Are mentors failing to fail underperforming student nurses? An integrative literature review Wray, Jane; Kennedy, Michelle; North, Hannah Introduction Ensuring that pre-qualifying nurses achieve proscribed standards for safe and competent practice is essential for nurse education and practice in the

Ensuring that pre-qualifying nurses achieve proscribed standards for safe and competent practice is essential for nurse education and practice in the UK (Nursing and Midwifery Council (NMC) 2015, 2017) and internationally (Clark et al. 2011; Zasadny and Bull 2015; Nursing and Midwifery Board of Australia 2018). Competency based education aims to generate a nursing workforce with the skills and flexibility to function within global markets (Foth and Holmes 2016) and fulfil employer requirements for graduates who are fit to commence employment without the need for extensive further training (Black et al. 2008). Competency based education focuses on what nurses need to know to function safely and effectively in the clinical environment (Foth and Holmes 2016) and can be defined as; ‘Broad clusters of general attributes which are considered essential for effective performance’ (Garside and Nhemachena 2013, p. 543).


Introduction
Ensuring that pre-qualifying nurses achieve proscribed standards for safe and competent practice is essential for nurse education and practice in the UK (Nursing andMidwifery Council (NMC) 2015, 2017) and internationally (Clark et al. 2011;Zasadny and Bull 2015;Nursing and Midwifery Board of Australia 2018). Competency based education aims to generate a nursing workforce with the skills and flexibility to function within global markets (Foth and Holmes 2016) and fulfil employer requirements for graduates who are fit to commence employment without the need for extensive further training (Black et al. 2008).
Competency based education focuses on what nurses need to know to function safely and effectively in the clinical environment (Foth and Holmes 2016) and can be defined as; 'Broad clusters of general attributes which are considered essential for effective performance' (Garside and Nhemachena 2013, p. 543).
Student nurses are required to demonstrate effective application of knowledge, judgement and skill (Paganini and Egry 2011) and achieve set learning outcomes to advance on their programme of study and ultimately qualify to practice independently (Gravina 2017). Assessment in clinical practice is centred on performance, with competency offering a mechanism for its measurement (Clark et al. 2011). However, assessing competence in practice is complex (Hunt et al. Mentors concerns regarding under-performing students and the phenomena of 'failing to fail' has been previously identified and discussed in the literature (Lankshear 1990;Duffy 2003;Scholes and Albarran 2005;Skingley et al 2007).
Duffys ' (2003) grounded theory study highlighted the issue of 'failure to fail' and identified reasons for this phenomena including; a lack of confidence, unwillingness to invest time required to fail a student, university over-turning the mentors' decision, and the emotional impact of failing a student. This study alongside others (Hawe 2003, Scanlan et al. 2001) informed the NMC's 'Standards to Support Learning and Assessment in Practice ' (2008). This framework aimed to enhance quality of learning in practice, address mentor preparation and support, and increase mentor awareness of their role. It was clear that mentors were accountable for confirming whether the student had or had not met the required standards for safe and competent practice (NMC 2015).
It is not clear however how successfully or consistently the NMC guidance (2008), available since 2008 has been implemented as the issue of 'failure to fail' continues to be discussed (Gainsbury 2010;Vinales 2015;Elliott 2016;Peate 2018). The revised Nurse Education and Assessment Standards in the UK due for implementation in 2019 (NMC 2018) have reignited this debate. These will inevitably impact upon the role of the mentor, including possible removal of the title 'mentor' and separation of the supervisory role from that of assessment.
Consequently, it is timely to review this issue and determine the nature of the current evidence base to inform future practice. This review posed the following question; 'What is the evidence that mentors are failing to fail underperforming student nurses?' This document is the Accepted Manuscript version of a Published Work that appeared in final form in British Journal of Nursing, copyright © MA Healthcare, after peer review and technical editing by the publisher. To access the final edited and published work see https: //doi.org/10.12968/bjon.2019.28.4.250. 4

Study Method
An integrated literature review was undertaken using a systematic approach using the following online-data bases; Medline, Scopus, PsycInfo, and Cumulative Index to Nursing and Allied Health Literature (CINAHL). The search was conducted between December 2015 and February 2017 and key words included; mentors, mentoring, students, nurses, nursing, assessment, competence, failure, failing, failure to fail. Boolean operators ('And', 'Or', 'Not') were used to combine or exclude keywords in addition to truncation (Craig and Smyth 2011). Inclusion and exclusion criteria were applied to focus the search and ensure inclusion of relevant papers (Aveyard 2014) (see Table 1).
INSERT were excluded as the evidence on this topic had already been reviewed, thus a contemporary perspective was sought. Database searching was combined to remove duplicates. Only a limited number of eligible papers were identified therefore, secondary searching was conducted via hand searching of journals and reference lists (Aveyard 2014). Following screening of titles, 12 relevant papers were identified. Further screening of the abstracts resulted in 7 further papers being excluded as not relevant to the review question or did not meet the inclusion criteria. Full text reading of the remaining 5 papers was undertaken prior to extraction of data. The search strategy and results are shown in Figure 1 (PRISMA diagram). Records used = 0

