Routine pre-operative testing before elective surgery is common.1 However, the importance of this testing before most procedures is considered to be low because it rarely changes management and it may cause harm to patients.2-10 Choosing Wisely is an international, health professional-led campaign which aims to reduce unnecessary and low-value patient care by encouraging medical colleges and speciality societies to identify clinical practices which should be questioned or avoided.11
Since the Choosing Wisely campaign began in New Zealand in 2016, over 150 evidence-based recommendations have been developed, including recommendations from the Australian and New Zealand College of Anaesthetists (ANZCA) to avoid pre-operative blood investigations, chest x-rays (CXRs) and spirometry prior to surgery for asymptomatic patients undergoing low-risk surgery.12,13 The effectiveness of Choosing Wisely depends on the dissemination and uptake of these recommendations.14
This study evaluates a brief Choosing Wisely campaign run in Christchurch Public Hospital (CPH) between August and October 2018. This campaign was in response to an internal audit of routine pre-operative CXRs in the surgical departments, which identified that two departments (general surgery followed by plastic surgery) ordered 91% of routine pre-operative CXRs. General surgery alone ordered 64% of all routine pre-operative CXRs.
The CPH campaign aimed to reduce the number of pre-operative CXRs by placing posters in the pre-operative admissions area, a piece in the chief executive officer (CEO) newsletter sent weekly to all staff, and an addition to the district’s online health pathways for pre-operative admissions. All of these methods communicated that CXRs should not be routine and should only be ordered if patients met certain criteria, where: “acute cardio-pulmonary disease is suspected, the patient is over 70 years old and has a history of cardio-pulmonary disease and hasn’t had a CXR or chest CT within the last six months, or if the patient is over 70 and has an unreliable history and is undergoing high-risk surgery”. These criteria are consistent with the Choosing Wisely ANZCA recommendation,13 and are supported by international literature.4–10
Despite the interventions to promote the Choosing Wisely recommendation, no apparent change in CXR ordering was observed. This qualitative study aimed to assess the effectiveness of the CPH campaign at spreading awareness of these guidelines, and to investigate what influences clinician behaviour in terms of pre-operative testing and how this behaviour could be changed. Although some literature exists internationally with regard to factors affecting clinician behaviour around pre-operative testing, to the authors’ knowledge a New Zealand study focusing on this issue has not been undertaken previously.15,16 Therefore this study aims to provide a New Zealand perspective, as well as support the international understanding of effective strategies for influencing clinician behaviour to reduce pre-operative testing.
Figure 1: The Choose Wisely poster used in the initial CPH campaign.
Based on information from the internal audit at CPH, doctors of varying seniority from the department of general surgery were selected for interview. Participants were recruited from a pool of 20 consultants, 22 registrars and 17 house officers who were identified retrospectively from rosters as being involved in pre-operative clinics from August to October 2018.
Doctors in each category were allocated a number. An online random number generator was used to generate five numbers for each category of seniority. Doctors corresponding to the numbers picked in each category were contacted by email to be recruited for interviews. When the doctors in a category declined to participate or failed to respond to repeat approaches, more numbers were randomly generated in order to pick further potential participants. Overall, seven house officers were contacted of whom five were interviewed while two did not respond; seven consultants were contacted of whom five were interviewed, one declined participation and one did not respond. Sixteen registrars were contacted of whom five were interviewed, four declined to be interviewed and seven did not respond. Of the 15 participants, seven were female and eight were male. Of the seven females, four were house officers, two were registrars and one was a consultant. At the time of the interview, one of the participants was found to have switched from a general surgery run to plastics surgery run during the intervention period. However, this participant would have been exposed to the campaign regardless and therefore was included in the study. All recruitment and interviews were conducted between 26 November and 12 December 2018.
Semi-structured one-on-one interviews were conducted in December 2018. Interviews covered four subject areas, 1) getting background information, 2) awareness and effectiveness of the Choosing Wisely campaign, 3) barriers to clinician behaviour change around pre-operative testing and 4) potential future interventions to change clinician behaviour. Interviews were audio recorded and conducted by a single interviewer using open-ended questions in order to reduce interviewer bias.
Data collected from the interviews were analysed using the Braun and Clarke thematic analysis methodology as a guide.17 Data were coded and then refined into major themes and sub-themes.
Figure 2: A mind map of the major themes and sub-themes.
