Clinical handover has been defined as the transfer of responsibility and accountability between one or more clinicians on a temporary or permanent basis.⁸ This official definition of clinical handover exists in national guidelines and more specifically in guidelines of professional obstetric practice.^{8 9} However, a specific legal determination for clinical handover is not mentioned in Australian law. Lack of communication in maternity care resulting in an adverse outcome has appeared in a previous Australian legal case. Failure of communication about a previous classical caesarean, which carried an increased risk of uterine rupture, resulted in long-term neurological disability for the affected child following an emergency delivery following uterine rupture in a subsequent pregnancy.^{10 11} Although clinical handover was not specifically mentioned in those terms, the concepts of the legal aspects of responsibility and accountability for the handover of relevant information (in this case, written documentation), involvement of the patient in the handover through informing them of relevant information to communicate to health professionals to assist in future decision making and the role in the recipient of the handover and the organisation where they work are illustrated by this case. Information transfer is implicit within the official definition of clinical handover and that the information is understood by the recipients.¹² Following on from this, liability was attributed to both the first hospital (where the first caesarean section occurred) and the second hospital (where the affected baby was born) for contributing in the communication breakdown that resulted in this adverse event.^{10 11}

Transfer of responsibility, accountability and information is part of a subset of reported functions and roles of this process in the literature.¹³ The unambiguous transfer of responsibility has been observed as one of the handover strategies to improve team coordination in work settings with high consequences for failure.¹⁴ Blurring of boundaries of responsibility have been linked to communication-related healthcare incidents, particularly associated with increasing numbers of clinicians and specialities involved in same patient care.¹⁵ The introduction of standardised handover protocols that involve clarification of time and/or designation of responsibility have been suggested as improvement strategies with varying success regarding unambiguous transfer of responsibility and accountability.^15–17

Apart from the official definition, other functions that have been reported include induction to the profession (eg, language, practices, enculturalisation), education, safety through cross-checking (eg, medication and charts), discussion on management issues, organisation, patient management, and discharge planning and socialisation (eg, social/emotional support and debriefing, stress relief, light hearted episodes).^18–22 Problems with this process have been associated with morbidity and mortality.^23–27 Inadequate handover has also been associated with problems with lack of coordinated team work (eg, duplication of tests).^{26 28 29}

Primary maternity care in Australia is undertaken by midwives, and doctors, including general practitioner (GP) obstetricians, obstetricians and obstetric trainee doctors under supervision. This is divided into public and privately funded healthcare.³⁰

Written and verbal information was provided and clinicians provided informed verbal consent before participation. Semistructured interviews (27 face-to-face and 6 written) and 18 focus groups took place facilitated by the first author. This format enabled exploration of themes that were raised at that time and enabled testing of themes that had developed in previous interviews and focus groups. Face-to-face interviews and focus groups were audio-recorded and transcribed.

Data were de-identified and themes were coded by the first author using the constant comparative method,³¹ which is founded on a grounded theory methodological approach.³² Themes were identified from transcripts and its presence was searched for in other interviews. A sample of three interviews and two focus group transcripts was independently coded by the second author for inter-reliability testing. When coding for presence and lack of presence of first author identified themes, a 74.04% agreement with a kappa of 0.34 on codes between the two authors was shown. Themes were discussed with the other members of the team. Where ambiguity or disagreement was identified, the transcripts were revisited and consensus achieved. NVivo software was used for data management.

The use of triangulation within focus groups and other transcripts, clarification of points during data collection with the participants and multidisciplinary review of coded data by other members of the research team provided rigour to this study. Further information about the background of researchers is outlined below.

The theme of responsibility and accountability was one of multiple themes that emerged from the analysis, and is reported in this paper.

In this hospital primary public maternity healthcare was undertaken by a combination of midwives, GP obstetricians and obstetric doctors (obstetricians and trainees) for antenatal outpatient care. In the intrapartum and hospital admission (antenatal and postnatal) setting, this primary maternity care was undertaken by midwives and obstetric doctors (obstetricians and trainees). Involvement by other specialities was based on the medical and psychosocial needs of the patient. Patients identified as lower maternity care risk were managed within one of two lower maternity risk care teams or within a family birth centre environment. In this group, care was predominantly undertaken by midwives with support from obstetric doctors, specialities and allied healthcare workers only as required. Those identified as high maternity care risk were managed in one of two maternal–fetal medicine units or smaller subspeciality units (eg, fetal management unit, diabetes unit, recurrent miscarriage unit, cervical surveillance unit, multiple pregnancy unit, women with physical and/or intellectual disability in pregnancy unit, drugs and alcohol in pregnancy unit, adolescent pregnancies unit). Birth was facilitated by midwives predominantly if they were spontaneous births and obstetric doctors if assisted vaginal birth or caesarean section were required. If interventions such as augmentation or continuous cardiotocographic monitoring were required and if patients were identified as increased obstetric, medical or psychosocial risk, obstetric doctors undertook primary management of these women.

Although not a major focus in this study, some of the clinicians also worked within the private healthcare system. Private maternity healthcare in Australia is predominantly undertaken by obstetricians. Private obstetricians practise either as solo practitioners, in an associateship (one or more obstetricians practising alongside each other but independently accountable from one another) or within a group practice where on-call cover is shared between the group of obstetricians. Referrals are made to other specialists as the needs arise. Solo obstetricians and those in associateship may also choose to share the work with other obstetricians for some on-call and when they are on leave from their practice. Within the private hospital environment, midwives and other specialists (as required) also provide care for the patient. However in this context, the obstetrician or their appointed covering obstetrician assumes obstetric responsibility and accountability for the patient.

