Handover, hand-off or sign-off in healthcare may be defined as: “The transfer of professional responsibility and accountability for some or all aspects of the care of a patient, or group of patients, to another person or professional group on a temporary or permanent basis.”1 Handover may be between shifts, between units, between secondary and primary care, or between different healthcare professions. Changing medical work patterns due to the European Working Time Directive (EWTD) or Accreditation Council for Graduate Medical Education (ACGME) duty hour restrictions mean that different teams of doctors now also look after the same group of patients over the course of any given day.

Evidence from other industries running continuous operations shows that the discontinuity of tasks and personnel which accompanies shift-working, can give rise to non-transmission or miscommunication of critical information.2 This can, in turn, lead to major incidents3 such as Piper Alpha4), to date the world’s worst offshore oil disaster in terms both of lives lost and impact to industry. Analysis of the Piper Alpha incident indicated that poor handover notes were a key causal factor.

Given these changes in medical working patterns, what we know from other industries,5 and their implications for patient safety, there has been a surge of interest in improving healthcare handover practices. Several bodies, including the Joint Commission National Patient Safety Goal, the World Health Organization High Five Initiative, the Australian Council for Safety and Quality in Healthcare and the Junior Doctors Committee of the British Medical Association, have published recommendations for optimal handover performance based on expert opinion.16 Unfortunately, there is evidence that these suggested standards for medical handovers are generally not met.78 Furthermore, US observational work indicates that the patient information communicated during medical handovers is inadequate for the task910 and this has been implicated in fatal delays and inappropriate patient care.11^–13

Handover at shift changes frequently falls to doctors in training due to the nature of their working hours and responsibilities. Handover standards may thus be improved by ensuring that skills are taught prequalification, at medical school.1 However, there are few reported studies of teaching handover to medical students.1415 Only 8% of US medical schools formally teach handover (“hand-off” or “sign-out”),16 and anecdotal evidence suggests this probably reflects UK practice.117 Thus, most newly qualified doctors are handing over patients with no training in how to do so. This means that new doctors must both acquire knowledge of handover and employ this knowledge immediately, without time to think or reflect—which is unlikely to lead to effective learning.18

However, empirical data to inform the content of teaching handover are lacking. Our study aimed to answer the research question: how can we best prepare new doctors to handover effectively?

METHODS

In the absence of prior UK empirical work, we selected qualitative methodology to help identify the issues which need to be addressed to achieve effective handover practice, and how to deliver teaching for optimal handover.19 We chose focus groups, as these are able to generate data regarding perceptions and beliefs but can also be particularly useful, as the group setting can make people more confident in sharing information.20 Further, focus groups allow participants to ensure that the topics under discussion are directed by participants themselves, not just by researchers’ agendas.21

Participants

All Foundation Year 2 (FY2), Senior House Officers (SHO) and Specialist Registrar (SpR: senior trainees) doctors working in Aberdeen Royal Infirmary (ARI), a large teaching hospital, in March 2007 were invited to take part in the study via an email circular from NHS Education for Scotland. Consultants who coordinate teaching for medical students and night nurse practitioners (ARI) were also invited to take part, as we felt their views would be useful given that doctors in training do not work in isolation. The dates and times of the focus groups were set in advance of the invitation going out, timed for the end of day shifts/immediately before night shifts for ease of participation. Those who expressed interest in the study were then contacted by the researchers.

Of relevance is the structure of teams in the UK. Intershift handovers involve, where possible depending on clinical demands and unit working patterns, the whole team of trainee staff, from senior trainees nearing Consultant status to Foundation Year 1 doctors (in their first year of work postqualification). Consultants may be involved depending on the working patterns of the specific unit. Trainee doctors’ hours are governed by the European Working Time Directive (EWTD) of 56 h per week at the time of writing.

Ethics permission for this study was granted by the North of Scotland Research Ethics Committee. Written, informed consent for data collection and publication of anonymised data was obtained from all participants. Participants were provided with a certificate of participation.

DATA COLLECTION

Five focus groups were carried out between March and June 2007. Participants were assigned to focus groups by grade/profession to encourage open discussion. All focus groups were carried out in ARI.

The same semistructured discussion guide was used for all groups. This was developed through a literature review followed by stakeholder (members of the local undergraduate medical degree management team and the Postgraduate Deanery) consensus. The topics covered included definitions of good and bad handover practice, what factors influence handover, what are the skills and abilities needed to carry out a good handover, can these be taught and, if so, how. The focus-group discussions were led by JC, who has experience in qualitative, focus-group research,22 and cofacilitated by at least one other author, all of whom have clinical experience of handover.

