Team functioning across different tumour types: Insights from a Swiss cancer center using qualitative and quantitative methods

Abstract Background Multidisciplinary care is pivotal in cancer centres and the interaction of all cancer disease specialists in decision making processes is state‐of‐the‐art. Aim To describe differences of MDTMs by tumour type. Methods Twelve multidisciplinary team meetings (MDTMs) with participation of different cancer disease specialists at a tertiary hospital were assessed by an exploratory sequential mixed method approach with interviews, observations and a survey to address the following five topics: organisational structure and supporting technology; leadership; teamwork; decision‐making, perceived value and motivation. Thirteen persons with different tumour specialities and levels of seniority were interviewed. The 12 MDTMs were observed twice by uninvolved persons and evaluated by the participating physicians with a survey. Results There were no systematic differences between MDTMs for different tumour types with the exception of the non‐disease specific type MDTM, which was the only one for which the organisational structure was not driven by an electronic tool. However, several factors could be identified that generally influenced the functioning of the MDTMs. In particular, the quality of decision‐making was highly dependent on the availability of case‐based information and the presence of relevant cancer disease specialists. Leadership and teamwork were rated as important and were comparable across the MDTM. Team participants' motivation and perceived value of MDTMs was high across all meetings. Conclusion MDTM at a single institution did not demonstrate disease specific characteristics. An effective MDTM, irrespective of the tumour type, can be successfully structured by technical means and a chairperson coordinating the interaction of cancer disease specialists to improve the decision‐making process.

Conclusion: MDTM at a single institution did not demonstrate disease specific characteristics. An effective MDTM, irrespective of the tumour type, can be successfully structured by technical means and a chairperson coordinating the interaction of cancer disease specialists to improve the decision-making process.
cancer management, behavioral science, cancer care, cancer education 1 | BACKGROUND Multidisciplinary team meetings (MDTMs); otherwise termed tumour boards or cancer conferences) are a key element for care planning by cancer disease specialists at cancer centres internationally. 1,2 Such teams typically include surgeons, oncologists, radio-oncologists, pathologists and radiologists (and in some settings cancer nurses) involved in the case-based discussion of a cancer patient and decision making about their care. The underlying premise of MDTMs is to facilitate cross-speciality collaboration and to elicit a multidisciplinary expert review of a patient's tumour and condition, which would not be available through a single cancer specialist. This multidisciplinary aspect of cancer care is well recognised and was recently underlined by an initiative of European representatives of cancer disease specialists. 3 The MDTM is therefore the main vehicle for the elicitation of expert clinical opinions on recommended patient care; and for the integration of these opinions into the decision-making process for each patient.
As observed in other areas of healthcare delivery, 4 the task of a MDTM of coordinating several cancer disease specialists within the multiple treatment options for an individual patient is complex and can be a challenge. 5,6 Cancer treatment processes managed by teams are subject to several internal and external influences. 7 Collaboration and teamwork are increasingly recognised as central aspects of a successful cancer patient care. 8 In the wider healthcare literature, the argument has been consistently made over the past decade that team functioning can be improved through analysis of a team, identification of any deficits and implementation of interventions. 9,10 How to specifically support the way cancer MDTMs function, however, remains less well understood. While there is a plethora of research on decision-making in organisations and teams 11 in medical institutions, 2,12 literature on teamwork and leadership in cancer MDTMs and its implications for patient care is only beginning to emerge. 13,14 This research has started to demonstrate that factors enabling MDTM decisionmaking are defined by both objective parameters and subjective factors.
Objective parameters include facets such as availability of radiology and pathology reports and the presence of all required cancer disease specialists at the MDTM. Subjective factors include aspects of team working such as the quality of team members' contributions to case reviews during meetings, and the team leaders' inputs. 15,16 One of the questions that to-date remains debatable is whether MDTM-working is broadly determined by the same human and organisational factors across all tumour types, or in contrast, whether successful MDTM-working modes are tumour-specific. Evidence on this is very limited. A national self-report survey study from the UK analysed data collected in 2009 on MDTM-working across different tumour types.
The study revealed that MDTMs in the UK were largely in agreement regarding their set up. 17 However, there were some significant crosstumour differences in a few areas, including meeting preparation, selection of patients for MDTM review, and the overall perceived usefulness of the team inputs into the decision-making. Further, this study revealed that haematology team members responded significantly differently compared to solid tumour experts. We are not aware of any further studies addressing specifically this question, which is important from the perspective of infrastructure setup and support offered to cancer teams in large cancer centres globally. This is an important evidence gap, which needs addressing before we consider potential team support or training interventions for cancer teams and MDTMs.
This study aimed to start addressing this gap by evaluating all MDTMs at a cancer centre in Switzerland. Our primary objective was to determine whether the studied MDTMs differed by tumour type regarding their supporting organisational and technological structures, leadership, and a wide range of human factorsincluding quality of team working, decision making and perceived value of the MDTM to cancer care. Our secondary objective was to add to the evidence base on MDTM organisation and support. In doing so, we aimed to produce evidence to inform Swiss cancer policy, as there is currently no formal guidance regarding how to structure and organise cancer teams in Switzerland.

