A qualitative study of the intra-hospital variations in incident reporting
University of Manchester, Applied Social Science, Manchester, UK
Objective. To determine the relationship between variations in hospital incident reporting and the corresponding attitudes and participation of medical professionals.
Methods. An in-depth qualitative case study using semi-structured interviews with hospital managers and clinicians. Twelve participants were theoretically sampled based on their involvement with clinical risk management and patient safety. Twenty-five medical physicians and four risk leads were selected from the specialist hospital departments of Obstetrics, Anaesthesia, General Surgery, Acute Medicine, and Rehabilitation. The data were analysed to develop a descriptive account of the intra-hospital variations in reporting and the associated attitudes of physicians.
Setting. The research was conducted in a single acute National Health Service Hospital Trust in the English Midlands.
Results. The qualitative data revealed significant variations in the intra-hospital organization of incident reporting between medical specialities that corresponded with the attitudes and participation of medical staff. Specifically, it was found that medical doctors were more inclined to report incidents where the process of reporting was localized and integrated within medical rather than managerial systems of quality improvement. Underlying these variations, it is suggested that medical reporting is more likely when physicians have greater control or ownership of incident reporting, as this fosters confidence in the purpose of reporting, in particular its capacity to make meaningful service improvements whilst maintaining a sense of collegiality and professionalism.
Keywords: culture, incident reporting, occupational variations, patient safety, qualitative
Address reprint requests to J. J. Waring, University of Manchester, Applied Social Science, Manchester, UK . E-mail: justin.j.waring{at}man.ac.uk
Accepted for publication June 4, 2004.
‘Patient safety’ has become a policy priority in Australia, the UK, and the USA [1–3]. Following the principles of Human Factors, this involves the implementation of proactive error management systems that can ‘learn’ about the threats to safety, matched by practices to ‘understand’ their underlying causes [4]. In the UK, this has involved the creation of the National Patient Safety Agency and the introduction of the National Reporting and Learning System throughout the National Health Service (NHS). This system requires all health care providers to establish new incident reporting procedures to enable organizational learning [2,5–7].
It has been recognized, however, that there are substantial barriers to staff participation in incident reporting and there is significant under-reporting by medical physicians [2,8,9]. It has been found that reporting can be inhibited by ‘individual factors’, such as uncertainties about the basis of reporting, ‘work factors’ including resource pressures, and ‘team factors’ associated with communication and occupational barriers [10]. It has also been found that reporting is inhibited by the professional hierarchies of health care, where lower grades of staff are discouraged from reporting to their superiors, especially for medical professionals [11]. Cole et al. [12] found that while doctors accepted the stated purpose of reporting, there remained misgivings about its practical contribution to service improvement. It was also found that doctors were sceptical about the idea of no-blame, and practical problems with time, resources, and the process of reporting also discouraged participation.
It is argued that the reluctance to report also stems from ‘cultural barriers’, especially ‘the fear of blame’ [2,4]. For Reason, this is rooted in the individualism of ‘Western culture’ and requires the creation of a ‘just culture’ that encourages reporting [4]. Professional cultures have also been shown to shape health care organizations [13]. Numerous socio-cultural studies of medical work have explored how issues of ‘perfection’, ‘clinical autonomy’, and ‘self-regulation’ influence medical attitudes and practice [14–16]. It has been found that there is a strong commitment to collegial forms of quality improvement, where issues of performance are addressed within ‘closed’ peer groups that serve to exclude non-professionals [17–19]. It is therefore important to appreciate how the cultural norms of medicine influence incident reporting.
The creation of a reporting culture, particularly amongst physicians, is central to the promotion of incident reporting in health care [2]. However, there is little in-depth qualitative research that explores the variations in incident reporting within medical specialities or engages with the cultural attitudes and practices of physicians working within hospital departments. Drawing on qualitative data, this paper describes the intra-hospital variations in incident reporting within five specialist medical departments, and gives an account of the associated attitudes and practices of physicians.
