Experiences of peer messengers as part of a professional accountability culture change program to reduce unprofessional behaviour: a cross-sectional study across eight hospitals

Keywords: hospitals, informal feedback, organisational culture, peer messengers, professional accountability, professionalism, speaking up, unprofessional behaviour.

Positive organisational cultures are central to delivering safe care to patients.¹ Unprofessional behaviour between staff, ranging from incivility to physical or sexual harassment, is damaging to organisational culture and negatively impacts the way health care professionals work and the outcomes of care provided.^2–8 Professional accountability programs are one form of organisational intervention designed to address unprofessional behaviour. The Promoting Professional Accountability Program developed by Vanderbilt University Medical Center has been at the forefront of these initiatives; it aims to promote senior clinician behaviour change via supportive policies, surveillance tools for capturing allegations against medical practitioners, training and a tiered model of intervention.^9–11 This model begins with informal conversations with clinicians who have demonstrated less severe unprofessional behaviour. A trained peer messenger (i.e. another doctor of similar professional standing) conducts these informal ‘cup of coffee conversations’ with colleagues to raise awareness and provide an opportunity for clinicians to reflect on their behaviour. If clinicians demonstrate continued patterns of unprofessional behaviour disciplinary action is taken.

There is some evidence that using peer messengers to provide this informal feedback is effective in reducing recurrent unprofessional behaviour.¹¹ However, the way in which messages are delivered and how messengers are supported are likely to impact the effectiveness and sustainability of such programs. Additionally, the role of hierarchy and the characteristics of messengers, including their level of seniority, length of time in the hospital and their own behaviour, may influence their perceived credibility and influence how messages are received. Despite increased implementation of professional accountability programs in hospitals, little research on the motivations and experiences of peer messengers has been undertaken.

A whole-of-hospital professional accountability and culture change program in Australia, called Ethos, was developed drawing on elements of the Vanderbilt program.¹² The Ethos program includes all clinical and non-clinical staff, and aims to reduce unprofessional behaviour amongst staff, encourage speaking up and improve hospital culture. If unable to speak-up in the moment, staff can report unprofessional behaviour by co-workers using an online messaging system (Fig. 1). These ‘messages for reflection’ are triaged and then assigned to ‘Ethos messengers’ (EMs) to deliver during an informal conversation with the subject of the submission. The purpose is to provide non-punitive feedback about how their behaviour was perceived and encourage individuals to reflect on their behaviour. The Ethos program was implemented sequentially in eight hospitals in three states between 2017 and 2020. We conducted a study to investigate the experiences of EMs on average 2 years after implementation of the Ethos program to identify factors important for the sustainability of the messengers’ role in the program.

Fig. 1.  Ethos reporting and feedback process.

A cross-sectional survey was conducted as part of a larger program of research evaluating the Ethos program.^8,12–14 The project was approved by the Human Research Ethics Committee of St Vincent’s Hospital Melbourne (HREC/17/SVHM/237).

All past and present EMs in eight hospitals (two in Victoria, three in New South Wales, three in Queensland) were invited to complete an anonymous survey between 16 October and 25 November 2020. EMs were sent an email invitation, followed by a reminder email.

EMs are recruited from all staff groups and identified as suitable for the role by the Ethos leadership and local management teams based upon a demonstrated commitment to hospital values, professionalism, integrity and respect among peers. EMs receive a day of training focused on the purpose and elements of the program, skills in how to role model safe and positive behaviours, when and how to use the Ethos messaging system, the messenger role, how to deliver high quality feedback, completing follow-up tasks and availability of support and resources. The training was developed based on simulation debriefing principles,¹⁵ evidence of peer delivered professionalism messaging¹⁶ and emerging evidence on the negative impact of incivility in the workplace on performance and safety.⁵ Specific components of the EM training include: how to create a framework for the message based on the objective components of the submission, specifically the perception of the reporter and how the event affected them; how to prepare for the meeting to optimise the privacy of the recipient and the efficacy of the message; how to manage the emotion of the conversation; and how to ensure support and welfare for a distressed recipient. EMs receive continued support from an Ethos manager and attend monthly messenger support group meetings.

‘Feedback for reflection’ submissions are ‘triaged’ by three to six trained staff on a weekly basis,¹² to determine the best approach to providing feedback to the person who is the subject of the submission. ‘Feedback for reflection’ submissions include behaviours that undermine patient and staff safety (e.g. intimidating behaviour, derogatory remarks or jokes). More serious misconduct or notifiable incidents are addressed by line managers or human resources, and clinical incidents or adverse events that put patients at risk of harm are reported in an incident management system (Supplementary Material S1). EMs receive the submission from the triage team within 7 days of submission. The aim is for EMs to be from the same professional group and level as the staff member identified in the submission. EMs have the option to decline the task or discuss details with a mentor or Ethos manager – this may be due to a previous relationship (e.g. close friend, dispute) with the subject of the submission, or to seek advice about formulating feedback. EMs then organise to meet with the staff member. They are trained to extract the objective issues from a submission, including the submitter’s perception of inappropriate behaviour and the impact that the behaviour has had on them, other staff, or patients. EMs do not investigate the events asserted in a submission, but rather provide feedback to allow the recipient to reflect and consider future behaviour. The delivery of feedback to the recipient is confidential and no information is recorded. After delivering feedback, EMs record messages as delivered and are encouraged to debrief with a mentor or fellow EM.