INSERT FIGURE 1 (PRISMA DIAGRAM)
Reason for exclusion = no relevance, based on failing dentistry students, social workers, foreign papers, based on assessment and competence of mentors, not failing to fail.
Three of the papers were qualitative exploratory studies ( Table 2 (Study Characteristics table).
INSERT  Three themes emerged from this study:-1) The complexity of assessing students.
2) The difficulty of assessing attitudes.
* Assessing clinical performance is complex, especially when the student performance is border-line.
* Mentors have difficulties with the assessment of students' attitudes.
* Deciding to fail/refer an assessment is stressful, but the response of the student makes it more difficult.
* Organisational, professional and personal factors impact on confidence when referring poorly performing students.
(+) Semi-structured interviews facilitated further exploration of issues initially raised in the focus group (some mentors were unwilling to discuss their sensitivities in a group) increasing validity.
(+) Interviews arranged at a time and place convenient for the participant's.
(+)All mentors had at least mentored at least three pre-registration students.
(+) A range of nursing professionals; ward nurses, district nurses and health visitors were involved, increasing validity. Five themes emerged from this study:-1) Failing a student is a difficult process.
2) Both academic and emotional support is required for students, preceptors and faculty advisors.
3) There are consequences for programs, faculty, and students when a student has failed a placement.
4) Occasionally, personal, professional, and structural reasons exist for failing to fail a student.
5) The reputation of the professional program can be diminished as a result of failing to fail a student.
(+) Credibility was achieved by comparing the analyses of experienced researchers.
(+) Used a wide range of informants increasing validity.
(+) Fittingness was enhanced by collection of data from different settings.
(-)Not all mentors had past experiences with a student who had failed or who was at risk of failing, therefore as they had not experienced being in that situation, this could affect credibility.
Brown.L, Douglas. V, Garrity.J, Shepherd What influences mentors to pass or fail students.

Quantative
Non-experimental design using a 29-item questionnaire to 1) Could not prove concern for failing the student were valid.
(+)large scale interview 1790 response rate out of 4341 mentors (participants) -across 6 health boards in 2) Felt unable to complete action plans and documentation objectively for fear of upsetting the student when wanting to fail them.

3) Mentors felt unsupported by link lecturers
(+) Wide selection of health professionals selected to participate in the study.
Scrutiny of each paper was undertaken to identify key findings, outcomes and limitations. Potentially relevant concepts and ideas were highlighted and these were then grouped identifying re-occurring themes and concepts within and across the 12 studies. Three main themes emerged; the mentors' relationship with the university, documentation, and psychological and emotional impact.

Discussion of themes
Mentors' relationship with the University.
In the UK, universities are responsible for ensuring mentors are adequately supported to support the student via specified roles such as link lecturers and reported the same concern however this was not informed by past experiences.In addition, although Jervis and Tilki (2011) noted that this theme emerged from their study findings it was underexplored in their paper. Lawson (2010), Larocque and Lahanga (2013) and Brown et al. (2012) also reported mentor concerns regarding the support they received from the university; a common theme highlighted in earlier research (Duffy 2003;Scholes andAlbarran 2005, Skingley et al. 2007). However, it is not clear whether all the 13 participants in Larocque and Lahanga's (2013) study had past experiences of mentoring a student who had failed or was at risk of failing. In contrast, Brown et al. (2012) found that only a small number of mentors in their study (144 out of 1790 participants) reported a problem with the relationship between mentors and the university. 68.2% (of 1790) mentors indicated that the academic practice partner (known as link lecturers in the UK) as being their first choice to approach if they encountered an under-achieving student. Mentors were also asked to rate the support they received from the university and although 9.5% (130) rated this as poor, on the whole (1243 out of 1373) mentors were satisfied with the support they received (22.7% satisfactory; 38.3% good, 29.5% very good). This mixed methods study accessed mentors from 6 different publically funded NHS health boards. Almost 23% (of 277) mentors who had passed a failing student reporting they had received poor support compared to 6% (of 1065) mentors who had failed a failing student.

Documentation when failing a student
Mentors must keep sufficient records to support and justify their decision of whether the student is fit to pass or fail (RCN 2007) and decisions made to fail a student must comply with professional guidance and document an appropriate action plan giving constructive feedback (NHS 2013). Stuart (2007) had previously highlighted that in some cases a decision to fail a student had been over-ridden by the examination board at the university due to poor documentation of the mentor. Brown et al. (2012) and Lawson (2010)  (2011) who also noted the pressure placed by students on mentors. These emotional demands were also cited earlier by Duffy (2003). It is likely that a poor relationship with academic partners would impact upon these anxieties (Brown et al. 2012), further undermining mentor confidence. Black et al. (2014) reported that some mentors were angry with other (previous) mentors for failing to address student performance earlier.