| Study design and setting
This was a descriptive mixed-method study, which used an exploratory sequential design including a number of data collection techniques to collect qualitative and quantitative data on MDTMworking. 18,19 The study was conducted at a single institution of a tertiary hospital in Switzerland. The study was reviewed and approved as a quality of care evaluation by the relevant hospital board; and all participants consented to the study data collection procedures (described below) prior to the data collection.

| Study sampling framework and procedure
At the time of study conduct, 12 MDTMs for different tumours were established at the hospital. All of these teams and the staff attending the MDTMs represented the study population and were eligible for inclusion in the study (see section 2.3 for details of the sampling done for the different study phases/methods).
The study was carried out in multiple phases including a characterisation of MDTMs and their practices, exploratory interviews with individual team members, real-time observations of MDTMs, and a self-report survey for all MDTM members. To address the primary objective, the types of data collected through the different methods are summarised according to the five main topics of interest (Table 1).

| MDTM characterisation
The main characteristics of MDTM including organisational structure and procedures in place (e.g., type of IT support, meeting frequency and leadership etc.) were collected by the first author (FH) for each of the hospitals' MDTM based on current daily practice. These aspects of MDTM-working can impact on the team interactions, hence were collated for a subsequent synthesis of the findings.

| Interviews with team-members
The exploratory interviews aimed at obtaining detailed information on the functioning of the teams from individual MDTM members and to develop a better understanding of the complex social processes and potentially controversial issues around MDTMs in order to inform the content of the quantitative survey. The semi-structured interviews were based on an existing interview guide used with UK cancer MDTMs and adapted for the purpose of this study. 20 Key questions addressed participants' opinions on MDTM attendance/role as participant, information presentation, case discussion, leadership, decisionmaking process, facilitators/barriers to effective team working, the motivation to participate and the value of the MDTM including their educational benefit (Table 1).
Interviewees were sampled purposefully to ensure representation across different specialties and hierarchy levels. Thirteen interviews with 10 heads of department/consultants and three trainees were conducted face-to-face covering the following specialties: pathology (N = 1), radio-oncology (N = 1), surgery (N = 3), oncology (N = 3), radiology (N = 1), haematology (N = 1), neurology (N = 1), dermatology (N = 1) and gynaecology (N = 1). An interviewer experienced in conducting semi-structured interviews (KR), who was not a member of any of the MDTMs conducted the interviews. Oral consent was given by the interviewees before the recording started. Each interview lasted between 30 and 45 min; all interviews were audiotaped and transcribed verbatim. Each transcript was analysed by two researchers independently using standard recommended analytic techniques. 21 In brief, the coding was thematically guided by the interview questions and complemented with an inductive thematic analysis. In addition, one of the researchers (MK) specifically searched for phenomena and processes known from research on decision making in groups, 22 multi-disciplinary team 23 and communication culture in organisations, 24 which may be implicitly expressed by interview participants. The identified themes were discussed among all authors to decide how they would subsequently be used to inform the content of the study survey. Verbatim quotes were extracted from the interviews to illustrate the identified themes.

