-2, original emphasis) Furthermore, Law argues that each order has its own ‘mode of ordering’ that affords different responsibilities and roles that may produce tensions when different orderings intersect. In this way, Law’s findings resemble the findings described in the above literature review, where the tension between order induced by quality development on the one hand and the existing clinical work on the other is described as inevitable. These findings are also in line with another of ANT’s analytical principles stating that entities are always contestable and fragile, precisely because their specific properties rely on the heterogeneous network that they are part of. If a new actor is introduced to the network or leaves it, the network will change its form. Or in other words, the black-box will open, and it will be possible to question the reality it depicted (Latour 1987). Berg discusses this apparent complexity and how it affects the ability to achieve a situation in - 83 - which a complex of individual practices “…function as a large formal machine, with formal subtasks efficiently being performed according to pre-set rules” (Berg 1997: 147). Berg uses the notions of distribution, drift and overflow to describe inherent instability in every network. Distribution refers to the relational understanding of agency described above; that actors do not do anything alone, but acquire their characteristics and capabilities in networks with other actors. The functioning of a given actor is not given by the actor in itself but is rather an effect of a chain of actions performed by an ensemble of heterogeneous actors. An actor, for instance a technology such as the electronic patient record, affects the work of other actors, but the functioning of the technology is also affected by the way other actors fill in information, work around the technology, uses the output information etc. Besides being a tool of information-sharing in the hospital, an electronic patient record can also be an important source of research data, and similarly the health professionals are tied to other networks where work should be coordinated. Each actor performs a small subtask in a larger chain of work, and because everyone only possesses partial knowledge of the intricacies of the whole chain of work, no one is in control (Ibid; 142-147). Based on this notion of distributed agency, Berg challenges the idea of stability, because – as he argues – each node in the network is tied up in another network, and hence drift is inevitable: Each actant does more than its specific position in the network calls for: since it is always tied up into other cross-secting networks, its concrete form will always overflow its definition in the particular network under study. (Berg 1997: 148). Actors in a network are not tied with unbreakable bonds and accordingly, each network has a tendency to behave unexpectedly or even fall apart. Consequently, it is not only interesting to study how networks are settled, but also how they are kept stable (Latour 1991, Law 1994, Latour 1997). - 84 - ANT in this analysis The overall aim of this dissertation is to broaden the analytical focus on quality development in a way that pays attention to quality development as more than a struggle between clinical work, professions and technologies of external regulation. In contrast, I wish to study quality work as situated and emerging processes in a context of varying demands of and resources for quality development. In order to do this, I draw on the relational understanding of orderings as achievements based on processes of translation described above. In many ways, quality work resembles laboratory work because of its explorative search for a sense of ‘truth’ about the world (or in this case the truth about quality), and an endless negotiation of what constitutes ‘good quality’ or ‘the level of quality’. Thus, I will analyse the specific quality development projects as processes of translation in which distinct definitions of quality problems and their solutions in terms of proposals of change emerge. From this also follows that I consider quality work in the specific form it takes in the health care organisations, as an achievement and a result of ongoing negotiations and compromises. Translation is a process before it is a result”, (Callon 1986): 224). Accordingly, I use translation as an analytical tool to describe and understand the empirical observations of the ongoing quality development processes with specific emphasis on the way actors are aligned and gain certain properties and capabilities. As I discussed in Chapter 4, quality development processes seem to be organised as temporary assemblages that draws on, intersect and/or engage in provisional alignments with medical work. Hence, I cannot consider quality development processes as the organisation of a network in transformation, but as an effort of coherence-making across different networks. As a consequence of this observation, I will also use the concept of translation to explore how quality work emerges in the intersections with processes of quality development as one kind of order, and clinical work as another kind of order. - 85 - Hence, with this thesis I wish to explore quality development a) as an emerging ordering of work and b) with particular emphasis on the coexistence and interrelation of other kinds of orderings in the hospitals. ARTICULATION WORK As a final element in this analytical framework, I wish to return to the vocabulary of work developed by Strauss and colleagues (Strauss et al. 1997), and specifically the concept of articulation work. Articulation work refers to work performed in order to create coherence among building blocks of the arcs of work and, accordingly, is a concept that provides an understanding of how a plurality of work practices can merge into a totality (Strauss 1985). Suchmann has further defined articulation work as the work that is required ‘[…] in order to bring together discontinuous elements – of organisations, of professional practices, of technologies – into working configurations’ (Suchmann 1996). Hence, articulation work constitutes a distinct type of work concerned with the fact that arcs of work do not automatically arrange themselves into coherent sequences. Strauss calls articulation work a “supra-type of work” and defines it as follows: Articulation work amounts to the following. First, the meshing of the often numerous tasks, clusters of tasks, and segments of the total arc. Second, the meshing of efforts of various unit-workers (individuals, departments etc.). Third, the meshing of actors with their various types of work and implicated tasks (Strauss 1985,8). Taking both the tradition of symbolic interactionism and an interest in the studies of human interaction as a point of departure, Strauss emphasises that cooperation between actors cannot be perceived as given, but rather as a result of an effort. Each and every actor will engage in the interaction based on their own perspectives on the subject matter, and if they wish to cooperate these actors will have to engage in an effort to align their actions with each other. They have to engage in a process of articulation through - 86 - which they can agree upon the purpose of their mutual actions, who does what, when and to what level of quality. In many situations, interactions are based on routines, for instance, as a result of frequent interactions or “standard operational procedures”. However, when new or unexpected things emerge (Strauss et al. 1997), or when new tasks or circumstances appear (Casper 1998, Fujimura 1987), the actors have to adjust or engage in negotiations to make new agreements that realign their actions within this new situation. Articulation work is performed by everyone whose work interrelates with other types of work. An important part of articulation work is communication and, thus, both the reception of information about the work of others and the provision of information about one’s own work to others. Some actors will only need to communicate with the actors closest to them, whereas others – often actors positioned in functions that encompass planning and coordination of wider structures of work – depend on a broader overview (Strauss et al. 1997, 152). In relation to the question of communication as a means of articulation work, the use of various technologies of communication has attracted the attention of many researchers (Schmidt, Bannon 1992, Fitzpatrick, Ellingsen 2013). For instance, in the study of articulation work in an airport, Suchmann has investigated the capabilities provided by technologies of communication to support the articulation of interrelated work that is separated by large distances (Suchmann 1996). Here, Suchmann describes how actors engaged in articulation work receive and translate messages about the work of others via computer screens, monitors and windows and subsequently translate these inputs into new messages that can be sent back and forth between other involved actors. However, the articulation workers’ insights are restrained by the available technologies of communication – they are not able to see everything – and the technologies of communication as providers of support in articulation work are only manifest in the articulation workers’ ability to use and - 87 - adapt them to the situation at hand. Furthermore, Suchmann suggests that the ability to ‘read the scene’ is a necessary trait of articulation workers, and that this requires ‘knowledge about past, present and future events’ combined with ‘timely communication’ (Suchmann 1997). The concept of articulation work is introduced in order to ascertain an empirical sensitivity to who and what becomes part of the negotiations, what their controversies are about, and through which sources and actions of communication they occur. Nevertheless, I do not consider articulation work to be negotiation and communication that is solely centred around direct interactions or technologies of communication. Rather, I pay attention to the quality coordinators’ ability to move information, and hence communication and negotiations, around the hospital organisation as a particular resource in the articulation work related to quality development. CONCLUDING REMARKS In this chapter, I have introduced my analytical framework. This framework was developed in order to study quality development as emerging connections of actors and as intersections of multiple orderings. In order to provide such an analysis, I will draw on Actor Network-Theory combined with the theoretical concepts of work and arcs of work, extending the work of