Journal of Psychiatric and Mental Health Nursing, 2014, 21, 797–805

Barriers to the reporting of medication administration errors and near misses: an interview study of nurses at a psychiatric hospital C . H AW 1 M R C P M R C P s y c h , J . S T U B B S 2 M P h a r m S G . L . D I C K E N S 3 RMN BSc ( Ho n s ) MA Ph D

MSc

&

1

Professor of Mental Health and Consultant Psychiatrist, 3Professor of Psychiatric Nursing and Research Manager, University of Northampton School of Health, St Andrew’s Academic Centre, King’s College London Institute of Psychiatry, 2Research Assistant, St Andrew’s Healthcare, Northampton, UK

Keywords: errors, medicine

Accessible summary

administration, medicine management, mental health nurses, near miss

•

Correspondence: G. L. Dickens St Andrew’s Academic Centre King’s College London Institute of Psychiatry St Andrew’s Healthcare Billing Road

•

Northampton Northants NN1 5DG UK E-mail: [email protected]

•

Accepted for publication: 31 January 2014 doi: 10.1111/jpm.12143

•

Medication administration errors and near misses are common including in mental health settings. Nurses should report all errors and near misses so that lessons can be learned and future mistakes avoided. We interviewed 50 nurses to find out if they would report an error that a colleague had made or if they would report a near-miss that they had. Less than half of nurses said they would report an error made by a colleague or a near-miss involving themselves. Nurses commonly said they would not report the errors or near misses because there was a good excuse for the error/near miss, because they lacked knowledge about whether it was an error/near miss or how to report it, because they feared the consequences of reporting it, or because reporting it was too much work. Mental health nurses mostly report similar reasons for not reporting errors and near misses as nurses working in general medical settings. We have not seen another study where nurses would not report an error or near miss because they thought there was a good excuse for it. Training programmes and policies should address all the reasons that prevent reporting of errors and near misses.

Abstract Medication errors are a common and preventable cause of patient harm. Guidance for nurses indicates that all errors and near misses should be immediately reported in order to facilitate the development of a learning culture. However, medication errors and near misses have been under-researched in mental health settings. This study explored the reasons given by psychiatric nurses for not reporting a medication error made by a colleague, and the perceived barriers to near-miss reporting. We presented 50 nurses with clinical vignettes about error and near-miss reporting and interviewed them about their likely actions and about their views and perceptions. Less than half of participants would report an error made by a colleague (48%) or a near-miss involving themselves (40%). Thematic analysis revealed common themes for both not reporting an error or a near-miss were knowledge, fear, burden of work, and excusing the error. The first three themes are similar to results obtained from research in general medical settings, but the fourth appears to be novel. Many mental health nurses are not yet fully convinced of the need to report all errors and near misses, and

© 2014 John Wiley & Sons Ltd

797

C. Haw et al.

that improvements could be made by increasing knowledge while reducing fear, burden of work, and excusing of errors.

Introduction Medication errors, including those made in prescribing, dispensing, and administration are a common and preventable cause of patient harm (Bates 1996). They are thought to affect 2–14% of all hospital inpatients, and result in 7000 deaths in the United States each year (Williams 2007). Medication administration errors accounted for 34% of errors in one large US study conducted in medical and surgical units (Bates et al. 1995). Rothschild and colleagues (Rothschild et al. 2007) examined rates of medication errors based on chart review and staff reports in a US 172-bed academic psychiatric hospital; they concluded that serious medication errors were as common in this setting as in general hospital settings. Studies in which nurses are observed by a researcher while administering medication indicate that administration rates are probably much higher than estimates made from error reports. Such studies conducted in general hospitals have yielded error rates varying between 3.5% and 27% of doses (see for example Barker et al. 2002). Although much medication error research has been conducted in general hospital settings, there is now a growing body of knowledge about their nature, frequency, severity, and causes in inpatient psychiatric settings, though much less is known about their occurrence in community psychiatry (Ito & Yamazumi 2003, Maidment et al. 2006, Procyshyn et al. 2010). In an observational study of medicine administration conducted on two wards for older psychiatric inpatients, the error rate was high at one in four doses (Haw et al. 2007). The most common error types were unauthorized tablet crushing, omission of a drug without a valid clinical reason, and failure to record administration of a medicine. Interestingly, none of the errors detected by the researchers were reported by nursing staff using the hospital’s medication error reporting system. Fortunately, most errors were of doubtful or minor significance. In addition to medication errors, consideration also needs to be given to so-called ‘near misses’. These are medication administration incidents which did not cause any harm to the patient but may have potential to cause harm, for example, almost administering the wrong dose of a medicine but becoming aware of the error prior to administration. A near miss can be conceptualized as an error that happened but did not reach the patient (Institute for Safe Medication Practices 2009). Near misses are considered to be a rich source of information that can inform error prevention strategies but, like medication errors, are thought to be grossly under-reported (Williamson 2009). 798

