Presentation on theme: "HEALTH TECHNOLOGY ASSESSMENT Has the UK got it right? Michael Drummond Centre for Health Economics University of York United Kingdom."— Presentation transcript:
HEALTH TECHNOLOGY ASSESSMENT Has the UK got it right? Michael Drummond Centre for Health Economics University of York United Kingdom
Disclosure of Interests Chair of the NICE Guidelines Review Panel on Primary Care Centre for Health Economics is one of the independent review groups for the NICE Technology Appraisal process Recipient of grants from the NHS R and D Consultant for industry
Outline of Presentation Structure of health technology assessment (HTA) in the United Kingdom (UK) Key features of the HTA process How do we do? Conclusions
Structure of HTA in the UK NCCHTA: – Based in Wessex. – Commissions a wide range of empirical and theoretical projects. – Administers contracts for NICE Technology Assessment Reviews (TARs). NICE: – Programmes in Technology Appraisals, Clinical Guidelines and Public Health. SMC: – Produces evaluations of all new medicines launched in Scotland.
Other Players in HTA in the United Kingdom NHS methodology programme Activities in Wales and Northern Ireland MRC and ESRC projects/fellowships Private research foundations Manufacturers of drugs and devices Health authorities
Key Features of the HTA Process Selection of topics Assessment procedures Stakeholder involvement Methodological development
Key Features of the HTA Process (continued) Developing a critical mass of skilled personnel Maintaining international links Implementation of HTA findings Transparency in decision-making
Cross-National Comparisons of Technology Assessment Processes Comparison of VATAP (USA), NICE (UK), CCOHTA (Canada) and AETS (Spain) Considered: (i) the reasons for the choice of topics, (ii) the types of technologies assessed, (iii) the methods of assessment and (iv) the outcomes of assessments Garcia–Altés et al, Int. J. Tech. Assess. Health Care 2004
Health Policy Issues dealt with in the Assessments TopicVATAPNICECCOHTAAETS Is the assessment the result of a formal prioritization process? Is the assessment the result of a political decision? Did the assessment include economic evaluation methods? Did any stakeholders participate in the assessment? Did the assessment include additional funding besides the own resources of the organization? Source: Garcia–Altés et al, Int. J. Tech. Assess. Health Care 2004
Selection of Topics In England the Department of Health sets NICE’s agenda. In Scotland the SMC considers every new drug. The NCCHTA and NHS Methodology Programme consult widely on topics, but then commission projects.
Selection of Topics: Observations Unlikely to be cost-effective to assess every new technology The mechanisms to select topics in England are not totally transparent Sometimes the choices have been strange (e.g. appraisals when a major trial will report in 1-2 years)
Selection of Topics: Observations (continued) Procedures are required to ensure appropriate use of technologies that are not appraised (e.g. drugs in the Netherlands). The approach followed by NICE allows groups of similar technologies to be compared. (This has both advantages and disadvantages.)
Assessing Drugs in the Same Class Together Most common with NICE in the UK, although there are a few examples from elsewhere (e.g. Canada). Main problem is the lack of head-to-head clinical studies. Puts the onus on manufacturers to demonstrate extra benefits to justify a premium price.
NICE’s Single Technology Appraisals A new ‘fast track’ procedure introduced in response to concerns over the time taken by NICE’s standard approach. So far applies to drugs, in the main cancer drugs. Will place more emphasis on analyses submitted by the manufacturer and incorporate less external review.
NICE’s Single Technology Appraisals May suffice in situations where the number of comparators is limited Raises further questions about the methods for prioritising topics Raises issues about burden of proof and responsibility for the results of the appraisal
Assessment Procedures The majority of HTA agencies undertake assessments in-house, although probably all commission some work outside (e.g. in Canada, CCOHTA spends 25% of its budget outside). In England, NICE places considerable emphasis on independent review by academic groups By-and-large the independent review groups apply ‘Cochrane-style’ methods.
Is Independent Review Cost-Effective ? Re-affirms the ‘arms length’ nature of HTA (assessment ≠ appraisal) More transparent and may help resolve disputes when multiple products are being considered, but: The Scots claim they reach the same decisions at a fraction of the (assessment) cost Systematic reviews place emphasis on RCTs as compared with other study designs Sometimes the economic model does not follow from the systematic review (‘a game of 2 halves’)
Methods Sample of all TARs published between January 2003 and July 2005. Data extracted on: –the TAR topic; –the clinical effectiveness measure; –the approach to pooling; –the measure of economic benefit; –the use, or non-use, of the systematic review in the model.
Results (1) 38 TARs published in the period studied, all of which contained a systematic review. Most of the economic evaluations were cost-utility analyses (32), reflecting NICE’s appraisal guidelines. The other studies were cost-effectiveness analyses (4) and cost-minimisation analyses (2).