| Real-time MDTM observations
For the real-time team observation during MDTM two assessment tools were applied, the Metric for the Observation of Decision-making (MDT-MODe) 20,25 and the ATLAS tool. 26 The MDT-MODe has been developed specifically to assess cancer teams' interactions in real-time and was validated for use across several tumour types. It allows a quantitative assessment of patient and disease related information discussed during the MDTM (patient history, imaging, pathology, comorbidity, psycho-social information, patient knowledge by the presenting physician) and also assesses the level of team working during a case review. The validated ATLAS tool has been developed to assess elements of team leadership in the context of cancer care with 12 chairing criteria (Table 1).
We Observations were analysed descriptively (means and standard deviations) for each criterion and according to MDTM. If available, we first calculated the mean of the two observers for each patient. If one observer had been unable to provide a rating for a specific element included in MDT-MODe or ATLAS, we included the single rating in the analysis. Means and standard deviations were calculated for each element scored. We also descriptively analysed the MDTM treatment decision taken for each patient (decision taken; decision suspended; or observers did not agree).

| Team-members survey
The self-report MDTM-members' survey was developed based on the themes identified in the interviews ( Table 1). The survey included items addressing teamwork (six items), leadership (five items) and decision-making (two items) in MDTMs. Participants were asked to rate if they agree with each statement on a scale ranging from 1 = not at all to 7 = absolutely. We further assessed motivation for participation in the meeting (single item, with eight response options) and perceived value of the MDTM a participant had just attended T A B L E 1 Topics addressed within the interview, observation and survey by qualitative and quantitative methods Note: Observation: A 1-5 scale to evaluate was used, being 5 the best answer on the scale with the following interpretation: high (4-5); moderate (2-3); low (0-1). Survey: Rating of 1-7 of a Likert scale was used.
(single item, with seven response options). The questionnaire was pilot tested with regard to feasibility and comprehension prior to data collection. Data collection took place once at the end of each meeting. Each survey question was analysed individually, with responses to each question summarised descriptively (means and standard deviations) and according to MDTM. There was no formal statistical testing of differences observed between MDTMsas we did not have any such hypothesis to formally evaluate. The study was therefore not designed or powered to evaluate cross-MDTMs differences.
In the final stage of the study, all data from the different sources were reviewed synthesised in order to produce a global assessment of MDTM working across tumour types.

| Exploratory interviews with team-members
Individual interviews revealed that participants of MDTMs were prepared for their case presentation. The standardisation of information presentation and the prioritisation of cases based on their complexity were factors seen as contributing to the efficiency of MDTMs: The diagnosis is shown, the underlying disease, secondary diseases. Histology, images, -X-rays, CT images, MRI's are presented. Everything that is done in terms of examination and diagnostics for this case is discussed together and interactively (I13, head of department) Interview participants considered the chairperson's ability to structure the discussion, prioritise cases, control time management, value individual opinions and leading through controversial discussions as important attributes of effective team leadership: Someone who bundles all the information that is being discussed and then makes the final decision and says: This is how we're going to do it now. But everyone has to agree (I3 trainee).
They also considered the expertise of each participant an important feature to successful MDTMs, but indicated that the meeting should not be dominated by opinionated individuals: The tumour board is only as good as the individuals who are on it. The more experience they have, the better the professional quality of the decisions (I2, head of department).
Problematic issues in MDTMs included hidden conflicts between chairperson and disease experts, unprepared experts, and case presentations that were inserted into the meeting late (hence lack of preparation for them) or had missing information.
Interview participants mentioned the ability to reach consensus, good camaraderie within the team, a psychologically safe atmosphere and a constructive culture of debate as facilitators for effective team working: One is allowed to speak freely and is not somehow cut off or 'Be quiet now'. Yes, it is actually a discussion with each other. At least in this tumour board it is a 'with each other' (I3 trainee).
In contrast, barriers identified by the interview participants included power struggles between specialities, the imposition of clinical opinions by individuals, different views on who is qualified to take a decision and decision-making occurring outside of the MDTM: The different points of view are problematic, or: Who has sovereignty over patients? The surgeon thinks differently than the oncologist or the gastroenterologist. And sometimes narcissistic personality structures come into play, which then make the decision no longer objective (I6, Head of department).
All interview participants shared the opinion that the attendance of cancer disease specialists from different disciplines is necessary for successful decision-making, but they also indicated that it might be a platform for demonstrating individual power: Now and then because of political infighting -so to speak the meetings take a bit long. It becomes more of a power show than really patient-centered (I7, Head of department).
Lastly, MDTMs were seen as a forum for clinical education. However, the motivation to participate was reported to be sometimes hampered by lack of time or conflicting clinical priorities: The potential for education is very high, but there are far too few trainees attending the meetings (I8, Head of department).