At the broadest level, the barriers that interfere with nurses reporting medication administration errors have been described as either organizational or individual (Leape 2002). The reasons nurses do not report medication administration errors have been studied in general hospital settings. In a survey of 1384 nursing staff working in 24 US general hospitals, factor analysis revealed four main factors to be responsible for the failure to report administration errors: fear, disagreement over whether or not an error had occurred, administrative responses to medication errors, and the effort involved in the reporting process (Wakefield et al. 1996). Fear (Osborne et al. 1999, Mayo & Duncan 2004, Chiang & Pepper 2006, Mrayyan et al. 2007, Koohestani & Baghcheghi 2009, Chiang et al. 2010, Hartnell et al. 2012) and administrative burden (Almutary & Lewis 2012; Chiang & Pepper 2006; Hartnell et al. 2012) are commonly reported as barriers to reporting across a number of studies. Interestingly, Ulanimo et al. (2007) found that the fear associated with reporting was largely about the potential reaction of managers and colleagues but was largely not related to a specific fear of disciplinary action or job loss. Other factors associated with self-reported failure to report errors in a large study in Taiwan included previous experience of making errors, congruence of attitudes about reporting errors made by self and by colleagues, underestimating error rate relative to colleagues, and poorer professional development (Chiang et al. 2010). Elsewhere, lack of encouragement by management to report (Sanghera et al. 2007), lack of feedback once reported (Armitage et al. 2010), and lack of knowledge about policy and procedure (Ulanimo et al. 2007) have been cited. An interesting finding by Baker (1997) suggested that one strategy adopted by nurses around error reporting involved them redefining or reclassifying errors as ‘not errors’ once action had been taken: for example an omitted medicine was no longer an error once the dose had been administered. In one study, nurses (and pharmacists) were more likely than doctors to report medication errors, and equally likely to increase their reporting rate as the severity of vignette scenarios increased (Sarvadikar et al. 2010). Despite increasing knowledge about barriers to reporting in general evidenced by the recent literature outlined above, we are not aware of any studies examining the barriers to reporting medication errors and near misses in inpatient psychiatric settings. Medication administration in inpatient psychiatry differs substantially from that in general hospitals, since in psychiatry use of intravenous drugs and infusions is virtually non-existent and most © 2014 John Wiley & Sons Ltd

Barriers to error reporting

drugs are administered orally, though some medicines are given intra-muscularly, for example depot antipsychotics. Psychiatric patients may refuse or conceal medication or may need considerable persuasion to take their medicines. Sometimes sedative medication has to be administered urgently in order to calm a patient and prevent violence to self or others (so-called rapid tranquillization). The aims of this study were to explore the reported reasons given by inpatient psychiatric nurses for not reporting a medication error made by a colleague. In addition, we also sought to determine the perceived barriers to near-miss reporting.

Method Setting St Andrew’s is a UK charity providing tertiary health care to around 900 inpatients with a wide variety of mental health problems. Almost all patients are referred by and funded by the National Health Service (NHS). Many of the patients have complex psychiatric needs, some exhibit challenging behaviour and most are cared for in conditions of low or medium security. Almost all patients are detained under the Mental Health Act of England & Wales 1983 (amended in 2007). The current study was conducted at the Northampton site among nurses working in adolescent and adult mental health care pathways.

Reporting of medication errors and near misses In the UK, the regulatory body for all nurses is the Nursing and Midwifery Council (NMC). The NMC has provided written Standards for Medicines Management (Nursing and Midwifery Council 2010). The guidelines state that: ‘As a registrant, if you make an error you must take any action to prevent any potential harm to the patient and report as soon as possible to the prescriber, your line manager or employer (according to local policy) and document your actions’ (p. 37: standard 24. Guidance contained in the document clarifies that this is to include near misses). Guidance on recording administration states ‘you must make a clear, accurate and immediate record of all medicine administered, intentionally withheld or refused by the patient, ensuring the signature is clear and legible’ (p. 7: standard 8). In the current study setting, the hospital’s medicine policy stipulates that all medication errors and near misses should be reported as soon as possible after the event by the person detecting the error/near miss and using an electronic error reporting mechanism called the Medi-Error system (see Haw & Cahill 2011 for a full description). This computerized reporting system consists © 2014 John Wiley & Sons Ltd