ANALYSIS

All focus groups were taped and transcribed verbatim, and anonymised. The transcripts were analysed inductively, using framework analysis,23 during which we determined content-related themes (what participants said). Framework analysis was used to classify and organise the data according to key themes (main and subthemes), concepts and emergent categories, which were used to examine the data for patterns and connections. Process-related themes such as the use of language and humour were not analysed.

Focus-group transcripts were initially analysed independently by each author who developed categories or themes as they emerged from the data.24 All authors then met to discuss and examine these independent analyses using the constant comparative method, where items were compared and contrasted, and themes negotiated and checked to establish analytical categories, or the “fit” between items, and to group data into categories and subcategories. The data were analysed using a paper (rather than a software) system.

RESULTS

Seventeen doctors and four night nurse practitioners (NNPs) took part (n = 21). One focus group was held with consultants (n = 3), one with SpRs (n = 5), one with Foundation Year 2 (FY2—previously called Senior House Officers) doctors (n = 3), one with NNPs (n = 4) and one with a mixed group of Foundation Year 1 (FY1—previously called Preregistration House Officers), FY2 and SHO doctors (n = 6).

Six (35%) male and 11 (65%) female doctors took part. All NNPs were female. Surgical and medical specialities were represented in the doctors who took part. All participants had clinical experience of handover in different hospital environments (eg, surgical and medical wards), in different hospitals (ie, their experience was not limited to just ARI), and had had been involved in handovers with, and/or as, new doctors. Only nurse participants had received formal training in how to handover. Consultants tended not to be involved in daily intershift handovers, which were usually registrar-led.

The focus-group discussions ranged from 40 to 65 min in length.

Five main themes relevant to the task of effectively handing over, and how to best teach handover, emerged. These are set out below.

Definition of handover

Handovers were defined as an exchange of relevant information, informing the next shift of necessary tasks and logistical information, transferring responsibility, prioritising and highlighting particularly unwell patients, and ensuring continuity of care. Junior doctors (FY2 and SHO) provided definitions that were task-focused compared with consultants and nurses, who talked about transfer of responsibility (box 1).

Experience of handover as a junior doctor

Junior doctors described their early experiences of handover negatively, as very stressful due to not knowing the environment, the patients, what was expected of them, and feeling overwhelmed by the amount of information given to them and how this was given (usually orally—see Systems factors later). Junior doctors felt they were not being provided with sufficient information about individual patient circumstances or which patients to prioritise, and this lack of a bigger picture could be stressful particularly when patients became unwell. Structured handovers (in terms of a set time and place) where at least some patient information was available in writing were strongly preferred (see box 2 and Systems factors later).

Junior doctors’ handover skills and attitudes

The data indicated that junior doctors had a relatively good grasp of the skills and knowledge required for effective handover (box 3). Participants proposed, with ease, the skills they saw as essential to effective handover. These included communication skills: active listening, not interrupting, introducing yourself, questioning (those who were handing over), team working/responsibility and communicating concisely. Communicating relevant information concisely was discussed in every focus group. Other important skills were prioritisation (identifying what was important information to handover), advance preparation of information to handover (to aid effectiveness) and having the skills, and confidence, to present information to many people. Prioritisation was reported as particularly important due to the quantity of information/patients being handed over.

However, there seemed to be difficulties putting these skills into practice. More senior doctors and nurses were critical of new doctors’ handover skills. They discussed that junior doctors often found it difficult to know what information was relevant and lacked professionalism (in terms of responsibility to colleagues to hand over the information required to provide the best care for patients) (box 4).

While more senior staff and nurses were aware that new doctors struggled with the complexity of handover, there seemed to be the view that handover skills were expected to develop “on the job” through experience and observation/role modelling. There was the view that the further a doctor progressed in training, the more they were able to prioritise and communicate critical information. However, in contrast, the data also suggested that experience and observation were not necessarily reliable methods of learning skills (box 3).

Systems factors

As stated above, the data indicated that junior doctors were aware of the skills required for effective handover (box 3), but not using these skills, or developing them further “on the job” seemed, in great part, linked to workplace or systems factors such as poorly structured handovers, large numbers of patients and no protected time for handing over. The latter meant that critical team members did not always attend handovers, or did not attend the whole session, due to workplace demands. Handovers could be chaotic and ad hoc, and lack obvious leadership. While feeling responsible for patients was seen as a positive factor which encouraged good handover, this was unsupported by shift working patterns. These systems issues seemed to be direct barriers to improving practice (box 5) and could be demoralising in terms of acting as barriers to change.

Teaching handover

All participants supported the concept of formal, teaching on handover before commencing work as a doctor. Participants proposed the skills they saw as essential to effective handover (see earlier). All groups discussed how handover should be written as well as oral (box 6).