Strauss and colleagues on medical work in a hospital. Thus, the three analytical chapters in this thesis all draw on the same analytical framework developed around different theoretical concepts; arcs of work, translation and articulation work. I will now move on to a description of the methods used to explore quality work in the two hospital departments. - 94 - comes to observe actors who are not aware that they are participating in research. In this study, some of the people in the quality coordinators’ networks got used to my presence, because we met on a daily basis, and they knew the purpose of my study. These were the closest colleagues of the quality coordinators, the department managers and to some extent also the clinical managers in the wards, with whom the quality coordinators continuously had meetings. However, the observation of people who were not aware of the purpose of my presence particularly took place in informal interactions where the quality coordinators met someone (e.g. a nurse, doctor or secretary from the wards), for instance in the hallways of the hospital, and where detailed and sometimes confidential information was exchanged. Then this person would leave, unaware of the reason for my presence. The quality coordinators made an effort to present me to those they interacted with, but sometimes the interactions were so brief that there was no time for further introductions. Even in meetings, where I also attempted to present myself to every participant and make them aware of the fact that I was a researcher and not taking the minutes (an impression that they easily could get, since I was writing notes constantly throughout these meetings). Some people arrived just before the meeting started and after the meeting had begun, and here I found it inappropriate to interrupt the meeting to present myself. Hence, there was almost always someone in those meetings who was not aware of who I was and what I was doing. In the surgical department, I tried to solve this problem by making a small poster to put on the information boards in the department with my picture and a short text describing my project. I am not sure whether it was circulated or noticed by anyone, though. Only on a few occasions did I sense that someone was uncomfortable with my presence. In one situation, I observed how a participant in a meeting sitting next to me covered her notes with her arm and almost turned her back to me, and in another situation I was confronted directly by a person who felt uncomfortable with me potentially becoming - 95 - part of my research reports. In this situation, I took the time to explain my methodological strategy and promised this person that I would not publish anything involving her without conferring with her in advance. ADDITIONAL SITES OF OBSERVATION In addition to observations of the quality coordinators’ work, I was given the possibility to participate in a number of meetings in the hospital’s quality coordinator network and in introduction courses in the two departments. I will briefly describe the observations that I made here, and then I will move on with describing how the formal interviews were conducted. Meetings in the quality coordinator network The quality coordinators from the 10 departments in the hospital were organised in a network with monthly meetings. The meetings were facilitated by the hospital’s quality manager and quality consultants from the department of HR & quality and were an occasion to share information on and experiences with quality development. Accordingly, these meetings addressed various themes of common interest, such as accreditation, information on and discussion of mandatory subjects for audits, audit processes, information on tools for pain measurement etc. I first attended one of these meetings in the summer of 2012, where I was invited by the hospital’s quality manager to present my study to the quality coordinators. This meeting took place just before the first round of accreditation of the hospital, and the forthcoming external survey was a major theme. The next meeting I attended was just after the external survey, in which the hospital had received the best possible score. Accordingly, the visit from the external surveyors was the key subject of this meeting, and in this way the content of these meetings in the quality coordinator network - 96 - followed the structure of the major common activities of quality development in the hospital. For six months, I was on the list of persons invited to these meetings, but I usually participated in the meetings in the periods where I was carrying out observations in one of the departments. These meetings provided me with an insight into the focus of attention in quality development outside the departments. Thus, I became aware of the way efforts of quality development were organised on the hospital level, and how the hospital related to the regional and national structures of quality development. Introduction courses In both departments, I was also invited by the quality coordinators to participate in an introduction course to become acquainted with the department’s wards and clinical functions. In the surgical department, a whole week was planned just for and I spent half a day in each of the department’s wards. Hence, I came to observe pre-examinations, a surgery and the work performed in the outpatient clinic and the bed units. In the medical department, I followed in a guided tour in the department along with new staff in the department. Similar to the field notes made in relation to the observation of the quality coordinators, I wrote notes whenever I had the possibility. I endeavoured to make my notes during the observation, but otherwise right after completion of the observations. These introductory courses were very informative as they provided me with an overview over the department’s wards and the primary clinical functions performed. Besides enabling me to find my way around the hospital on my own, these courses also made me able to ask more precise questions during observations and interviews, because I was able to refer to details I had observed in the wards; for instance specific clinical procedures, the physical surroundings and furnishing, or material and technological artefacts. The - 97 - quality coordinators also, referred to these courses during our conversations (‘As you might notice during the introduction…’), and I was thus able to place their explanations and elaborations in a context. These courses, especially the course in the surgical department, were also a way for me to become acquainted with the different versions of ‘quality’ that I elaborated on in Chapter 4. I was, off course, exposed to these different versions because I presented myself as a researcher interested in quality and quality development. This piece of information almost immediately prompted the actors I observed to show me what procedures they performed in order to document or assess quality, share their thoughts on what constituted or became obstacles to high quality, or even asked me if I considered their work as being of high quality. I was also guided through their overall clinical tasks as they explained to me what they did along the way or after the patients had left the room, in the same way as the quality coordinators did. FIELD NOTES Throughout the periods of shadowing, I wrote notes. However, because I both observed while on the move and in situations where I was a more passive observer (for instance in meetings) my note-taking technique differed depending on the specific situation. I always carried a notebook and a pen with me, so I could take notes in ‘real time’. This was feasible in situations where I had the ability to sit down, for instance when observing a meeting or when I was sitting in the quality coordinators’ offices. In these situations, I was able to capture details in the wording of the conversations, make notes on body language and on what the physical surroundings looked like. However, on some occasions I stopped taking notes because private matters were discussed and then just wrote notes summarising the contents of the conversation afterwards (e.g. nurse enters office, she closes the door, says that she is about to resign, new job). In this way, I made a note of the interaction (with whom and about what) but left out the details (although as I have already noted; no one ever asked me to stop taking notes, so in that sense I was the - 98 - person censuring the field notes). In other situations, we were on the move, and note taking was harder to perform standing or walking. In these situations, it was not possible to make detailed notes in a notebook, so I made brief notes afterwards – for instance when sitting in the quality coordinators’ offices or after I had left them when the day had ended – and supplemented these notes with details from my memory when writing up the field notes. All notes was transferred to the computer immediately after the periods of observation, and here I often supplemented with additional details that I remembered. It was also in this process that I began commenting on or asking questions about the empirical material in a ‘comments and thoughts’ section placed after the transferred notes. These comments and questions varied in content, some of them being reminders to request supplementary material, others being more analytical and even sometimes referring to theoretical concepts. However, their genesis did not follow any particular or structured pattern, and they were just noted as they came to my mind during the process of transfer. Below I will provide a further description of what I did with the empirical material and what guided my choice of empirical material. Before that, however, I will turn to the additional and supplementing empirical material that I collected during the field study. INTERVIEWS As stated in the section on the components of the participant observations, I was not only observed but also engaged in informal conversation with the quality coordinators as well as the actors I met while following the quality coordinators around the hospital. These conversations could be described as unstructured interviews (Neyland 2008: 112, Justesen, Mik-Meyer 2012: 52) that did not follow a pre-planned structure or set of themes. Rather, they occurred along the way, when I became curious about something I observed or when the quality coordinators or others felt inclined to explain what they were doing and I asked additional questions. - 99 - Besides the ‘commented walks’ and ‘go-along interviews’ that took place during the observations, this study also encompassed formal