| External observers' ratings of MDTM
The observer analysed 153 patient cases, with a minimum of four and a maximum of 36 patient cases per MDTM and 8-72 observations, respectively. Ratings from both observers were available for 137 of these patient cases. MDT-MODe was the tool used to score patient discussions.
The ratings for patient and disease related information regarding patient history and patient being known by the presenting physician resulted in high mean scores. Imaging was always present, whereas pathology reports were presented with large variation across the MDTMs. The discussion of psycho-social information and patient's preferences were scored low in all of the observed meetings ( Table 3).
The observations of experts' contribution from and radiology revealed largely divergent results for different MDTMs. Most striking was the result for soft tissue and sarcoma MDTM with excellent scores (mean ratings above three) for the contribution of radiologists and pathologists. Other MDTMs with presence of these disciplines scored lower (mean rating 1.5 for gastrointestinal neoplasm MDTM) but reached a mean score above three for radiologists.
In 125 (82%) patient cases in the observed MDTMs a clinical treatment decision was reached. Suspended decision (unclear or missing) were reported in 10 cases (7%). Six out of 10 cases (7%) with suspended decisions were in the non-disease-specific MDTM. In 119 of 137 cases with two observations the observers were in agreement on whether a decision was taken or not (Tables 3 and 4).  Note: aparticipation of surgeon, medical oncologist, radio-oncologist, radiologist, pathologist. bparticipation of surgeon, medical oncologist, disease related specialist as neurologist, dermatologist, haematologist. cparticipation of surgeon, medical oncologist; no presence of radiologist, no pathologist.

| Survey results
Of the available population of 183 team-members in attendance of the meetings, we received 181 surveys (response rate = 99%).
The survey participation reflects a typical constitution of MDTM: Seniority levels were represented with 8% heads of department, 32% senior consultants, 30% junior consultants and 30% trainee physicians. Participants' specialties were as follows: 40 medical oncologists, 79 surgeons, 10 radio-oncologists, nine gastroenterologists, four haematologists, three neurologists, and two dermatologists, laboratory experts included nine radiologists, and 10 pathologists.  However, there were some differences seen between experts for several responses: Medical oncologists chose taking better diagnostic and T A B L E 3 Observer ratings for patient and disease related information in means and standard deviation