Four themes and 17 sub-themes emerged from the interviews (Figure 2). The main themes included: awareness of Choosing Wisely; thoughts around pre-operative testing; barriers for changing clinician behaviour; and tools and strategies for reducing unnecessary pre-operative testing.
Awareness of Choosing Wisely
This theme can be broken down into two sub-themes: what clinicians knew about Choosing Wisely and how they acquired knowledge of Choosing Wisely.
Ten out of 15 participants had heard of Choosing Wisely before the interview. In these participants, their understanding of Choosing Wisely centred around picking the evidence-based investigations for their patients.
“It’s a practice of how to choose tests wisely and weighing up the risks and benefits in regards to studies and putting it through to our own everyday use.” –House officer 4
In terms of the usefulness of the CPH campaign in spreading awareness, the poster was seen by eight out of 15 participants, however only four of these had consciously recalled its message. The newsletter was only seen by five out of 15 participants of whom only two had read the full article, while only two out of 15 had seen the revised Health Pathways guidelines.
Therefore, few participants had both seen the campaign media and understood its message. For those interviewed, the CPH campaign had not been effective in spreading awareness of Choosing Wisely.
Thoughts around pre-op testing
This theme included opinions on current pre-operative testing and factors which influence it. Opinions included the presence of too many unnecessary tests with a need to minimise them, a need for patients to have blood tests for baseline, and need for flexibility based on the clinical scenario.
“There probably is a degree of unnecessary tests done I would say. When you’re in preadmissions, you're just obviously not really thinking too much about that side of things.” – House officer 2
Clinicians described factors which influenced their decisions around pre-operative testing. These included: the age of patient, type of surgery, medical history of patient, if patient was operated under the breast unit protocol and if pre-operative tests would be preferred by other clinicians involved in the patient’s care.
“The things that would be relevant to me would be the age of the patient, whether their surgery was involving the chest or diaphragm and whether they had known diseases of the lungs or heart that I needed to know about before their surgery.” – Consultant 1
“I think its preset by a consultant, whatever the consultant wants.” – House Officer 1
Barriers for changing clinician behaviour
Several barriers for changing clinician behaviour were raised. These included: lack of knowledge around new evidence, staff being busy and not having enough time, mental automation around practices, guideline-related barriers, lack of communication between clinicians, traditional practice of ordering tests and worry around patient safety if not tested.
“…every department is so busy, there’s this whole functioning in silos approach. So cross departmental or inter-departmental communication, interaction is actually very minimal.” – Consultant 2
Ordering tests as they are a part of traditional practice and the “what if?” scenario were recurrent barriers.
“In my opinion a lot of the times it’s what if something happens? What if we find something?” – Registrar 1
“People have been doing things a certain way for a number of years. They get used to it. There is also just an intrinsic fear that you will miss that one person and umm it will become a media storm and also you will cause patient harm. So I think its constantly fighting against the human… what’s the word superstitions and a bit of humanity in us all of trying to do the right thing for our patients.” – House officer 4
Guideline-related barriers raised included: the presence of too many guidelines, issues in accessing these guidelines and the reluctance of some clinicians to read posters or consult electronic guidelines. It was suggested that house officers do require more guidance in the clinics.
“Communication, I think there needs to be clear guidelines for the house surgeons in pre-admissions clinic.” – Consultant 5
Tools and strategies for reducing unnecessary pre-operative testing
Participants suggested several tools and strategies that could help in reducing unnecessary pre-operative testing. These included: auditing pre-operative testing decisions, providing adequate guidelines/protocols, having an appointed Choosing Wisely promoter in departments to provide reinforcement, having evidence-based educational presentations for all clinicians, having team meetings when new staff members join the team to improve communication, improving house officer orientation and/or removing house officers from pre-operative clinics.
The strategy most commonly suggested was educational presentations in departments. These presentations should be aimed at clinicians at all levels, be evidence-based and include up-to-date local statistics of unnecessary testing. Participants also suggested these presentations should introduce the protocols which clinicians were expected to use.
“It’s good to have the data there to back up the practice as opposed to just people saying that it’s probably not useful, so the data is necessary. And then it’s just a matter of bringing it to the audience that needs to hear it… I just feel because there are so many meetings, there is an opportunity to put it in a meeting.” –Registrar 1
“A bit more talking about it, in terms of having lectures and orientation.” –House officer 3
Many clinicians also expressed the need for house officer orientation, especially around pre-operative testing because it can be difficult for a junior doctor to get accustomed to a new role and its expectations.