semi-structured interviews (Justesen, Mik-Meyer 2012). The two periods of observation were initiated by a semi-structured interview with the quality coordinators that was structured around the following four key themes: x The quality coordinator’s professional background x Past and current tasks and positions in the department x Important co-workers x Methodological and technological resources. The main purpose of these interviews was to gain an overview over the way quality development was (formally) organised in the departments, and to learn about the quality coordinators’ background and what they regarded as their main tasks and responsibilities in the department. These interviews were performed in the quality coordinators’ offices and lasted approximately 2 hours. As a supplement to these interviews, as well the observations, I was also regularly in email contact with both of the quality coordinators, where I requested information on some of the processes I observed and participated in during the periods of observation. Some of these continued after I left the departments, but became a subject of my analysis and hence I was also continuously in contact with the quality coordinators so as to stay informed on the status these particular processes. Furthermore, I performed semi-structured interviews with head nurses and chief physicians from the departments and the hospital’s quality manager. These interviews regarded the motives behind the chosen organisation of quality work and the perceived role of the quality coordinators in the departments and the hospital, respectively. The - 100 - insights from these interviews are primarily used to inform the context descriptions of this thesis. All interviews followed an interview guide and were transcribed verbatim. I have now provided a description of the methods used and the sources of the empirical material of this thesis. In the following, I will explain how I analysed this material and how the choices of empirical material were made. FROM FRAGMENTED FIELD NOTES TO COHERENT ANALYTICAL STORIES Shadowing has been emphasised for its ability to reveal how empirical events are interconnected and interdependent, even though they are dispersed in time and space (Czarniawska 2007, McDonald, Simpson 2014, Czarniawska 2008, Gill, Barbour & Dean 2014, Czarniawska 2014). I tend to agree with this observation, but the interconnectedness and interdependence did not reveal itself immediately. The terms ‘walking with’ and ‘following’ that I used in the above when describing what I did may give the impression that I observed coherent events that followed one another in a logical progression. However, during the fieldwork I experienced the process more as a criss-crossing and jumping from one spot to another in a pattern I was not always able to understand. When I transferred my field notes from my notebooks to my computer, I typically noted the main events of the current day in the beginning of the document (see excerpt from field notes in Figure 5). These notes are also illustrative of a typical day for the quality coordinator; the many different themes and projects in which they were engaged, the physical places that they went to and the people with whom they interacted. For me as an outsider, the immediate impression was that I was witnessing an incoherent and fragmented mess. - 101 - I think a prominent reason for this feeling of being lost in fragments was that a study planned to involve shadowing – and participant observation in general – is actually not planned at all, at least not from the researcher’s perspective. Rather, it is unpredictable and uncertain, because you have to go along and make do with the empirical course provided for you (van der Waal 2014, Gill, Barbour & Dean 2014). In the case of this fieldwork, where I stepped in and out of the studied field according to the planned strategy, I witnessed many different projects and series of activities that where starting up, ongoing or finalised during my observation period, but I was very rarely present for the whole life cycle of such projects or consecutive activities. Additionally, what I observed was determined by the quality coordinators’ movements within the organisation. Hence, this feeling of only seeing fragments was supplemented by a feeling of witnessing a highly selective empirical body of material. As I have already claimed, the clear advantage of following the quality coordinators was that it enabled me to be present in situations of informal interactions and when mundane activities were performed. Situations that may have been lost from the quality coordinators’ minds’, or situations that would have been considered unimportant in a formal interview situation. I was, mainly in the beginning of the fieldwork, tempted to reframe my empirical strategy and leave my original idea of following the quality coordinators, and instead identify and follow some specific quality development processes with a beginning and an end. Yet with this study I did not aim to study the fate of individual quality development projects or their effects. Rather, I was interested in how quality work was constituted through particular connections of actors in the departments; what kind of work did this require? Obviously, the quality coordinators were a large part of this work, so I resisted the temptation to change the empirical strategy and stayed true to my original plan. Furthermore, if fragmentation was a central premise of the quality coordinators’ work I did not want to erase this from my observations, even though it caused me trouble when trying to find my own sense of order in a messy body of material. - 102 - - 103 - Figure 5: Overview over the quality coordinators’ activities during a week of observation (surgical department) Monday 8:45-11 am: Meeting of the department’s quality consultants – Sofie, Anne, Iben (the department’s quality consultants) and Lene (the quality coordinator) 11-11:30 am: Walked to surgical wards and the outpatient clinic to find out where to place Kaizen boards 11:30 am-12 pm: Lunch 12-12:30 pm: Lene and Iben talked about an assignment that Iben is going to take on as part of a course in project management. 12:30-14 pm: Preliminary meeting about a project, ‘Relational Coordination’. 14-16 pm: Lene’s office. Preparations for the next day’s meetings. Tuesday 9 -10 am: In Lene’s office: emails, paperwork etc. Lene showed me the hospital’s intranet, and we talked about the way quality work is organised in the department. 10-11 am: Weekly meeting with Christina (consultant from dept. of HR & quality) about the project on relational coordination. 11-12 am: In Lene’s office: I have some questions about the project on relational coordination, Lene explains the process. Emails, paperwork etc. 12-12:30 pm: Lunch 13 pm: Meeting with Karen and Anders (ward sister and consultant physician in the surgical ward) and Chistina about the project ‘R elational Coordination’. 13-14 pm: Meeting with Louise and Flemming (the department’s head nurse and chief physician): status on present quality work. 14:30-17 pm: Workshop with former patients on experiences from surgery (Lene was absent due to illness in her family, Sofie (qualit y consultant in surgical department) participated instead). Wednesday 8:30-9 am: Meeting on ‘Relational Coordination’ – Flemming, Louise, Christina, Sofie and Lene. 9-10:30 am: Summarising after the meeting in Lene’s office – Christina, Sofie and Lene. 10:30-11 am: Short meeting of the department’s quality consultants – Sofie, Anne and Lene. 11 am-14 pm: Lene and Sofie prepared a forthcoming workshop on procedures for preexaminations and an information meeting for the staff in the department on the project on relational coordination. Thursday 8-8:30 am: Information meeting in the surgical ward on the project on relational coordination. 8.30-9 pm: Karen’s office – small-talk about the meeting. Evaluation - did it go all right? (Lene, Sofie, Karen, Christina, Louise). 9-11 am: In Louise’s office. Fl emming and Louise are dissatisfied with the way the project on relational coordination is managed (Lene, Flemming and Louise). 11-11:30 am: Meeting about relational coordination (Flemming, Niels, Louise, Christina, Sofie and Lene). Lunch The rest of the day: Lene prepared for meetings and events in the coming week. Several people stopped by to ask or discuss pending issues. Friday 8:30-12 am: Lene’s office: as the day before. - 110 - ‘OPERATION JOINT FUTURE’ During the fieldwork performed in the surgical department, I became interested in a particular quality development project referred to as ‘Operation joint future’. Initially, this process caught my attention because it made a lot of ‘noise’ due to ongoing controversies that threatened the project’s viability. Later on, I came to see it as an example that could provide more generalizable answers to the question of how actors became connected to specific quality development processes. Hence, I engage in an analysis of the initial phase of this project, in which I explore the intricacies of formulating the project’s purposes and the attachment of key actors to the project. I begin this chapter by describing the reactions and controversies related to two versions of the purpose ‘Operation joint future’. Following this, I summarise some of the significant actions taken to involve significant actors in the project and construct a network that is stable enough to exist throughout the lifetime of the project. Here, I discuss how not only direct confrontations between the implicated actors but also the use of constructions of perceived preferences change the course of the project. Finally, I elaborate on the instability of quality work as a result of its interference with power relations and hierarchies in the hospital. ‘OPERATION JOINT FUTURE’ I encountered the project ‘Operation joint future’19 when it was in its initial phase, and hence in a phase where the project’s purpose was about to be defined, concurrently with attempts to engage managers and staff from different parts of the hospital organisation. 