Towards culture change in the operating theatre: Embedding a complex educational intervention to improve teamwork climate

Abstract

Background: Changing teamwork climate in healthcare through a collective shift in attitudes and values may be a necessary precursor to establishing a positive teamwork culture, where innovations can be more readily embedded and sustained. A complex educational intervention was initiated across an entire UK Trust's surgical provision, and then sustained. Attitudes towards teamwork were measured longitudinally to examine if the intervention produced sustainable results.

Aims: The research aimed to test whether sustaining a complex education intervention to improve teamwork would result in an incremental, longitudinal improvement in attitudes and values towards teamwork. The intervention's larger aim is to progress the historical default position of multi-professional work to authentic inter-professional teamwork, as a positive values climate translates in time into behavioural change defining a safety culture.

Method: Attitudes were measured at three points across all surgical team personnel over a period of 4 years, using a validated Safety Attitudes Questionnaire with a focus on the ‘teamwork climate’ domain. Pre- and post-intervention ‘teamwork climate’ scores were compared to give a longitudinal measure as a test of sustainability.

Results: Mean ‘teamwork climate’ scores improved incrementally and significantly following the series of educational interventions, showing that practitioners’ valuing of teamwork activity can be improved and sustained.

Conclusions: Longitudinal positive change in attitudes and values towards teamwork can be sustained, suggesting that a deliberate, designed complex intervention can shape a safety climate as a necessary prerequisite for the establishment of a sustainable safety culture.

Introduction

Surgical iatrogenesis and effective teamworking

Unacceptably high levels of medical iatrogenesis describe a global phenomenon (WHO 2008), with surgical environments affording the highest risk for adverse events (Sarker 2003). For example, the UK House of Commons Health Committee report (2009, p. 5) showed that 11 people are seriously harmed every day during surgery carried out within the UK National Health Service (NHS); and that major surgical errors had risen in the UK by 28% in 5 years, where ‘The NHS lags unacceptably behind other safety-critical industries, such as aviation, in recognising the importance of effective teamworking and other non-technical skills’ as major causes of error.

Poor teamwork is a symptom of a structural faultline in medicine and surgery, where persistent autocratic practices shape inflexible work hierarchies that stifle effective communication (Bleakley et al. 2011). This historical legacy places little emphasis upon both collaborative and affective aspects of work and frustrates the possibility of team members lower on the hierarchy (such as nurses) speaking out against perceived poor or inappropriate practice. Patient safety outcomes can, however, be affected ‘by improving communication and teamwork skills among health professionals’ (AHRQ 2008, p. 5), where an estimated ‘70–80% of healthcare errors are caused by human factors associated with poor team communication and understanding’, and 50% of such errors can be avoided through improving team-based communication (Xyrichis & Ream 2008, p. 232). Improved quality of clinical teamwork has been shown to correlate with lower patient mortality and improved work morale (West et al. 2002; Wheelan et al. 2003), yet collaborative teamwork is not the norm across clinical teams generally (Giddings & Williamson 2007), and surgery in particular (Healthcare Commission 2007).

Traditional hierarchies generate climates of monologue (telling and informing) rather than dialogue (asking, conversing and debating), and this pattern may be the norm for surgeons in operating theatres (OTs), setting an atmosphere and climate (Bleakley et al. 2012). A third of surgeon-led communication exchanges may potentially jeopardise patient safety, including tension-provoking statements in particular (Lingard et al. 2004). In contrast, where surgical teams work collaboratively and communicate effectively through dialogue, they counter the potential for small problems to accumulate and escalate into crisis (Catchpole et al. 2007).

Climate change precedes and forms culture change

Although such studies of teamwork tell us a good deal about the ‘anatomy’, or makeup, of effective teams, they do not tell us about the ‘physiology’ of effective teamwork – how team process is initiated and maintained. For example, it is surely important that team members value working together, where the sum of their attitudes towards teamwork is positive and sustained over time.

The sum of a workforce's attitudes can be seen to form a values ‘climate’ (Genn 2001). Attempting to change the behaviour of a workforce through educational intervention or introduction of a protocol is often met with resistance, where ingrained cultural habits are challenged and there is no prior will to embrace change. Behavioural culture change requires a prior change in attitudes and values that can be measured through appropriate attitudinal scales.