of a series of drop-down boxes and free text which takes 10–15 min to complete once the reporting nurse is familiar with the reporting process. The Medi-Error system sits within the widely used RiO (CSE Healthcare Systems) electronic notes system and does not require a separate log-in. Medi-Error reports are automatically copied to pharmacy and the relevant senior nurse manager. The nurse manager may make further inquiries about the medication event, interview nursing staff, and take remedial action which might include referring the nurse concerned to the in-house Safe Medicines Management training course. In some cases the manager may suspend the nurse from administering medication until they have attended the course and have demonstrated their competence. Rarely would any disciplinary action be taken against a nurse since the organizational aim is to encourage a no-blame culture regarding errors and their reporting, most errors being multi-factorial in cause and rooted in system failures.

Study approval The study was part of a larger training needs analysis designed to ascertain baseline information about current nursing medicines management practice. We ascertained that NHS research ethics approval was not required for the study. However, the authors conducted the study within the codes of standards, performance, and ethics of their respective regulatory bodies. The study design and methodology was reviewed and approved by senior nursing management and by the head of clinical effectiveness. Interviews were conducted confidentially.

Recruitment process The potential participants were all of the nursing staff on the adolescent and adult male and female inpatient wards. We first contacted ward managers and, after providing them with written information about the study, sought their permission to approach staff. Potential participants had to have administered at least one drug round within the last 6 months. They were provided with an information sheet about the study and any questions they had were answered by a member of the study team.

Interview schedule The interview schedule was devised by the study team. Initially, C.H. used clinical experience and studied the NMC Standards for Medicines Management (2010), the Code of Practice (Department of Health 2010) and the hospital’s relevant policies to develop two clinical vignettes. The vignettes were discussed by a multi799

C. Haw et al.

disciplinary reference group (including J.S and G.D.), and the feedback was used to amend them. The schedule was piloted with six nurses, and the results discussed by all three authors and a nursing development officer with responsibility for medicines management training. Modifications to the schedule were then made in the light of these discussions, and a final format for the interview schedule was agreed. Each interview took 30–40 min in total; some aspects of the interview were not about medicines errors and are not reported on in this paper. The interview was designed to elicit participants’ free response about vignettes and therefore we did not ask them to respond to predetermined categories. Participants were given general prompts and cues, for example ‘is there anything else you would do in this situation’, in order to elicit maximum information. Vignette 1. You are doing the 8am medication round. You notice that the nurse who did the 10pm medication round yesterday has failed to sign for an item on a patient’s chart – Olanzapine 10 mg nocte.

Each participant was asked: (1) Is this a medication error? (2) Would you report it as such? (3) If you would not report this error why not? Vignette 2. You have two young men on your ward. One is called John Hunter and the other Jonathan Hunt. John Hunter is on Olanzapine 20 mg nocte and Jonathan Hunt is on Quetiapine 600 mg nocte. You are doing the 10 pm medication round and you almost give John Hunter’s medication to Jonathan Hunt.

Each participant was asked: (1) Is this a near miss? (2) Would you report it as such? (3) If you would not report this why not?

Data analysis Participants’ responses were transcribed verbatim and later typed up. Responses were subject to a descriptive analysis using qualitative methodological process, namely a theoretical thematic analysis as described by Braun & Clarke

(2006). We adopted a realist standpoint, eschewing a more interpretative reading of the data, and sought to understand participant’s responses in a straightforward way where what they said reflected their experiences and opinions. Initial coding of lower level themes was undertaken by C.H. who assigned a preliminary code to each response. These were validated by G.D. and, in cases of disagreement, coding was discussed by all three authors and consensus achieved. We then went back to the original transcripts in order to check that our codes matched the data and also to quantify the frequency of each theme in the data set. At this point, themes were grouped together into emergent high level themes that were each agreed by the three authors. We were informed by previous research in the area but were also alert to newly emergent themes given the lack of research particularly in psychiatric settings.

Results Details of participants Of the 52 nurses who were approached, 50 (96%) agreed to be interviewed. The remaining two cited pressure of work as the reason for their non-participation. Most (n = 34; 68%) were staff nurses or senior staff nurses, the rest being ward managers or their deputies (n = 16; 32%). Participants had been qualified as psychiatric nurses for a median of 7 years (range 0.3–38 years) and 40 (80%) administered at least one medicine round on all or most days when on duty.

