Accurate pre-operative imaging plays a vital role in patient selection for surgery and in allocating stage-appropriate therapies to patients diagnosed with pancreatic cancer (PC). This study aims to: (1) understand the current diagnosis and staging practices for PC; and (2) explore the factors (barriers and enablers) that influence the use of a pancreatic protocol computed tomography (PPCT) or magnetic resonance imaging (MRI) to confirm diagnosis and/or accurately stage PC.
Semi-structured interviews were conducted with radiologists, surgeons, gastroenterologists, medical and radiation oncologists from the states of New South Wales (NSW) and Victoria, Australia. Interviews were conducted either in person or via video conferencing. All interviews were recorded, transcribed verbatim, de-identified and data were thematically coded according to the 12 domains explored within the Theoretical Domains Framework (TDF). Common belief statements were generated to compare the variation between participant responses.
In total, 21 clinicians (5 radiologists, 10 surgeons, 2 gastroenterologists, 4 medical and radiation oncologists) were interviewed over a four-month-period. Belief statements relevant to the TDF domains were generated. Across the 11 relevant domains, 20 themes and 30 specific beliefs were identified. All TDF domains, with the exception of social influences were identified by participants as relevant to protocol-based imaging using either a PPCT or MRI, with the domains of knowledge, skills and environmental context and resources being offered by most participants as being relevant in influencing their decisions.
To maximise outcomes and personalise therapy it is imperative that diagnosis and staging investigations using the most appropriate imaging modalities are conducted in a timely, efficient and effective manner. The results provide an understanding of specialists’ opinion and behaviour in relation to a PPCT or MRI and should be used to inform the design of future interventions to improve compliance with this practice.
Citation: Maharaj AD, Evans SM, Zalcberg JR, Ioannou LJ, Graco M, Croagh D, et al. (2020) Barriers and enablers to the implementation of protocol-based imaging in pancreatic cancer: A qualitative study using the theoretical domains framework. PLoS ONE 15(12): e0243312. https://doi.org/10.1371/journal.pone.0243312
Editor: Eugene J. Koay, MD Anderson Cancer Center, UNITED STATES
Received: August 31, 2020; Accepted: November 18, 2020; Published: December 17, 2020
Copyright: © 2020 Maharaj et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting Information files.
Funding: The authors of this study would like to acknowledge the Australian National Health and Medical Research Council (NHMRC) for the Pancreatic Cancer Registry for Quality Improvement (PCR4QI) grant [grant number APP1125395].
Competing interests: The authors declare no conflict of interest.
Accurate pre-operative imaging plays a vital role in patient selection for surgery and in allocating stage-appropriate therapies to patients diagnosed with pancreatic cancer (PC) . The recommended method of assessing operability is to use a high-quality multi-phase computed tomography (CT) scan that examines the abdominal area in the arterial and portal venous phase. Such a CT scan can determine the proximity of the tumour to major vascular structures and the presence of locally advanced disease or intra and extra-abdominal metastases [2, 3]. In 2012, the Society of Abdominal Radiology and the American Pancreatic Association released a consensus statement describing a standardised reporting template for the accurate staging of PC to improve disease management. This statement was authored by a multi-institutional group of experts comprising radiologists, gastroenterologists, and hepatopancreatobiliary surgeons . For accurate disease staging, it was recommended that all patients with no obvious metastatic disease or local invasion at initial routine CT, undergo a repeat examination with a dedicated pancreas protocol multiphase computed tomography (PPCT) prior to endoscopy, biliary stenting or invasive tissue sampling [4–6].
More recently, evidence-based clinical practice guidelines recommend that PPCT be used to assess the extent of the disease across all stages of PC [7–9]. Where there is clinical suspicion of PC, a PPCT is also considered to be the primary imaging modality to diagnose and stage the extent of the disease within one single session [10, 11]. The pancreas is anatomically intertwined with critical vascular structures: specifically, the celiac artery, hepatic artery, superior mesenteric artery, portal vein and the superior mesenteric vein. A PPCT evaluation of the vascular involvement is highly predictive of the extent of vascular involvement which determines operability and overall survival . In addition, a PPCT is also required for accurate planning and delivery of targeted radiotherapy, which requires clear delineation of the tumour in relation to normal pancreatic parenchyma and surrounding normal structures.
