1. The Quality & Outcomes
Framework – triumph or
tragedy?
Steve Gillam
26.10.11

2. On this day…
 „Mr Atlee is a very modest man.
Indeed, he has a lot to be modest about.‟
 „I‟m just preparing my impromptu remarks.‟
 „If this is a blessing, it is certainly very well
disguised.‟
 „A pessimist sees the difficulty in every
opportunity; the optimist sees the opportunity in
every difficulty.‟

4. Outline
 Background
 Methods
 Main findings
 Impact of QoF
 Implications

5. Background
 International literature on pay for performance (P4P)
 Introduced in 2004 in the UK
 >£1billion per annum
 22% GP income
 Largest natural experiment in P4P in the world
 Precursor schemes, e.g. PRICCE

6. Domains for quality indicators in QOF 2010
 Clinical
 Secondary prevention of coronary
heart disease  Organisational
 Cardiovascular disease: primary  Records and information
prevention  Information for patients
 Heart failure  Education and training
 Stroke & TIA  Practice management
 Hypertension  Medicines management
 Diabetes mellitus
 COPD
 Epilepsy
 Hypothyroid  Patient experience
 Cancer  Length of consultations
 Palliative care  Patient survey (access)
 Mental health
 Asthma
 Dementia
 Depression  Additional services
 Chronic kidney disease  Cervical screening
 Atrial fibrillation  Child health surveillance
 Obesity  Maternity services
 Learning disabilities  Contraception
 Smoking

7. QOFability – ideal indicator is
 Acceptable
 Attributable
 Feasible
 Reliable
 Sensitive to change
 Of predictive value
 Relevant

8. Methods
 Systematic review of all published research till
end august 2011
 Medline, EMBASE, CINAHL, PsycINFO, Health
Business Elite, Health Management Information
Consortium, British Nursing Index, Econ Lit
 575 research papers identified; 124 selected for
review

9. Main findings
 Health care gains
 Population health and equity
 Cost effectiveness
 Impact on providers and teams
 Patients‟ experience

10. Health care gains
 Real but modest gains in some areas, e.g. asthma, DM
(?trendlines)
 Better recording in QOF areas but not untargeted areas
 No definitive improvement in outcomes, except possibly
epilepsy/DM admissions
Doran et al. N Engl J Med 2009;361:368-78.

11. Population health and equity
 Inequalities related to deprivation slowly narrowing
 Reductions in age-related differences for CVD/diabetes
 Variable effects for e.g. gender related differences in
CHD
Lancet 2008;
372: 728–36
Dixon, Khachatryan & Boyce. The public health impact, In Gillam & Siriwardena (eds) The
Quality and Outcomes Framework, Radcliffe, Oxford 2010.

12. High risk individual and population based
strategies for prevention (Rose)
Identify and treat Shift the whole
those beyond a population
threshold for risk distribution of risk
factor factor

13. QOF scores nationally (% total points) and
changes in exception reporting rates 2004-2009
Limited evidence of ‘gaming’ but
does ER reduce QOF’s impact on
neediest populations?

14. Costs and effectiveness
 No relationship between pay and health gain
 Limitations to modeling, e.g. omit costs of
implementation
 Cost effectiveness evidence studied for 12 indicators in
the 2006 revised contract with direct therapeutic effect
(Fleetcroft et al). 3 most cost-effective indicators were:
 ACEI/ARB for CKD
 Anticoagulants for AF and
 Beta-blockers for CHD

15. Costs and effectiveness
 Modest mortality reductions modelled - potential saving
of 11 lives per 100,000 people per year across all
indicators (Fleetcroft et al).
 Average indicator payments ranged from £0.63 to
£40.61 per patient; the percentage of eligible patients
treated ranged from 63% to 90% (Walker et al).
 Improvements in performance required for QOF
payments to be cost-effective varied by indicator from
less than 1% to 20% (Walker et al).

16. Impact on providers and teams
 Changing structures, roles and staff – nurse-led care
 Greater use of information technology
 Restratification: „chasers‟ and „chased‟
 Emphasis on the biomedical
 Commodification of care
 Narrative of „no change‟
Checkland & Harrison. Impact of QOF on practice organisation and service delivery. SocSciMed, 2008.

17. Checkland & Harrison. Impact of QOF on practice
organisation and delivery. Soc Sci Med, 2008.
 „Every day I come in I check (performance)… I‟m a chaser… You
have to chase yourself though. You‟ve no credibility if you don‟t
deliver.‟
 „Some patients will come to you and they‟ll plead with you: „please
don‟t give me any tablets, I‟ll bring my bp down, I‟ll do
everything…but we‟re saying to them: „well look, we‟ve checked it
three times now and it remains raised, you‟re clinically classed as
hypertensive, we follow these guidelines and this is what we should
be doing with you.‟
 „All I think QOF did was make it a bit more organised and that. I
don‟t think it was anything new.‟

18. Patients’ experience
 Little research on patient related/reported impact
 Continuity and relationships affected
 Fragmentation of care
 Little explanation provided to patients
“A slim, active 69-year-old patient attending for influenza vaccine
was faced with questions about diet, smoking, exercise and
alcohol consumption. There was no explanation for why these
questions were asked; they seemed irrelevant to having
a „flu vaccine. Blood pressure and weight had to be recorded and
a cholesterol test organised. A short appointment lasted almost
15 minutes without the patient having the opportunity to ask a
question about any aspect of „flu vaccine.”

19. Summary - QOF balance sheet
 Better data recording and analysis
 Modest health benefits for individuals and populations
 Narrowing of inequalities in processes of health care
 Improved team-working
 Opportunity costs unknown, e.g. impact on preventive care
 Unintended consequences: on workforce, professionalism
 Scientific bureaucratic medicine and the McDonaldisation of care
 Re-defined meaning of „quality‟

20. Implications – ways forward
 Limit expansion but expand local discretion
 Options
 Leave indicators unchanged and anticipate higher achievement
each year
 Add new indicators or conditions
 Select from a larger set of evidence-based measures
 Remove measures once agreed level achieved
 Rotate measures

22. 