We set out to provide a platform for ongoing change in the culture of OTs across an entire surgical complex within a UK teaching hospital and two small satellites, promoting inter-professional collaboration as the future default position for teamwork in the face of a current, historically-determined default position of hierarchical uni- and multi-professionalism (Bleakley et al. 2004, 2006). However, we recognised that our complex educational intervention, aimed at changing teamwork practices, must be reflected first in a sustained, collective change in climate (the sum of attitudes and values), where what is valued are patient safety principles (democracy, teamwork, collaboration and respect) and practices (briefing, debriefing and close call reporting).

Practitioners focusing on potentially divisive profession-specific issues, rather than overarching collaboration, frustrate the opportunity for building what is described as ‘collaborative intentionality capital’ – social capital reflecting commitment to common purposes in a work team (Engeström 2004; Beyerlein et al. 2005). Building such collaborative capital is not easy – it demands a common will to democracy and the translation of that will into collaborative, patient-centred practices, such as pre-list briefing and post-list debriefing (Allard et al. 2007, 2011).

Our concern in this study is to report on the sustainability of an educational intervention designed to improve teamwork and set a supportive values climate in surgical settings. Although a previous study (Bleakley et al. 2006) demonstrated the initial impact of such an intervention in significantly changing measured attitudes towards teamwork, could this be sustained? In this study, we report a longitudinal dimension to that initial study.

Table

Practice points

Good teamwork reduces surgical error. Surgical culture's historical legacy of hierarchy frustrates effective teamwork. Modelling good teamwork requires a designed and sustained educational intervention. Establishing collaborative teamwork, as a culture change, requires a precondition of attitude change. Sustaining an educational intervention can result in an incremental, positive valuing of teamwork and safety by practitioners.

Methods

Design

In the ongoing, collaborative inquiry introduced earlier in the text, we have been interested in measuring and tracking changes in attitudes towards safety through iteration, or longitudinal use, of a validated Safety Attitudes Questionnaire (SAQ; Sexton et al. 2006). Such changes have been related to the introduction and iteration of a complex educational intervention designed to improve teamwork, described in the following section.

The genesis and development of our educational intervention have been reported elsewhere (for example Bleakley et al. 2004; Allard et al. 2007, 2011), as have the effects of the first round of the intervention (Bleakley et al. 2006). In short, the intervention includes three strands: data-driven iterative education in human factors for all OT personnel; establishing a local, reactive close call (near miss) incident reporting system; and developing team self review (briefing and debriefing) across all surgical teams. Educational support has included feedback of data through face-to-face meetings, a newsletter and an open-access website (www.ttrm.co.uk) that has acted as a repository for data analysis.

The anglicised version of the SAQ for OTs was administered to the entire surgical cohort to provide a baseline score (round 1). The educational intervention was then introduced to complex A (general theatres) but not to complex B (orthopaedic and trauma theatres). The SAQ (round 2) was delivered again to all personnel just over 1 year later (Bleakley et al. 2006). The same educational intervention used in complex A was then introduced to complex B soon after the SAQ round 2 was completed. More than 2 years later, the SAQ (round 3) was delivered to all personnel as the complex intervention was sustained.

Our experimental design was as follows: pre- and post-intervention scores for complex A (taken from SAQ rounds 1 and 2) were combined with pre- and post-intervention scores for complex B (taken from SAQ rounds 2 and 3). This provided overall pre- and post-intervention scores on ‘teamwork climate’ for all OT staff but across two locations. Our focus is then on pre- and post-intervention scores for both OT complexes (SAQ returns 1 and 2 for complex A and SAQ returns 2 and 3 for complex B). Additional returns (SAQ return 1 for complex B and SAQ return 3 for complex A) are then omitted from our analysis. A univariate linear model was used for preliminary analysis of ‘teamwork climate’ scores, containing ‘location’ (complex A or B), and ‘pre- or post-intervention’. Comparison of variance for pre- and post-intervention ‘teamwork climate’ scores across locations (fe = 0.407 and p = 0.524) supports the use of a combined ‘location’ score to analyse the cumulative effect of the intervention on all OT staff in the study.