 |
Methods |
|---|
 Top Methods ResultsDiscussionConclusionReferences |
|---|
The research was conducted between 2001 and 2003 as one component of a larger ethnographic study of clinical risk management and incident reporting. The setting was a medium-sized NHS District General Hospital in the English Midlands. The initial method of data collection involved non-participant overt observations within the management structures of the hospital. These were conducted over a period of 6 months to understand the function of the various hospital departments, with particular focus on the management of risks. The second phase of the research involved face-to-face interviews with managers and clinicians from across the hospital. In total, 42 interviews were conducted in the hospital, utilizing two sampling strategies. Initially, ‘theoretical sampling’ [20] guided the selection of 12 participants, where inclusion was based upon their theoretical relevance to the study rather than randomness or statistical rigour. The theoretical basis for selection was the respondents’ direct involvement in the management and administration of the hospital-wide processes of quality improvement, risk management, or incident reporting (Table 1). A second phase of 25 interviews was then conducted based on a representative sample of specialist medical staff (consultant-grade doctors) working within five clinical departments, Acute Medicine, Anaesthesia, General Surgery, Obstetrics, and Rehabilitation, inclusive of each Clinical Director. In addition, four ‘risk leads’ were selected from these departments based on their involvement in the processes of incident reporting, with one shared between the directorates of Acute Medicine and Rehabilitation. Each interview was prefaced by a review of the study’s aims and all respondents were made aware of the ethical considerations before acquiring their consent. The interviews were semi-structured following a broad thematic guide that enabled participants to discuss issues in considerable depth (Table 2). The interviews lasted between 40 minutes and 2 hours, with an average of approximately 1 hour. The interviews were tape recorded and then transcribed verbatim.
|
|
The qualitative data were imported into the computer package Atlas ti [21] for the purpose of analysis by the author. Atlas ti was used to ‘code’ the data through the detailed inspection of interview transcripts to identify descriptions, cases, and views that occurred and reoccurred in the interviews [20]. These instances of talk were then labelled according to their content and relevance to the study, for example the code ‘reporting process’ was associated with all sections of talk that concerned making, returning, and receiving a report. A key feature of this process was the constant comparison [20] of codes between participants, and between and within codes to identify contradictions, conceptual boundaries, and theoretical relationships. For example, the sections of talk labelled under ‘reporting process’ were compared between the different medical departments, and also with other codes such as ‘form design’, ‘confidentiality’, or the role of the ‘risk lead’. Through the identification of codes and their relationships, empirically grounded descriptions and categories were identified. The results presented below draw on these data, substantiated by observational findings, to provide descriptive accounts of the intra-hospital variations in reporting and to explore the corresponding attitudes of participants; example quotations are given.
|
Results |
|---|
|
TopMethodsResultsDiscussionConclusionReferences |
|---|
The hospital-wide incident reporting system
As with the majority of acute hospitals in the NHS, the study hospital has operated an incident reporting system since the mid-1990s, associated with participation in the Clinical Negligence Scheme for Trusts (CNST). This hospital-wide system was administered by a clinical risk management department, which was a subsection of the hospital’s overall corporate management. Two risk coordinators were responsible for the day-to-day collection and processing of incident reports; one would proof-read reports and enter the information into a computer database, while the other would make an initial analysis. The process of analysis followed the hospital’s documented risk management policy. Firstly, reports were aggregated for the production of trend data to identify recurrent safety issues. Secondly, reports were graded along two scales for frequency and severity, leading to five distinct risk scores.
The risk coordinators stated that they communicated these incident data to their manager, the Deputy Director of Corporate Manager, and the hospital’s Clinical Risk Management Committee. The Deputy Director had various responsibilities for hospital performance, quality improvement, and litigation, and provided leadership for the analysis and investigation of incidents and dissemination of risk management strategies throughout the hospital. The Committee provided an opportunity for managerial and senior professional representatives to participate in clinical risk management decision-making. Observational and interview data indicated that this group’s function primarily involved investigating and discussing high-risk incidents (graded 4 or above), informing corporate decision-making and providing strategic guidance for the management of clinical risks within the hospital.
The risk coordinators claimed that incident reporting forms were distributed to each department and ward in the hospital. In accordance with hospital policy, it was expected that all staff should make a report at a suitable time following the detection of an adverse event or near miss. Reports should then be returned directly to the hospital risk coordinators or, in some departments, via a local risk lead. They stated that the hospital-wide incident reporting system received approximately 3500 forms a year. However, it was also stated that there were considerable variations in reporting across the hospital and that the majority of forms were returned by nursing and midwifery groups, whilst there was significant under-reporting for other occupational groups, especially medical doctors.
Acute medicine and rehabilitation
For the purposes of this paper, physicians within the hospital departments of Acute Medicine and Rehabilitation are discussed together because of their common experiences, close working relationships, and shared resources in clinical risk management. The interview data indicated that these physicians were generally aware of the hospital-wide incident reporting system, with an understanding of the processes of reporting, the role of the corporate risk management function, and the potential contribution of risk management to service improvement. It was also found that these two departments made use of a shared risk-lead. This person had responsibility for collecting all incident reports made on the Acute Medicine and Rehabilitation wards, developing a list of recurrent incidents, and forwarding reports to the hospital risk coordinators. This person was described as a contact point to assist in local planning and decision making, particularly through providing incident summaries at departmental management meetings.