Magnetic resonance imaging (MRI) is considered by some to be equivalent to a CT in detecting and staging PC . However, recent evidence recommends the addition of MRI as an adjunct to detect the presence of liver metastases, rather than a replacement for PPCT . The liver is the most common organ affected by metastasis, and establishing the accurate extent of liver disease can help avoid futile attempts at surgical resection . It has been suggested that all patients with PC deemed resectable should undergo liver MRI to complement CT evaluation, but this is not always feasible due to limited resources and access to MRI .
A set of quality indicators for PC was developed by Australian clinicians in 2018 and compliance with these indicators is reported by the Upper Gastrointestinal Cancer Registry (UGICR) [14, 15]. Preliminary analysis of registry data found that 29% (n = 64/224) of patients with potentially resectable and 43% (n = 54/125) of patients with locally advanced disease (without metastases) did not undergo either a documented PPCT or MRI. In a disease where surgery is complex but remains the best chance for long-term survival in patients with localised disease, incomplete or sub optimal pre-operative staging can result in adverse outcomes, such as planned procedures and surgery being abandoned intraoperatively or a margin-positive resection, resulting in poorer survival .
Given the likely benefits of PPCT and MRI, and the evidence for variation in these practices, it is important to understand the reasons why patients with PC may or may not be receiving this care. Here we explore the barriers and enablers to the implementation of a PPCT or MRI for diagnosis and/or staging of PC. In so doing, we will inform tailored knowledge translation and quality improvement interventions which address modifiable barriers and enhance enablers of PPCT and MRI use.
This study aims to: (1) understand the current diagnosis and staging practices for PC; and (2) explore the factors (barriers and enablers) that influence the use of a PPCT or MRI to confirm diagnosis and/or accurately stage PC.
This study is underpinned by the theoretical domains framework (TDF). The TDF was developed through international collaboration between behavioral scientists and implementation researchers. The framework consolidates theories relevant to behavioral and psychological processes determining the influences on healthcare practitioners’ behavior. Constructs from these theories are grouped into domains (Fig 1) and provide a theoretical lens to identify the determinants of behavior [17–19].
Step one: Describing ideal practice—focus group
A focus group was convened comprising clinical experts involved in the treatment and management of PC to understand the steps involved in the use of PPCT and MRI in a clinical setting. Following a systematic approach for using the TDF to implement evidence-based interventions into practice , the group considered ‘who needs to do what differently, when, where and how’ and developed a decision tree which clearly articulated the clinical decisions and processes involved in appropriate execution of PPCT and MRI. This decision tree formed the reference, or ‘ideal practice’, of the interview questions; these were grouped into domains according to the TDF.
Step two: Understanding the barriers to and enablers of ideal practice- semi-structured interviews
Medical specialists involved in the treatment of patients with PC (radiologists, surgeons, medical and radiation oncologists and gastroenterologists) from public and private hospitals, within metropolitan and regional areas, across the states of New South Wales (NSW) and Victoria were invited to participate in semi-structured interviews underpinned by the TDF. A stratified purposeful sampling strategy was used to recruit participants, followed by snowball sampling . This initially involved inviting by email clinicians contributing to the UGICR and clinicians identified as specialising in PC on hospital websites . In addition, those interviewed were asked to nominate other specialists for researchers to contact. Participants who opted-in to the study provided consent through email. Non-responders were sent a reminder email.
An interview guide was developed exploring factors influencing the implementation of protocol-based imaging (CTs and MRIs) in the management of all patients diagnosed with PC. Open-ended questions were asked to cover each domain of the TDF. The interview schedule was initially piloted with two surgeons (DC and CHCP) and a nurse specialist (TD), following which it underwent further refinement. The interview schedule is provided as S1 File.
Two researchers (ADM—lead interviewer and SEG) conducted interviews either in person or using Zoom video conferencing (https://zoom.us/). With participants’ consent, interviews were recorded on video and audio using Zoom conferencing software, transcribed verbatim, de-identified and imported into NVIVO 12.0 Plus qualitative software (QSR International Pty Ltd. Version 12, 2018) for analysis.
Interview transcripts were thematically coded according to the 12 domains of the TDF. Overarching themes perceived to influence the implementation of protocol-based imaging and associated belief statements were then generated. A belief statement was defined as “a collection of responses with a similar underlying belief” that determined the relevance and role of the domain in influencing the behaviours around protocol-based imaging [17, 22].