Lower level themes Vignette 1: Barriers to reporting a medication administration error made by a colleague. In total, 26 (52%) participants said that they would not use the Medi-Error system to report a missing signature error made by a colleague. The reasons they gave for not reporting the error are given in Table 1. Some participants gave more than one reason.

Table 1 Reasons why nurses would not report a medication error made by a colleague (n = 26)1 High-level theme Excusing

Fear Knowledge Burden

n (%) 27 (90)

7 (27) 12 (46) 3 (12)

Sub-theme

n (%) 2

Passing the buck – e.g., ‘I would tell my line manager and they can decide what to do’ It was just a one off and won’t happen again. If it did happen again then I would report it There could be a good reason why they made the error e.g., ‘It was just a momentary oversight’ It is a very common problem and therefore not worthwhile reporting Empathy and wanting to support and not punish a colleague Not wanting to be labelled a ‘tell-tale’ or appearing rude Not wanting to cause conflict or upset the nursing team. Unaware of the Medi-Error reporting system and procedure Uncertainty as to whether or not the event is an error Pressure of work and too much effort involved in the reporting process

7 (27) 6 (23) 6 (23) 4 (15) 4 (15) 5 (19) 2 (8) 3 (12) 2 (8) 3 (12)

Some participants gave more than one reason thus % total > 100%. 2 N.B., under hospital policy the nurse discovering the error has a duty to complete a Medi-Error form. 1

800

© 2014 John Wiley & Sons Ltd

Barriers to error reporting

Table 2 Reasons why nurses would not report a near-miss (n = 30)1 High level theme Excusing

n (%) 5 (17)

Fear

18 (60)

Knowledge

11 (37)

Burden

15 (50)

1

Sub-theme1

n (%) (%)

No harm done to the patient It was just a one off and won’t happen again Forgetfulness Fear of loss of status, stigma and/or trust with peers Fear of disciplinary action including dismissal Fear of litigation Problems to do with the definition of a near-miss Unaware of the Medi-Error reporting system Pressure of work and too much effort involved in the reporting process Not a worthwhile activity

3 (10) 1 (3) 1(3) 9 (30) 8 (27) 1 (3) 9 (30) 2 (7) 10 (3) 5 (17)

Some participants gave more than one reason thus % total > 100%.

Seven (27%) said that they would tell other senior staff about the error but not officially report it themselves, for example ‘I would report it to the nurse manager and doctor and leave it to them’ (N.B. hospital policy dictates that the person who discovers the error should report it using the Medi-Error system). Six (23%) said that the omission was just a one off and probably would not happen again, though most clarified that if it was a repeated problem then they would report it (e.g. ‘I would give people the benefit of the doubt for the first omission and not report. If it happens again then do’) and another six said there might be a good reason for why the nurse had made the error (e.g. ‘Sometimes people give valid reasons for not signing a medicines card’). Other themes were concerned with not wanting to upset the nurse or nursing team (e.g. ‘if you become too hard on colleagues it can destroy your relationship with them’ and ‘I would feel guilty going behind their back, especially if they are a higher grade than you’) or not wanting to be labelled as someone who got other people into trouble (e.g. ‘you don’t want to be a tell-tale’). Some cited pressure of work or lack of awareness about how to report an error as reasons for not reporting it. A small number (n = 4, 15%) said that this type of error was very common and therefore not worthwhile reporting (e.g. ‘It’s such a faff’ in reference to the reporting process). Vignette 2: Barriers to reporting a near miss. A total of 21 (42%) participants would not report a near-miss event involving themselves, and a further 9 (18%) were uncertain. The reasons they gave are given in Table 2. Some participants gave more than one reason. The four most commonly cited barriers to reporting were: pressure of work and the effort required to use the Medi-Error reporting system (n = 10) (e.g. ‘It’s tedious, the paperwork. There is no time to have a cup of tea even’ and ‘Hassle, the time it takes to do an event form. If we did all the paperwork we are expected to do I would never see patients’); not being certain as to exactly what constituted a near miss (n = 9) (e.g. ‘So it almost happened. If it was a proper error and it wasn’t reported that would be an error’ and ‘Technically I © 2014 John Wiley & Sons Ltd

have not made a mistake’); fear of loss of face with their peers (n = 9) (e.g., ‘Being scared that other people will be thinking you are not competent’ and ‘How do the other team members take it if you get labelled as the one who makes mistakes’; and fear of disciplinary action (n = 8) (e.g. ‘I would be fearful of the consequences and being disciplined or punished’ and ‘I did a near-miss years ago and was not allowed to do medicines for three months’. Less common but nonetheless interesting themes included reporting a near miss as not being a worthwhile activity (n = 5) (e.g. ‘You could probably do something more useful’ and ‘They don’t see it as worthwhile’); and that no harm had been done to the patient (n = 3) (e.g. ‘People might see it as an error, though no harm was done to the patient’ and ‘No negative outcome from it’).

