In the following, we explore key dimensions of team functioning from our conceptual model, particularly those related to meeting inputs and processes. We then explore data on whether bringing groups of health and social care professionals together in this way appeared to offer additional benefits for coordinating and delivering care.
Matching Attendees and Case Needs
Some staff groupings tended to be present at all meetings: administrators, GPs, social workers, (including managers), and nurses, (including Clinical Nurse Specialists and Community Matrons). However, a sub-set of Pioneer 2 meetings differed (attended by nurses, occupational therapists and physiotherapists, administered on a rotating basis). Mental health workers were present at Pioneer 1 meetings. CVS attendees were present in Pioneer 1 and 2 meetings. Where GPs attended, numbers varied. In most meetings, they remained throughout, in others leaving after case presentation. Administrators (where present) sometimes contributed in depth to meeting discussions.
Attendance by professional groupings appeared partly to be a product of caseload profiles. MDTs in both sites sought to prevent hospital (re)admission or manage post-hospital discharge, and barring one, were primarily managed from GP practices but cases discussed during meetings at the Pioneer 2 sites seemed to be older people, with fewer problems related to housing, substance misuse or learning disability. Despite the fact that attendance appeared to be shaped by the types of cases to be discussed, sometimes key staff were not present whose expertise matched the requirements of the cases: for example, none of the meetings had direct input from local authority housing services and this was particularly notable in the Pioneer 1 meetings, where chronic housing problems were often discussed. The frequent absence of a representative from the local NHS acute trust at Pioneer 1 meetings was commented on by one GP. Learning disability services and substance misuse services were not represented at any of the meetings, which may have been helpful. When two substance misuse workers attended simply to introduce their service at a meeting in Pioneer 1, one became involved in productive discussions about patients, demonstrating the value of being able to access a wider range of expertise.
“[The drug and alcohol worker] immediately jumped in to say, ‘Oh, I know that patient,’ or, ‘You could refer that patient to me.’ So, they were immediately acting as a potential resource for the team, which seemed interesting, and I think the staff that were there found it really helpful.” (Researcher 1, Team Analysis Meeting 1).
Sharing Intelligence
Access to the internet, IT hardware (PCs, laptops, tablets) and software (databases containing information about patients and their carers) appeared to be critically important in meetings, given the functions of the MDTs. In Pioneer 2 meetings, designated administrators and clinical staff primarily accessed the relevant databases. In the Pioneer 1 sites, in addition to primary care databases, social services databases, were heavily relied upon due to the nature of patients’ needs or service use. These databases often appeared to contain the most detailed information about patients and their carers but were only accessible to the social workers. We observed that attendees from different organisations often could not access each other’s systems (either for technical or information governance reasons): the social services, CVS and mental health databases in Pioneer 1 were entirely separate and required workers from these organisations to be present to access them. There was one highly inefficient system that prevented senior nurses from accessing two required databases on the same device at once. Also, visiting staff had to bring their own equipment, which was variable in quality, whereas GPs often used PCs onsite.
Reliable access was also often variable, with examples of difficulty logging on to systems or slow/unreliable connections, particularly in Pioneer 1. We observed that access to information was determined to some extent by the quality of the IT connections and/or hardware available. Accessing and sharing information from different systems—a key function of the meetings – appeared to consume an inordinate amount of time, especially in the Pioneer 1 meetings.
“There were so many different systems that they were trying to access in the meetings and they still didn’t have access to all the ones they needed.” (Researcher 3, Team Analysis Meeting 1).
Meeting Processes
Leadership and Chairing
Meetings varied in the extent to which explicit chairing was observable. In Pioneer 2 and some of the Pioneer 1 meetings, formal chairing, often performed by GPs, nurses or sometimes social workers, was evident. In other Pioneer 1 meetings, chairing was more informally shared between two or more attendees (most usually a GP, social worker and/or senior nurse).
“At the [name removed] one that we attended, where the social worker was chairing, it was a very obvious chairing, […] He’d bring it to a point where he’d say, ‘Okay, so what are the actions?’ and that tended to focus minds. […] It wasn’t quite as formal in the other two meetings that we attended, and in fact, after one meeting [we researchers] discussed who was the chair there.” (Researcher 1, Team Analysis Meeting 1).
Cohesion, Participative Safety and Conflict Resolution
Meetings appeared to be characterised by uniformly collegial, professional and respectful interactions. We observed humorous exchanges and friendly banter. Attendees were on first-name terms and interacted in ways that suggested established working relationships. An exception was a meeting where the atmosphere was perceived by both researchers to be tense.
We observed that some CVS attendees in the Pioneer 1 meetings contributed infrequently to discussions, although the reasons for this were unclear. We did not observe lesser participation by particular groups in any of the Pioneer 2 meetings, although certain professionals appeared to be more vocal or more involved in chairing, as described. We observed no major distinctions of status or hierarchies affecting meeting dynamics.
