A ‘Communication and Patient Safety’ training programme for all healthcare staff: can it make a difference?
- Correspondence to Dr Peter Lee, Clinical Governance Unit, Level 1 Melbourne St Office, Queensland Health, PO Box 48, Brisbane, Queensland 4001, Australia;
Contributors Peter Lee, as lead author, takes responsibility for the integrity of the work as a whole from inception to published article. Kellie Allen, the second author, made substantial contributions to the concept, the design and the data acquisition and interpretation for our programme, as well as providing critical review of the article. Michael Daly, the third author, also made substantial contributions to the concept and design of the programme, the interpretation of the data, and provided significant critical review of the article. All three authors then gave final approval of the version to be published.
- Accepted 3 October 2011
- Published Online First 18 November 2011
Communication breakdown is a factor contributing to most cases of patient harm, and this harm continues to occur at unacceptable levels. Responding to this evidence, the Metro South District of Queensland Health (Australia) has developed a communication skills training programme titled ‘Communication and Patient Safety’. The three modules, each lasting 3½ h, cover both staff-to-patient and staff-to-staff communication issues, and an unusual feature is that clinical and non-clinical staff attend together. Following positive evaluation data from our initial pilot programme (involving 350 staff in a single hospital), the programme was expanded to all five hospitals in the district, and has now been completed by over 3000 staff. The results show that despite the significant time commitment, participants find the courses useful and relevant (Kirkpatrick level 1), they learn and retain new material (level 2), and they report changes in behaviour at individual, team and facility levels (level 3). Although it remains a challenge to obtain quantitative data showing that training such as this directly improves patient safety (level 4), our qualitative and informal feedback indicates that participants and their managers perceive clear improvements in the ‘communication culture’ after a workplace team has attended the courses. Improving ‘communication for safety’ in healthcare is a worldwide imperative, and other healthcare jurisdictions should be able to obtain similar results to ours if they develop and support interactive, non-didactic training in communication skills.
- patient safety
- culture change
- safety culture
- human factors
- quality improvement
- quality improvement methodologies
- quality measurement
- continuous quality improvement
- healthcare quality improvement
Appalling rates of patient harm have been reported for half a century,1 2 and healthcare's poor safety record has been discussed publicly for over 15 years.3–5 Yet reduction in this harm is ‘frustratingly slow’,6–8 and safety may even be getting worse for patients, not better.9
One avenue to address this ‘wicked problem’10 of harm is to focus directly on communication. Published data11–13 now confirm the experience of all clinicians that communication breakdowns are involved in the great majority of adverse patient outcomes. Improving communication skills has been a key objective in bringing ‘crew resource management’,13–15 particularly teamwork training,16 17 to healthcare settings. Recent publications describe promising results in general medical units,18 in operating theatres, and intensive care units.19 However, there have been few programmes that specifically target an improvement in communication skills for all staff (leaders and frontline staff, clinical and non-clinical), not only within teams or ‘microsystems’20 but also at an institution-wide level.
To address these issues, we developed a ‘Communication and Patient Safety’ (CAPS) training programme, which has now been completed by over 3000 healthcare staff in southern Queensland, Australia.
A pilot communication training programme consisting of three modules, each lasting 3½ h, was attended by approximately 350 staff from four nominated units (emergency, outpatients, maternity, and special care nursery) in ‘hospital 1’, an outer-suburban 300-bed hospital. Participants included administrative, operational, medical, nursing, and allied health staff.
Table 1 outlines the course content for this pilot. Topics were chosen after extensive consultation with healthcare staff, along with reviews of previous undergraduate25 26 and postgraduate21 27–29 interventions. The format and style were based on our previous delivery of highly interactive patient safety training,29 30 with skilled facilitators using a mix of practical exercises, video clips, small-group discussion and other adult learning techniques.31
Following the success of the pilot training (tables 2 and 3 below), the programme continued unchanged, and was expanded to all five hospitals in the district. We have now delivered over 500 3½ h modules, with over 3000 staff completing the 10½ h programme. Feedback to senior management has been so positive that four out of the five hospitals in our district have requested CAPS training for every employee.