High-level themes Thematic analysis revealed that the main barriers to reporting both errors and near misses given by participants in this study can be summarized under four headings: excusing, knowledge, fear, and burden. Excusing This theme encompassed sub-themes in which nurses identified reasons not to report errors or near misses. This usually involved excusing others’ behaviour in relation to error reporting and their own in relation to near-miss reporting. There was one exception: passing the buck described nurses excusing their own behaviour when not reporting errors made by another nurse. Errors or near misses could be excused on the basis of outcome (‘no harm caused’), frequency (this could be rarity or commonality), or perceived cause (‘forgetfulness’ or ‘momentary oversight’). Fear This theme encompassed a number of varied concerns including both official (‘disciplinary action’, ‘litigation’) 801

C. Haw et al.

and unofficial sanction (‘loss of face with colleagues’, ‘stigma’). Some nurses did not want their actions regarding medications errors or near misses to be the cause of conflict within the team. Knowledge The knowledge theme comprised sub-themes referring to knowledge deficits in two key areas: lack of knowledge of process (not knowing how to report an error) and uncertainty about definitions (e.g. What is an error? What is a near miss?). Burden Finally, the theme of burden encompassed sub-themes that emphasized the already pressured nature of the job and the lack of available time to report each incident. Some participants noted that reporting incidents was not worthwhile.

Discussion Through the use of semi-structured interviews, we explored whether mental health nurses would be willing to report medicines errors made by other nurses or near misses made by themselves. Slightly less than half of participants said they would definitely report a medication error made by a colleague (24; 48%) or a near miss made by themselves (20; 40%), despite the fact that all but a handful were aware of the official hospital mechanism for medication error reporting. Thematic analysis revealed four high level themes that accounted for the explanations given by nurses for failing to report such incidents (Fear, Knowledge, Burden, and Excusing). Fear is a commonly occurring theme reported by nurses in research about error reporting conducted in general medical settings (Wakefield et al. 1996, Osborne et al. 1999, Mayo & Duncan 2004, Chiang & Pepper 2006, Mrayyan et al. 2007, Koohestani & Baghcheghi 2009, Chiang et al. 2010, Hartnell et al. 2012). It is perhaps therefore unsurprising that fear-related issues were the most commonly cited reasons for not reporting a selfperpetrated near miss. It is more surprising, however, that respondents reported fear given the precise nature of the study vignette in which the nurse successfully identifies the similarity of patient’s names and therefore avoids making a more serious wrong drug/ wrong patient error. The reporting of a near miss in this scenario would, we contend, in real life, be a real opportunity for other nurses to learn and to provide a basis for a bespoke intervention to reduce the likelihood of future reoccurrence. Fear-related reasons for not reporting a near miss involved both fear of official sanction (disciplinary action) and of unofficial sanction 802

(loss of face, stigmatization by peers); this finding has a precedent in the research literature in general medical settings (Ulanimo et al. 2007). In relation to reporting a colleague’s error fear was only the third most commonly cited reason for failure to report and was largely focused on fear of unofficial sanction (being seen as a tell-tale or disrespectful, or causing team conflict). These particular fear-related sub-themes do not appear to have been reported in the wider literature on barriers to error reporting. The reasons for this are unclear but may reflect values held to be important by mental health nurses such as maintaining good relationships with colleagues and being supportive (Haque et al. 2002, Stacey et al. 2011). An alternative explanation is that nurses’ responses in our study reflect that the clinical vignettes we offered them described scenarios with relatively little potential for harm. Accordingly, themes generated from responses about this particular scenario should not be assumed to be applicable in all circumstances including when errors or near misses are potentially severe. Knowledge-related themes are very commonly reported in studies of barriers to medicines administration reporting in general medical settings (Wakefield et al. 1996, Ulanimo et al. 2007, Chiang et al. 2010). Thematic analysis of interview data in the current study revealed a similar knowledge theme comprising sub-themes about knowledge of process and systems (how to use the Medi-Error reporting system) and knowledge of definition (when is an error an error? and when is a near miss a near miss?). The most commonly reported knowledge deficit in the current study was about the definition of a near miss, while knowledge deficits about reporting process and systems were less commonly reported. This suggests that the Medi-Error reporting system is generally well understood, but that more effort should be made to create a shared understanding around the importance of reporting all incidents. Burden of reporting is also commonly reported in the related literature (Almutary & Lewis 2012; Chiang & Pepper 2006; Hartnell et al. 2012), and it is unsurprising to find that this is also an issue in mental health inpatient settings. Mental health wards are busy and unpredictable environments with multiple administrative and clinical demands vying for the attention of nursing time; mental health nurses may routinely spend less than half of their working time in direct patient contact (Sharac et al. 2010). Nurses commonly report facing a ‘paperwork overload’, a recent survey of UK nurses found that 17.3% of all hours worked by NHS nurses in the United Kingdom were spent on non-essential paperwork (RCN 2013). The current study indicated that nurses more commonly reported burden issues in relation to reporting near misses than in relation to reporting a colleague’s error. This might suggest that near-miss reporting, or at least a near miss like that © 2014 John Wiley & Sons Ltd