Introducing the educational intervention to our previous ‘control’ group – a second cohort of OT practitioners (complex B) – investigates the transferability of the intervention. Importantly, we were bound ethically to offer the educational intervention to cohort B, having demonstrated its worth in our initial ‘experimental’ group complex A.

Outcome measure

The outcome measure is drawn from the SAQ, widely used in healthcare settings for organisational benchmarking purposes (Sexton et al. 2006) and in this study adapted and anglicised in collaboration with the principal author for use in UK OTs, and used with that author's permission. ‘Teamwork climate’ is one of six domains from the full SAQ. The other domains are ‘safety climate’, ‘working conditions’, ‘perception of management’, ‘stress recognition’ and ‘job satisfaction’. The ‘teamwork climate’ domain of the SAQ describes ‘perceived quality of collaboration between personnel’ (Sexton et al. 2006).

In the original SAQ, the ‘teamwork climate’ domain has seven questions. In anglicising the questionnaire in a pilot study, two questions were rejected as potentially confusing, thus basing our ‘teamwork climate’ analysis on five questions. Each question requires a response on a five-point Likert scale: disagree strongly (1point), disagree slightly (2), neutral (3), agree slightly (4) and agree strongly (5), where scores for the five questions are combined to give a total ‘teamwork climate’ score. The lowest possible score is then five, with 25 the highest. Mean scores are then calculated across cohorts.

The anglicised items used from the ‘teamwork climate’ domain of the original SAQ are:

1. Decision making in the OT utilises adequate input from relevant personnel.

2. This hospital encourages teamwork and cooperation amongst its personnel.

3. Disagreements in the OT are resolved appropriately (i.e., not who is right but what is best for the patient).

4. The doctors and nurses in this study work together as a team.

5. Nurse input about patient care is well received in the OT.

The domain questions rejected were:

• In the OT in this study, it is difficult to speak up if I perceive a problem with patient care.

• I am frequently unable to express disagreement with consultants.

Both of these questions created widespread misunderstanding amongst personnel. Senior surgeons and anaesthetists perceived the first question as redundant. The second question contains a double negative. Removing the two confusing questions in the ‘teamwork climate’ domain does not alter the significance of the findings reported.

Ethics approval

Ethics approval was sought and granted from the Ethics Committee of the Trust. The study was treated as an audit of an educational intervention. All data returns were anonymous and confidential to the study. Staff group and theatre complex origins were indicated on the questionnaire return.

Results

The equality of means for pre- and post-intervention ‘teamwork climate’ scores were assessed using an independent sample t-test. This contains ‘teamwork climate’ as the dependent variable and ‘pre- or post-intervention’ as the independent variable. When the two samples are compared, the difference between scores is significant (p = 0.034), suggesting a longitudinal or sustained effect of the educational intervention in the absence of any other significant changes in work patterns across the surgical provision (see Figures 1 and 2).

Figure 1. Longitudinal changes in mean teamwork climate scores across two complexes. Notes: Mean teamwork climate SAQ scores compared across two complexes (A and B) over 3+ years. Vertical axis: estimated marginal means of teamwork climate scores. Horizontal axis: batch 1 = baseline scores; batch 2 = 1 year later; batch 3 = 3 years later.

Figure 2. Mean teamwork climate scores for each location and timing of intervention.

Questionnaire return rates were 73% (round 1), 68% (round 2) and 53% (round 3) of whole OT populations (see Table 1 and key). It is predictable that numbers of returns would decline due to familiarity with the questionnaire. However, on inspection, SAQ returns were not biased to particular staff groups and closely replicated population profiles for OT personnel.

Table 1  SAQ response rates

Total numbers of respondents are shown in Table 2.

Table 2  Total numbers of respondents

	Total	Study total
Pre-intervention	187 (3 missing)	N = 184
Post-intervention	164 (6 missing)	N = 159

OT practitioners classified as ‘missing’ (see Table 2) are respondents who answered twice, or failed to complete, one or more teamwork questions. ‘Teamwork climate’ scores were removed for these respondents. SAQ returns remained representative of personnel groups and interpretable.

