Caoimhe Nic a’ Bháird, Isla Wallace , Dr Penny Xanthopoulou and Professor Rosalind Raine
Department of Applied Health Research, University College London, 1-19 Torrington Place, London WC1E 7HB
Aim
Using Consensus Methods
To develop a set of feasible recommendations for
improving multidisciplinary team meetings (MDMs) for
patients with chronic diseases
Phase 1
findings
• Formal consensus methods are structured facilitation
techniques that explore levels of consensus among a
group of experts by synthesising opinions (Campbell et
al., 2003). They are designed to minimise some of the
limitations associated with group decision making.
• Their main purpose is to define levels of agreement,
particularly where there is an insufficient or
contradictory evidence base (Jones and Hunter, 1995).
Background
Relevant
research
literature
Relevant
policy
68 potential recommendations for improving
MDMs
• This is particularly important in the field of health care,
where clinical practice varies widely, and clinicians are
often faced with uncertainty about the value of
different options (Murphy et al., 1998).
• MDMs are widely established in the NHS and have
been endorsed by the Department of Health as the core
model for managing chronic diseases (Department of
Health, 2004; 2007).
• Three main approaches have been used in health
research since the 1950s:
o the Delphi Method
o the Nominal Group Technique
o the RAND appropriateness method
• It is believed that MDMs ensure higher quality
decision making and improved outcomes. However,
the evidence underpinning the development of MDMs
is not strong and the degree to which they have been
absorbed into clinical practice varies across conditions
and settings.
Expert panel rates
recommendations
in two rounds
• In practice, formal consensus studies often adopt
elements from each of these methods to optimally
address specific research objectives (Murphy et al.,
1998). We adopted such an approach to develop a set
of recommendations that are both desirable and
feasible.
• We conducted a large mixed-methods observational
study of multidisciplinary teams (in Cancer, Heart
Failure, Memory, and Mental Health services) to
investigate current MDM practice and examine the
determinants of effective MDM decision making
(Phase 1).
1st round:
No interaction
2nd round:
Panel meeting
Final recommendations are those that are
rated highly and with a high level of
agreement among the expert panellists
• Drawing on this data, we applied consensus methods to
develop recommendations to improve MDM decision
making and effectiveness (Phase 2).
Phase 2: Developing
recommendations for
improvement
Figure 3. Phase 2: Developing recommendations
• The steps of Phase 2 are illustrated in Figure 3.
Phase 1: Examining current
MDM practice
Phase 1 involved the collection and analysis of
qualitative and quantitative data sources to identify areas
for improvement in MDMs. This process is illustrated in
Figure 1.
Review of medical
records of 3184
Team Climate
Inventory
patients to assess MDM
decision
implementation, clinical
data and
sociodemographic
details
Questionnaire
completed by 161 team
members
Interviews with 53
team members
• We established a panel of 16 expert stakeholders
including patient representatives, clinicians and
policymakers.
• MDMs are widespread across the NHS. This study, the
largest of its kind, compared MDTs in a range of
clinical specialties, allowing us to develop
recommendations that are generalisable to a broad
range of disease types.
• A Questionnaire Pack was sent to the panelists,
containing a summary of research and policy and the
recommendations relevant to each of the 16 themes.
• The study developed recommendations both from
empirical evidence and the experience of a diverse
group of expert stakeholders.
• Panellists privately rated the feasibility and desirability
of each recommendation on a scale of 1 to 9, where 1
indicated strong disagreement with the
recommendation and 9 indicated strong agreement
(Round 1 ratings, see Figure 2a).
• This process allowed us to identify recommendations
that are both desirable and feasible, making it more
likely that they can be practically implemented.
•
Observation of 370
MDMs of 12 teams
(qualitative field notes
and quantitative data on
MDM and decision
characteristics )
16 themes
identifying
potential
areas for
improvement
Conclusions
• Drawing on the Phase 1 findings and a review of the
relevant research and policy literature, we developed a
list of 68 potential recommendations.
Interviews with 20
patients and carers
A meeting was then convened to discuss the ratings,
focusing on areas of low consensus among panellists.
The discussion was chaired by the principal
investigator, and was designed to ascertain whether
discrepant ratings were due to real clinical
disagreement or to misunderstandings.
Further information
• At the meeting, each panellist received a second,
personalised version of the questionnaire pack showing
the distribution of all panellists' first round ratings,
together with his/her own ratings (see Figure 2b).
1.
Lack of clarity regarding purpose of MDMs
2.
Attendance and participation in MDMs
3.
Chairing the MDM
4.
Administrative support and the coordinator role
5.
Agreeing which patients should be discussed
6.
Preparing and presenting cases
7.
Discussing comorbidities in the MDM
8.
Discussing patients holistically
2 (a)
9.
Incorporating patient preferences into discussions
MDT discussions should result in a
documented treatment plan for each
patient discussed
13. The role of research and evidence in MDMs
14. Teaching as a function of MDMs
15. Recruitment to trials in MDMs
16. Monitoring the quality of MDMs
Figure 1. Phase 1: Examining current MDM practice
•
Follow the Dept. of Applied Health Research on Twitter @ucl_dahr
•
visit tinyurl.com/improvingmdts or scan the QR code on the right
References
• The extent to which each respondent agrees with each
statement (i.e. 7-9 on the Likert scale), and the extent
to which respondents agree with each other will be
used to generate a final list of recommendations.
11. Patient attendance at MDMs
12. Providing feedback to patients
email c.bhaird@ucl.ac.uk
• Having discussed each theme as a group, panellists
individually rated each recommendation a second time
(Round 2 ratings).
Potential areas for improvement identified in Phase 1
10. Patient awareness of MDMs
•
Strongly …………...………
Disagree
1 2
2 (b)
3
4
5
6
Strongly ………………..……
Disagree
Strongly
Agree
7
Don’t
know
8 9
Strongly
Agree
Don’t
know
MDT discussions should result in a
documented treatment plan for each
patient discussed
1
1
2
1
2
3
4
5
Department of Health. (2007). Cancer Reform Strategy. London: Central Office of Information.
Department of Health (2004). Improving chronic disease management. London: Author.
Campbell, S., Braspenning, J., Hutchinson, A. & Marshall, M. (2003). Research methods used
in developing and applying quality indicators in primary care. British Medical J
ournal, 326, 816-819.
Jones, J. & Hunter, D. (1995). Consensus methods for medical and health services research.
British Medical Journal, 311, 376.
Murphy, M., Black, N., Lamping, D., McKee, C., Sanderson, C., Askham, J. & Marteau, T.
(1998). Consensus development methods and their use in clinical guideline
development. Health Technology Assessment , 2, i.
3
2
7
6
7
8 9
Figure 2. Rating response scales for Round 1 (a) and
Round 2 (b)
Acknowledgments
The researchers would like to thank Natalie Austin-Parsons, Sophie Bostock, Mike Galsworthy,
and Khadija Rantell. Guidance was provided throughout by the study co-applicants and Expert
Advisory Group members: Julie Barber, Anne Lanceley, Alex Clarke, Gill Livingston, Archie
Prentice, Jane Blazeby, Dave Ardron, Miriam Harris, Michael King, Susan Michie, Simon
Gibbs, and Ewan Ferlie. We would also like to acknowledge the support of the Comprehensive
Clinical Research Network.