Barriers to error reporting

described in our case vignette, may be viewed by some as a non-essential activity. Alternatively, the perceived importance of reporting such incidents may simply fail to reach a level at which the nurse deems it worthwhile to complete a form which on average will take 10–15 min. While knowledge of the Medi-Error system did not appear to prohibit reporting (see Knowledge above), it may be that it is still perceived as a burden. If this is the case, then we should expect to see greater willingness to report if the process is further streamlined. We also speculate that lack of encouragement by management to report (Sanghera et al. 2007) and lack of feedback once reported (Armitage et al. 2010) may contribute to some nurses’ conception of near-miss reporting as a burden although neither of these negative issues was mentioned by nurses in the current study. Excusing was an interesting emergent high level theme whose constituent sub-themes were very frequently cited by those nurses who would not report a colleague’s error. Sometimes the hypothetical colleague was excused on the basis of empathy; we speculate that this may be an alternative expression of the fear theme with concern about sanction projected onto the nurse who has made the error. However, quite commonly the error-making nurse was excused on the basis of the surmised reason for the error (‘momentary oversight’) or on the basis of the frequency of errors. The error might be excused either because similar errors are very common (and presumably by extension perceived as relatively harmless), or on the basis that the nurse’s own record of errors was previously unsullied (‘just a one-off’). This process of excusing errors differs somewhat from Baker’s (1997) conception of the reclassification of errors to ‘not errors’ through remedial action. In addition, some nurses also wished to excuse themselves from responsibility by passing the buck to a more senior colleague. A number of sub-themes within the high-level excusing theme have not to our knowledge been previously reported in widely accepted classifications such as that by Wakefield et al. (1996). In particular, belief that the error was just a one off and would not happen again, that there could be mitigating circumstances, and an expressed belief that this type of error is so common that it is not worth reporting.

Clinical implications In the United Kingdom, there has been a considerable focus on preventing adverse events for at least the past decade. The Department of Health (2010) report ‘An organisation with a memory’ identified that all incidents involving actual or potential injury, loss or damage should be reported immediately’ (p. 50) and decried that there is too often a ‘blame culture’ and not a ‘learning culture’ (p. 77). While © 2014 John Wiley & Sons Ltd

Excusing Errors

Reporting Burden

Medication Event Reporting Knowledge

Fear

Figure 1 Intervention focus areas to improve medication event reporting in mental health nursing

there have been substantial improvements in recent years our study suggests that the situation has not been solved entirely, and/or that nurses have not been sufficiently convinced that their efforts to act in accordance with learning culture values will be rewarded. Therefore, further efforts should be made to encourage a ‘no-blame’ learning culture within an organization in which the reporting of errors is encouraged and seen as a positive action. An understanding of medication error reporting by mental health nurses that is informed by the findings of the current study could assist in achieving the desired culture. Simply, we propose (see Fig. 1) that educationalists, managers, and practitioners should develop teaching, policy, and practice that challenges and reduces excusing behaviours, that reduces unrealistic fear about event reporting, that reduces the burden associated with reporting, and increases knowledge of errors and near misses and the importance of learning from them. Paradoxically, the success of such interventions might be measured in terms of an increase in reporting as indicated in the diagram. Currently, we are conducting a programme of work to review training and policy in light of these recommendations and will report on its content and success in due course. In brief, in relation to fear although sometimes an error will highlight the need for some additional medicines administration training, only in very rare instances should it result in disciplinary action, since most errors have multiple and systemic causes. Efforts should be made to ensure that this culture is instilled among all those involved in the process of medicines management including doctors and pharmacists. In relation to burden nurses and others need to work with information technology providers to make the errorreporting systems as quick and easy to use as possible. Previous studies in the current setting (Haw & Cahill 2011) have demonstrated that introduction of electronic MediError reporting can increase incident reporting. This process needs to be further refined so that, within a learning culture, reporting can be made so simple that it is both the expected norm and simple to achieve. This in turn has implications for training (knowledge) and nursing staff will 803