Interactions were collegial and largely consensus-based, which raised questions for us about the extent of challenge in the meetings. For example, we saw no overt examples of conflict. However, in instances where a degree of challenge did take place, this appeared to add to the quality of the discussion.
“In one of the meetings I attended there was a mental health worker that just asked questions, like continually challenging the discussion and the debate that was happening, which was really helpful because it really unpicked what was going on.” (Researcher 6, Team Analysis Meeting 2).
Sense-Making
We also paid attention to the deliberation processes in meetings, and were able to identify a consistent pattern of:
-
Case presentation – an attendee introduced the patient, giving salient facts and reasons for referral or addition to/presence on, the caseload.
-
Information seeking/sharing – attendees searched their individual databases or drew on personal knowledge to identify what was known about the patient and any informal carers. This was a major element of the process.
-
Narrative construction – attendees shared information to create a historical reconstruction and collective view of the patient’s current needs, challenges, or situation. This included information such as demographics, health and care needs and problems, living and family circumstances, service usage, formal and informal care and support being provided, barriers and facilitators to working with the patient, and likely acceptability of potential service/treatment options to patients and carers.
-
Solution seeking – suggestions and proposals to address the needs identified were considered, often involving sharing information about each other’s services and other local services that might be appropriate.
-
Decision-making and task allocation – actions were agreed and allocated, including further assessments and referrals, whether to remove the patient from the caseload or review at a later date.
This was not always a linear process, especially if the patient was already on the caseload, where there was less focus on narrative construction and more on ‘status update’. The patient databases were especially important in establishing the narrative and what actions had already taken place outside the meeting.
Pace and Depth
Time spent discussing cases varied considerably. In Pioneer 1 meetings in particular, extensive deliberations were observed (though cases were often also discussed rapidly). However, in one MDT in Pioneer 2, case discussion was very rapid, to the extent that following the substance of deliberations was often challenging. In Pioneer 1, we observed that participants sometimes struggled to discuss all intended cases in the time allotted, with deliberations becoming more hurried towards the end of meetings.
Who Decides?
Making care co-ordination decisions was central, principally about whether to retain or discharge a patient from the caseload and, if retained, necessary actions to provide appropriate care and support. Criteria for adding a particular patient to the caseload or, in many cases, for discharging them, were not readily transparent, though this may be related to our partial view of administrative processes.
Where decisions were clear, we frequently observed decision-making by key dyads and triads, which varied in composition depending on professional groups present and the nature of the case. GPs, social workers and nurses were frequently central to these dyads/triads but other configurations were seen, particularly in Pioneer 2, where GPs were not present and nurse-physiotherapist-occupational therapist configurations were observed to be making care-related decisions.
Some deliberations were extremely rapid: on receiving some information from an attendee, the professional(s) from the service(s) best placed to act or meet the identified need, made a swift decision about patient management actions and the discussion moved promptly to the next case. In other cases, there was lengthy and detailed deliberation. In some of the Pioneer 1 meetings, perhaps due to less structured chairing, responsibility for moving the discussion to a conclusion was sometimes unclear.
“It sometimes wasn’t clear that a decision had been reached or indeed precisely what the decision was. There was quite a lot of deferral and not much discharge, although that is different in different places.” (Researcher 3, Team Analysis Meeting 2).
However, in other Pioneer 1 meetings, GPs appeared to determine when a discussion was concluded, possibly because they had ongoing, ultimate responsibility for the patient, while other agencies were involved only if they had a relevant service to offer and the patient was eligible (e.g., CVS, mental health services, social services).
“While other organisations’ representatives somehow are used to the fact that if they cannot do anything for the patient, it’s not their job, it’s always the GP’s job to some extent.” (Researcher 5, Team Analysis Meeting 2).
Deferring Difficulty
Requests for further assessments and tests, especially for new or returning patients, could be an important precursor to decision-making, but we also observed instances where further assessment appeared to be used to defer difficult decisions in cases where the course of action was unclear.
“The sorts of actions they committed to, most of them, I don’t want to use the term ‘kicking the can down the road,’ but it was requesting additional assessments or more information.” (Researcher 2, Team Analysis Meeting 1).
Differences in the caseloads between the two sites may have been significant in decisions to defer. In Pioneer 2, the caseload was predominantly frail people, over 75 with multiple complex health conditions and related social care needs. This group was also prevalent in Pioneer 1 meetings’ caseloads. However, their caseloads also included more younger people and those whose problems included homelessness, substance misuse, mental illness and learning disability. Here, we sometimes observed professionals struggling with what appeared to be highly complex and at times, distressing cases, where problems were severe and enduring, multiple options for intervention had been exhausted and services had difficulty engaging with patients.