Despite the general support for risk management and the existence of a local risk-lead, the interviews with these physicians demonstrated a common disinclination to participate in the hospital-wide reporting system. Specifically the system was regarded as nurse-led, dealing with ward issues and the work of non-medical groups.
‘Well, the nurses use it to report things like needle-sticks and falls . . . they go to the risk-lead.’ (Acute Medicine 2).
Given the apparent support for incident reporting, two enthusiastic doctors working in these departments had attempted to create an alternative reporting system to encourage localized medical reporting. Unlike the hospital-wide system, this involved an alternate confidential incident form that was designed to be easier to use than the hospital form and more appropriate for medical work, to the extent that it enabled open text recording and the use of medical terminology. Incident reports were then collected by a designated physician, who extracted information about location, time, and the context of the event to be used in clinical audit processes.
‘We asked every physician to carry a chart in their pocket and just record every event. . . . It was a little thing, that size [A5] and it had four boxes: patient’s ID, date, description of the event.’ (Rehabilitation 1)‘We keep it all anonymous but we talk through what were the issues about why this didn’t go as well as it could have and try to think about how you can make it better for the future.’ (Acute Medicine 3)
The interview data indicated that unlike the hospital-wide incident reporting system, this alternate approach had more support from physicians, primarily because it was regarded as more relevant to the improvement of medical services. This view was illustrated with reference to the role of the designated physician who was regarded as a professional peer, who provided purposeful feedback for service management. Despite some initial success with this approach, it was stated that the time and resource demands of managing the process were too great, whilst hospital management were eager to establish a single system across the hospital and end such localized approaches. In consequence, this small trial in medical reporting was abandoned after 3 months.
Anaesthesia
The interview data indicated that anaesthetists were generally supportive of incident reporting as a mechanism for improving the quality of their work. Like those working in Acute Medicine and Rehabilitation, it was also evident that these physicians remained sceptical about the hospital-wide reporting system and were generally disinclined to participate in this approach. Alternatively, these physicians preferred a localized ‘in-house’ speciality-based reporting system.
‘I think we are all aware that the information is useful for us as clinicians, but if somebody else gets hold of it . . . whether it’s going to be a big brother thing looking down upon you, not that we’re not responsible, with clinical governance we all have to be responsible for what we do, but as I say the error side of reporting is more what we do in-house and keep it in-house as opposed to referring it outside our department.’ (Anaesthetist 2)
Unlike the hospital-wide system, it was stated that the in-house reporting system was established and coordinated by the anaesthetic professional body (Royal College) and managed locally by a designated physician. Furthermore, it used a specialized incident form that could be completed and returned anonymously to the designated physician. This person collated the data and produced incident trends in a similar way to the hospital’s risk coordinators. This information was then used within local quality improvement and audit procedures, contributing to national professional education and development.
‘One of the consultants’ jobs is to analyse incidents regularly. Any incidents that occur are reported and then we have regular morbidity and mortality meetings once a month, looking at a series of incidents that have occurred, plus we have regular directorate [department] meetings where we can feedback.’ (Anaesthetist 2)
For the anaesthetists there was considerable support for their in-house incident reporting. From the interview data it was evident that these physicians took considerable pride in this approach because they considered it to be more advanced and purposeful than the hospital system, especially in informing national professional education and local service planning. It was also apparent that the confidential nature of incident reporting provided a degree of protection for physicians and promoted the notion of learning rather than blame.
‘And as I say, Anaesthesia has independently done this for many years before the hospital was more interested, risk management is the thing these days but we have been doing it for years and years and years.’ (Anaesthetist 1)
General surgery
The interviews with the surgeons revealed a marked discrepancy with the other physicians involved in the study. It was found that they were comparatively unfamiliar with the incident reporting system used in the hospital, uncertain about its purpose or contribution to quality improvement, and some surgeons actually expressed hostility towards the idea of reporting incidents, particularly to hospital management. Importantly, it was found that the hospital-wide reporting system remained their only available reporting mechanism, but given the apparent absence of support it was not surprising to find that there was little willingness or experience of participation.
‘There is no point in reporting incidents unless something is done about it, and that is one of the problems that we perceive in our practices, you know you fill in all these forms and waste a lot of time and nothing gets done about it in the end.’ (Surgeon 2)‘I knew that if there was a major incident that we would have to report it, but I wasn’t aware of the forms’ (Surgeon 4)
Despite such antipathy towards incident reporting, the surgeons did acknowledge the importance of communicating their safety concerns, but this involved writing a letter or making a telephone call to the Clinical Director. Moreover, the surgeons were supportive of other schemes to promote quality, including the Confidential Enquiries into Peri-operative Deaths, Morbidity and Mortality Committees, and local clinical audit systems, which were regarded as collegial and educational in focus, providing opportunities for professional development whilst remaining under occupational direction. In contrast, the interview data indicated that the hospital incident reporting system remained disassociated with such forms of quality improvement.