C. Haw et al.

need training on how to access and use the electronic Medi-Error reporting system so that they are confident as to how to report errors. In order to support the development of a learning culture, they may also benefit from regular refresher courses in medicines administration. Mentoring new nurses about medicines management, including the avoidance and reporting of administration errors, has been proposed as a useful way of improving medicines administration skills (Murphy 2012). Finally, our study has found that excusing describes a set of beliefs and behaviours that may provide a barrier to appropriate reporting of medication events. These should be supportively challenged in practice, in education, training, and policy.

Study limitations Our study has a number of limitations. First, it was conducted in a non-NHS specialist psychiatric inpatient setting and cannot be assumed to be generalizable to other psychiatric settings. However, all qualified nursing staff have been trained alongside (or met equivalent standards to) NHS colleagues, and the profile of these services is not dissimilar to NHS and other independent sector providers of medium and low secure services. We devised our own interview schedule which used only two brief vignettes and did not

References Almutary H.H. & Lewis P.A. (2012) Nurses’ willingness to report medication administration

employ a validated tool. However, the topic of barriers to error reporting in psychiatry has been relatively underresearched, and an exploratory design was appropriate in these circumstances. The clinical vignettes utilized in the study described incidents which were not immediately life threatening, and it should not be assumed that nurses’ responses would generalize across levels of severity.

Conclusion The current study is the first in the literature to describe the barriers that psychiatric nurses experience related to the reporting of medication errors and near misses. Like nurses in general hospital settings, they sometimes lack knowledge of precise definitions of near miss and error, fear the consequences of reporting errors, find reporting processes to be cumbersome and with sometimes little tangible benefit. Nurse leaders and managers need to persuade nurses that they can develop a true learning culture where every error or near miss is viewed as an opportunity and not a burden.

Acknowledgment We wish to thank the nursing staff at St Andrew’s, Northampton who so willingly gave of their time to participate in this study.

tion. ADE study group. The Journal of the

Haw C. & Cahill C. (2011) A computerized

American Medical Association 274, 29–34.

system for reporting medication events in psy-

Braun V. & Clarke V. (2006) Using thematic analy-

chiatry: the first two years of operation. Journal

sis in psychology. Qualitative Research in Psy-

of Psychiatric and Mental Health Nursing 18,

chology 3, 77–101.

308–315.

errors in Saudi Arabia. Quality Management

Chiang H.-Y. & Pepper G. (2006) Barriers to

Haw C., Stubbs J. & Dickens G. (2007) An obser-

in Health Care 21, 119–126. doi: 10.1097/

nurses reporting of medication administration

vational study of medication administration

QMH.0b013e31825e86c8.

errors in Taiwan. Journal of Nursing Scholar-

errors in old-age psychiatric inpatients. Interna-

Armitage G., Newell R. & Wright J. (2010)

ship 38, 392–399.

tional Journal of Quality in Health Care 19, 210–216.

Improving the quality of drug error reporting.

Chiang H.-Y., Lin S.-Y., Hsu S.-C., et al. (2010)

Journal of Evaluation in Clinical Practice 16,

Factors determining hospital nurses’ failures in

Institute for Safe Medication Practices (2009)

1189–1197.

reporting medication errors in Taiwan. Nursing

ISMP survey helps define near-miss and

Outlook 58, 17–25.

close call. ISMP Medication Safety Alert!

Baker H. (1997) Rules outside the rules for administration of medication: a study in New South

Department of Health (2010) An Organisation

Acute Care Edition, September 24th. Available

Wales, Australia. Image Journal of Nursing

with a Memory. Report of an Expert Group on

at: http://www.ismp.org/newsletters/acutecare/

Scholarship 29, 155–158.

Learning from Adverse Events in the NHS. The

archives/Sep09.asp#24

Stationery Office, London.

2013).

