asking for advice

[b.n...@cphc.keele.ac.uk]

Dear all,
I want to ask you advice about the use of the NMP model in a
particular proposal i.e. whether this is possible or not.
I am working with a colleague on a project proposal that aims to
explore whether, and if so how, PCTs are prioritising musculoskeletal
conditions within their overall strategy, and commissioning plans.
Thus, we are primarily interested in policy and strategy, and focus on
the implementation of the Musculoskeletal Framework. We want to use as
two points of conceptual reference the diffusion model (Greenhalgh
et.al. 2005) and the NMP. However, we are not examining the level of
clinician-patient interaction i.e. what happens in clinical practice.
Instead, we focus on how the PCT policy priorities shape the context
and resource allocation, and how they can shape routine clinical
practice. Is it possible to only look at the first two constructs?
This means that we 'translate' them as follows: first, 'contextual
integration' that refers to the capacity of an organisation to
understand and agree the allocation of control and infrastructure
resources to implement a complex intervention, and this will be
relevant in terms of understanding priority setting processes.
Second, 'skill-set workability' that refers to the formal and
informal divisions of labour and the mechanisms by which knowledge and
practice are distributed, and this may explain the commissioning and
implementation of MSK care pathways.
Does this approach make sense - or is it rubbish? I would like to hear
your feedback if you have any thoughts on this.
Thanks
Pauline

[Carl May]

I don't know what others think of this. It's something we have been
thinking about in terms of the ehealth work we are doing. My view is
that I don't believe that there's an obligation to attend equally to
all four constructs of the model but you may find it helpful to hold
on to interactional workability and relational intergration. This is
because the implementation of a 'pathway' seems to me to inevitably be
something that is worked out down the line in the mediation of
knowledge and practice amongst networks of professionals (relational
integration) and in their actions with specific patients
(interactional workability). So, I'm guessing that there would be some
'overflow' between the exogenous factors that you are interested in,
and the endogenous ones that you are not. These will become apparent
when GPs start talking about where the pathways take their patients.

Linda Gask has been using the model looking at governance and mental
health services, I wonder if she has a view about this.... Linda?

Carl

[Linda Gask]

Pauline- your ideas fit very much with what I have been working on trying to use the NPM to interpret the data from our clinical governance project collected over 5 years. I do not have data at the level of the patient-professional interaction- but do have interviews with managers about policy interpretation and governance with respect to the implementation of the NSF for mental health in primary care (contextual integration) and with the workers responsible implementation of the policy- the CG and mental health leads (skill set workability). The failure of the process to normalise leads me to the possible conclusion that this is actually due at least in part to the lack of clarity of process and pathways at the other two levels (mental health care is not itself actually normalised in primary care- so activity relating to its quality improvement at an organisational level is probably doomed!) in addition to the lack of stability in organisational structure (constant change in PCTs) but I don't have data at the level of GPs and patients.

Hope this makes some sense?

I am in the process of writing it so I am trying to find out what I think as I type the words!

Linda

[Pauline Ong]

Dear Carl,
Thank you so much for this. At least you don't think that I am off the
pace. I will include your thoughts about the two other constructs more
explicitly. I wasn't intending to ignore them, but the emphasis would be
on the first two.
Best wishes,
Pauline

--
Pauline B.N.Ong
Professor of Health Services Research
Primary Care Musculoskeletal Research Centre
Primary Care Sciences
Keele University
Keele
ST5 5BG
Tel: 01782 584708
Fax: 01782 583911
Email: b.n...@keele.ac.uk

[Pauline Ong]

Dear Linda,
Thank you so much for this advice. It looks like you have addressed
similar issues as we are intending to do with the Musculoskeletal
Framework, and I guess we will come up against the same problems.
If we get the grant I will certainly catch up with you to find out how
you have studied the Mental Health NSF implementation.
Best wishes,
Pauline

--

[scott....@sunderland.ac.uk]

I find this very interesting. In applying the NPM to give explanation
to the likely normalisation of a complex intervention we do have to
take account of the agents (GPs, patients) the objects
(musculoskeletal framework) and the collectives (GP practices, GP
contract, PCT, national guidelines QoF etc). Each of these agents are
accounted for in each of the four constructs. To selectivey use only
some constructs would serve to weaken that expanation/theory. There
are two challenges: 1 how much weight is given to each construct; and
2 how to produce an evaluation to populate the NPM in its entirety.
The latter may not be possible in many small projects but inference
may be drawn and fed into the NPM from other published evidence
together with ones own study.

My experience of applying the model to the normalisation of open
access HSG for the initial management of infertility in general
practice showed that contextual integration was the most influential
construct to predict normalisation - almost the first hurdle to leap
before describing the others. I do have data at the level of the
patient and GP which has given detailed explanation why open access
HSG has been adopted and not normalised.

I think you are right to propose a study looking at 'contextual
integration' of MSK care pathways i.e. does it fit with the practice,
PCT and NHS and confer an advantage to these collectives. Can it, will
it be resourced etc? Historically the 'carrot' approach has had a lot
of success (with the 1990 GP contract and GP QoF contract). Example:
GP injections of tennis elbows became normalised (classic MSK work)
following the 1990 contract, however many GPs did do this before the
contract and for a minority it was normalised due to other factors in
other constructs. So PCT policy priorities (which are often differing
interpretations of national guidance) do shape the context and
resource allocation, and can shape routine clinical practice but do
not give the full picture. Skill set workability, do the skills exist
for it to happen in general practice? In the two constructs you omit,
the questions might be, do GPs believe it falls within their remit and
value the MSK pathway and it is the right thing to do? These
constructs become more important when GPs are not forced into a course
of action through the carrot or stick approach of contextual
integration and contractual obligations.

The more I look at the model the more I feel that weighting is
important but difficult to define. Perhaps weighting of the constructs
is contextual. I think I'll stop there!

In summary I feel that your proposal to populate the first two
constructs will give an explanation of those constructs and inclusion
of the endogenous factors will complete the explanation/theory of
normalisation or not.

Hope I've understood where you are comming from and this is helpful.
Best wishes
Scott

[Pauline Ong]

Dear Scott,
Thanks so much for your helpful thoughts. I have now included all four
constructs but still propose to focus on contextual intergattion and
skills set workability. Your ideas help me to think through the other
two constructs in more detail, and the idea of weighting might be a way
forward - however, I am also daunted by the thought of how to do this.
If we get the money I will spend more time working through these issues.
Best wishes,
Pauline

--