Suggestions for teaching methods included simulated handovers and exercises to develop the skills necessary for effective handover. Using real clinical information was seen as important. Participants emphasised that teaching simulations should reflect real life practice (eg, handing over many patients, handover in an imperfect environment) to be transferable to the work environment on qualification.

Encouraging clinical staff to communicate the importance of handover and to role model effective handover was also seen as important, as was providing opportunities for undergraduate medical students to observe a handover. Reflection on simulated or observed handovers was also considered an important learning tool.

DISCUSSION

We have identified that new doctors have a narrow definition of handover (hand-off or sign-off), focusing on tasks which require completion by the end of the shift. New doctors have insight into the skills and attitudes needed for safe and effective handover but the reality of handing over multiple patients under difficult circumstances (eg, no protected handover time) results in a task focus and stressful early experiences of handover. New doctors came to handover with no specific training14 and thus were unprepared in terms of having a basis for building skills. Experienced medical and nursing staff seemed to believe that new doctors would learn handover on the job, but evidence of this was lacking. Rather, workplace demands and poor systems precluded the use of the clinical handover as a teaching and learning opportunity.

While the skills and organisational factors identified in this study support those suggested in policy documents based on expert panel views,16 our empirical approach adds to existing knowledge in two ways. First, it emphasises the necessity of teaching handover first in the undergraduate curriculum as a means of preparing new doctors for the workplace. Second, in terms of highlighting that handover is not just about skills: medical students must also be encouraged to reflect on the role of professional attitudes in effective handover.25 It also seems clear from these data that new doctors need to be aware of the systems in which they will work and the difficulties these may involve for handover.

With reference to cognitive learning theory, providing opportunities to learn how to handover before handing over in a clinical environment will support the development of “readiness to learn”26 and provide students with the scaffolding required to continue their learning “on the job,” or in context,27 more effectively. Teaching should focus on supporting learners in the purpose as well as the skills of handover.28 Simulated handover situations, designed to mimic the complexity of real clinical situations and systems, could help learners apply their knowledge to problems (eg, many sick patients to handover) and barriers (eg, an unstructured handover), and work out how to cope with the problem.29

It is clear from the present study, and the wider literature, that training is only one aspect of improving handover: systems issues such as protected time for handover and the use of structured handover tools must also be in place.16 Designing and implementing effective handover systems must be given high priority by healthcare organisations.2 For example, if it is expected that trainee doctors should learn handover skills “on the job,” then a system for providing timely and constructive feedback on their contribution to handover must be instituted.

An obvious strength of this study is that we included doctors at every stage of training, as well as consultants and night nurse practitioners, in the data collection in order to assess the differing expectations of junior trainee doctor handover. The skills mix of the authors meant that each focus group was cofacilitated by people with experience and knowledge of running focus groups and clinical medicine, thus enabling more in-depth discussion than may have been achieved if facilitation had been carried out solely by a non-medical researcher. Furthermore, initial, independent data analysis by medical and non-medical researchers provided different perspectives on some data, then discussed fully before final themes and subthemes were agreed, thus leading to a more robust analysis. One potential weakness of this study is that the data were collected in one geographical locality and one, very large, teaching hospital. However, all our participants had experience of studying or working in other hospitals, and their discussion was not limited to local or current experiences. The small number of participants in some focus groups could be seen as limiting the discussion and hence the information gathered. However, group members had worked or trained together, and seemed comfortable discussing their views and experiences. Finally, the focus of this study was on how we can best prepare new doctors for handover. This focus limited exploration of the systems that training could not address (but which seemed extant in the participant responses).

This study has research implications. Qualitative research does not stake a claim to generalisation in the way quantitative research does. Thus, replicating this study in diverse locations would be helpful in terms of illuminating the generalisability of the findings. However, future qualitative work may wish to take a balanced approach to discussion which allows full exploration of both training and systems as potential influences on handover. Supplementing focus groups with individual interviews to target any groups felt to be under-represented may also be beneficial. Longitudinal studies are required to identify if teaching has an impact on actual practice rather than simulated practice,29 and to explore further the interaction between individual and systems factors. Given that stress at work has been consistently related to lack of control of the working environment,30 as seems to be the case with new doctors and handover, a study exploring levels of stress31 associated with different levels of preparedness for handover may also be timely.32

In conclusion, the results of this study will inform the development of optimal teaching of handover, based on learning theory, aimed at helping medical students handover more effectively when they start working as doctors. However, training must be considered with respect to the human as part of the healthcare delivery system: training new doctors for handover will not be effective if this is in isolation from implementing robust systems for handover.