Perspectives on Health Care Decision Making

Appropriate perspectives for health care decisions Simon Walker Centre for Health Economics, University of York

Acknowledgements Karl Claxton Susan Griffin Stephen Palmer Mark Sculpher

Overview Allocation of resources Public sector agency Economic evaluation Purpose of health care Fixed budgets and opportunity costs Health care costs displace other aspects of value too An example: Lucentis for diabetic macular oedema Accountable deliberation Policies with multisector impact Conclusions

Allocation of resources Basic economic problem: Resources scarce but wants infinite Economics is the study of the allocation of these scarce resources Market extolled as optimal method for allocation However, most societies allocate a proportion of resources to be allocated by the public sector. Key question- How should social choices about provision of goods in the public sector be made?

Public sector agency Within public sector, responsibility for resource allocation typically split between different departments Each department has a distinct budget and remit (set of objectives) However, public policies and interventions often impact beyond the main focus of activity Example 1: a new arthritis treatment which allows a patient to return to work has wider economic benefits Example 2: a school meals programme has health benefits but imposes costs on the education system Key- Impact on outcomes and costs which are beyond the remit of the decision maker involved

Economic evaluation The comparative analysis of alternative courses of action in terms of both their costs and consequences (Drummond et al) Two core questions: What is of value? Normative- reflects what we consider to be better or worse What is forgone? Fact- e.g. with a fixed budget if we fund a new more expensive treatment, something else must be displaced 1. 2.

What is purpose of health care? When considering value it is important to think about what is the purpose of the good being provided. For health care, is it: Health Welfare based on individual preferences Wider social welfare

Moving beyond health care If costs and benefits fall on different sectors, more outcomes than just health to consider However, there is no consensus on how to trade off different arguments How much consumption would we give up for a unit of health? How many units of health would we trade for a unit of education? Could impose a social welfare function (a function which defines what is better or worse across all possible states) But- possible that some important arguments cannot be specified.

Fixed budgets and opportunity costs Many sectors are subject to fixed budget constraints (at least in the short term) These have implications for what we forgo if new demands are made on those budgets (i.e. the opportunity cost). Cost-effectiveness thresholds are estimates of the cost at the margin of an output being displaced (question of fact)

Two sector example: Health and Wider social benefits Consider a new health care intervention Intervention will have: Impact on health of patients Impact on wider social benefits Costs to the health care budget 2 questions Question of value: What is our willingness to trade off health for wider social benefit Question of fact: What will be displaced if we fund the new treatment

The UK NHS- Health care costs displace other aspects of value too How much and what type of health and for whom? Life years and quality of life effects By age, gender and ICD code Wider social benefits Net production effects of a change in health Marketed and non market production Net of marketed and non marketed consumption

Health care costs displace health Cost per death averted Cost per life year Cost per QALY (mortality effects) Cost per QALY Qol associated with LYs - 1 Norms Based on burden Qol during disease - 0 0 Based on burden YLL per death averted - 4.5 YLL 4.5 YLL 4.5 YLL 12.7 QALY QALYs per death averted - 4.5 YLL 3.8 QALY 11 PBCs (with mortality) 105,872 23,360 28,045 8,308 All 23 PBCs 114,272 25,214 30,270 12,936 From Claxton et al

What are the expected health consequences of 10m? Change in spend Additional deaths LY lost Total QALY lost Due to premature death Quality of life effects Totals 10 ( m) 0.45 0.76 0.46 0.32 0.33 0.19 0.60 0.46 0.77 0.68 0.21 1.79 0.10 0.19 0.09 0.29 0.20 0.36 0.09 0.35 0.30 1.01 51 3.74 22.78 13.37 2.62 0.72 0.67 1.21 2.25 0.00 0.01 0.36 2.83 0.04 0.05 0.03 0.00 0.24 0.39 0.04 0.03 0.00 0.00 233 37.5 116.0 16.1 24.7 5.3 5.0 6.5 3.3 0.0 0.4 1.7 12.8 0.2 0.2 0.1 0.0 1.1 1.8 0.2 0.2 0.0 0.0 773 26.3 107.8 229.4 43.9 15.7 60.6 109.1 10.6 0.0 0.2 21.8 95.3 0.7 4.2 14.0 6.8 1.9 23.2 0.8 0.7 0.0 0.0 150 24.4 73.7 10.1 16.2 3.6 3.2 4.3 2.1 0.0 0.2 1.1 8.3 0.1 0.2 0.1 0.0 0.7 1.2 0.1 0.1 0.0 0.0 623 1.9 34.1 219.3 27.7 12.1 57.3 104.8 8.5 0.0 0.1 20.7 87.0 0.6 4.1 13.9 6.8 1.2 22.1 0.7 0.6 0.0 0.0 Cancer Circulatory Respiratory Gastro-intestinal Infectious diseases Endocrine Neurological Genito-urinary Trauma & injuries* Maternity & neonates* Disorders of Blood Mental Health Learning Disability Problems of Vision Problems of Hearing Dental problems Skin Musculo skeletal Poisoning and AE Healthy Individuals Social Care Needs Other (GMS)