To a lesser extent, some women choose to have private care with independent practising midwives providing homebirth support, which may be supported by a GP obstetrician. Private planned homebirths have the opportunity for transfer to the public healthcare system when increased obstetric risk and/or complications arise.

Official clinical handover in this public and private maternity care environment was predominantly uni-disciplinary for shift and in private healthcare, on-call/leave cover. In the public hospital where this study was based, maternity shift handover was face-to-face. Multidisciplinary presence at some public maternity shift handovers occurred at this time with obstetric doctors giving handover at these meeting corresponding to obstetric doctor shift change. Clinicians of differing seniority were present at these handovers. There were two separate birth suites, one family birth unit, two designated postnatal and one designated antenatal ward in operation at the time of this study in the hospital. The birth suites contained labouring and birthing patients, those requiring more one-to-one care for acute antenatal and postnatal conditions and for some emergency antenatal assessments. The family birth unit provided some low-risk maternity patients antenatal, intrapartum and postnatal care predominantly by a core group of midwives with support from other clinicians as required. They practised group shift midwifery handover away from the patient. Subsequent one-to-one midwife handover was performed later between off-coming and on-coming midwives. One birth suite which was designated to have lower maternity risk patients (but also had higher risk patients when there was high bed occupancy) had a policy in encouraging this one-to-one handover to occur at the bedside in the presence of the patient.

Observational research by clinicians in their working environment exists in published literature. Particularly in ethnographic research there are examples of clinicians undertaking observational work while also undertaking clinical duties in the same area.^{21 34} The impetus for our group studying maternity handover had its roots within the earlier clinical work of the first author. Having a local cultural and technological awareness of the context of study data through the first author's personal working experience in this hospital facilitated: (i) a native understanding of the language used by the participants; (ii) an exploration of concepts raised within the interviews and focus groups; and (iii) provided a conceptual basis for the development of understanding of this area from comparison and testing of perceptions provided by participants in this study.

However, in this study all interviews and focus groups were done outside the immediate work environment and at times where the researcher was identified as being outside clinical duty. This was to encourage the participants to engage with the interviewer as much as possible in her research role rather than her clinical role. This definition between working and clinical roles was also to decrease the element of bias that may occur in blurring of this boundary. In this study, frankness displayed in the transcripts particularly from other specialities and junior obstetric doctors reflected the ability of participants to focus on the interviewer's research rather than the clinical relationship to the participant with the interviewer.

In total were 7 clinicians in this study who spontaneously identified the transfer of responsibility and/or accountability as one of the roles or functions of handover. However, when asked specifically about this function, the other clinicians agreed that it was indeed connected to handover. Once introduced to this concept, they spoke about the transfer of responsibility and accountability in terms of complete transfer as well as shared and ongoing personal responsibility and accountability (box 1).

Some clinicians in this study described the transition of responsibility and accountability for patients as occurring once communication of information needed for the safe transfer had taken place. Some also recognised that this did not occur until there was acknowledgement or a sign of understanding of this information from the on-coming clinicians.¹²

Other clinicians took a more temporal and physical approach to handover by describing it in terms of the rostered shift or when a clinician physically left the work environment. Shift work is something that is well understood by those in the nursing and midwifery profession with handover being seen by some as almost a ritualised practice.²¹ However, this is a newer concept to the medical profession with the transition from the historical personal continuity of care to the development of the modern ‘safe hours’ shift work phenomenon.³⁵ The establishment of the modern ‘safe hours’ for medical staff recognises that clinicians do not work at their best or even safely when fatigued. Issues with cognition, response time, attention and increased rates of errors are associated with the traditional longer shifts and on-call duty.^36–39 With the advent of shift work for medical staff, it has become imperative for doctors to understand when it is time to hand over care of a patient and to become more familiar with team communication.³⁵

Some clinicians in this study experienced confusion or blurring of boundaries of responsibility and accountability. This is was when clinicians still felt responsibility and accountability for patients once a handover had taken place. There were times when this ambiguity in responsibility and accountability was observed to be beneficial. Examples given by participants were when staff completing a shift called back to add or correct information from an inadequate handover or responding to an emergency when on-coming staff were unavailable. In these cases, the impression of ongoing responsibility and accountability provides system resilience; in other words an informal back-up strategy when there is a system issue.⁴⁰

However, confusion or blurring of responsibility and accountability following a handover can lead to problems. Role confusion, uncertainty of allocation of work and identification of the appropriate person for reporting and inquiry for the status and management for patients may be the result of such ambiguity which may lead to clinical incidents.^{15 41} There is also the potential for significant psychological impact on a clinician feeling ongoing responsibility and accountability, real or misinterpreted, for an unexpected, unwanted and/or adverse outcome. It may result in the clinician changing their behaviour, questioning whether to continue to work in the profession and/or negatively impact on a clinician's psychological well-being.^{42 43} It is important that clinicians not to feel falsely responsible or accountable in such cases and receive the appropriate support and counselling as required. However, such support may not routinely occur.^{42 43}

There was the potential of the interview and focus group facilitator's working relationship with the participants to create bias in responses. This was minimised by interviewing outside the immediate working environment and at times when the clinician was not identified to be on clinical duty. By drawing these boundaries it facilitated engagement of the participants with the interviewer in their research rather than clinical role.

The study of clinicians of different specialities and seniority who undertake maternity care in one hospital enabled triangulation of data was also one of the strengths in this study. This provided interesting insights of perceptions of roles, relationships and work within the maternity team. However, by localising the research to one study site limited the scope to one hospital. Other studies are required to see whether these findings are also seen in other similar working environments.