19 In Danish: ‘Operation Fælles Fremtid’. This name makes use of the double meaning of the word ‘operation’, referring both to the place – the surgical department – where the project was initiated and the project as a dedicated and coherent set of activities. - 111 - As mentioned above, the project ‘Operation joint future’ took place in the surgical department, more specifically in the surgical ward. The quality coordinator, Lene, described this case as an extreme case, because of the large number of stakeholders that took an interest in the project and challenged the ability to formulate a cohesive purpose that everyone could work towards together: It was a challenge that a very large group of managers were enrolled in the working group of the project – and a lot of them were also in the steering committee, which was already far too big. (…) The result was that, with all these people, it was very difficult to manage both the working group and the steering committee effectively, because everyone had their own agendas. (Quality coordinator) Though somewhat extreme, this example can serve to illustrate how the project developed through the interactions of various actors with different perspectives on how to improve quality, and actors protecting their authority over certain domains of work. ‘Operation joint future’ was initiated in the autumn of 2012 and was part of an overall strategy in the hospital of becoming acquainted with the concept of ‘Relational coordination’ as a method of quality improvement (see Figure 7). The concept of relational coordination was developed by the American researcher Jody Gittell (2009), who has been able to show a causal relationship between relational coordination and medical performance. Thus, the concept has travelled the world as a promising framework for both understanding and improving health care services. This particular project in the surgical department was initiated by the hospital administration, which, as many other Danish health care institutions, had become interested in the findings of - 112 - Gittell. The surgical department’s head nurse had also expressed her interest in the concept and signed up the department’s surgical ward (in the forthcoming OP) as ‘guinea pigs’ in a pilot project. Figure 7: ‘Relational coordination’, Gittell’s definition (Gittel 2009) The precursory history of managers from the surgical department’s interest in relational coordination was linked to at least two current challenges in OP. One reason for setting up the project was a pressing need – formulated by the department’s head nurse and chief physician as well as the quality coordinator – of improving the department’s psychosocial working environment. According to them, the surgical ward had been - 113 - adversely affected by several mergers within the last few years and was suffering from hostility between the old fractions and between nurses and doctors: bullying and unwillingness to cooperate were characteristic of the working environment. This had become even more pertinent after the death of a patient during surgery. This incident was under scrutiny as a severe adverse event and was treated with great confidentiality. However, the implicated doctor’s unwillingness to listen to one of the nurse’s warnings about the patient’s condition during surgery was part of the story told20. This incident added to the history of a disharmonious working environment in the ward, and to the relevance of engaging with the framework of relational coordination. In the forthcoming outline of some of the key events of the project, I will show how not everyone believed in the causal relationship between ‘relational coordination’ and quality. This led to unwillingness to be involved in the project, and this lack of involvement of some actors conditioned the decision to include a goal of effectiveness by increasing the departments’ yearly number of surgeries from 2000 to 3000. This goal had already been set by the hospital directors, but the practical solutions to the implementation of this goal were yet to be found. This goal also met resistance from the departments’ staff, who were sceptical about feasibility of increasing the rate of operations without increasing the staff, but the goal was included to give the project a ‘harder edge’ and supplement the ‘soft’ concept of ‘relational coordination’. However, as we shall see this additional goal not only led to the connection of reticent actors to the project as expected, it also led to new interests and new resistances among other actors. 20 This incident was being analysed in a so-called root cause analysis and was treated with great confidentiality. Hence, the incidence was not recounted to me in detail. I did not ask for further information on the incident either. First of all, because the handling of adverse events was not a key interest of mine, secondly, because the case was described to me as a sensitive matter that had generated a lot of distrust in the department, especially after someone had leaked the incident to a journalist. I reasoned that the more I knew about the case, the greater would be the risk of being a subject of distrust, and I did not require more details than what I was able to learn from the newspapers and what was briefly revealed to me in conversations or during meetings. - 114 - CONNECTING DOCTORS TO THE PROJECT: FROM REJECTION TO CRITICAL PARTICIPATION The validity and importance of the concept of ‘relational coordination’ and the claimed correlation with quality was not embraced and accepted by everyone. This became obvious at a morning meeting in OP, where all the staff was summoned to be introduced to the project. The introduction was given by the project leader; a consultant from the department of HR&Q (Christina), and the management team from OP; the consulting physician (Anders) and the ward sister (Karen). The doctors and nurses of OP were gathered in the lunch room, where Karen and Anders took turns to explain about the project. Karen explained about Jody Gittell’s emphasis on the many relations between actors in modern health care (Figure 8, lefthand section), and she explained how the project would pay attention to these relations and how to improve them. Then she showed the seven dimensions of ‘relational coordination’ ([Figure 8, right-hand section). Figure 8: Figure shown in the ward sister’s slides at the information session Source: http://rcanalytic.com/) - 115 - Anders took over and showed a hand-drawn poster that illustrated a state he called ‘crazy time’ (not illustrated here), a state in-between organizational changes, where the organisation is about to unlearn old habits and learn new. He explained how this slide had been presented to him when he participated in the preparations for the project, and it made him accept the importance of the project: that a merger of five wards into one necessitated a common goal that could help them work in the same direction. He also presented the two purposes of the project: 1: To increase efficiency (reduce the time between operations [called turnover time] and reduce the number of cancelled operations) 2: To improve quality and the working environment (describe all the vital processes during turnover time, and improve the ‘relational coordination’ index [Figure 8, righthand section] in the ward). Karen explained about the methods (e.g. a week of observation performed by Christina, Lene and others, questionnaires etc.). Additionally, she explained how they defined ‘turnover time’ as the time slot between the final suture made in Patient A till the doctor put the scalpel to Patient B. This meant that procedures related to cleaning and preparation of the operational theatre, transportation of patients to and from surgery and anaesthesia was in focus in this project. A nurse commented on this: I think this project is a gift. As it is, we are only paying attention to the time of operation and hence only the doctors’ work. Everything else that is done in this ward is invisible. For instance, we have to refill the cupboards continuously [with sponges, bandages etc.], otherwise it would be impossible to perform an operation. This comment gave rise to a comment from one of the doctors, who said that she could not understand why they, the doctors, needed to be included in the project, when the - 116 - subject of concern was ‘turnover time’. Often she was only summoned when the patients were ready and the operation could begin, and hence she was not even present during turnover time. Karen replied that it was more often the case that they [nurses, patients and the anaesthetic staff] had to waited for the doctors in the operation theatre, because the doctors were taking their rounds in the bed unit in-between operations. As she formulated it: This is actually ‘a relation’ [referring to Gittell’s web of relations], although the doctor isn’t present during turnover time. The doctor said that she accepted this explanation. Christina took over and explained about the questionnaire, which would be circulated among the staff in order to measure the level of ‘relational coordination’. Several of the doctors objected to this as a valid method, saying things like: “How can we possibly formulate questions that will validly resemble such vaguely defined concepts regarding feelings?” and “What, exactly, is the purpose of the project, and how can we measure whether or not the project is a success?” Christina continued to referr to Jody Gittell’s research and her findings of a linear relationship between ‘relational coordination’ and improved surgical performance (Figure 9). - 117 - Figure 9: The relationship between ‘relational coordination’ and surgical performance (Gittell 2009: 31) One doctor, sitting with his arms crossed, was especially eager to understand [or to detect weaknesses in] the project setup, and persistently asked how they could be sure that the measures of effect could be ascribed to the project, and never seemed completely satisfied with the answers he received. In this meeting, at least two types of resistances against the project and the then formulated purposes could be detected. The first type of resistance was related to the perceived relevance of the choice of ‘turnover times’ as a key concern in the project, and was expressed by the doctors. The nurse considered this choice as an opportunity to raise awareness of the nurses’ contribution to a smooth running surgical performance and an otherwise invisible type of work (Star, Strauss 1999). Based on a similar reasoning, one of the doctors rejected this purpose as relevant for them to pursue, because their primary - 118 - task was the performance of surgery and because they were typically absent during ‘turnover time’. Accordingly, the nurses as a group immediately involved themselves in the project, whereas the level of the doctors’ involvement to the project was lower. The doctors’ reluctance was, however, also related to the disbelief in the validity of the findings of a relationship between quality and ‘relational coordination’ in general, but also in scepticism towards the ability to detect any effects of this particular project. This reaction, however, did not come as a surprise, and there