Other aspects of value gained and displaced Wider Social Benefits (net production) M05 E11 M45 F30 F20 J45 M81 G35 J43 G40 L40 Rheumatoid arthritis Diabetes Ankylosing spondylitis Depression Schizophrenia Asthma Osteoporosis Multiple sclerosis Emphysema and COPD Epilepsy Psoriasis 30,034 27,421 26,190 23,489 22,697 20,100 17,910 15,482 14,525 14,245 11,890 11,611 8,138 6,912 6,284 5,234 3,102 2,888 351 - 513 - 2,430 - 6,949 - 8,061 - 10,602 - 13,211 - 14,395 - 16,752 - 21,568 - 23,382 - 24,813 - 32,709 - 36,067 - 53,860 Displaced E66 C53 K50 J30 G20 C50 G30 A40 F03 I64 C18 C61 C64 I21 I26 J10 C90 C92 C22 C34 C25 Average of displaced QALYs Obesity Cervical cancer Irritable Bowel Syndrome Allergic rhinitis Parkinson's disease Breast cancer Alzheimer's disease Streptococcal septicaemia Dementia Stroke Colon cancer Prostate cancer Kidney cancer Acute myocardial infarction Embolisms, fibrillation, thrombosis Influenza Myeloma Myeloid leukaemia Liver cancer Lung cancer Pancreatic cancer

An example Appraisal of ranibizumab (Lucentis) for diabetic macular oedema 2011 Retinal thickness 400 subgroup before PAS Additional costs = 3,506 per patient Incremental cost-effectiveness = 25,000 per QALY 23,000 eligible patients each year Attributes Investment Disinvestment Net effects Lucentis for diabetic macular oedema ( 80m pa) Expected effects of 80m pa Deaths 0 -411 -411 Life years 0 - 1,864 -1,864 QALYs 3,225 - 6,184 -2,959 Burden of disease QALY loss 2.68 2.07 0.61 Wider social benefits Consumption QALY equivalent ( 60,000 per QALY) 85.2m 1,420 - 49.8m - 830 35.4m 590

How should we decide? Restrict to health and health care Net health benefits = 3,225 6,184 = - 2,959 QALYs A single societal perspective Ignore the constraint Net costs = 80m - 85.2m = - 5.2m Account for the constraint (but not displaced WSB) Net health loss = -2,959 QALYs Wider social benefits = 85.2m Worthwhile if consumption value of health < 28,800 per QALY Account for displaced wider social benefits Net health loss = -2,959 QALYs Net wider social benefits = 85.2m 49.8m = 35.4m Worthwhile if consumption value of health < 11,900 per QALY

Accountable deliberation Multi sectoral perspective Identify where cost fall and benefits accrue Any particular SWF will be disputed e.g., use of market prices Other arguments difficult to specify Reflect the implications of current constraints Where opportunity costs will actually fall Social values implied by current arrangements Account for other aspects of value displaced Approve technologies that reduce health and wider social benefits Health care perspective Excludes some aspects of value But also excludes the opportunity cost too Could be zero sum or worse

Policies with multi sector impact What if another government sector is also impacted Need to account for opportunity costs on their budget as well For example, free school meals Impact of policy Better educational performance? Better health? Wider social benefits? What is displaced? Better educational performance? Better health? Wider social benefits?

Conclusions What we gain and lose as a result of the introduction of a good are questions of fact: Direct benefits of the good Benefits of other things which are displaced How we then go about valuing those benefits to see whether the introduction of the good is beneficial is much more challenging and controversial.