A random sample of transcripts (n = 3, 10%) were analysed independently by two researchers, who then met to discuss the generation of themes and coding guidelines. Any discrepancies were discussed with another researcher for resolution. The full summary of results was independently reviewed and presented to the same focus group as Step One above over two meetings for feedback and interpretation of findings.
Twelve clinical experts were invited to the initial focus group and nine participated. The focus group comprised three surgeons, two medical oncologists, a radiation oncologist, a palliative care specialist, a nurse specialist, and a nurse care coordinator. The decision tree, presented in S1 Fig, clearly articulated the clinical pathway leading to a PPCT or MRI, and the appropriate precursors and outcomes. This was agreed by the focus group as the evidence- and consensus- based ideal clinical practice.
A total of 49 specialists were approached across the two states over a four month period (August–December 2019) and 21 (40%) participants were interviewed. Participant characteristics are summarised in Table 1. Most were male (n = 20, 95%); 71% were from metropolitan settings, and 62% were from Victoria. The case volume of surgeons varied by state and rurality. Surgeons in the state of Victoria and in regional areas commonly saw between one and three PC cases per week, compared to five to ten PC cases seen by surgeons working in NSW and metropolitan regions. This difference is likely related to the model of centralised care in NSW. Clinicians working in other disciplines from both states reported seeing from zero to two cases per week.
Factors influencing the implementation of a PPCT or MRI
Across the relevant domains, 20 themes and 30 specific beliefs were identified. Table 2 summarises the main belief statements, categorised according to the domains of the TDF. Each belief statement is demonstrated by representative quotes.
All TDF domains, with the exception of social influences were identified by participants as relevant to protocol-based imaging using either a PPCT or MRI, with the domains of knowledge, skills and environmental context and resources being offered by most participants as being relevant in influencing their decisions.
There was variation within the domain of knowledge about the differences between a PPCT or MRI and when each modality should be used for diagnosis and/or staging, even between radiologists. Some participants viewed the modalities as equivalent. However, the majority stated that there were clear differences outlining when they would use each. PPCT was viewed as the cornerstone for the staging of PC by the majority of specialists (beliefs about consequences) and most believed that if a normal single-phase CT scan or MRI showed clear metastatic disease, then there was no need for further protocol-based imaging (beliefs about consequences). Many viewed the MRI as a more challenging modality than a CT scan for a number of reasons. Some specialists described that the MRI required patients to cooperate for a longer period of time, generated anxiety in patients which could require sedation (emotions), and half the participants deemed that they were less accessible and more costly to the patient as they did not receive a rebate within the Australian context, (environmental context and resources). While there was a general awareness of some evidence-based clinical practice guidelines relevant to protocol-based imaging, not all were familiar with the recommendations within the guidelines and many did not know of any specific research. A further belief that emerged within this domain was the notion that organisational protocols were more relevant to practice than the guidelines itself (knowledge).
A number of surgeons and other specialists stated that they were not aware of, nor did they follow, any protocols or referral pathways to facilitate the implementation of protocol-based imaging. However, those who did follow clear, well laid-out protocols mentioned uniform management across the board, no delays in imaging and a higher volume of surgery undertaken that was concentrated amongst a limited number of surgeons (behavioural regulation). Many specialists believed that their organisation had the capacity to manage PC by providing the necessary resources including surgeons, radiologists, oncologists, intensive care and multidisciplinary team (MDT) services. Yet, a number of specialists mentioned that the volume of PC cases determined the focus within the organisation. For example, some described themselves as a “pancreatic centre of excellence” due to the high volumes and focus on PC cases managed within the organisation, whilst a minority alluded to the fact that within their organisations they were the only surgeon managing patients with PC and did not have a specific framework to streamline processes for patients (environmental context and resources) which may affect the implementation of a PPCT or MRI.
Overwhelmingly, having access to expert radiologists or radiology practices specialising in pancreatic radiology (i.e. “a good quality radiologist whom you trust”) was viewed as essential for the diagnostic and staging modalities by all craft groups (skills). Related to this was the strong belief that quality and interpretation of imaging provided by specialist radiologists play a large role in providing confidence to clinicians in the diagnosis and staging of PC (beliefs about capabilities).