were several attempts to accommodate this, for instance by formulating and re-formulating the projects purpose and careful selection among arguments in order to predetermine their imvolvement. In the following section, I will pay specific attention to the way in which Lene, among others, made active use of predictions about the doctors’ interests and motivations in the attemps to involve them in the project. Making the doctors’ indispensable to the project In a meeting between Lene and Christina a few days before the morning meeting in OP, they discussed how they should present the project to the staff in the ward. At the same time, they were looking for a set of slides they had received from various workshops on ‘relational coordination’21. Lene: I don’t know how Anders and Karen have shared the presentation [in the upcoming morning meeting], but Anders has asked Sofie [quality consultant in the surgical department] to make the slides. Christina: I don’t know … should we have more theory. How much more? I think we should present some of Jody Gittell’s slides [shows Lene the slides (Figure 7 and Figure 8)], Maybe we need some more text? 21 Prior to the project in the surgical department both Lene and Christina had participated in workshops held by the hospital, in which the concept of ‘relational coordination’ had been presented. - 119 - Lene: No, then we will lose them [the staff]. We should avoid too much text. For some of them it will be interesting, but for others it won’t. It is a bit too ‘academic’ to show both methods and theory. I think the figures say enough. Later the same day, Sofie had made a set of preliminary slides, which they presented to Karen and Anders. Here, Karen and Anders decided to adjust the slides a little in order to emphasise, very concretely, how the project was relevant for the staff and what the project was about. Additionally, Pernille wanted to include information about the overall structure of the project and how it was part of a larger strategy in the hospital. All in all, this morning meeting was carefully planned so that the perceived interests (Latour 1999a, Latour 1991) of the doctors were considered in order to align them with the overall project. The result of this planning was, as decribed above, that ‘relational coordination’ was presented as a solution to what the consultant doctor, Anders, referred to as ‘crazy times’. In using this image he was referring to the challenges related to the many mergers. Karen, the ward sister, presented her notions of the challenges of being in an organizational reality where the provision of good services was embedded in multiple relations, and together they framed ‘relational coordination’ as a relevant object of development in order to improve the services formulated in the two goals of the project. This was additionally supported by the quality consultant, Christina, who referred to Jody Gittell’s findings of a relationship between ‘relational coordination’ and surgical performance. Though the doctors’ maintained their scepticism by discrediting both the methodology of Jody Gittell (How can one measure feelings?) and the methodology of the project (How can we detect the effects of this project?). In this way, they not only rejected the particular purposes, but the entire underlying network of the project, i.e. research results, arguments presented as graphs etc. Overall, the level of the doctors’ attachment to the - 126 - accordingly that a supplementary purpose was necessary in order to engage the doctors in the project and move on to the next phases of the project. In this particular project, Lene referred to Anders’s first response to a purpose that only regarded ‘relational coordination’; how he had been against it and how she considered this response to be indicative of the response expected from the doctors, who according to Lene generally: “pay attention to the site of operation”. Clearly this conclusion was not only based on Anders’s reaction, but also on similar encounters with the doctors of the department. These encounters painted a certain picture of the doctors’ primary motivations, which enabled Lene to determine the doctors’ motivation in relation to perceptions of relevance. Karen’s change in attitude to the project brought about by the reformulated purpose challenged the stability of the project, but her reaction came as a surprise. The explanation for this reaction was given immediately after Karen had expressed her change in attitude, rather than before, and hence had to be handled as the project progressed. This raises the question of why this reaction was not handled just as carefully as the doctors’ perceived motivations to engage. According to Lene, the main reason for the particular problems of this project was the lack of consideration of, not only the doctors’, but all the participants’ possible reactions to the details of the project. However, there may be an additional explanation, rooted in the actors’ concurrent affiliation with several networks, of which ‘Operation joint future’ was only one. The staff groups of OP and the anaesthetic department were also nodes in the network related to surgery. In the current organisation, OP was managed as part of the surgical department and had staff from the anaesthetic department coming in as ‘visitors’ to perform their particular tasks in the surgery. - 127 - With the invitation of the managers of the anaesthetic department to participate in ‘Operation joint future’, on equal terms with the managers of the surgical department and OP, they were potentially given a more powerful role, as participants in a process were the organisation of OP was to be re-defined. Ironically, the inclusion of turnover times was initially welcomed by both Karen and the other nurses, who clearly saw the project as a way to increase the awareness of their specific contributions to successful operations. However, the inclusion of the anaesthetic department on equal terms with OP was perceived as a threat, and accordingly Karen’s reaction to the changed conditions of the project was unrelated to the project content that she had previously accepted. Instead, her changed motivation to participate was overshadowed by a perceived threat to her position in the hierarchy between OP and the anaesthetic department. To paraphrase Berg (Berg 1997), her attachment to ‘Operation joint future’ changed due to overflow from her simultaneous attachment to and position in the broader hospital organisation. In this way, the reformulated purpose not only changed the doctor’s motivation to participate but also led to Karen’s resistance. The project came to rely on the participation from the staff of the anaesthetic department and fanned the flame under the old issue of authority in the surgical ward. What stands out as a paradox is that the anaesthetic department initially refused to participate, because they felt threatened by the surgical department, and that Karen’s attachment to the project was weakened for the exact same reason, when the anaesthetic department agreed to participate. Whereas the doctors shifted from being adverse to critical participants, Karen shifted from being a fully engaged to a sceptical participant22. CONCLUDING REMARKS 22 The project progressed, and neither Karen nor the anesthetic department left ‘Operation joint future’. However, as Lene explained in the quote at the beginning of this chapter, it was not considered an easy project to manage. - 128 - In this chapter, I followed the initial phase of the project ‘Operation joint future’ in the surgical department, with a special emphasis on the interdependencies of the project’s formulated purposes and the attachment of significant actors with the project. By taking a closer look at the doctors, nurses and managers’ reactions to different versions of these purposes, I identified different reactions to the project related to 1) the actors’ belief/disbelief in the underlying ontologies of the project, 2) the actors’ perception of the relevance/irrelevance for their own key tasks in the department or 3) the actors’ feeling of a pending threat to their present position in the department. The latter reaction was not directly related to the specific purposes of ‘Operation joint future’, but grew out of the project’s interference with existing hierarchies in the broader hospital organisation that rekindled old controversies and threatened the stability of the network related to the project. In this case, quality development interfered directly with an existing, though contested, order, by introducing a new constellation of already related actors competing for the authority to manage the ward. Thus, the development of quality work cannot be perceived as something that emerges in a vacuum. Quality development is closely related to the organisation in which it is embedded, and in the development of the individual quality development projects, these relations have to be taken into consideration. In ‘Operation joint future’, the formulations and re-formulations of purposes were based on the quality coordinators’ (and other implicated actors’) predictions about these actors’ motivation to engage in the project. These predictions were based on systematisations of earlier encounters with these key actors, but it was not possible to foresee every reaction. As such, the project’s progression was dependent on the project managers’ ability to deal with many different actors and their various (and changing) reasons to engage/dis-engage in the project. I will return to this particular aspect of the - 129 - management of quality work in chapter 10. In the forthcoming chapter, I will turn the significance of quality data as a component of quality development. - 130 - QUALITY DATA: CONSTRUCTIONS AND THEIR USE In the previous chapter, I explored the simultaneous construction of the purpose of quality development and attachment/detachment of significant actors and groups of actors to/from the project ‘Operation joint future’. Additionally, I showed how, besides connecting and disconnecting actors, this purpose also interfered with existing controversies and changing motivations that framed what could be dealt with as an object of quality improvement. In this chapter, I will also explore how connections between actors emerged and enabled certain quality problems and quality agendas to emerge and stabilise. This time, however, I will pay particular attention to the methods and technologies of quality assessment and the practices related to the construction of quality data and the following utilisation of these, either as initiators of local reflection on quality problems or