‘Most of the more important events people get to hear about anyway, so most major events will be reported by other means and will usually come out to an M and M [Morbidity and Mortality] forum and that’s probably how you will get most out of it being discussed.’ (Surgeon 5)
The lack of surgical enthusiasm and participation in reporting was recognized by the departmental Clinical Director who expressed a desire to promote change amongst his colleagues and had recently introduced a new risk-lead to encourage reporting and promote local control of the process.
Obstetrics
Of the medical specialities involved in the research, the obstetricians were clearly the most enthusiastic and supportive of incident reporting. It was stated that for >10 years, doctors working in this speciality had been involved in forms of risk management and had high levels of reporting. Despite the desire to develop a localized incident reporting form, these doctors were accustomed to using the hospital-wide incident reporting system. It was reported that much of the support for risk management in obstetrics had originated through national professional education and guidance, within the context of increased litigation pressure, but it was also stated that reporting has made a positive contribution to local service development.
‘Within the profession, it’s within the profession, and also to be fair, the hospital, the insurance scheme, the CNST, has driven this forward. A lot of things that we aim to meet are their standards, and the higher the standard the lower the insurance.’ (Obstetrician 4)‘We tend to report most things, we are very open.’ (Obstetrician 2)
Participation in reporting was supported through several developments in the localized working practices of this department. This included specialized training and induction courses, the proactive leadership of the Clinical Director and the introduction of a risk-lead. As with the anaesthetists, this person had a key role in collecting incident reports and providing feedback to individuals and local management groups before sending the forms to hospital risk management. In particular, incident data contributed to forms of quality improvement, including clinical audit and clinical governance, as well as being reported to the corporate level of the hospital.
|
Discussion |
|---|
|
TopMethodsResultsDiscussionConclusionReferences |
|---|
The qualitative data reported reveal significant variations in the organization of incident reporting within the specialist departments of the hospital. Importantly, it was found that these variations corresponded with the support and participation of specialist physicians. The reporting systems exhibited distinct histories, leadership, and coordination arrangements, and levels of integration within both hospital-wide processes and localized management procedures. In those departments where medical attitudes were more positive and participation common, for example Anaesthesia and Obstetrics, it was evident that reporting was instigated or coordinated by departmental medical staff and that incident data were used within localized medical procedures for service improvement. Conversely, medical support for and participation in reporting was found to be low in General Surgery, Acute Medicine, and Rehabilitation, excluding their localized experiments, because the only available method of reporting was organized from outside the department by hospital managers. It appears, therefore, that physician support for and participation in reporting was greater where reporting was based within rather than outside the medical department.
Given the evident importance of departmental professional control in the development, coordination, and application of incident reporting, three prominent and interrelated themes can be proposed as influencing medical attitudes and participation: confidence, purpose, and collegiality. The concepts of confidence and trust permeated the data, with particular reference to the anonymous and confidential handling of reports. It may certainly be the case that the fear of blame discourages medical reporting [4], but more significant was an apparent distrust in the activities of non-medical groups. The interviews indicated that the hospital-wide system was regarded as managerial, non-medical, and ‘faceless’; correspondingly, there was less confidence in this approach. Conversely, the departmental systems were administered by designated professional peers working alongside clinical staff in whom a degree of confidence could be invested. A related factor was the perceived purpose of incident reporting. For the hospital-wide system there were few positive accounts of how incident reports contributed to service improvement, reinforcing any distrust in this system. The departmental systems, however, were consistently described as contributing to service and professional development, and were regarded as making meaningful contributions to the work of those who filed reports. Drawing together these two themes is the crucial issue of professional control and collegiality. It was evident that physicians had confidence in the processes and purpose of reporting where it was linked to existing departmental procedures, rather than being an external or managerial intervention into medical work. This is because it ensured professional direction and control, and linked reporting to meaningful service improvements.