It was acknowledged by around half of the participants that the pathway to the diagnosis and staging of PC is complex and can involve numerous disciplines (social/professional role and identity). An identified barrier to a timely PPCT or MRI is that referral from primary care to the most relevant specialist (beliefs about capabilities) can be intercepted or delayed by referral to other disciplines such as gastroenterology, specialists that may not necessarily undertake a PPCT or MRI as the first step. Conversely, in the metropolitan private sector, a potential barrier identified was the disincentive to undertake full imaging if an opportunity arose to take a patient with potentially resectable cancer straight to surgery (motivation and goals).
For all participants, the motivation to undertake a PPCT or MRI was to provide good quality care (motivation and goals). However, a belief reported by a minority that emerged as a barrier to PPCT or MRI from the domain of emotions was the potential for nihilism, especially in terms of management of older patients who were otherwise well enough to undertake surgical resection. A further belief reported by a few was that in some cases patients themselves were a barrier to optimal care and did not wish to know their diagnosis. In this context imaging may not be undertaken, placing value on the patient perspective (emotions).
Related to the patient perspective, a few participants mentioned distance to radiology services as a potential barrier for patients in regional or remote areas, while others recognised that often these patients were resigned to the fact that they would have to travel. Whether these decisions were overlayed as a consequence of patients’ preference or a result of clinical therapeutic nihilism is difficult to estimate (environmental context and resources). Related to patients with PC living in regional areas was the belief that “having a succession plan, an age range of clinicians… with a policy of working together”, and establishing sound practices, was a driving force that would enable services to remain in the regional vicinity and not be lost to the cities (nature of behaviours). It was further highlighted by regional specialists that a quality PPCT would detect the more complex cases such as those needing vascular resections, and these would then need to be referred to surgeons based in the metropolitan areas with expertise in this type of resection.
Appropriate implementation of a PPCT or MRI reduces the risk of an adverse outcome in patients who are diagnosed with non-metastatic PC. Proper staging including PPCT/MRI allied with assessment of co-morbidities enables informed personalisation of therapy to a surgery first, neoadjuvant, chemotherapy or supportive pathway, minimising futile interventions. Behavioural research can help identify the factors that influence the implementation of this best practice. To our knowledge this is the first study to explore the barriers and enablers to the use of PPCT or MRI for diagnosis and/or staging in PC using the TDF as a guiding framework. The use of the TDF allowed a comprehensive evaluation of specialists’ behaviour with respect to this practice, using a systematic approach. It provides a basis from which to tailor future interventions aiming to overcome barriers and harness enablers, improving uptake of PPCT or MRI and therefore potentially improving outcomes for patients diagnosed with PC. Further, our approach to this study was based on methods implemented by Graco and colleagues which have been important in categorising the opinions of specialists . However, in our study we opted not to include the frequency of belief statements to avoid important beliefs being overlooked.
Current practice shows that diagnosis of PC is often reliant on preliminary imaging such as a single-phase CT scan or an endoscopic ultrasound (EUS), followed by referral to appropriate specialists who may then order a PPCT to stage PC preoperatively if appropriate (S1 Fig). Knowledge and skills were revealed as important domains in this study, as it was not widely understood that there is a hierarchy approach to the application of PPCT or MRI. We did not anticipate this gap in knowledge or awareness related to the recommendations within clinical practice guidelines and the differences in perspectives on the information provided by a PPCT compared to MRI. The PPCT has higher accuracy in identifying the extent of PC, locoregional extension, vascular invasion, distant metastases and resectability, whereas the MRI has some advantages over CT in detecting small tumours, distant metastases, especially to the liver, and isoattenuating tumours . Therefore, MRI is best reserved for diagnosis rather than staging of PC, particularly as highlighted in our study, it is a more challenging modality both from a clinical and patient perspective.
Pre-operative staging using a PPCT, ideally reported using standardised templates, provides confidence and enhances surgeons’ ability to better select patients for surgical resection . A practice of concern was the motivation to undertake surgery without the necessary preoperative staging by some surgeons. Without proper staging using a PPCT, the likelihood of achieving complete curative resection is at risk and patients who may have benefited from neo-adjuvant therapy may miss out due to surgery undertaken too early . Organisations, should have systems in place to ensure patients with clinical suspicion of PC undergo proper staging to better select patients likely to benefit from surgery with curative intent. A potential safeguard is the mandatory presentation of patients to MDT meetings prior to management decisions being reached, especially where considerable multidisciplinary expertise is essential for optimal care and all the more so in smaller/low volume centres . The use of protocols and algorithms for investigation and staging will be more likely to ensure uniformity and equity of access to care .
