Quality improvement in healthcare organisations requires structural reorganisation and systems reform, and the development of an appropriate organisational “culture”.1 Although there is variability in how organisational culture is conceptualised,2 it is generally understood to be a shared set of implicit and explicit beliefs, attitudes, values and norms of behaviour that allows the members of the organisation to communicate and work effectively together.3 Beliefs and attitudes that are thought by experts to be conducive to quality improvement in hospitals include:

• understanding of healthcare as a complex system;

• recognising the importance of coordination of healthcare processes;

• a positive attitude towards medical error;

• adhering to the concept of continuous improvement;

• a central preoccupation with the patient’s welfare.

Several studies have described organisational cultures, and explored associations between different cultural “types” or “attributes” and quality-related outcomes.4^–9 For example, Marshall and colleagues10 found that “cultural traits” such as the value placed on a commitment to public accountability, a willingness to work together and learn from each other, and the ability to be self-critical and learn from mistakes were associated with greater implementation of clinical governance in general practice. However, it is still unclear which set of shared beliefs and values is the most effective in fostering quality improvement.11 This is in part because many studies to date have been methodologically weak, and also because study instruments tend to use predetermined, researcher-defined attributes of interest. Few studies have explored inductively the underlying beliefs and values that guide attitudes and behaviours. Such issues are better addressed using qualitative research methods, which are particularly useful for identifying issues that are most meaningful to the target group and the language used by the target group to talk about these issues.12

The objective of our study was to explore the ideas held by hospital-based doctors and nurses about quality in healthcare, in particular their perceptions about the main barriers to achieving quality healthcare, factors that facilitate the delivery of quality healthcare, and notions of responsibility for ensuring quality healthcare. We were particularly interested in whether the beliefs and attitudes of practitioners (we use the term to refer to both doctors and nurses) corresponded to notions promoted by expert-based quality improvement models.

METHOD

Study setting

We conducted this study at University Hospitals of Geneva, a public hospital group in Geneva, Switzerland. This hospital system consists of an acute care hospital, a psychiatry hospital, a geriatric hospital, a rehabilitation facility, and a long-term care facility. It has about 2200 beds, and employs approximately 1200 doctors and 3600 nurses across 12 departments. Since 2003, the hospital’s quality and safety commission has been working to implement a participatory quality improvement process following the European Foundation for Quality Management model.13 An important element of this approach is the development of a “culture of quality” through training and quality improvement projects conducted by the clinical services.

Study design

The purpose of our study was to explore what doctors and nurses think about quality. Because our aim was to identify the range of ideas rather than measure the frequency or distribution of those ideas, we chose a qualitative methodology.

We used a non-probability, “maximum variation” sampling strategy. Such samples aim to theoretically represent the study population by maximising the scope or range of variation in the subject of study.14 We felt that the main sources of potential variation in practitioners’ ideas about quality (and possible sources of quality subcultures) were: their clinical specialty and their profession and relative position in the hospital hierarchy (frontline patient care vs supervisory responsibility). Therefore, we selected respondents to represent the major variations in these dimensions. We limited our sample to doctors and nurses because quality management activities are primarily aimed at these practitioners. Table 1 shows the sampling framework that guided our selection of respondents.

To recruit study participants, we first prepared a list of potential interviewees with the characteristics of interest in each specialty (profession, hierarchical position, sex, professional experience), and then contacted them by telephone to ask whether they would be willing to participate in the study. In general, the first person contacted who agreed to be interviewed was included in the study. Because the purpose of our interviews was to explore in depth the ways in which healthcare workers think about quality, we purposefully selected individuals who were willing and interested in participating in the study.15 We planned to interview an initial set of 20 informants, analyse the results, and continue interviews until we reached theoretical saturation—that is, until no new concepts or themes emerged from the analysis.16 However, redundancy of concepts and themes occurred early on in the analysis process, after about 10–12 interviews had been conducted and analysed. Therefore we did not conduct more interviews after the initial round.

Data collection

The interviews explored the practitioners’ definitions of quality healthcare, perceived barriers and facilitating factors with regard to providing quality healthcare, and ideas regarding responsibility for ensuring quality healthcare. An interview guide was developed (box 1), consisting of open-ended questions followed by prompts. Respondents were encouraged to narrate their personal experiences and to talk about issues or topics most relevant to them.