The pilot training was evaluated in conjunction with the University of Queensland and Griffith University32 (approval no. 2008/137 from the Princess Alexandra Hospital Human Research Ethics Committee), using an 83-item previously validated questionnaire, covering work relations, handovers, errors and patient safety. This questionnaire was distributed to all staff in the four trial units of hospital 1 before the training, and resent 3 months after its conclusion to those staff who had completed at least one module. To form a control group, the same questionnaire was also distributed at identical times to staff in the equivalent units in two neighbouring hospitals (hospitals 2 and 3), which were not yet receiving the CAPS training.
Evaluation of pilot training—hospital 1
Table 4 shows the questionnaire response rates for the pilot training evaluation.
The questionnaire responses on a seven-point Likert scale (1=strongly disagree, 4=neutral and 7=strongly agree) were analysed by analysis of variance. A selection of the results is given in table 2.
Ongoing evaluation—hospitals 1–5
One question on the course feedback questionnaire asks each participant to rate on a six-point scale ‘Overall, was this a good use of your time?’ Even over three different time periods and three different institutions, 80–90% of participants consistently indicated that it was a ‘very good’ or ‘excellent’ use of their time (table 3).
Staff opinion surveys recently conducted at hospital 1 show a steady improvement in several measures during the 2008–10 CAPS training period (table 5).33 While correlation does not prove causality, the opinion of the acting chief executive officer was “We have improved significantly and this is in no small way attributable to the CAPS programme that you have rolled out in our area! I would like to thank you once again for your commitment to our organisation.”
Qualitative data on learning and behaviour change
CAPS programme participants continue to report examples of learning and behaviour change. Many relate to adoption of the ‘SBAR’ (situation–background–assessment–recommendation) communication tool,11 21 but other tools such as graded assertiveness,22 23 briefings, and verbal de-escalation of aggression are also being learnt and used:
A paediatric consultant at hospital 2 insisted that his juniors fill out an SBAR sheet before calling him at night. As well as improved clarity in the calls, he recorded a 17% drop in the number of calls, as juniors clarified the problems for themselves.
At least 20 nursing units have now changed their nursing handover to an interactive SBAR format after attending the training.
The entire district of five hospitals is now moving to implement SBAR as a standardised and expected mode of communication.
A community health team has now changed their computerised patient history and handover system so that instead of free text, staff are required to complete four separate fields in the SBAR format.
Over 30% of the work units trained have asked for unit-specific ‘Above and Below the Line’ sessions. This is a transparent way of promoting acceptable team behaviour, widely used in the education sector.34
Many informal reports indicate that the training can have a profound effect on individuals and teams. For example:
S—I was an audience member; B—My previous communication on charts was unstructured; A—this session was very beneficial; R—I
will attempt to use SBAR in clinical practice. (Medical officer, hospital 1)
I have just conducted a PA&D (review) with one of my admin staff who was previously seen to be less than enthusiastic at times.
What a difference the CAPS training has made. I was blown away when she started referring to her training and how it has helped
her understand the important role that she has within our team. (Admin services coordinator, hospital 3)
There is already a noticeable change in communication and staff morale here at (hospital 5), and I feel the CAPS programme
had a big part to play in this. (Nurse unit manager, hospital 5)
I use the new skills I have learned in everyday life as well as in the hospital. It has changed the way I look at life. (Nurse
manager, hospital 1)
‘Communication and Patient Safety’ has proved highly popular with staff and managers, despite the considerable time commitment required. But does it work? A common framework used to address this question is Kirkpatrick's four ‘levels’ of evaluation for training—reactions, learning, behaviour and results.35
The feedback data shown in table 3, along with many informal comments, indicate that staff find the courses useful and relevant. Meta-analyses36 indicate that these ‘utility’ reactions have greater positive association with learning (level 2) and behaviour change (level 3) than simple ‘affective’ reactions (ie, whether the participants enjoyed themselves).
The results show that participants learn from the programme, as evidenced by the decreased agreement with ‘Errors are a sign of incompetence’ (table 2), the rapid take-up of the SBAR tool, and the large number of requests for more training. The exact extent of this learning cannot be quantified from our present data, however, and would require larger sample sizes than our pilot evaluation provided.