Results (2) In 15 cases, the clinical data were not pooled in the systematic review, owing to: (a) heterogeneity; or (b) the limited number of studies. In 5 cases, a new model was not constructed, because: (a) the cost implications were thought to be trivial; (b) lack of evidence; or (c) the existence of satisfactory models provided by industry.
Results (3) In cases where there was both a pooled estimate from the review and an economic model: –The review was always used when survival was the endpoint (e.g. in the cost- effectiveness studies); –It was less often used when the endpoint was QALYs (e.g. in the cost-utility analyses).
Reasons For Not Using The Systematic Review In Cost-utility Analyses Preference-based QoL measures are only occasionally used in clinical studies The summary measure of clinical effectiveness does not facilitate the calculation of QALYs
Examples New drugs for epilepsy (adults): –clinical effects assessed in terms of total or partial reduction in seizures; –These were classified as partial and total response and a utility value to each state. Drug therapy for attention-deficit hyperactivity disorder (ADHD): –effectiveness measure was points change on the Connors Hyperactivity Scale; –economic evaluation used response/non-response and assigned a utility to each state.
Stakeholder Involvement Stakeholders can include manufacturers, professional organisations, health authorities, academic groups and patient organisations. All HTA agencies have some stakeholder involvement, but NICE is probably at the all- inclusive end of the spectrum. Stakeholder involvement is resource-intensive but may (i) lead to better assessments; (ii) reduce the number of appeals and (iii) lead to better implementation of HTAs.
Are Cochrane-Style Reviews Always Helpful? Respite care for frail older people: an appraisal of effectiveness and cost-effectiveness (Mason et al, in press) Number of hits (in the literature review)12,927 Papers retrieved 379 No. of potential studies 171 Potential economic evaluations 41 Actual number of economic evaluations (EEs) 22 Number of EEs of respite care meeting criteria for inclusion 5 Number of such studies undertaken in the UK 1
Methodological Development No doubt that HTA processes in the UK have stimulated methodological development; e.g.: mixed treatment comparisons; probabilistic models Several methodology TARs The NHS Methodology Programme has a good track record in funding projects relevant to HTA
Developing A Critical Mass Of Skilled Personnel The steady funding for NICE TARs has enabled academic units to build a critical mass of skilled personnel. Training fellowships are available from the MRC, ESRC and the NHS R and D The UK is probably better placed than most countries
Maintaining International Links We could probably learn more from comparing and contrasting the different approaches for drugs in England and Scotland. On the international level, the UK seems to be fairly well-connected. After Canada, the UK has the most members of HTAi (the international society).
Implementation of HTA Findings No health care system does this particularly well. Characteristics of the broader health care system greatly influence the opportunities for implementation, e.g. whether funding is largely general, or tied to the use of particular technologies. Restrictions in use (to most common guidance from NICE) pose more implementation challenges than simple yes/no decisions.
Evidence On The Implementation Of NICE Guidance Sheldon et al, BMJ 2004; 329 (30 October) Audit Commission Report, September 2005 Conn, F. Thesis, LSE, 2006 Daily Mail – every issue!
NICE Outcomes Hospital use of Paclitaxil & DocetaxelOrlistat & Alzheimer drugs Source: Sheldon et al, BMJ, April 2005
NICE Outcomes NICE commissioned report on 28 guidance areas covering 33 (1/3 of all recommendations to start of 2005) recommendations to assess impact using IMS data and UK Hospital Pharmacy Audit data 30% recommended for 1 st line use; 57% recommended for 2 nd /3 rd line use or specific patient groups; 12% (#4) rejected (but one entered a risk sharing agreement) Source: NICE commissioned Report from Abacus International, 2005
Implementation Of NICE Guidance Variable, by technology and location The formal requirement, to implement TAs within 3 months, probably has an impact Biggest problem is finding funding within a resource-constrained system. (The new tariffs will not reflect NICE guidance)
Implementation Of NICE Guidance (continued) Local professional involvement and good financial systems are important Almost half of GPs are welcoming of NICE NICE is perceived (by GPs) as being independent of industry but not of government
Implementation Of HTA Findings: what can be done? Develop an implementation plan for each HTA Produce more advice on what to discontinue, as well as what to adopt Link funding streams more closely to guidance (although not easy in the NHS) Increase the monitoring of the adoption of guidance
Transparency in Decision-Making In general all HTA increases transparency NICE is among the most transparent of HTA agencies The use of commercial-in-confidence data in technology appraisals can reduce transparency (Drummond, Applied Health Economics and Health Policy, 2004)
Conclusions Overall we do not do too badly in the UK Although existing procedures for assessment, involvement of stakeholders and production of guidance could be streamlined, they are basically sound Priorities for selection of topics and mechanisms of implementation of findings need more attention In resource-constrained systems, critics often confuse the message with the messenger