Interviews were conducted in French by VK, AC and PH. To ensure comparability across interviews and interviewees, training was conducted by PH before the start of data collection and focused on the general order and formulation of interview questions and the types of probes used. The interviews lasted 30–45 minutes.

Analysis

A thematic analysis of the verbatim transcripts was conducted by an anthropologist (PH), a psychologist (AC), a nurse (VK) and a doctor (PC). The analysis was done in several stages.17 The first stage involved reading all transcripts and marking any passages that seemed of particular importance or relevance. The team met weekly to discuss the transcripts and to compare and contrast impressions. Slight differences in opinion occasionally arose regarding the meaning of transcript passages, and usually were a reflection of the differences in the knowledge of the analysts about an interviewee’s profession or work context. Consensus was reached through general discussion of clinical and administrative tasks and organisational processes, in order to put interviewees’ comments in context. On the basis of these discussions, the anthropologist wrote summaries for each interview, and developed an initial list of codes from the key phrases and themes that were identified by the team.

In the second stage, team members individually hand-coded the transcripts, and then met to discuss and resolve any discrepancies. In the third stage, the anthropologist coded the transcripts with agreed codes using the qualitative data analysis software MAXqda (VERBI Software Consultants, Sozialforschung, Germany). Coded segments were then read code-by-code to identify sub-themes. The list of codes evolved to reflect emerging concepts and categories, and the transcripts were re-coded several times. Data display tables (key codes by respondent) were then created to facilitate the examination of similarities and differences across respondent subgroups and to identify common themes. In particular, we compared themes across specialities, and between doctors and nurses. A diagram was then developed to illustrate the relationship between key themes in the data.

We attempted to maximise the validity of our findings by “triangulating” different professional perspectives during the analysis process.18 We also explored the credibility of our interpretations and conclusions by conducting a “peer debriefing” with an expanded group of seven hospital colleagues interested in quality issues.

RESULTS

We interviewed 21 doctors and nurses from five clinical specialties. All the practitioners who were approached initially agreed to participate in the study, but because of difficulty in setting up an appointment with the surgeons who were contacted, the first two available surgeons were interviewed. Two psychiatrists were interviewed because of change in availability of a respondent. Four of the 11 doctors were women, and seven of the 10 nurses were women. The number of years at the current hospital ranged from 2 to 24 years, with an average of 10.5 years; only 4 respondents had worked <5 years at the current hospital (all between 2 and 4 years).

There was a—perhaps surprising—degree of similarity in the respondents’ ideas about quality, across specialties, between doctors and nurses, and within professional hierarchies. We present the key themes that emerged in each of the four areas explored in this study, and discuss any relevant variations. Illustrative quotes are presented in boxes 2–5 (N: nurse; D: doctor), and fig 1 provides a diagrammatic summary of results. All quotes were translated from French by PH.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1 Diagrammatic representation of study results.

What is quality healthcare?

When asked to define good quality healthcare, both doctors and nurses placed emphasis on what they referred to as the “human” aspects of care (box 2). Good patient–provider communication was considered a key aspect of quality medical care because it allows practitioners to develop a partnership with patients and to identify and respond to patients’ expectations. Only about half of the doctors and nurses spontaneously mentioned the importance of more technical aspects of care (eg, timely and effective diagnosis and treatment). These were generally presented as a necessary component of quality care but insufficient on their own. Quality was seen as the intersection between the technical and human aspects of care.

Both doctors and nurses felt that quality assessment had an important subjective component. Evaluation criteria depended on whose perspective was being considered (the patient’s or the practitioner’s), and what aspects of care were considered most important. Because patients and practitioners can have different ideas and expectations, both subjective and objective measures are needed to assess the quality of care. Ideally, quality healthcare should result in the satisfaction of both the patient and the practitioner.

What factors contribute to quality healthcare?

When asked about what contributes to quality healthcare, both doctors and nurses focused on the personal and professional qualities of practitioners (box 3). Technical competency, achieved through training and supervision, coupled with personal motivation and goodwill, were considered to be fundamental to the provision of quality healthcare.

