Partially Wrong? Partial Equilibrium and the Economic Analysis of Infectious Disease Interventions

Abstract:     
Health economic evaluation (as generally practiced) is a partial equilibrium analysis. Although any change in health care activity (such as investment in a health programme) will generate ripples throughout the health care sector as well as the wider economy, the analyst effectively seals off part of the economy (when adopting a societal perspective), or part of the health care sector (when adopting a health care payer's or sector's perspective) by invoking 'ceteris paribus'. This approach is often defensible based on the argument that a change in an individual's health status has only a small impact beyond the costs and benefits incurred by that individual, their employer, health care providers, insurer, family and friends (i.e. people and instances considered most "relevant" in traditional health economic evaluation). The extent to which all of these affected parties should be accounted for is disputable (eg, should QALY losses to caregivers be considered?). Furthermore, costs resulting from the patient's illness, but incurred by other third parties (eg, caused by a reduction in demand for certain services outside of the health sector, or a decrease in health care services capacity) are assumed to be small and are therefore ignored. Similarly, a change in health care resource allocation for a single health care intervention is usually assumed to have little impact on economic activities outside the health care sector.

When (gradual or sudden) shifts in the prevalence of health problems cause the impairment of health system functioning, then this will affect patients who do not have the disease in question, as well as those that do. For example, during a pandemic of influenza, the available health care facilities are likely to be severely under capacitated. Ensuing delays of treatment of "third patients" would not be valued within a 'standard' economic evaluation. However, these postponements do represent an opportunity cost for the patients involved. We argue that these losses could be valued using utility scores for ill-health states over the extended duration of non-treatment. The presence of a health threat can also influence the general expectations and behaviour of consumers and investors, and therefore have an impact far beyond the direct reduction in productivity from sick patients (or their family). For instance, for SARS, it has been observed that by far the largest economic impact occurred by reduced engagement in non-vital consumption activities, as well as postponement of consumption of other goods. Using revealed behavioural changes (and the associated economic impact) during past outbreaks, and stated behaviour and risk perceptions from surveys, along with broad macroeconomic data, we argue that shocks to the wider economy of various outbreak scenarios can be estimated by macroeconomic modelling.

Ignoring the impacts we describe here, as would be the case under the standard framework, may produce misleading results when assessing the cost-effectiveness of interventions to avoid or mitigate outbreaks of international concern.

Keywords: public health, economic evaluation, vaccination, SARS, pandemic influenza

JEL Classification: I12, H41

working papers series