An online survey using Qualtrics (Supplementary Material S2) was developed to investigate the experiences of EMs. Respondents provided descriptive information (e.g. time employed at hospital, professional role, time as an EM, number of messages delivered), and were asked their level of agreement (5-point scale: strongly disagree to strongly agree) to 14 statements about their experience as an EM and the program more broadly. Two further questions asked about their capacity to provide timely feedback and how receptive recipients were to messages (always, most of the time, about half the time, sometimes, never). Four open-ended questions asked about motivations for accepting the role, challenging and rewarding aspects and suggestions for changes to the program. The survey was piloted with a small group of clinicians and refined prior to administration.

Descriptive statistics were calculated using Excel. The ‘somewhat’ and ‘strongly’ agree and disagree response options were combined to form ‘agree’ and ‘disagree’ categories respectively. We developed the coding scheme for the open-ended questions through iterative review of a sub-sample of responses. Codes were refined by consensus, applied to all responses and major themes identified.

Of an estimated 145 EMs across the eight hospitals, 60 responded to the survey (response rate: 41.4%). Seven completed less than 20% of survey items. All data were retained for analysis.

Characteristics of respondents are shown in Table 1. Eighty percent (n = 40) had been an EM for more than 12 months. Over half (57.7%; n = 30) had delivered 1–5 ‘feedback for reflection’ messages, 17.3% (n = 9) had delivered 6–10, 17.3% (n = 9) none and 7.7% (n = 4) 10–20 messages.

Table 1.  Characteristics and experiences of Ethos messengers (n = 60).

Most agreed that they had received enough training (90.1%; n = 48) and access to support (84.9%; n = 45); have the skills (90.1%; n = 48) and enough time to fulfil the responsibilities of the role (70.0%; n = 30); and that their responsibilities are clearly defined (84.9%; n = 45) (Table 2). Approximately 72% (n = 31) of respondents indicated that they always or most of the time were able to deliver feedback in a timely fashion and that recipients were always or most often receptive (53.5%; n = 23) (Table 3). Most agreed that their hospital was committed to Ethos (73.6%; n = 39), that they were satisfied with the way the program is being managed (76.9%; n = 40) and believed that it would lead to a decrease in unprofessional behaviour in their hospital (55.7%; n = 29).

Table 2.  Respondents’ perceptions of being an Ethos messenger and the Ethos program.

Table 3.  Respondent perceptions of the feedback for reflection process.

Comments from 55 EMs were categorised into three main themes and an ‘other’ category: A ‘belief in the philosophy of the Ethos program’ (Table 4; Q1) and a desire ‘to contribute to a positive hospital culture’ were commonly articulated (Table 4; Q2). EMs reported a range of ‘Individual reasons’ for taking on the role including that they were the right person for the job or that it was an opportunity for professional development (Table 4; Q3). Less common ‘other’ reasons included being nominated and because they valued patient and/or staff safety (Table 4; Q4).

Table 4.  Exemplar quotes for open-ended questions.

Three themes were identified from 53 responses. Many reported ‘practical issues’, which made it difficult to deliver feedback related to shift work, competing commitments and finding a suitable time and place to meet (Table 4; Q5). The second theme was ‘delivering feedback for reflection effectively’, with many EMs reporting difficulty in ensuring that feedback was communicated meaningfully and without causing distress to the recipient (Table 4; Q6). A third theme related to ‘recipient responses to feedback’ involved challenges when recipients had a negative response to feedback (Table 4; Q7). ‘Other’ challenges included a negative impact of being an EM on relationships with colleagues (Table 4; Q8).

Fifty-three EMs provided responses from which we identified three common themes and an ‘other’ category. Most EMs reported a combination of rewarding aspects including ‘having a meaningful conversation’ that provided the opportunity for reflection and constructive conversations about behaviour with colleagues (Table 4; Q9). ‘Having staff embrace the reflection process’ incorporated positive comments about seeing recipients reflect, understand and want to improve their behaviour (Table 4; Q10). EMs also described how their role was ‘contributing to positive culture change’ (Table 4; Q11). ‘Other’ rewarding aspects included experiencing gratitude from recipients and being part of a team and valued in the workplace (Table 4; Q12).