objects of external legitimisation. Earlier in this thesis (in Chapter 4), I referred to the quality coordinators’ position in the department as a position without any formal authority and the challenges that this led to in the field of quality development, which inherently concern change and control. Through the focus on the practices involved in the construction and utilisation of quality data, however, it is also possible to show how this authority is enabled (though still challenged) through the quality coordinators’ maintenance and management of quality data, as a repository of knowledge that needs to be maintained and managed. DATA AS ACHIEVEMENTS Before I turn to the analysis, I will briefly reflect upon ‘data’ as an object of research. These reflections were fuelled by a somewhat surprising conclusion made by one of the quality coordinators during the fieldwork. I participed in a meeting between the quality - 131 - coordinator in the medical department, Hanne, and a ward sister, and the ward sister in the acute clinic expressed concern over excess waiting times23. This concern caused Hanne to make the following suggestion: Should we spam the system with adverse events in relation to prolonged waiting time on the test results? I mean, it is a bit odd that it is the culture among the doctors and the waiting time for test results – things that you can’t do anything about – that influence the waiting times in the ward The suggestions were rejected by the ward sister, who explained that they were already in negotiations about how to solve the problem. She was not prepared to take things that far yet. However, it was clear from this situation that the provision of reports on adverse events served as a potential method to make a strong argument for a perceived problem. On our way back to Hanne’s office, I asked her if they used reports on adverse events as a deliberate method to raise awareness about quality problems. This was confirmed by Hanne, and she told how a while back they had used this method deliberately to raise awareness of a shortage of pressure relief mattresses24. The Facility service had bought too few mattresses, but after several reports on the matter they had agreed to lease some more, and now the problem had been solved. According to Hanne, the same applied for a situation where the rubbish had not been collected frequently enough: The rubbish sacks piled up in one of our wards, and they split open. The nurses walked around in rubbish, and dragged it around everywhere. By reporting it as an adverse event every time we experienced this problem, we succeeded in attracting attention to the problem, and now the rubbish is collected more frequently. Described in this way, adverse events were given a rather prominent and unquestionable role in the attempts to involve actors, in this case the facility service, in quality 23 This empirical example is treated in more detail in the next chapter, ‘A care pathway for medical infections’. 24 Used to avoid pressure ulcers among patients that are bed bound. - 132 - development initiatives. The reason why this was a surprising statement was that I had observed in several other empirical situations how quality data was more a subject of negotiation than a firm statement of a quality problem that everyone immediately aligned with. Theoretically, this is also somewhat surprising because, as Latour reminds us, data are not something that ‘are’ but something that ‘become’; an achievement, in relying on actions and interactions: “One should never speak of ‘data’ – what is given – but rather of sublata, that is, of ‘achievements’” (Latour 1999b: 42). Accordingly, he and other authors have shown how data are a result of intense labour, where the studied objects are rendered constant and comparable through successive stages of sampling, collection and structuring of objects that transform them into words or texts, and hence a claim of truth; a scientific fact (Latour 1999b, Latour 1987, Latour, Bastide 1986). In this process, locality, particularity, materiality, multiplicity and continuity are lost. It constitutes a reduction of the richness and details of the studied object, but also leads to the advantage of it gaining a format (for instance a text: a scientific paper, a report, a set of numbers etc.) that allows for comparability, standardisation, text, calculation, circulation and claims of relative universality (Latour 1999b:70-71). As a continuation of Latour’s points, Power (1997) argues in his work on the ‘audit explosion’ that notions of evidence also require acceptance from what he calls a community of observers: “ ‘[A]uditability’ (…), is often constructed in the interaction between auditor, auditee and official knowledge in an active process of rendering auditable” (Power 1997Ibid: 70). From this point of view, it is not possible to make claims a priori as to whether a ‘valuation technique’ is reliable. If audits are considered effective and are widely trusted - 133 - in the network in which they are embedded, it is an effect of consensus and support of both the technique, the user of that technique (the auditor/the valuator) and the knowledge that is produced (Ibid: 78-82). Thus, it is not only the construction of data but also the construction of a consensus about techniques, routines and expertise that solidifies the usability of, for instance, audits as a means of evaluation. I am not emphasising the above empirical example in order to shed doubt on Hanne’s conclusions. I have no reason to believe that the problem was not solved through the ‘spamming’ with adverse events. Instead, I consider it an occasion to inquire into the translation processes that allow quality data to be used as strong arguments used to align actors around quality development processes. Consequently, in the following I will focus on the technical and situated practices related to documentation and quality assessment and the processes of translation by which clinical and administrative practices are inscribed into mobile entities, which can be displaced in time and space, saved and indexed in order to accommodate different purposes and agendas. The first part of this chapter will deal with how images of the department’s level of quality emerge from practices of documentation. The second part of the chapter explicates how quality data as mobile entities can be used to initiate reflection and negotiations on quality problems, and accordingly quality development processes. Finally, in the third part of the chapter I will explore how the quality data function as a repository that allows for the emergence of different agendas in quality development, and how this authorises the quality coordinators’ position in the departments. TRANSFORMING CLINICAL WORK INTO QUALITY DATA Documentation is an inevitable and large component of quality work. On the one hand, documentation is related to a quest for transparency of the delivered services and can be - 134 - positioned as part of an agenda of rationalisation that emphasises accountability and performance measurement as important levers of evaluation and regulation (Wiener 2000, Power 1997). In health care quality development, this is most obvious in relation to the accreditation programmes, which in some parts of the world have developed as part of a funding and regulation agenda (Wiener 2000). As such, documentation is initiated to provide regulators and surveyors with an insight into the provided services that enable them to make regulative decisions or allocate resources. On the other hand, the quest for documentation is also described as a means to internal quality control and development. For this purpose, documentation is described as the basis of so-called Continuous Quality Improvement (CQI), where the health care institutions themselves identify problems and solutions, and develop indicators to assess their impact (Friis 2014, Wiener 2000) using the principles of the PDSA cycle. Viewed in this way, the purpose is not the control from actors outside the hospital, but processes of adjustments in the delivered services, which are initiated and executed by the individual health care institutions. This is also the perspective put forward in the method section to DDKM (Institut for Kvalitet og Akkreditering i Sundhedsvæsenet 2013). In contrast, DDKM and accreditation were introduced into Danish health care as part of a professional and political debate about the development of a national model of quality development, rather than being part of a reorganisation of financial or regulatory systems25 (Knudsen, Christiansen & Hansen 2008, Knudsen, Fuglholm & Kjærgaard 2004). Other types of documentation practices, such as the reporting of adverse events in the Danish patient safety system, have been introduced in a similar manner under the 25 Although there were no formal sanctions coupled to DDKM, there have been examples of managers responsible for quality being made redundant as a consequence of less favourable accreditation results. Also, a general belief was that bad performance could tilt the balance and lead to closure of departments or hospitals in relation to regional or national re-organisation of the hospital sector (Interviews and conversations during fieldwork). - 135 - banners of self-surveillance and organizational learning (Pedersen, Mogensen 2003, Zinck Pedersen 2013). Regardless of the underlying purpose, these quests for documentation raise the question of what methods, technologies and practices are needed for the required information to be provided. Documentation is not a new practice in health care, where for instance the use and development of patient records constitutes a resource for coordination among health care staff engaged in the same care trajectories, based on continuous documentation (Vikkelsø 2005, see for instance: Berg 1999, Berg 1996). Obviously, in the existing technology, only parts of the original state-of-affairs are retained (the clinical work and organisation), as the primary purpose these technologies is coordination in relation to the treatment and care of the individual patient. Thus, the quest for documentation as a component of quality development leads to a requirement of new types of information, new technologies and new methods tailored to conceal and provide information that can be used specifically for quality development purposes. These are treated below. Procedures of documentation: concealing clinical practices In an interview with the quality coordinator from the surgical department, she showed me a list entitled ‘sources of data’ (see Figure 10). It was a long list, and as I learned along the way, many