Although not viewed as an impediment to receiving a PPCT or MRI, several factors resulted in delays to receiving appropriate imaging. Such factors included referrals to other specialists, the distance travelled to reach radiological services, as well as the technical details of the actual scans, accessibility of the reports and finally the quality of the scan and interpretation of the radiological report. A cross-sectional study assessing presentations to general practitioners’ (GP) before a cancer diagnosis found that PC was associated with the highest risk of presenting multiple times before being referred to an appropriate specialist . Further research is required to understand the timeliness of referrals, the impact of distance on protocol-based imaging and the quality of imaging in primary care. Participants in this study believed that radiologists and radiology services that specialise in pancreatic radiology produce imaging and reports of higher quality.
An opportunity also exists for radiologists to play a more proactive role in contacting the primary provider to hasten the implementation of a PPCT in a single session if they become suspicious of pancreatic aetiology. Therefore, further research would explore: (1) the GP’s understanding of the importance of referring patients to specialist radiology services; and (2) radiology centres taking responsibility for ensuring that patients with a suspicion of PC receive the best available service (requiring the necessary expertise in pancreas radiology).
Within the domain of environmental context and resources, it was widely understood that without access to PPCT and skilled radiologists, hospitals with a low volume of cases may be disadvantaged compared to high-volume tertiary metropolitan-based centres. A meta-analysis reviewing the relationship between hospital volume and outcome confirmed the association between higher hospital volume and lower post-operative morality and lower length of stay .
This study provides detailed insights into the perceptions of specialists involved in deciding whether a patient has a PPCT or MRI and whether it is undertaken. The breadth of the specialists involved in the project across two states in Australia provides a novel and broad view of barriers and facilitators.
Our study had some limitations. Radiologists who had agreed to participate in the focus group were unable to attend on the day of the meeting. We had less participation from disciplines other than surgeons in NSW, and this limits our ability to report on a range of perspectives from this state. In a similar respect we may not have captured all perspectives from individual disciplines due to lower participation. For example, only one radiation oncologist was interviewed in this study. This may bias the findings of our study. Further, a well-known concept in qualitative research is data saturation. We interviewed a heterogenous population. However, data was collected until no new information presented on the barriers and enablers within the 12 TDF domains .
Our study highlights the importance of the TDF domains of knowledge, skills and environmental context and resources in understanding the diagnosis and staging practices for PC.
Clinical practice guidelines that are often systematically developed, based on the most current evidence, have the potential to improve health outcomes and provide consistency of care . Deviation from guidelines can be a perceived barrier if organisational protocols or procedures do not align with international, peer-reviewed, evidence-based guidelines. Organisational protocols should also be revised and kept up-to-date with current evidence . Given that a barrier to the routine implementation of PPCT is the lack of knowledge, we recommend the dissemination of knowledge to clinicians, and mechanisms to profile such guidelines through forums that provide educational opportunities.
There is evidence that audit and feedback, and reminders can attain positive behaviour change . Further, an earlier review highlighted a lack of health services research examining the influence of guidelines in PC management . This study adds important literature on the use of clinical pathways to facilitate the implementation of guidelines by identifying practices and beliefs that prevent evidence-based care. An important consideration is the role of MDT meetings as a reminder for undertaking PPCT or MRIs, providing access to a range of relevant disciplines, especially specialist radiologists and a forum for receiving feedback. While MDT meetings are endorsed as best practice in cancer care, they are not mandatory and this can limit the consistency in care received by patients .
It is imperative that diagnosis and staging investigations using the most appropriate imaging modalities are conducted in a timely, efficient and effective manner. This qualitative study used a knowledge translation approach and psychological theories in health to explore factors associated with implementing the appropriate imaging investigations for diagnosis and/or staging in PC. The results provide an understanding of specialists’ opinion and behaviour in relation to a PPCT or MRI and should be used to inform the design of future interventions to improve compliance with this practice.