Staff are changing what they do as a result of this training. The pilot evaluation results (table 2) indicating decreased agreement with ‘I will suffer negative consequences if I report a patient safety problem’, and increased agreement with ‘Disagreements in this hospital are resolved appropriately’, are self-reports, rather than observed behaviour change. However, as illustrated above, we receive frequent reports about behaviour changes such as the implementation of the SBAR communication format, use of ‘Above and Below the Line’ posters to curb unacceptable behaviour, and use of techniques of verbal de-escalation of aggression. While this study cannot quantify the extent of this behaviour change, observational studies to measure the implementation of the SBAR technique are planned.
The fact that the chief executives of four hospitals have requested CAPS training for every employee is also not a direct measure of changes in staff behaviour, but nevertheless indicates their perception of the programme's beneficial effects.
One ‘proxy measure’ of a safety culture is staff satisfaction37; while our data from one hospital on staff opinion changes (table 5) appear positive, that survey was not specifically targeted at CAPS training participants, and the results should be viewed as suggestive only.
Although aspects of the ‘work culture’ may be hard to quantify numerically, the CAPS training is acknowledged by many to be a culture-changing programme, and we are currently seeking ethnographic expertise to analyse this further.
Limitations of the data
Larger-scale survey and observational studies are required to quantify the changes in learning and behaviour that are being reported to us. Ultimately our programme aims to reduce actual harm to patients; however the variability inherent in measuring rates of patient harm across multiple units, multiple professional groups, and multiple institutions made analysis of patient outcomes beyond the scope of our programme. Other proxy measures of improved safety such as increased patient satisfaction, and reduced staff turnover, were also not studied here, but would be important variables in further analysis at Kirkpatrick's level 4—results.
Our findings show that when staff attend interactive communication skills training in inter-professional groups, they find the material useful and relevant, they learn the communication tools covered, and they adopt new and more positive behaviours (Kirkpatrick levels 1, 2 and 335). As they do this, they report informally that they feel culture change happening around them, although our data cannot quantify these improvements in ‘results’ (Kirkpatrick level 4).
As emphasised by Neily et al,19 once culture change is started through effective training, it needs to be carefully maintained through both ‘top-down’ and ‘bottom-up’ support, and consistent leadership is essential.37 As well as promoting ‘leadership walkarounds’,38 the CAPS programme is following the initial sessions with short refresher courses for frontline staff. As illustrated by Panozzo,39 however, the adoption of new processes can be surprisingly difficult, and the path to genuine culture change is not smooth.
To sustain change, local ‘champions’ are essential. The CAPS programme has trained over 30 frontline staff (from varied backgrounds) to become part-time co-presenters of the material, making them a valuable local resource on communication issues.
For patients, the potential benefits of the CAPS programme extend well beyond a reduction in adverse events. Communication practices profoundly influence the experience of patients, for good or for ill—the tone of just four 10 s audio clips can predict a US surgeon's malpractice claims history.40 If healthcare staff communicate well, patient satisfaction increases, which directly relates to improved patient outcomes.41 Effective communication saves duplication and misunderstanding, thereby saving frustration, time and money.
Continuing high levels of patient harm, and the knowledge that this harm is largely related to communication failures, has given an urgency to this endeavour. While it may prove difficult to demonstrate quantitatively that patient safety improves as a direct result of communications training, our results on other dimensions (reactions, learning and behaviour change) indicate that well facilitated courses can have positive and durable effects. Other healthcare systems could readily find similar benefits from introducing broad-based non-didactic communication skills training.
Some components of the training material in the CAPS courses are derived from the ErroMed Group's ‘Human Error and Patient Safety’ training,42 and are used by Queensland Health under licence (see also Lee et al30). The evaluation research was conducted by Dr Bernadette Watson (University of Queensland) and Dr Liz Jones (Griffith University) with the assistance of Tara Becker. Sincere thanks must go to the many senior hospital executives in the Metro South District of Queensland Health, whose leadership and support has been essential to the success of the programme.
Funding Entirely funded by The State of Queensland (Queensland Health).
Competing interests All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare that (1) No financial support has been received for the submitted work other than from their employer; (2) PL has undertaken training for the ErroMed Group in the previous 3 years, otherwise there are no relationships with commercial entities that might have an interest in the submitted work; (3) their spouses, partners, or children have no financial relationships that may be relevant to the submitted work; and (4) There are no non-financial interests that may be relevant to the submitted work.
Ethics approval Ethics approval was provided by Princess Alexandra Hospital (Brisbane) Human Research Ethics Committee, ref 2008/137.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Data available on request from the corresponding author.