Practitioners’ ability to effectively communicate and collaborate with other health professionals was also considered key to providing good-quality care. Interestingly, the nurses, but not the doctors, talked about the importance of consensus in a healthcare team. The nurses felt that it was easier to provide good-quality care to patients when nurses and doctors agreed on what should be done and how. Examples of both optimal and suboptimal quality reflected the importance of sharing patient information, discussing any difficulties encountered, and showing respect for different professional roles and competencies. Clinical guidelines were rarely mentioned, and material and technical resources were mentioned only in the sense that they were a necessary but insufficient element of quality care.

What are the obstacles to quality healthcare?

Whereas quality was thought to depend primarily on individual practitioners, the major obstacles to ensuring quality care were seen to lie at the systems level (box 4). There was a sense of frustration among both doctors and nurses that they could not provide the quality of care they felt they were capable of and motivated to provide due to institution-level decisions that affected their work conditions. A key theme in the interviews with both doctors and nurses was the impact of inadequate staffing and increasing administrative tasks on practitioners’ ability to care for patients, and in particular their availability and motivation to ensure the more interpersonal aspects of care.

How can quality of care be ensured?

Practitioners’ suggestions for ensuring quality of care were directly related to their understanding of quality as something that depends primarily on individual practitioners and yet is constrained and threatened by institutional factors outside their control. Both doctors and nurses in our study emphasised two main strategies for ensuring quality (box 5). Probably the most salient issue for both doctors and nurses was that of institution-level support for practitioners in the form of increased staffing and reduction in administrative tasks so that practitioners could adequately focus on their activities related to patient care. In addition, several practitioners (two doctors and five nurses) talked about the importance of trust and non-interference on the part of their hierarchies. Finally, nearly all doctors and nurses said that quality depends on continuous, non-judgmental self-reflection and evaluation, at both the individual and team level, in order to identify problems and implement solutions to prevent their reoccurrence. Surprisingly, however, only two individuals (both nurses) mentioned the hospital incident reporting system, and they were somewhat wary of the practical utility of such a system for their own work.

DISCUSSION

Our results suggest that doctors and nurses working at this large Swiss teaching hospital consider quality of care to be primarily the responsibility of individual practitioners. The quality of care mainly depends on practitioners’ technical and interpersonal skills. In contrast, the healthcare institution is seen primarily as a source of obstacles to good-quality care—providing insufficient resources and imposing an excessive administrative burden. Practitioners have to overcome these systemic obstacles to be able to do their job. This corresponds largely to a traditional ethos of professionalism in medicine and nursing, where the individual’s skills, dedication to the patient, autonomy and responsibility are key to achieving the desired outcome. In our opinion, while such an understanding of quality may be sufficient for patient–provider encounters in doctors’ offices, it does not take into account the complexity of most hospital care.

Practitioners versus quality experts

In general, the views of practitioners were narrower than current expert frameworks for quality management (table 2). For instance, among the dimensions of quality of healthcare considered by experts to be important,1920 only patient-centredness and (technical) appropriateness figured prominently in the practitioners’ discourse. Our respondents’ image of quality—located at the intersection of the human and the technical aspects of healthcare—echoes the views of quality experts such as Donabedian who describe healthcare as consisting of both a technical task and an interpersonal exchange between doctor and patient.21

Other dimensions, however, including efficiency, safety and equity, were mentioned only rarely, if at all. We were particularly surprised that almost no one mentioned patient safety or incident reporting in their discussions of quality and quality control, even though incident reporting systems have been designated as a hospital priority and were implemented in several of the hospital departments. However, our findings echo those of Waring23 who found that doctors preferred a collegial and blame-free approach to quality improvement, in which problems are addressed within closed peer groups.

With the notable exception of customer focus, the study respondents did not address the core principles of quality management proposed by professional bodies such as the European Foundation for Quality Management13 and the International Organization for Standardization22 (table 2). Several respondents cited the importance of effective communication between health professionals, which reflects the systemic nature of healthcare, and others mentioned the need for self-reflection and evaluation, two ingredients of continuous quality improvement. Other principles of quality management, such as the focus on effective processes of care, were not mentioned.

The lack of a systemic, process-oriented vision among the practitioners reflects their perception of their relationship with the institution. The practitioners felt that the role of the institution should be to support and facilitate their clinical activities, and not the other way around. In contrast, many quality experts would consider practitioners to be only one (albeit central) component of a complex system which has the ultimate responsibility for ensuring quality care.

