Transcript Using this template - British Hypertension Society

Hypertension
Cost effectiveness
Mark Caulfield
For the British Hypertension Society
The William Harvey Research Institute and Barts
and The London NIHR Biomedical Research Unit
Queen Mary University of London
Launch: Wednesday 24th August 2011
NICE clinical guideline 127
The guideline
2004: National Institute for Health and Clinical Excellence
(NICE) hypertension guideline
2006: Drug model developed as part of pharmacological
update to guideline
2011: Model updated during second guideline update
National Collaborating Centre for Chronic Conditions
Model overview
Population
• Patients with essential hypertension seen in primary care
• Base case patient: 65-year-old men and women with 2%
CVD
risk, 1% HF risk and 1.1% diabetes risk
Comparators
• no intervention (NI)
• thiazide-type diuretics (TD)
• calcium-channel blockers (CCB)
• beta-blockers (BB)
• ACE inhibitors/angiotensin-II receptor antagonists
(ACE/ARB)
Perspective and time horizon
• UK NHS perspective, lifetime horizon
Model overview: model structure
Markov model
Health states chosen to represent disease
(simplification/data dependant)
• Transitions between health states affected by the
effectiveness of treatments over time (Cycles)
Model inputs: drug
effectiveness
•Systematic clinical review and meta analysis
•Head to head studies off different drug classes
reporting various clinical outcomes
UA
Thiazide-type
diuretics (D)
0.893
MI
0.78
0.796
0.855
0.85
Diabetes
0.985
0.808
1.137
0.77
Stroke
0.69
0.656
0.851
0.69
Heart failure
0.53
0.731
0.761
0.65
Death
0.91
0.883
0.939
0.9
Outcome
Calcium-channel
Betablockers (C)
blockers (B)
0.881
0.984
ACEi/ARB
(A)
1.01
Studies included in systematic review
•ALLHAT
2002
•MRC-0
1992
•ANBP2
2003
•NICS-EH
1999
•ASCOT
2005
•ONTARGET
2008
•CORD
2009
•PHYLLIS
2004
•ELSA
2002
•SHEP-P
1985
•HAPPHY
1987
•SHEP
1991
•INSIGHT
2000
•STOP-H2
1999
•INVEST
2003
•SYST-EUR
2000
•JMIC-B
2004
•Tedesco
2007
•LIFE
2002
•VALUE
2004
•MIDAS
1998
•VHAS
1998
•MRC
1985
Model inputs: drug costs
Drug used in model
ACEi
Ramipril (10mg)
ARB
Losartan (100mg)
BB
Atenolol (100mg)
CCB
Amlodipine (10mg)
DD
Bendroflumethiazide
(2.5mg)
UK generic list prices (BNF)
2011 Yearly drug
2006 Yearly drug cost
cost (EUR)
(EUR)
£21 (€26)
£30 (€49)
£26 (€32)
£217 (€287)
£13 (€16)
£13 (€17)
£19 (€24)
£70 (€92)
£12 (€15)
£17 (€22)
Interpreting cost effectiveness results
•Clinical and cost effectiveness
•Cost effectiveness = Costs and health outcomes and
getting the most health gain from the resources available
•Health is measured in QALYs
•Life years x QoL (Utility) = QALY
Incremental costeffectiveness ratio
(ICER)
Difference in costs
=
=
Difference in QALYs
Cost per QALY gained
Health economics of HTN therapy 2006
Health economics of HTN therapy 2011
Uncertainty - sensitivity analysis
•Varying individual parameters to test sensitivity of
model results
• Risk of CVD events (HF, MI, UA etc...)
• Effectiveness of drugs
• Cost of drugs and events
• Side effects of drugs
•Model remained fairly robust to changes
•In a few extreme scenario analyses, other drugs
became cost effective (mainly diuretics)
Interpretation
•Treatment of hypertension is highly cost effective
•Based on UK, generic drug prices
•CCBs are the most cost effective option with an ICER
below £2,000
•The results are more robust than in the 2006 model
•As commonly used anti-hypertensives become generic
it is now cheaper to treat hypertension than to do nothing
Acknowledgements
Kate Lovibond
Ralph Hughes
Prof Bryan Williams
Leo Nherera
Joanne Lord
2006 hypertension (update) GDG members
2011 hypertension (update) GDGmembers
British Hypertension Society
National Institute for Health and Clinical Excellence