hours of work were spent on maintaining each of the data sources to make them available as inputs to the negotiations on quality. The original list was not only long, it also encompassed both technologies and methods to collect and contain information on different practices in the departments, and technologies and methods to transform this information into quality data. Thus, in order to gain an overview over the presented data sources I have - 142 - contact person27 , and Lilian checked the front page of the records to see if this information was documented. In all five records, the contact person was indicated by the insertion of a small card, with the initials of the responsible physician written on it, placed in a small folder on the front page. However, Hanne consistently marked that question with an ‘N’ (for No). Lilian protested, referring to that information already being present on the card. “Well, that’s not good enough”, Hanne explained, “it has to be registered inside the patient record, otherwise we cannot be sure that it will be referred to in the electronic record”. After answering the questions for all the patient records, Hanne took the now completed questionnaire back to the office and entered the answers into SurveyXact (an electronic questionnaire system), from where she was able to calculate the percentage of the degree of compliance with the quality standards and generate a graphic illustration comparing this audit with previous audits. However, as early as during the audit she concluded that the front page of the patient record had to be changed, to make it more clear how information on the contact persons should be documented. In an email to me some months later, Hanne explained how they had made slight changes to the audit procedure: In the latest two audits, we have entered the answers to the audit questions directly into SurveyXact. The questions look like those I have attached (Figure 11). We have 2-3 computers going, so that we are able to see the patient’s electronic record (the medication module) and enter the answers straight away. We are carrying out an audit of 20 patient records for each of our management areas. We are aiming for continuity among the persons involved in the audits, i.e. persons who have a managerial or clinical 27 The provision of a contact person among the departments’ health care staff features as a formulated standard in DDKM. The contact person(s) should be responsible for assuring coherence in the treatment and care of the patient, and for informing patients and relatives about the health care service provided. (Institut for Kvalitet og Akkreditering i Sundhedsvæsenet 2013, 2.3.2) - 143 - responsibility and are able to understand the questions and respond to errors in practice. SurveyXact generates a report straight away. But they also learn a lot by looking in the patient records, and the result is always that they talk to persons or groups immediately after the audit. Both praise and criticism is given. One thing is the large amount of time spent on continuous documentation; another thing is what purpose this documentation were said to serve. Hanne refers to it as a something that those responsible for this particular practice could act upon. In a similar way, the quality coordinator from the surgical department explained why she considered the list of data sources as a crucial tool for quality development: This is the one [the list of data sources, figure 10] that I find really good, because it forms the basis of our quality work. These are the data available to us. There are various audits based on patient records, there are the results from the national survey of patient experiences that we receive every spring, and the workshops with former patients that may serve as a background for asking: “Should we do something?”. The data sources are referred to as ‘the basis’ of quality work, because they provide an overview over the ‘available data’ that can be used as indications of a quality problem. The majority of data sources were related directly to specific quality standards. Thus, quality could be shown as a percentage of compliance with these quality standards and graphical illustrations of compliance over time. On several occasions, I observed situations where the quality coordinator presented the results of, for instance, an audit to clinical managers or some of the clinical staff, because she could see a drop in compliance with a specific standard compared to the previous audit and asked them to explain or at least reflect upon what the reasons could be. The continuous documentation became representations of ‘quality’ and hence indications of areas in which quality development was needed, and their presence became an occasion to ask “should we do something”, as phrased by the quality coordinator in the above quote. - 144 - If we return to the audit event in the medical department, the purpose of noting the patients’ contact persons is clearly related to the quality standard, and it is difficult to consider the registration of the patient’s contact person as an activity detached from quality development. However, the documentation of contact persons cannot be considered as quality data until it has been mediated, and hence translated, by the audit procedure. When the information about the contact person appears as notes in the patient record, it can potentially be used for several purposes, for instance coordination in relation to individual patient trajectories in the clinical wards (Berg 1999). However, to become quality data, – and be used as a source of insight into quality development for the quality coordinators or managers – it had to be transformed from a particular and situated event into cumulative patterns of practices. Hence, the documentation of contact persons became quality data through the mediation of the audit questionnaire, and this is the point at which it can be used to pose the question: “should we do something”. In the audit situation described above, the audit result immediately led to the decision to try to improve the documentation practices by the altering the front page of the patient records. Similarly, in Hanne’s clarification on the new audit procedure in the medical department we see how the audits provided clinical managers with a particular insight into clinical work. An insight that allowed them to assess whether the quality level was satisfying or not and confront clinical staff with this assessment. Accordingly, if documentation is the concealment of clinical practices, quality data are the concealment of the assessments of these same practices, which can be used to engage actors in reflection upon quality problems and solutions. In the following, I will continue along the path of quality data construction and look at how they are used to initiate reflections about quality and to persuade managers and staff to engage in and align with quality development processes. - 145 - “SHOULD WE DO SOMETHING”: MOVING QUALITY DATA TO ARENAS OF REFLECTION Earlier in this chapter, I implied that the construction of quality data through transformations of the local practices in the departments into documentation and quality was highly important, because it allowed for both spatial and temporal displacement as well as storage. Regarding the ability of displacement, I will return to yet another of Latour’s conceptualisations. A predominant theme in Latour’s texts is related to the way in which science is linked to society. Latour’s overall point is that science cannot be disconnected from the rest of the society, and that we cannot fall back on explanations relating to ‘the nature of things’ when making studies inside the laboratory and sociological explanations when making studies outside. If there is a connection between science and society it is because someone has put an effort into its establishment (Latour 1999b, Latour 2005, Latour 1987). Hence, Latour introduces the concept of ‘mobilisation’ as a way to account for the efforts made to make ‘the world’ mobile: ‘it is a matter of moving towards the world, making it mobile, bringing it to the site of controversy, keeping it engaged, and making it available for arguments’ (Latour 1999b: 101). The primary aim of making the world mobile is to make alliances; to make other people interested. Accordingly, the (scientific) ‘facts’ must be able to leave the laboratory or the original site from where they were conceived. This requires an effort, however. Following the same argumentation, quality data did not result in a process of quality development by themselves; they had to be actively taken out of the offices of the quality coordinators to become part of an argumentation on the need for quality development. These efforts will be covered in the following. Displacing quality data and making them available for argumentation - 146 - One afternoon Hanne was taking her round in the department wards. Hanne had promised to stop by the cardiac outpatient clinic, where a ward sister needed advice, and in the geriatric ward Hanne wanted to find the ward sister because she needed information on the status of a guideline they were working on. Hanne generally used appointments such as these as an occasion to walk through the rest of the wards as well. On this particular occasion, Hanne looked into the nurses’ rooms and the ward sisters’ offices in the various departments, small-talked a bit and then proceeded to the next ward. In the corridor in one of the wards, she ran into a secretary. Hanne stopped and initiated a conversation about the rehabilitation plans that the doctors made prior to the discharge of patients: Do you know anything about the rehabilitation plans? I have received a report from the municipalities on how we are performing, and it says that we are not quite managing to deliver our plans at the day of discharge, as we are required to. The rehabilitation plans served as a tool of communication between the hospital and the municipalities about the patients’ discharge from hospital to their homes. The rehabilitation plans described the hospital doctors’ recommendations as to the future care of the patients, and was an indication of what services the municipalities should offer the patients in order to continue their rehabilitation at home. The hospitals (the regions) and the municipalities worked according to standards for the cooperation and distribution of tasks between the institutions within these two politically and administratively separate organisations. Here, a standard stated that the rehabilitation plans should be sent on the same day as the discharge, so that the municipalities would