“I think there is a lot of room for improvement in start of run orientations in all respects… from a house surgeon’s point of view it can be really clumsy and painful learning the practicalities of a run and so to have some sort of designated teaching time about preadmissions that included things about testing, that could be helpful.” – House officer 5
Regular auditing of pre-operative testing requests and learning from this was another strategy frequently mentioned by the participants.
“Yeah I think auditing will be useful.” – House officer 1
Participants expressed the need for guidelines/protocols to be clear and specific to certain surgeries, be approved by all consultants and be accessible both as physical copies in clinics and electronically. Another suggestion was to integrate guidelines into test ordering forms for better accessibility.
“You need a consistent approach from a consultant level and an agreed consistent approach because as soon as you get different people doing different things, if you haven’t got consensus, you are never going to be able to do that at a junior staff level.” – Consultant 3
The presence of guidelines or recommendations in a healthcare setting does not guarantee that these will be followed. A systematic meta-review by Francke et al found adherence rates of healthcare practitioners with guidelines varied between 27 and 67%.18 This review found that cessation of guideline implementation strategies often led to adherence rates returning to baseline levels.18 Not enough is known about what influences clinicians to change their behaviour when they become aware of a guideline. Clinician compliance with guidelines may be affected by a variety of factors, including characteristics of the guidelines, the implementation strategies used and factors relating to the clinician, patient or environment.18,19 Clinician-related barriers can include lack of awareness, lack of familiarity, lack of agreement, lack of self-efficacy, lack of outcome expectancy and inertia of previous practice.19 These findings are consistent with the results of this CPH research.
This small, qualitative CPH study reveals a number of useful insights into the behaviour of clinicians around pre-operative testing, and the usefulness of common strategies to raise awareness and encourage uptake of guidelines in this area.
Clinicians expressed their concern around there being too many guidelines in the clinics, which caused uncertainty about which guidelines were to be followed. Other important factors highlighted by the participants were the need for guidelines about pre-operative testing to be specific and be agreed upon by all the consultants in the department. In relation to the effectiveness of posters, the results show that doctors may notice them but they were unlikely to read them. Time pressures in a busy clinic contribute to this. In other instances clinicians may not need to refer to these guidelines if they have had experience dealing with related cases. These ideas suggest explanations for the low success of the CPH campaign in spreading awareness of Choosing Wisely.
Many of the barriers in this study are interlinked. For example lack of knowledge around new evidence together with staff being busy can lead to mental automation of ordering tests regardless of whether the tests are warranted.
Similarly, time pressures can lead to lack of communication between clinicians. This may lead to uncertainty around ordering tests, especially among junior staff who are often fearful of the ‘what if?’ scenario and may be unable to communicate their uncertainty with senior staff, who are often not present in pre-admissions clinics. This highlights the issue that house officers require more support in the clinics.
Interestingly, there was no mention of overdiagnosis or overtreatment by any of the clinicians interviewed. These can result from incidental findings during testing where a condition which may not be causing any harm is found and consequently treated. There is extensive literature available internationally which discusses the drivers of overdiagnosis and the need to reduce overdiagnosis.20–23 Minimising overdiagnosis and overtreatment to provide better medical care is also a goal of the Choosing Wisely initiative.11 Hence it may be worth exploring this issue in future studies.
Lack of communication can also influence pre-operative testing decisions made by staff members who are new to the department/team and are uncertain of the protocols. Therefore it would be beneficial to have team meetings when a new staff member joins the team to improve communication and clarify any doubts they may have.
The clinicians interviewed do believe there is a need to reduce unnecessary testing. Several tools and strategies identified through this research could be applied to change clinician behaviour around testing. This is supported by evidence from previous research into initiatives to reduce unnecessary pre-operative testing. An initiative in the US created a standardised process for pre-operative clinics which were led by nurse practitioners and physician assistants.18 Compared to the standard physician-led clinics, 4% of tests were unnecessary tests in the intervention clinics compared to 23% in the standard clinics.15 Based on this information, a case could be made for removing house officers in New Zealand from pre-operative clinics, with testing instead done by nurse practitioners according to protocols decided by senior medical staff.
