1. Governance and Public Policy: a NICE example John Brazier Professor of Health Economics, ScHARR, University of Sheffield, UK With thanks to Matt Stevenson for up to date slides ESRC Seminar on The Politics of Wellbeing Seminar 3 – Governance and Public Policy, 23rd September 2014

2. The problem Resources constrained Demands/needs >resources Choices need to be made, but how?

3. Cost-difference + - Effect-difference + - I II III IV Cost-effectiveness threshold (e.g. £20,000 per Cost-effectiveness decision rule for new intervention

4. NICE Technology Appraisals Technology appraisals are recommendations on the use of new and existing medicines and treatments within the NHS (National Health Service) Recommendations (made by an appraisal committee) are based on reviews of the clinical evidence and economic evidence - the incremental cost per quality adjusted life years (QALYs) over and above existing treatments The NHS is legally obliged to fund and resource medicines and treatments recommended by NICE's technology appraisals if their doctor believes it is clinically appropriate

5. Quality-adjusted life years (QALYs) Source: Drummond et al, 1997

6. Efficacy data from a trial

7. How can we convert these data into QALYs?

8. Results: Putting clinical data together with quality of life

9. The BIG questions 1.How should health be described? 2.How should it to be valued? 3.Who should value it?

10. EQ-5D By ticking one answer in each group below, please indicate which statements best describe your own health state TODAY. Please tick one 1. Mobility I have no problems in walking about  I have some problems in walking about  I am confined to bed  2. Self-care I have no problems with self-care  I have some problems washing or dressing myself  I am unable to wash or dress myself  3.Usual Activities I have no problems with performing my usual activities  (e.g. work, study, housework, family or leisure activities) I have some problems with performing my usual activities  I am unable to perform my usual activities  4.Pain/Discomfort I have no pain or discomfort  I have moderate pain or discomfort  I have extreme pain or discomfort  5.Anxiety/Depression I am not anxious or depressed  I am moderately anxious or depressed  I am extremely anxious or depressed 

11. How does the Appraisal committee use this stuff?

12. NICE Appraisal Committees Four independent advisory committees ≈ 30 Committee Members drawn from:NHS, Patient / Carer Organisations, Academia, pharmaceutical (and medical devices) industries The advice of the AC is independent of vested interests – Those with conflicts of interest (intervention or comparator) cannot attend

13. Two forms of appraisals Single Technology Appraisal (STA) - a single technology for a single indication Multiple Technology Appraisal (MTA) - normally covers more than one technology, or one technology for more than one indication

14. STAs Evidence provided by the manufacturer Critiqued by an Evidence Review Group (ERG) who assess the: clinical evidence; mathematical model; validity of results produced; and interpretation of the results. Amend model as necessary ERG’s are encouraged to produce an ‘ERG most plausible’ incremental cost per QALY gained (henceforth ICER). Not uncommon to see higher ICERs suggested by the ERG Typical duration: 13 weeks from ERG receiving manufacturer’s submission to AC meeting

15. Scoping a formal scope is released which is consulted on, specifying Intervention Population Comparators (Note that interventions which are widely used in the NHS but not licensed can be a comparator – bevacizumab in macular degeneration) Outcomes Economic analysis

16. The Methods Guide This document provides the ‘reference case’. This should be adhered to with explicit reasons provided if there is deviation

17. Cost per QALY thresholds Most plausible ICER ≤ £20,000 : Typically recommended > £20,000 and ≤ £30,000 : ??? (Certainty, Innovation, quality of life insufficiently captured) > £30,000 : Typically not recommended Exceptions: Those treatments that meet the end of life (EoL) criteria Empirically-based data suggest true threshold could be <£13,000 per QALY gained

18. Logistics of an AC Presentations are made by NICE AC members Clinical experts and patient representatives attend to provide evidence The manufacturer(s) attend to answer question provided by the AC and to highlight factual inaccuracies The majority of the meeting is undertaken in public although the final decision is made in private

19. Process Common discussion points Extrapolation of immature data Generalisability of trial populations to population to be treated Quality of life data…… aim for consensus, where the decision is contentious it may require a ‘secret’ vote Post-decision Consultation process with stakeholders Appeals procedure – reviewed by panel

20. NICE Recommendations 1 st March 2000 to 31 st July 2014

21. Political Pressure

22. Considerable political pressure EoL – little empirical basis for this Cancer Drugs Fund – A Government initiative that undermines the NICE process Value based pricing. DH attempt to introduce explicit pricing largely scuppered Contentious negative decisions may be delayed Turning Tide?? Charities criticise manufacturer over price of Trastuzumab Emtansine

23. Disinvestment NICE undertake very few disinvestment appraisals Disinvestment in technologies currently bought are left to the funders If the ICERs of displaced technologies are low, positive decisions may be harming societal health

24. Other NICE committees Highly specialised technologies (very rare conditions) Diagnostics NICE Clinical Guidelines NICE Public Health Vaccines Last three not mandatory funding for positive recommendations

25. Discussion points Process aims to provide transparent, independent, and legally challengeable advice to NICE – this system has many years of experience BUT: What do you think about this degree of quantification? Do you like the ‘economics’ framework….? Can it be adapted to well-being through well-being adjusted life years (WELBYS?) Can you see it working in other areas of Government – particularly for costly national policies? Do you think having a technocratic arms length organisation is helpful in making these choices?

26. Thank you! Any questions?