Discussion
The issue of 'failing to fail' features regularly in the UK literature and is informed by on-going concerns regarding assessment of student nurse competence and the implications for professional practice and patient safety (Burden et al 2017, Black et al. 2014, Clark et al. 2013. This review has shown that although there is some evidence regarding this topic and valid concerns regarding this issue in clinical practice, overall the quality and extent of the existing evidence does not fully explain this particular phenomena. Nor does it account for the substantive number of mentors who are confident in their assessment of students. Hunt et al. (2012) found that a larger number of students failed theory when compared to practice (a ratio of 4:1) and concluded that this offered some support to the argument that mentors were failing to fail underperforming students. However, assessing competence in practice is complex and subjective occurring in a high pressured environment in which individual mentor judgements may not lend themselves well to competency based assessment processes (Burden et al. 2017). Mentors have substantive professional knowledge and expertise in assessing students in clinical practice and this must be recognised and valued by their partner universities (National Nursing Research Unit (NNRU) 2013). When mentor decisions are over-turned or challenged by the university this can impact upon trust and respect, damage relationships and undermine confidence. Addressing perceived lack of support and enhancing confidence in decision-making is key to sustaining this relationship; issues also identified in a previous review (of literature 2003-11) by Elliott (2016). Mentors must liaise with university representatives if a student is at risk of being failed or there are concerns regarding competence (NHS 2013). This is essential to ensure that the psychological and emotional issues attached to failing a student nurse does not discourage some mentors and that additional support and guidance with decisionmaking is provided. Both Black et al. (2014) and Jervis et al. (2011) indicated that this may be an area where practical support for mentors could be provided by peers and, or the university partner.
Although failing a student nurse in practice may be difficult, it is a necessary requirement of mentors and the majority remain confident in their ability to effectively undertake this role. Mentor concerns regarding failure to fail may be a consequence of other factors including mentor experience, relationship with the student and competing clinical demands (Elliott 2015). It is important that new mentors are supported in the early stages of their mentoring career until confidence and competence is established. Being an experienced nurse may not necessarily equate with 'experienced mentor', it could be argued that these are distinct roles and the skills and competencies required of mentors are different. The quality of the mentor may well depend upon whether this particular role has been chosen or not (Andrews et al. 2010;NNRU 2013). Since the guidance was issued by the NMC in 2008, assessment has become a more substantive component of the mentor role (Andrews et al. 2010) and this role is increasingly being delivered within the context of on-going workforce retention issues and increased workloads and demands (Buchan et al 2017).
Future students in the UK will be supervised in clinical practice by any registered health or social care professional (NMC 2017) as opposed to the current situation where students are supervised and assessed by a registered nurse with mentor status. The RCN (RCN 2017) has supported the NMC proposals and the benefits of inter-professional learning whilst highlighting the need for safeguards to ensure practice supervisors have the necessary knowledge, skills and experience to observe and supervise the student in practice (RCN 2017). Of concern however, is that the current requirement of a national training level for mentors and the maintenance of a mentor register is no longer advocated. Whilst any registered health care professional can supervise a student nurse, assessment will be solely the responsibility of a registered nurses undertaking one of two roles; 'nominated practice assessor' or 'nominated academic assessor' (NMC 2018). Roles and responsibilities will need to be clarified (RCN 2017). There does appear to be some merit in the splitting of these roles as it has long been recognised that combining supervision and assessment, with the responsibility for both traditionally falling to one individual mentor, does cause tension for many (Neary 2000).
This review has indicated that further focus is required on documentation. For universities, establishing stronger supportive relationship with their partners in practice is fundamental to safe and effective student learning and assessment and this should include proactive and pre-emptive approaches such as 'early warning systems' to identify potential difficulties. 'Failing to fail' does warrant further research and investigation that explores the range and extent of this phenomena in a systematic way. Key to further research in this area is ensuring that the perspectives of students and academic staff are also included in this research and debate. As this review focused on primary evidence (rather than other literature reviews), few research papers were identified to inform the review. Equally, using the term 'mentor' may well have limited the search to UK only studies as other terms may be more culturally appropriate for supervisory and assessment arrangements in non UK countries.

Conclusion
There is some but limited evidence that the phenomena of 'failing to fail' a student nurse in the UK but arguably this is not as pervasive as previously suggested. Whilst this may be a concern and challenge for some mentors in practice, the majority are competent and comfortable in the decisions they make regarding student competence. Many universities have established and effective approaches to shared decision making regarding student competence in practice that are embedded in policy and practice. These should be and can be sufficient to address the assessment of competence in practice.