European Conference on Health Economics – Dublin, July 2014

Last week saw the European Conference on Health Economics (ECHE) hosted by Trinity College Dublin.  CCHSR’s Ed Wilson was there.

International conferences are always a great opportunity to find out what your colleagues are up to in different corners of the world.  ECHE was no exception.  The venue was Trinity College Dublin, home to the beautifully illuminated 1200-year old manuscript of the four gospels known as the Book of Kells (pictured), itself housed in the equally breath-taking 18th century Old Library.  Dublin, of course is famous for a certain alcoholic beverage, but less well known is that the brewery is the source of not only stout but also T.

In between some intensive ‘networking’ in various establishments in the city, I attended sessions on diverse topics such as whether and how to capture non-health outcomes in an economic evaluation, choosing between alternative approaches to model the costs and effects of treatments, and techniques for measuring and valuing health care resource use.  A particularly interesting session looked at evaluating incentive structures in primary care.

Peter Smith (Imperial College London) began by observing that any health care system contains incentives, for better or worse, and they can act either on the structure of an organisation (having the right services in place), on processes (making sure medics do the ‘right thing’) or outcomes (rewards based on health gain, readmission rates etc).  Around half of OECD countries have some form of pay for performance scheme in primary care, and most of these concentrate on rewarding processes.  In the UK, we have the Quality Outcomes Framework (QOF).

Colleagues of mine in CCHSR are far better informed on QOF than I but briefly, this is a scheme implemented in 2004 where about 20% of a GP practice’s income is determined by incentives.  GPs earn points for achieving goals in clinical and organisational domains, and points mean prizes.

So how well has it worked?  Well, there didn’t seem to be much of a step change in outcomes: all quality measures were on an upward trajectory anyway and there was no obvious ‘kink’ in the curve after 2004.  A particular criticism was that the targets were initially set too low, and GPs were being rewarded for doing what they always did anyway.  As a result some targets were ‘toughened up’ from their original levels.  There were also fears of gaming: massaging of figures to boost points, but the incidence of this appeared reassuringly low.

QOF appears to have had a number of additional side benefits, specifically speeding up the computerisation of GP practices with consequent benefits for information flow between care providers.  There is also a better focus for GPs and clearer definition of what is expected of them.

However, hard evidence on changes in effects and costs is trickier to come by.  The only area where a significant effect could be traced to QOF was in secondary stroke prevention where improvements in GP care appeared to lead to a reduction in both the incidence of repeat events and in hospital costs.

The second paper by Steve Thomas (Trinity College Dublin) reflected on some of the initiatives in place and planned for the Irish health care system.  Around 40-50% of an Irish GP’s income comes from capitation payments (a fixed payment per head of population, irrespective of the care provided).  However 41% of all income is from out of pocket, fee for service payments from patients (it costs around €50 to see a GP in Ireland, which may be covered by health insurance, depending on the policy an individual takes out).

A capitation based funding system encourages GPs to do less: to raise thresholds for referral, to not order that extra diagnostic test.  It also encourages a focus on prevention and health promotion.  On the other hand fee for service encourages over-provision of care:  the more the GP does, the more he/she gets paid, and a cure is most definitely more profitable than prevention.

So with an Irish GP’s income split more or less 50/50 between these two alternatives, what is the best way forward?  Well firstly the contracts could be simplified (there are currently no fewer than 65 separate allowances a GP can claim for various services, although arguably QOF is not dissimilar in this respect).  Secondly, fee for service could be restricted to just preventative care such as immunisations, and chronic disease management.  The exact details of the GP contract are yet to be confirmed, but stakeholder responses so far are perhaps not the most supportive, with accusations of ‘coercion’ to accept the ‘morally and economically flawed proposals’.

However, the reforms do provide other opportunities to address areas such as chronic disease management programmes, to incentivise provision of health services in more deprived areas, and to invest in research and education around primary care such as the SPHERE initiative.

In summary, any contract renegotiation will always be fraught with turf wars and conflicts of interest and incentives, particularly in the Irish context due to the public and private mix of payment mechanisms.  Added to this is a good deal of fear, loss of good will and ‘austerity fatigue’.  However, the reforms also provide an opportunity to incentivise provision of preventative care with primary care teams taking a leading role.

So given the current state of play in the UK and Ireland, how do we go about evaluating whether one system or another is effective and which is the more cost-effective?  Stirling Bryan (University of British Colombia, Vancouver) provided some suggestions by looking at the economic evaluation of complex interventions.

A ‘simple’ intervention is something like a drug: all that is required is for the patient to take the drug at the right time.  It is clear what was done to the patient, and any change in outcomes can probably be attributed to the effect of the drug.  However, a complex intervention is characterised by many interacting components requiring a number of specific behaviours on behalf of those both delivering and receiving the intervention to succeed. An overall package of care, or care pathway is just such a complex intervention.  Reform of reimbursement contracts also meets this definition.

Evaluation of a complex intervention is made more difficult because if any one of the agents do not change their behaviour as intended, the intervention may not work, or may lead to unintended consequences.  As a result, the Medical Research Council has guidance recommending a phased, iterative approach where the development is as important (if not more so) than the final evaluation in order to ensure a strong, effective and generalisable intervention.  Researchers are required to know and show where they are in that process.

Prof Bryan concluded with a plea for health economists to end their ‘addiction to adoption’ (that is, focusing only on whether a single component of care is cost effective or not, along the lines of a NICE Technology Appraisal), and instead engage with all stages of the development and evaluation of complex interventions to provide a more useful and comprehensive evaluation of their cost-effectiveness.  This would provide a more useful and comprehensive guide to policy makers as to whether their reforms will do more good than harm.

This entry was posted in Blog and tagged , , , . Group: . Bookmark the permalink. Both comments and trackbacks are currently closed.
  • The Cambridge Centre for Health Services Research (CCHSR) is a thriving collaboration between the University of Cambridge and RAND Europe. We aim to inform health policy and practice by conducting research and evaluation studies of organisation and delivery of healthcare, including safety, effectiveness, efficiency and patient experience.