| Discussion
Our study met both of its objectives. Using multiple methods and sources of data, we were able to characterise in detail the struc- and other areas that seem to be lacking. Lastly, we were able to produce a detailed analysis for a Swiss cancer centre that is directly comparable on how cancer teams work and how they can be supported. Overall, we studied descriptively 12 MDTMs within a sin- which appeared to also be the case in our study. 28 The further potential of clinical decision support technology in MDTMs for assistance, preparation, data collection and documentation of decision is well documented. 29 Further to the technology, we also found that some MDTMs had an established and clearly defined chairpersonwhich again was seen as facilitative of the overall case review and an active coordinator of the meeting flow. In light of these findings it follows that IT support is part of the infrastructure host a hospital should provide for the cancer teams. Also, within the teams, a designated chairperson is needed.
Second on leadership, we observed that who chaired the MDTMs was typically defined by the hierarchy of the experts in attendance, such that chairing was typically undertaken by the moist senior person present. Anecdotally, this is not an uncommon practice (in healthcare meetings overall, we believe). In this study, the views of MDTM members in the survey were very positive, and this pattern was confirmed by high scores on the observational assessment of leadership during the meetings. The emerging picture is that, in the centre we studied, leadership of MDTMs seemed to be well-applied and received. We are unable to comment on whether an alternative model to team leadership might work better, as such a comparison was not available. A recent study revealed that, in naturally occurring cancer MDTMs, in which experts took turns to speak during the meeting, the pace of speech was very fast, such that the natural pauses (i.e., opportunities to speak) were few. 30 The risk this presents for a fast-paced MDTM is that the conversation might be dominated by some vocal individualsa risk that we noted in our interview analyses in this study too. These on facilitating case reviews and involve experts into the discussion to ensure their contribution to the decision-making process. 33 We would propose to our colleagues who regularly attend (and chair) MDTMs to consider and trial this alternative strategy, and report their experiences with it. Added to this, the evidence suggests that professional development and support to MDTM current chairs should be offered by cancer centers to enable them to undertake their role effectively.
Third, on participation and involvement, we obtained an interest-

| Strengths and limitations
The study has several strengths. We chose to combine qualitative and quantitative methods to reflect both an external view on an MDTM (MDTMs characterisation and observation methods) and an internal view with information coming directly from team-members (interview and survey methods). We applied validated observational tools (MDT-MODe and ATLAS) specifically designed for MDTMs and applied by trained observers, which overall worked well in the Swiss clinical setting. The joint application of these methods and analysis of the findings allowed us to produce a comprehensive picture of the studied MDTMs.
Limitations included that the study was carried out at a single institution, hence we cannot be certain to what extent the findings reflect an institutional perspective. The generalizability of our observations to other Swiss cancer centres or elsewhere remains to be tested through multi-institutional studies. Moreover, due to the purely descriptive nature of the study, we are not able to comment on how well the MDTMs studied performed in relation to an external performance benchmark. Furthermore, we took a sampling approach that maximised breadth, which means we analysed a wide range of cases and numbers of participants across the different MDTMs; this limits our ability for cross-group or tumour-specific statistical comparisons due to small subgroup samples sizes. Methodologically, we report findings descriptively as the study was not designed, or powered, to evaluate specific inter-group differences. We did not carry out member-checking as part of the interview analysis or a peerdebriefing as part of the observational data analysis. Lastly, the crosssectional nature of the study at a single time point meant a limited sampling window, so we cannot account for the contribution to team functioning of different leaders, experts and caseloads over time. The final results of the MDTM study will be presented to the participants for further development and improvement of the structure and team interaction.

| Future directions
To our knowledge this is the first descriptive study addressing in depth the issue of MDTM working practices across numerous tumour types and the first study of this types in Switzerland. Overall, we found a coherent pattern, suggesting that policies and guidelines to support MDT functioning sufficiently address the needs and ways of working of all types of MDTMs in Swiss cancer centres. This study is largely replicating a pattern found in MDTMs globally. 8 From a research perspective the tools that this study applied, can be used to carry out comparative analyses between cancer centres in Switzerland and beyond. 41 From a practice perspective, the tools used to characterise MDTMs may also be useful for cancer teams to analyse their own work patterns and identify opportunities for improvementfor example, through the formal appointment of a skilled MDTM chairperson to facilitate the case reviews. The analysis of the MDTM's that we have carried out can be fed directly into practical recommendations for MDTMs to implement locally. We have articulated some of these alreadyincluding a defined chairperson, and consideration of allocating the role without reference to clinical seniority. A simple checklist can be used to work up cases prior to MDTM presentation and hence improve the information coverage of clinical comorbidities. From a hospital perspective, offering consistent IT support and professional training in meeting chairing and leadership skills are areas the hospitals should invest in to support their cancer teams. Lastly, further consideration should be given to information about the patient's circumstances and wishesunderstanding of the flow of the cancer pathway is needed to identify where this can be done best. We hope that the study offers a baseline for both discussions with health policy makers about advancing the delivery of cancer care in Switzerland and for improvements to be considered to improve routine daily practice in cancer care. 42