Although our concern in this article is with the ‘teamwork climate’ domain of the SAQ, where our argument concerns improvement in teamwork as a necessary precursor to the development of a self-sustaining surgical patient safety culture, we will briefly note that there were improvements in mean scores on all but one of the other SAQ domains, but these increases were not significant. ‘Stress recognition’ did reduce significantly. As well as the positive shift in ‘teamwork climate’ there was a similar, although statistically insignificant, trend in improvement in the SAQ ‘safety climate’ domain following the educational intervention. Figure 3 displays comparative effects of the intervention on standardised domain scores (mean = 100) pre- and post-intervention.

Figure 3. Comparison of standardised domain scores pre- and post-intervention (mean = 100) across the six SAQ domains (‘job satisfaction’, ‘stress recognition’, ‘perception of management’, ‘working conditions’, ‘safety climate’ and ‘teamwork climate’).

Discussion

In a previous article (Bleakley et al. 2006), we concluded that:

Positive, unidirectional changes in attitudes towards inter-professional teamwork can be established through a structured educational intervention. The aggregate of such attitude change can create a new ‘milieu’, or ‘habitat’ (Genn 2001) that, we suggest, offers the conditions of possibility for the emergence of safer practice and improved patient care.

In this study, the results suggest that, beyond initiating or ‘establishing’ such change, an improvement in collective attitudes towards teamworking is sustainable. Our findings are supported through other data. Individuals who report that they regularly engage in briefing also express a positive shift in attitudes to teamwork in comparison with those who report less contact with briefing (Allard et al. 2011).

Studies across entire hospital complexes (West et al. 2002), and in intensive care units (Wheelan et al. 2003), have shown a relationship between improved clinical teamwork and patient outcomes. However, they do not specifically articulate how improving teamwork leads to better patient outcomes. We are now beginning to understand the dynamics of surgical teamwork that may prove beneficial for patient safety. For example, emphasis upon collaborative production of social capital addresses the historical resistance to the importance of non-technical factors such as good communication and teamwork.

Democracy and horizontal patterns of communication (networks) replace autocracies and supplement meritocracies based on relative complexity of technical competence. This dynamic also produces the identity of the ‘inter-professional’, transcending the default and dominant identity of ‘(uni)professional’, and challenges the rhetorical tactics by which professional silo boundaries are maintained and unproductive identity differences are crystallised (Lingard et al. 2002; Bleakley 2006).

As far as the wider findings of other SAQ dimensions are concerned, that a significant improvement in attitudes towards stress recognition was shown does not necessarily suggest that surgical team members were becoming insulated against recognising stress, but possibly the opposite – that they were dealing with stress. Finally, an increase in uptake and support of teamwork practices designed to improve patient safety in the study cohort may indicate early stages of a transition to a sustainable patient safety culture.

Limitations to the study

The primary limitation is that we report changes in practitioners’ attitudes, values and perceptions, rather than changes in practitioners’ behaviour and performance, or in patient outcomes, as a result of the complex educational intervention. However, we did not set out to look at either performance or patient outcomes, but to examine the conditions of possibility for a shift from climate to culture change.

A further limitation to the study is that complex A received a supplement to the educational intervention that was not available to complex B. We were unable to gauge the impact of this intervention on its own. Early in the project, we employed (part-time for 6 months) a theatre nurse who facilitated the use of briefing and debriefing. This extra intervention may have served to increase the trend in improvement in teamwork climate for complex A.

Conclusion

Positive attitudes to teamwork, improved through a structured, complex educational intervention, can be sustained. This builds a safety climate based on collaborative intentionality that may in turn transform the historical, default position of uni- and multi-professionalism into authentic inter-professionalism as a shift in surgical team culture.

Acknowledgements

We would like to thank Colin Pritchard for help with statistical analysis of data, and Bryan Sexton for collaboration in producing an anglicised version of the SAQ. The wider project has recently received support through European Social Funding, who have also recognised the wider Cornwall Theatre Team Resource Management (TTRM) project as a centre of excellence in researching clinical teamwork.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the article.