A UK initiative to reduce unnecessary pre-operative investigations for elective procedures achieved high compliance by implementing a staff training program around unnecessary investigations and providing targeted written information in clinics.16
In our study, evidence-based educational presentations in departments and better house officer orientation/education were the strategies most commonly recommended by participants. Considering the previous literature, it is justified to believe that these strategies could be effective in reducing unnecessary pre-operative testing. Likewise there may also be value in having clearer guidelines which are available both in hard copy and electronically.
Although auditing testing decisions could be useful as a behaviour change tool, some clinicians in this study mentioned that this sort of intervention would need to be undertaken with care. Clinicians were concerned that the aggregate nature of audits would not take into account the unique aspects of individual cases. Another issue mentioned by clinicians was that it could be seen as a ‘blame game’. Clearly clinician concerns about the use of auditing and reporting of results would need to be addressed if this strategy is to be effective at influencing their behaviour.
There are a number of important learnings from this study, which can be used to inform future efforts to change clinician behaviour around the ordering of investigations, and also other Choosing Wisely initiatives to reduce unnecessary care.
Firstly, poster and electronic newsletter campaigns to raise awareness among clinicians appear to have limited effectiveness—both in terms of reaching the intended audience and in terms of changing their clinical practice. This is an important finding, as many Choosing Wisely initiatives are based around using similar methods to raise awareness and change behaviour of health professionals and patients.
Secondly, it is clear that changing clinical guidelines alone will not effect change in clinician behaviour. While there is a need to increase clinician awareness of evidence-based recommendations to reduce unnecessary testing, there is also a need to go beyond this in convincing clinicians to put these into practice, and removing obstacles to their uptake. As mentioned by participants in this study, this includes making the recommendations relevant to their practice, through sharing local audits of test ordering, providing space for discussion and agreement from all senior staff, with clear communication of expectations to current and new junior staff. In addition, a number of factors reinforce the tendency to over-order unnecessary tests, including juniority, fear of reprisals/mistakes and lack of time. These barriers also need to be addressed for the strategies above to be successful.
One major limitation of this study is the uncertainty whether our data is generalisable to clinicians in other places. This limitation stems from the qualitative nature of this study and its small sample size. As this study was restricted to CPH, it would be important to repeat a similar study in other regions of New Zealand to confirm whether clinicians around the country have similar thoughts and feelings around pre-operative testing and how inappropriate testing could be reduced.
Other limitations of this study include the lack of a quantitative component for evaluating the effectiveness of the CPH campaign; and the aspect that qualitative research and findings are subjective and can be influenced by the views of the researchers themselves. Hence if this study were to be repeated with different researchers, there could be discrepancies in the findings between these studies.
Alternatively the greatest strength of this study also stems from its qualitative nature. This being the ability to evaluate clinician reasoning around pre-op testing. Furthermore, this was done using Braun and Clarke thematic analysis, which is a robust and time-tested framework.
Conducting semi-structured opposed to structured interviews allowed gathering of richer quality data because interviewees had more room to express themselves and highlight points which otherwise may have been missed. Additionally, both the interview and analysis phases were conducted by the same researcher, minimising any information bias which could have arisen in translation. The final yet significant strength of this research is that it was fairly cost and time effective to conduct. Hence resources would not be a barrier if this study were to be repeated elsewhere.
Figure 3: Summary of findings and recommendations.
- Clinicians believe that there is a need to reduce unnecessary pre-op testing.
- CPH Choosing Wisely campaign media had limited effectiveness in spreading awareness.
- Creating/changing guidelines is not enough. Must address barriers to behavioural change to implement change in practice.
- Significant barriers include lack of communication between clinicians, lack of knowledge around protocols among junior staff, mental automation and traditional practices.
- Useful strategies for future campaigns include evidence-based educational presentations, improving communication between clinicians and conducting local audits.
Changing clinician behaviour to reduce the ordering of unnecessary pre-operative investigations requires a number of barriers to be addressed. The findings of this study suggest that a strategy that relies on revising guidelines and raising staff awareness alone is likely to be of limited effectiveness. In addition to increasing clinician awareness of evidence-based recommendations to reduce unnecessary testing, other strategies may be needed to support behaviour change, including sharing local audits of test ordering, providing space for discussion and agreement from all senior staff, and clear communication of expectations to current and new junior staff. Factors which reinforce the tendency to over-order unnecessary tests, including juniority, fear of reprisals/mistakes and lack of time also need to be addressed for behaviour change strategies to be successful.