We identified two main themes and an ‘other’ category from 53 EM responses relating to how the program could be strengthened. Many EMs reported that the program should be better promoted and that education should be available to all staff (Table 4; Q13). ‘More support for Ethos messengers’ was highlighted and related to the desire of EMs for more regular meetings with the Ethos manager and EM team (Table 4; Q14). Some EMs suggested there should be further support for recipients of feedback and that staff were unsure whether the Ethos program was effective and that evidence of effectiveness should be sought (Table 4; Q15).

This study examined the experiences of peer messengers who play a key role in the delivery of a professional accountability program by providing informal feedback to staff about their unprofessional behaviour. Our findings demonstrate that staff will volunteer for a peer messenger role and can be appropriately trained and supported in their roles to deliver feedback. Additional support to address challenges, including support arranging meetings between EMs and recipients and effectively communicating information to recipients, was desired by messengers.

Despite 70% of respondents indicating that they had enough time to fulfil the responsibilities of the role, a frequent issue raised in the open-ended questions was finding an appropriate time when both the EM and message recipient were on site and not constrained by work commitments. EMs are required to balance their role as peer messengers with their professional roles. Busy work schedules could jeopardise their ability to schedule a time to have a meaningful conversation. An evaluation of the ‘Freedom to Speak Up Guardians’ in the United Kingdom, who have a similar role to those of EMs, found that time scarcity negatively impacted the ability of Guardians to effectively address concerns raised by hospital staff about safety which were often associated with bullying and harassment.¹⁷ Guardians are a point of contact for hospital staff with concerns about risk, malpractice, or wrongdoing. Yet, similar to EMs, this is an additional role and thus they are expected to juggle competing priorities.¹⁸ We did not examine the time taken up by the EM role, nor did we explore whether demands are equally distributed among EMs, both of which should be investigated in future research.

The delivery of feedback and bad news can be a complex, multi-step process.¹⁹ Around a third of EMs indicated that recipients were ‘sometimes’ or ‘never’ receptive to messages. Many made specific comments about the challenges of communicating with peers who lacked insight and were unable to reflect on their behaviour. EMs also reported concerns about causing distress to recipients. Studies of delivering negative performance feedback have found that managers feel uncomfortable and worry about negative reactions from employees,²⁰ and employees feel threatened by bad news because it can impair their self-esteem and increase uncertainty about the future.²¹ However, training that includes principles of fairness when feedback is delivered has been found to significantly reduce negative responses and increase motivation to improve performance in recipients.^22,23 Fairness training incorporates five principles from Leventhal (1980):²⁴ consistency (procedures are the same across people and time), bias suppression (there is no bias or self-interest), accuracy (accurate information is provided), representativeness (the needs of all involved are acknowledged) and ethicality (moral and ethical standards are followed).²⁴ Training that teaches peer messengers how to enact relevant fairness principles during the informal feedback process may mitigate distress in ‘feedback for reflection’ recipients.

Studies have found that the process of peer feedback is effective in promoting behaviour change in hospital staff.^9,25–27 For example, it has been shown to sustain high levels of hand hygiene adherence and improve surgeon professionalism and communication.^25,27 In one study, Pichert et al.¹¹ used a database of patient complaints from 16 medical centres in the United States to identify physicians associated with the highest number of patient complaints. Peer messengers met individually with ‘high-risk’ physicians to informally make them aware of their risk status and asked them to reflect on why patients made complaints. Between 2005 and 2009, 178 peer messengers completed 1371 meetings with 373 physicians. Most physicians’ risk scores improved over time (64%). However, for 19% of physicians their risk scores did not change and for 17% risk scores worsened. This lack of improvement by a proportion of physicians is consistent with this survey’s findings that around a third of message recipients were not receptive to feedback.

Strengths of this study include the use of qualitative data to supplement and extend the close-ended survey questions and the range of respondents, from different professional backgrounds, from eight hospitals in three states. The survey consists primarily of positively framed items, which limited the control of social desirability bias. The experience of EMs varied with 30 participants reporting they had delivered between one and five ‘feedback for reflection’ messages, nine had not delivered any and eight did not provide this information. This mixed level of experience should be considered when interpreting the findings. Details of the characteristics of the EM population are not available and thus it is not possible to determine if responders were representative of all EMs.

Our findings indicate that modest training and support can provide peer messengers with the skills required to deliver informal feedback to colleagues about unprofessional behaviour. Enhancing training, particularly in relation to delivering difficult information, was identified by EMs as central to the ongoing effectiveness and sustainability of their role. Training a cadre of EMs, at all organisational levels, who convey at times unwelcome messages, builds capacity in speaking-up skills across the organisation and is likely to enhance the safety culture.²⁸ Further studies are required to examine the influence of EM characteristics on their perceived credibility and how messages are received, as well as investigation of if and how active engagement in the feedback process contributes to the effectiveness of accountability programs in reducing unprofessional behaviour.

Supplementary material is available online.

The data that support this study will be shared upon reasonable request to the corresponding author.

The authors declare no conflicts of interest.

National Health and Medical Research Council Partnership Project Grant (1134459) in partnership with St Vincent’s Health Australia.