This research therefore shows that positive attitudes towards reporting are more common where physicians have confidence in the collection and application of data, where it has a meaningful purpose in service improvement, and thus where the control and coordination of reporting is within rather than outside medical practice. The implications of this research for policy are important, as priority is currently given to the creation of a shared ‘culture of reporting’ within hospitals [2,5,6]. The feasibility of this ambition is questioned where reporting systems are established and managed outside the localized working practices of physicians and by non-medical groups. Of significance here are those socio-cultural accounts of medical professionalism that emphasize the importance of collegiality, to the extent that this fosters the internalized evaluation of performance to the exclusion of non-professional groups [15–19]. However, it is necessary to question the merits of uni-professional reporting given the interdisciplinary basis of health care delivery and the need for service-wide learning.
|
Conclusion |
|---|
|
TopMethodsResultsDiscussionConclusionReferences |
|---|
The research has shown that physicians are inclined to report when there is confidence in the processes and purpose of reporting, where reporting has a meaningful contribution to service development, and typically where these are satisfied through collegial forms of incident reporting. For medical incident reporting to be encouraged across the NHS it may be necessary to introduce reporting from within the profession rather than through non-professional organizational processes.
|
References |
|---|
|
TopMethodsResultsDiscussionConclusionReferences |
|---|
- Wolff A, Bourke J. Reducing medical error: a practical guide. Med J Aus 2000; 173: 247–251.
- Department of Health. An Organisation with a Memory. London: The Stationery Office, 2000.
- Institute of Medicine. To Err is Human: Building a Safer Health System. Washington DC: National Academy Press, 1999.
- Reason J. Managing the Risks of Organizational Accidents. Aldershot: Ashgate, 1997.
- Department of Health. Building a Safer NHS for Patients. London: The Stationery Office, 2001.
- National Patient Safety Agency (NPSA)/Department of Health. Doing Less Harm. London: NPSA, 2001.
- NPSA. Seven Steps to Patient Safety. London: NPSA, 2003.
- Barach P, Small S. Reporting and preventing medical mishaps: lessons from non-medical near miss reporting. Br Med J 2000; 320: 759–763.
[Free Full Text] - O’Neil A, Petersen M, Cook E, Bates D, Thomas H, Brennan T. Physician reporting compared with medical-record review to identify adverse medical events. Ann Intern Med 1993; 119: 370–376.
[Abstract/Free Full Text] - Vincent C, Stanhope N, Crowley-Murphy M. Reasons for not reporting adverse incidents: an empirical study. J Eval Clin Pract 1999; 5: 13–21.[CrossRef][Web of Science][Medline]
- Lawton R, Parker D. Barriers to incident reporting in a health care system. Qual Safety Health Care 2002; 11: 15–18.
- Cole J, Pryce D, Shaw C. The Reporting of Adverse Clinical Incident—Achieving High Quality Reporting: the Results of a Short Research Study. London: CASPE Research, 2001.
- Helmreich R, Merritt A. Culture at Work in Aviation and Medicine. Aldershot: Ashgate, 2001.
- Leape L. Error in medicine. In: Rosenthal M, Mulcahy L, Lloyd-Bostock L, eds. Medical Mishaps. Buckingham: Open University Press, 1999.
- Freidson E. The Profession of Medicine: a Study in the Sociology of Applied Knowledge. Chicago, IL: University of Chicago Press, 1970.
- Lupton D. Medicine as Culture. London: Sage, 1994.
- Arluke A. Social control rituals in medicine: the case of the death round. In: Dingwall R, Heath C, Reid M, Stacey M, eds. Health Care and Health Knowledge. London: Croom Helm, 1977.
- Bosk C. Forgive and Remember. London: University of Chicago Press, 1979.
- Rosenthal M. The Incompetent Doctor. Buckingham: Open University Press, 1995.
- Strauss A, Corbin J. Basics of Qualitative Research: Grounded Theory Procedures and Techniques. London: Sage, 1990.
- Atlas ti. Scientific Software Development, 1997.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  D O Farley, A Haviland, S Champagne, A K Jain, J B Battles, W B Munier, and J M Loeb Adverse-event-reporting practices by US hospitals: results of a national survey Qual. Saf. Health Care, December 1, 2008; 17(6): 416 - 423. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. J. Holden and B.-T. Karsh A Review of Medical Error Reporting System Design Considerations and a Proposed Cross-Level Systems Research Framework Human Factors: The Journal of the Human Factors and Ergonomics Society, April 1, 2007; 49(2): 257 - 276. [Abstract] [PDF]
|
|
|
|
 |
|
 L. Freestone, S. N. Bolsin, M. Colson, A. Patrick, and B. Creati Voluntary incident reporting by anaesthetic trainees in an Australian hospital Int. J. Qual. Health Care, December 1, 2006; 18(6): 452 - 457. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M. Hudelson Culture and quality: an anthropological perspective Int. J. Qual. Health Care, October 1, 2004; 16(5): 345 - 346. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||