Barker K.N., Flynn E.A., Pepper G.A., et al. (2002)

(accessed

14

July

Medication errors observed in 36 health care

Haque M.S., Nolan P., Dyke R., et al. (2002) The

Ito H. & Yamazumi S. (2003) Common types of

facilities. Archives of Internal Medicine 162,

work and values of mental health nurses

medication errors on long-term psychiatric

1897–1903.

observed. Journal of Psychiatric and Mental

care units. International Journal of Quality in

Bates D.W. (1996) Medication errors. How common are they and what can be done to

Health Nursing 9, 673–680. Hartnell N., MacKinnon N., Sketris I., et al.

Health Care 15, 207–212. Koohestani H.R. & Baghcheghi N. (2009)

(2012) Identifying, understanding and overcom-

Barriers

ing barriers to medication error reporting in

administration errors among nursing students.

Incidence of adverse drug events and potential

hospitals: a focus group study. BMJ Quality

Australian Journal of Advanced Nursing 27,

adverse drug events. Implications for preven-

Safety 21, 361–368.

66–74.

prevent them? Drug Safety 15, 303–310. Bates D.W., Cullen D.J., Laird N., et al. (1995)

804

to

the

reporting

of

medication

© 2014 John Wiley & Sons Ltd

Barriers to error reporting

Leape L.L. (2002) Reporting of adverse events. New England Journal of Medicine 347, 1633–

a comprehensive review. CNS Drugs 24, 595–

interaction on psychiatric inpatients wards:

609.

a literature review. International Journal of

RCN (2013) Nurses spend 2.5 million hours a

1638. Maidment I., Lelliott P. & Paton C. (2006) Medi-

week on paperwork – RCN survey. Available

How do nurses cope when values and practice

cation errors in mental healthcare: a systematic

at: http://www.rcn.org.uk/newsevents/press

review. Quality and Safety in Healthcare 15,

_releases/uk/cries_unheard___nurses_still_told

409–413.

_not_to_raise_concerns (accessed 29 August

Mayo A.M. & Duncan D. (2004) Nurse percep-

Nursing Studies 47, 909–917. Stacey G., Johnston K., Stickley T., et al. (2011) conflict? Nursing Times 107, 20–23. Ulanimo V.M., O’Leary-Kelly C. & Connolly P.M. (2007) Nurses’ perceptions of causes of

2013).

medication errors and barriers to reporting.

tions of medication errors: what we need to

Rothschild J.M., Mann K., Keohane C.A.,

know for patient safety. Journal of Nursing

et al. (2007) Medication safety in a psychiatric

Journal of Nursing Care Quality 22, 28–

Care Quality 19, 209–217.

hospital. General Hospital Psychiatry 29, 156–

33.

Mrayyan M.T., Shishani K. & Al-Faouri I. (2007)

Wakefield D.S., Wakefield B.J., Uden-Holman T.,

162.

Rate, causes and reporting of medication errors

Sanghera I.S., Franklin B.D. & Dhillon S. (2007)

et al. (1996) Perceived barriers in report-

in Jordan: nurses’ perspectives. Journal of

The attitudes and beliefs of healthcare profes-

ing medication administration errors. Best

Nursing Management 15, 659–670.

sionals on the causes and reporting of medica-

Practice in Benchmarking Healthcare 1, 191–

Murphy M.M. (2012) Mentoring students in medicines management. Nursing Standard 26, 51–56. Nursing and Midwifery Council (2010) Standards

tion errors in a UK Intensive care unit.

197. Williams D. (2007) Medication errors. Journal of

Anaesthesia 62, 53–61. Sarvadikar A., Prescott G. & Williams D. (2010)

Royal College of Physicians Edinburgh 37,

for Medicines Management. NMC, London.

Attitudes to reporting medication error among

Osborne J., Blais K. & Hayes J. (1999) Nurses’s

different healthcare professionals. European

Williamson S. (2009) Reporting medication errors

perceptions: when is it a medication error?

Journal of Clinical Pharmacology 66, 843–853.

and near-misses. In: Medication Safety: An

Nursing Administration 29, 33–38.

doi: 10.1007/s00228-010-0838-x.

Essential Guide (eds Courtenay, M. & Griffiths,

Procyshyn R.M., Barr A.M., Brickell T., et al. (2010)

Medication

errors

© 2014 John Wiley & Sons Ltd

in

psychiatry:

Sharac J., McCrone P., Sabes-Figuera R., et al. (2010)

Nurse

and

patient

activities

and

343–346.

M.), pp. 155–171. Cambridge University Press, Cambridge.

805

Copyright of Journal of Psychiatric & Mental Health Nursing is the property of WileyBlackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.