- 1. Zins M, Matos C, Cassinotto C. Pancreatic Adenocarcinoma Staging in the Era of Preoperative Chemotherapy and Radiation Therapy. Radiology. 2018;287(2):374–90. pmid:29668413
- 2. Burge ME, O’Rourke N, Cavallucci D, Bryant R, Francesconi A, Houston K, et al. A prospective study of the impact of fluorodeoxyglucose positron emission tomography with concurrent non-contrast CT scanning on the management of operable pancreatic and peri-ampullary cancers. HPB (Oxford). 2015;17(7):624–31. pmid:25929273
- 3. Kartalis N. CT and MRI of pancreatic cancer: there is no rose without a thorn! Eur Radiol. 2018;28(8):3482–3. pmid:29796919
- 4. Al-Hawary MM, Francis IR, Chari ST, Fishman EK, Hough DM, Lu DS, et al. Pancreatic Ductal Adenocarcinoma Radiology Reporting Template: Consensus Statement of the Society of Abdominal Radiology and the American Pancreatic Association. Gastroenterology. 2014;146(1):291–304.e1. pmid:24355035
- 5. Gandy RC, Barbour AP, Samra J, Nikfarjam M, Haghighi K, Kench JG, et al. Refining the care of patients with pancreatic cancer: the AGITG Pancreatic Cancer Workshop consensus. Med J of Aust. 2016;204(11):419–22. pmid:27318402
- 6. Liles JS, Katz MHG. Pancreaticoduodenectomy with vascular resection for pancreatic head adenocarcinoma. Expert Rev Anticancer Ther. 2014;14(8):919–29. pmid:24833085
- 7. Sohal DPS, Kennedy EB, Khorana A, Copur MS, Crane CH, Garrido-Laguna I, et al. Metastatic Pancreatic Cancer: ASCO Clinical Practice Guideline Update. J Clin Oncol. 2018;36(24):2545–56. pmid:29791286
- 8. Khorana AA, Mangu PB, Berlin J, Engebretson A, Hong TS, Maitra A, et al. Potentially Curable Pancreatic Cancer: American Society of Clinical Oncology Clinical Practice Guideline Update. J Clin Oncol. 2017;35(20):2324–8. pmid:28398845
- 9. Balaban EP, Mangu PB, Khorana AA, Shah MA, Mukherjee S, Crane CH, et al. Locally Advanced, Unresectable Pancreatic Cancer: American Society of Clinical Oncology Clinical Practice Guideline. J Clin Oncol. 2016;34(22):2654–68. pmid:27247216
- 10. Australian Gastro-Intestinal Trials Group (AGITG). A consensus statement on definition of surgical standards for pancreatic cancer. 2015.
- 11. Lee ES, Lee JM. Imaging diagnosis of pancreatic cancer: a state-of-the-art review. World J Gastroenterol. 2014;20(24):7864–77. pmid:24976723
- 12. Zhang L, Sanagapalli S, Stoita A. Challenges in diagnosis of pancreatic cancer. World J Gastroenterol. 2018 May 21;24(19):2047. pmid:29785074
- 13. Jeon SK, Lee JM, Joo I, Lee DH, Ahn SJ, Woo H, et al. Magnetic resonance with diffusion-weighted imaging improves assessment of focal liver lesions in patients with potentially resectable pancreatic cancer on CT. Eur Radiol. 2018:1–10.
- 14. Maharaj AD, Ioannou L, Croagh D, Zalcberg J, Neale RE, Goldstein D, et al. Monitoring quality of care for patients with pancreatic cancer: a modified Delphi consensus. HPB (Oxford). 2019;21(4):444–55. pmid:30316625
- 15. Maharaj AD, Holland JF, Scarborough R, Evans SM, Ioannou L, Brown WA, et al. The Upper Gastrointestinal Cancer Registry (UGICR): a clinical quality registry to monitor and improve care in upper gastrointestinal cancers. BMJ Open 2019; 9:e031434. pmid:31575580
- 16. Torgeson A, Garrido-Laguna I, Tao R, Cannon GM, Scaife CL, Lloyd S. Value of surgical resection and timing of therapy in patients with pancreatic cancer at high risk for positive margins. ESMO Open. 2018;3(1):e000282. pmid:29387477
- 17. Atkins L, Francis J, Islam R, O’Connor D, Patey A, Ivers N, et al. A guide to using the Theoretical Domains Framework of behaviour change to investigate implementation problems. Implement Sci. 2017;12(1):77. pmid:28637486
- 18. Michie S, Johnston M, Abraham C, Lawton R, Parker D, Walker A. Making psychological theory useful for implementing evidence based practice: a consensus approach. Qual Saf Health Care. 2005;14(1):26. pmid:15692000