be informed simultaneously with the patients’ transfer to their homes. The secretary answered that she was responsible for sending the rehabilitation plans as soon as the doctors had finished them: - 147 - Secretary: Normally, I send them before 2.30 pm. The rehabilitation plans that are completed after office hours I send the following morning. Hanne: But it is only 66% of the plans that we have sent on the same day. Secretary: But how can we do things differently. My impression was that we were doing just fine. Hanne: You must not take this as personal criticism, but I think I have to present this problem at your [the department secretaries’] next meeting. It sounds like we’ll have to ask the doctors to finish the rehabilitation plans earlier, so that you are able to send them before you leave. And of course another possibility is that we try to change the phrasing of the standard, so instead of ‘Plans are sent the same day’ it could read ‘Plans are sent the same day or the next’. In this situation, we see how quality data represent a shortcoming in a particular part of the clinical work, which is described by a quality standard and documented through the registrations of timely deliveries of rehabilitation plans. The quality coordinator brings this information with her – not in a physical form, but as a percentage based on the municipalities’ records on the matter – and presents it to the secretary responsible. In the resulting meeting, this specific piece of information is made available for reflection and argumentation. Specifically, the secretary argues that if there is a problem this must be due to other implicated actors, and based on this Hanne reasons that it is either the practices of the doctors or the standard formulation that need to be changed. Hence, in this case the quality coordinator does not leave with a problem solved, but with identifications of new sites and actors (the secretaries meetings, the doctors and the standard) to which the quality coordinators can bring her information on a potential quality problem. This situation resembles the audit situation described earlier in this chapter, where the patient’s contact persons were incorrectly documented in the patient records. Through - 148 - the doctors’ protests against Hanne’s conclusion of flawed practices of documentation and Hanne’s insistence on the same, an inappropriate structuring of the pre-printed fields to be filled out in the patient records was revealed. Due to this insight, Hanne was immediately able to present the problem of documentation to those responsible for the patient record format through a displacement of the audit results. So, are these acts of displacement merely a question of moving information around? The answer is obviously no, because these conversations initiated by displacements of percentages or other types of quality data also caused shifts in attention, or in other words a translation of a percentage that indicated a problem into a more detailed understanding of where the problem was rooted. The complexities of the practice that the quality data represented were re-introduced in a process were the situated details of the practice were unfolded by the secretary in the case of the rehabilitation plans and by the doctor in the case of the contact persons. These details did not necessarily reveal concrete solutions, but they appointed new actors as responsible, and hence relevant to engage in further reflections about the solution to the specific problem that had been illuminated through the quality data. Progress was ensured, but the question, “should we do something” was still open for reflection. Negotiating the validity of data In the situation described above, the secretary’s main argument was related to disbelief in the presented percentage. This disbelief was translated into an explanation emphasising the doctors and the responsibility for the quality standards, and hence a provisional stabilisation of the data validity. However, the dubious validity of quality data was a common objection put forward by clinical managers or staff when confronted with a potential quality problem. One type of response was related to lack of documentation: “we may not have documented it, but I am sure we are doing it anyway”. On - 149 - other occasions, responses were related to a disbelief in the methods or technologies of documentation, and hence to the ‘quality’ of the quality data. Accordingly, when the question of validity remained open then the question of whether the quality data resembled a ‘real’ quality problem would also remain open. Hence, the question of validity was not left unresolved, but was addressed actively by both quality coordinators and , as we shall see in the following section, the staff members themselves. Lene, the quality coordinator in the surgical department, was called to a meeting with a nurse, Kirsten, from one of the department’s bed units and a consultant, Pernille, from the department of HR & quality. Kirsten, who was responsible for the implementation of the PDAs28, had discovered some irregularities in the printouts from the system that registered the nurses’ use of the PDAs. The nurses were supposed to use the PDAs as a safety precaution, to make sure that the right medication was given to the right patient. When the nurses, in the medication room, administered the medication to a given patient, they were to print a label with a barcode on that matched a barcode on the patient’s wristband. When handing out the medication they were to use the PDAs to scan the barcodes on both the label and the wristband to ensure that medication was handed out to the right patient. Both prints of labels and scans of wristbands were registered in an electronic system and could be used as indications of medication administration and medication hand-out, respectively. According to the data available to Kirsten, there were more registrations of medical administrations than medication handouts, especially in weekends. This indicated that the nurses did not use the PDAs properly, and hence that they acted against the standards of safe medication. Kirsten expressed surprise in the findings, because she had been under the impression that the nurses used the PDAs: 28 Personal Data Assistant – handheld PC. - 150 - I am actually surprised by these numbers, because when I look into the medication room the chargers where the PDAs are placed when there are not in use are usually empty. It is also my impression that the patients remind the nurses to scan their wristbands when they hand out medication and that this has helped them to remember to use them. Kirsten only expected the nurses to be reluctant to use the PDAs in relation to PN medication (medication dosed ‘as required’ in contrast to fixed doses). Otherwise, she suspected the registrations to be invalid, not least because she had discovered some registrations made in weekends when the bed unit was closed29. Kirsten and Lene agreed that they had to find out whether the data were invalid or whether the nurses should be reminded once again to use the PDAs. Pernille explained that they could look at the time lapsed from the time of administration to the time of hand-out if they wanted to know whether the PDAs were used at the patient’s bedside or in the medication room. If the timeframe was a matter of seconds, then the PDAs had evidently not left the medication room and the nurses would probably not have used it to ensure the correspondence between medication and patient. If the timeframe was longer, this was not clear proof of correct use, though it would indicate that the PDAs could have been used outside the medication room. Furthermore, Pernille asked if the weekend registrations they had discovered could have anything to do with the referrals of patients to other wards: If the patients are referred to another ward in weekends – and the medication is handed out from there – could it be the case that this medication was still registered in the ward that the patient came from? Kirsten and Lene found that explanation implausible. The patients still required a great deal of medication in the days following discharge, and hence they were sent home with 29 In the surgical department, the average duration of hospitalisation was no longer than a few days. Hence, one of the two bed units was closed in weekends, and those patients who could not be discharged were referred to the other department. - 151 - medicine that would cover the first days. So, the nurses printed labels covering several days in one go. Generally, it was Kirsten’s impression that the nurses were very good at printing several labels at the time, for instance when patients were discharged from the hospital. Provided with this information, Pernille became able to provide an alternative explanation to the mismatch between registrations of administration and hand-out of medication. There is a fault in the system that only allows one registration of medication hand-outs even though it covers the administration of several doses of medication. I think this is why there is a decrease in the registration of hand-outs in weekends or other days where a large number of patients are discharged. I recommend that you look at these registrations per month instead of per week, because this would even out the fluctuations that you see over a week. Because of these failures in the registration system, you should not aim for 100% compliance, 60% would be good enough. Try to keep an eye on the ‘discharge days’ and whether it is on these days you find these fluctuations. Otherwise, try to focus on the areas where you suspect that the PDAs are not being used, for instance in relation to PN medication. When Pernille left the meeting, Lene and Kirsten continued discussing how to interpret and explain the PDA registrations. Despite the fact that they ranked higher in the use of PDA than other departments, they were not satisfied. Kirsten said that she did not trust the data and suggested that they went through the registrations and looked at the time lapse between administration and hand-out of medication, as Pernille had suggested. They wanted to create a faithful representation of how the nurses used the PDA, and additionally came up with the idea that they should urge the nurses to make a manual registration when they made an administration of medication covering several days. Lene would make a large poster with the headline “Administration of several doses”, and hang it up in the medication room, so that the nurses could leave a mark every time they administered and handed out several doses of medication to the patients. In that way,