Abstract
There is misunderstanding about both the meaning and the role of cost-effectiveness thresholds in policy decision making. This article dissects the main issues by use of a bookshelf metaphor. Its main conclusions are these:
It must be possible to compare interventions in terms of their impact on a common measure of health.
Mere effectiveness is not a persuasive case for inclusion in public insurance plans.
Public health advocates need to address issues of relative effectiveness.
A ‘first best’ benchmark or threshold ratio of health gain to expenditure identifies the least effective intervention that should be included in a public insurance plan.
The reciprocal of this ratio – the ‘first best’ cost-effectiveness threshold – will rise or fall as the health budget rises or falls (ceteris paribus).
Setting thresholds too high or too low costs lives.
Failure to set any cost-effectiveness threshold at all also involves avertable deaths and morbidity.
The threshold cannot be set independently of the health budget.
The threshold can be approached from either the demand-side or the supply side – the two are equivalent only in a health-maximising equilibrium.
The supply-side approach generates an estimate of a ‘second best’ cost-effectiveness threshold that is higher than the ‘first best’.
The second best threshold is the one generally to be preferred in decisions about adding or subtracting interventions in an established public insurance package.
Multiple thresholds are implied by systems having distinct and separable health budgets.
Disinvestment involves eliminating effective technologies from the insured bundle.
Differential (positive) weighting of beneficiaries’ health gains may increase the threshold.
Anonymity and identity are factors that may affect the interpretation of the threshold.
The true opportunity cost of health care in a community, where the effectiveness of interventions is determined by their impact on health, is not to be measured in money – but in health itself.
It must be possible to compare interventions in terms of their impact on a common measure of health.
Mere effectiveness is not a persuasive case for inclusion in public insurance plans.
Public health advocates need to address issues of relative effectiveness.
A ‘first best’ benchmark or threshold ratio of health gain to expenditure identifies the least effective intervention that should be included in a public insurance plan.
The reciprocal of this ratio – the ‘first best’ cost-effectiveness threshold – will rise or fall as the health budget rises or falls (ceteris paribus).
Setting thresholds too high or too low costs lives.
Failure to set any cost-effectiveness threshold at all also involves avertable deaths and morbidity.
The threshold cannot be set independently of the health budget.
The threshold can be approached from either the demand-side or the supply side – the two are equivalent only in a health-maximising equilibrium.
The supply-side approach generates an estimate of a ‘second best’ cost-effectiveness threshold that is higher than the ‘first best’.
The second best threshold is the one generally to be preferred in decisions about adding or subtracting interventions in an established public insurance package.
Multiple thresholds are implied by systems having distinct and separable health budgets.
Disinvestment involves eliminating effective technologies from the insured bundle.
Differential (positive) weighting of beneficiaries’ health gains may increase the threshold.
Anonymity and identity are factors that may affect the interpretation of the threshold.
The true opportunity cost of health care in a community, where the effectiveness of interventions is determined by their impact on health, is not to be measured in money – but in health itself.
| Original language | English |
|---|---|
| Place of Publication | York, UK |
| Publisher | Centre for Health Economics, University of York |
| Pages | 1-22 |
| Number of pages | 22 |
| Publication status | Published - Dec 2015 |
Publication series
| Name | CHE Research Paper |
|---|---|
| Publisher | Centre for Health Economics, University of York |
| No. | 121 |