Furthermore, the practitioners perceived the institution as an important source of barriers to quality care, rather than the purveyor of quality improvement tools and processes. The perceived lack of institutional support for practitioners affects their motivation as well as their physical and psychological availability for patients, and imposes limits on the quality of care that they can provide.

Comparison with other studies

Our study, which was conducted in a general hospital context, expand on the findings of several other studies. A study of perspectives on quality in women’s healthcare found that both doctors and nurses were particularly concerned about interpersonal aspects of care and patient satisfaction, and perceived systems support for practitioners’ work as key to providing quality care.24 A study in five acute care hospitals in England found that junior doctors considered good medical expertise, time with patients and patient satisfaction to be important. The doctors in that study also mentioned clinical audit and accessibility of treatment as important, factors which were not mentioned by the practitioners in our study.25

Strengths and limitations of this study

This was a small, qualitative study that aimed to explore the range of variation in doctors’ and nurses’ ideas about quality care. Qualitative methods allowed us to explore practitioners’ perspectives, identify the issues that are most meaningful to them, and explore how the specific context in which they work affects their ideas about quality.12 A quantitative survey method would be needed to measure the frequency and distribution of these ideas. As a next step to this study, our research group plans to develop a survey instrument that would incorporate quality-related ideas identified in this study, as well as components of quality as defined by quality experts. Such an instrument will allow us to describe the “culture of quality” at our institution, identify any subcultures, and measure over time any changes in quality-related notions of staff.

Because our study was conducted at a single hospital group, the results may not be generalisable to other settings. The study was also limited in that it explored only the views of doctors and nurses. We made this decision deliberately, so that we could focus on those groups most directly concerned with quality improvement activities. However, hospitals are complex organisations and their efficient functioning depends on a broad range of clinical and non-clinical staff. It is possible that the ideas and attitudes about quality differ in other professional groups not included in the study and that their attitudes have implications for the success of quality improvement activities.26

Implications and conclusion

Our study suggests that a consensual model of quality may exist among doctors and nurses working at this hospital, although different research methods would be necessary to confirm this.27 For better or worse, this shared mental model bears little resemblance to models of quality developed by quality experts. This may have local consequences, as the poor alignment between the views of practitioners and those of quality experts may hamper the implementation of the quality management model of the European Foundation for Quality Management, selected as the blueprint for quality by this hospital.

This study leaves open the question of whether anything can or should be done to change the local “culture of quality”. As Scott and colleagues point out in their review of theory and practice in the area of healthcare culture change,11 both the definition of organisational culture and the relationship between culture and organisational performance have not been adequately clarified.

Different perceptions of organisational culture lead to differences in opinion about whether, and how, organisational culture can be controlled. One possibility is that the culture of quality is a byproduct of institutional quality improvement projects and activities (among other things), and that it will evolve if and when a global quality management model is implemented. Alternatively, the culture of quality can be seen as a lever of change, in which case it may or should be subjected to intervention. Such a perspective at our institution has led to the development of several strategies aimed at influencing organisational culture related to quality, including undergraduate and postgraduate training of healthcare workers in key concepts, values and practices related to quality; demonstration projects that highlight and promote key quality-related activities such as root cause analysis for incidents; and the promotion and rewarding of peer-initiated and managed quality projects through an annual “quality prize”.

While research has suggested that teams with shared mental models tend to plan, coordinate and perform better than teams that do not have shared mental models,28^–30 cause and effect remain unclear. Furthermore, exploration of the literature regarding models of quality in healthcare confirmed that even among experts there is no consensus about the precise attributes of a “culture of quality” that would be most conducive to the best quality care. However, regardless of the specific aims of planned culture change, many common obstacles may be encountered. According to Scott et al,11 these include:

• a lack of ownership of the change process on the part of employees;

• the complex nature of cultural transformation, and the need for realistic timeframes and multilevel strategies;

• external influences that work to counter culture change efforts, such as professional organisations that influence training and internalisation of professional core values;

• lack of appropriate leadership styles;

• competing and overlapping professional subcultures within the organisation.

Currently, there are no clear prescriptions for organisational culture change, and any effort to influence organisational culture must pay attention to and address these potential barriers.