- 19. Maharaj AD, Evans SM, Zalcberg JR, Ioannou LJ, Graco M, Croagh D, et al. Barriers and Enablers to the implementation of mulitdisciplinary team meetings: a qualitative study using the theoretical domains framework. BMJ Qual Saf Epub ahead of print: [25 Nov 2020]. pmid:33247002
- 20. French SD, Green SE, O’Connor DA, McKenzie JE, Francis JJ, Michie S, et al. Developing theory-informed behaviour change interventions to implement evidence into practice: a systematic approach using the Theoretical Domains Framework. Implement Sci. 2012;7(1):38. pmid:22531013
- 21. Palinkas LA, Horwitz SM, Green CA, Wisdom JP, Duan N, Hoagwood K. Purposeful sampling for qualitative data collection and analysis in mixed method implementation research. Adm Policy Ment Health. 2015; 42(5):533–44. pmid:24193818
- 22. Francis JJ, Stockton C, Eccles MP, Johnston M, Cuthbertson BH, Grimshaw JM, et al. Evidence-based selection of theories for designing behaviour change interventions: using methods based on theoretical construct domains to understand clinicians' blood transfusion behaviour. Br J Health Psychol. 2009;14(Pt 4):625–46. pmid:19159506
- 23. Graco M, Berlowitz DJ, Green SE. Understanding the clinical management of obstructive sleep apnoea in tetraplegia: a qualitative study using the theoretical domains framework. BMC Health Serv Res. 2019;19(1):405. pmid:31226999
- 24. Tummers WS, Willmann JK, Bonsing BA, Vahrmeijer AL, Gambhir SS, Swijnenburg R-J. Advances in Diagnostic and Intraoperative Molecular Imaging of Pancreatic Cancer. Pancreas. 2018;47(6):675–89. pmid:29894417
- 25. Brook OR, Brook A, Vollmer CM, Kent TS, Sanchez N, Pedrosa I. Structured Reporting of Multiphasic CT for Pancreatic Cancer: Potential Effect on Staging and Surgical Planning. Radiol. 2014;274(2):464–72. pmid:25286323
- 26. Horlait M, Baes S, Dhaene S, Van Belle S, Leys M. How multidisciplinary are multidisciplinary team meetings in cancer care? An observational study in oncology departments in Flanders, Belgium. J Multidiscip Healthc. 2019;12:159–67. pmid:30863082
- 27. Feldman MK, Gandhi NS. Imaging Evaluation of Pancreatic Cancer. Surgical Clinics. 2016;96(6):1235–56. pmid:27865275
- 28. Lacey K, Bishop JF, Cross HL, Chondros P, Lyratzopoulos G, Emery JD. Presentations to general practice before a cancer diagnosis in Victoria: a cross‐sectional survey. Med J Aust. 2016;205(2):66–71. pmid:27456447
- 29. Hata T, Motoi F, Ishida M, Naitoh T, Katayose Y, Egawa S, et al. Effect of Hospital Volume on Surgical Outcomes After Pancreaticoduodenectomy. Ann Surg. 2016;263(4):664–72. pmid:26636243
- 30. Saunders B, Sim J, Kingstone T, Baker S, Waterfield J, Bartlam B, et al. Saturation in qualitative research: exploring its conceptualization and operationalization. Qual Quant. 2018;52(4):1893–907. pmid:29937585
- 31. Woolf SH, Grol R, Hutchinson A, Eccles M, Grimshaw J. Clinical guidelines: potential benefits, limitations, and harms of clinical guidelines. BMJ (Clin Res ed). 1999;318(7182):527–30.
- 32. Kim FJ, da Silva RD, Gustafson D, Nogueira L, Harlin T, Paul DL. Current issues in patient safety in surgery: a review. Patient Saf Surg. 2015;9:26–. pmid:26045717
- 33. Johnson MJ, May CR. Promoting professional behaviour change in healthcare: what interventions work, and why? A theory-led overview of systematic reviews. BMJ Open. 2015;5(9):e008592. pmid:26423853
- 34. Gagliardi AR, Soong D, Gallinger S. Identifying Factors Influencing Pancreatic Cancer Management to Inform Quality Improvement Efforts and Future Research: A Scoping Systematic Review. Pancreas. 2016;45(2):161–6. pmid:26752254
- 35. Slavova‐Azmanova NS, Johnson CE, Platell C, Bydder S, Saunders CM. Peer review of cancer multidisciplinary teams: is it acceptable in Australia? Med J Aust. 2015;202(3):144–7. pmid:25669477