“Almost all of them [patients] […] had multiple, interrelated, complex life difficulties that didn’t neatly fit into any known category. What I thought I witnessed was a combination of people trying to, as you say, puzzle out, ‘What could we usefully do?’ and, ‘Is there a limit to what we should be expected to do?’” (Researcher 4, Team Analysis Meeting 1).
We questioned whether it was always clear in what circumstances MDTs could reasonably decide to withdraw from case management and reallocate cases that had apparently exhausted the MDT process.
“What they did in those sorts of cases was they kept them on the caseload because they didn’t know what else to do, and it seems to me that there needs to be some sort of discussion about how are they given permission to say, ‘We’ve run out of road here,’ and ‘What’s the [next] response to those patients?’.” (Researcher 3, Team Analysis Meeting 1).
Patient-Centredness
We explored whether patient-centredness was evident as part of the deliberations. The CVS organisation present in the Pioneer 1 meetings attended as a service provider but it was unclear whether it also performed a formal patient advocacy function. Although patients were never present, we noted examples of substantial patient-centredness in the discussions, both explicitly, in considering needs and preferences, and implicitly, through the process of constructing the patient’s narrative.“To what extent was patient-centredness a feature of the meetings? To a very large extent. I felt that whenever they were proposing something, they were immediately thinking about whether this would be suitable for that patient, whether it would work, so it was an integral part of the conversations.” (Researcher 2, Team Analysis Meeting 1).
Added Value?
We explored the issue of whether there was ‘added value’ from bringing together professionals in MDT meetings, further to what might have been achieved by them working separately. The meetings appeared to offer an effective alternative way to share information about patients.
“These meetings provide a forum or a setting in which you could have discussions that wouldn’t be appropriate to have over an email, for example, discussing about a case and someone giving them the background, which is not available on any of their systems. I felt that was a definite added value that would make them decide or reach decisions they wouldn’t have otherwise, knowing the additional background.” (Researcher 2, Team Analysis Meeting 2).
The ability to consult databases and give ‘real-time’ updates to inform decision-making was another observed benefit.
“One thing I would add to the added value question is the immediacy of what was happening in the room, which they wouldn’t have got if they phoned or emailed. They could get an immediate response and update, which was often, I felt, really beneficial to patient care.” (Researcher 6, Team Analysis Meeting 2).
Mutual understanding of decisions made by individual services regarding planning or patient care was enhanced by having attendees from different services and organisations present.
“They were contrasting the patient-centred narrative with their experience of the services, trying to identify what misfired in the encounter between the need and the potential supply, […] [asking] ‘Why do we have a note saying […] that they didn’t meet the criteria?’” (Researcher 5, Team Analysis Meeting 2).
Also, for patients that services had difficulty engaging, or cases that were difficult to resolve, attendees could discuss who had the best rapport and likelihood of engaging them to achieve potential solutions.
“Another instance [of added value] was a discussion about who should approach the patients to get them to do something that they are not keen to do. […] Everyone felt it was in the patient’s best interests but who has the best rapport with the patient to do that? I can’t imagine them exchanging emails saying, ‘You ask…” (Researcher 3, Team Analysis Meeting 2).
Another form of added value for participating services was shared identification and management of risk in often complex cases and uncertain circumstances.
“In part, the decision making relates to that, […] what they’re concerned about as well are things like the risks, ‘If we don’t do X, Y might happen.’” (Researcher 1, Team Analysis Meeting 2).
Sharing clinical learning or information about how organisations and the local H&SC economies functioned, and how to navigate this was also a beneficial consequence of MDT meetings.
“Different organisations explaining to others how they operate was a definite added value, and occasionally the medical background that the GP was explaining, which helped other organisations understand why they need to provide certain support, that was a definite added value.” (Researcher 6, Team Analysis Meeting 2).
However, we noted scope to enhance shared learning by making more space for reflexivity within meetings. For example, Pioneer 1 meetings had a standing agenda item to discuss what had been learned but this was often overlooked or dealt with in a perfunctory way at the end of meetings. However, when shared learning occurred, it could reportedly have benefits beyond the immediate MDT meeting.
“[A GP] said to us afterwards that one of the key components for her, I guess in terms of added value, was that she was learning what was available and what people could contribute […] and that she would take that learning back to her own practice.” (Researcher 6, Team Analysis Meeting 2).
The format of the MDT meetings also appeared to enable participants to support one another practically and emotionally.
“They weren’t working in isolation. They were able to bring and receive support, probably also emotional support for dealing with quite difficult, in some cases, patients with really complex mental health and social needs, as well as physical.” (Researcher 6, Team Analysis Meeting 2).