PubMed Health⌕ Search

Biomedical subjects

Paul Contoyannis

Publications and source records attributed to Paul Contoyannis.

3 recordsLinked to original sources

Estimating the price elasticity of expenditure for prescription drugs in the presence of non-linear price schedules: an illustration from Quebec, Canada.

The price elasticity of demand for prescription drugs is a crucial parameter of interest in designing pharmaceutical benefit plans. Estimating the elasticity using micro-data, however, is challenging because insurance coverage that includes deductibles, co-insurance provisions and maximum expenditure limits create a non-linear price schedule, making price endogenous (a function of drug consumption). In this paper we exploit an exogenous change in cost-sharing within the Quebec (Canada) public Pharmacare program to estimate the price elasticity of expenditure for drugs using IV methods. This approach corrects for the endogeneity of price and incorporates the concept of a 'rational' consumer who factors into consumption decisions the price they expect to face at the margin given their expected needs. The IV method is adapted from an approach developed in the public finance literature used to estimate income responses to changes in tax schedules. The instrument is based on the price an individual would face under the new cost-sharing policy if their consumption remained at the pre-policy level. Our preferred specification leads to expenditure elasticities that are in the low range of previous estimates (between -0.12 and -0.16). Naïve OLS estimates are between 1 and 4 times these magnitudes.

Aged↗

Using simulation-based inference with panel data in health economics.

Panel datasets provide a rich source of information for health economists, offering the scope to control for individual heterogeneity and to model the dynamics of individual behaviour. However the qualitative or categorical measures of outcome often used in health economics create special problems for estimating econometric models. Allowing a flexible specification of the autocorrelation induced by individual heterogeneity leads to models involving higher order integrals that cannot be handled by conventional numerical methods. The dramatic growth in computing power over recent years has been accompanied by the development of simulation-based estimators that solve this problem. This review uses binary choice models to show what can be done with conventional methods and how the range of models can be expanded by using simulation methods. Practical applications of the methods are illustrated using data on health from the British Household Panel Survey (BHPS).

Bayes Theorem↗

Socio-economic status, health and lifestyle.

The role of lifestyle in mediating the relationship between socio-economic characteristics and health has been discussed extensively in the epidemiological and economic literatures. Previous analyses have not considered a formal framework incorporating unobservable heterogeneity. In this paper, we develop a simple economic model in which health is determined (partially) by lifestyle, which depends on preferences, budget and time constraints and unobservable characteristics. We estimate a recursive empirical specification consisting of a health production function and reduced forms for the lifestyle equations using Maximum Simulated Likelihood (MSL) for a multivariate probit (MVP) model with discrete indicators of lifestyle choices and self-assessed health (SAH) on British panel data from the 1984 and 1991 Health and Lifestyle Survey (HALS). We find that sleeping well, exercising, and not smoking in 1984 have dramatic positive effects on the probability of reporting excellent or good SAH in 1991, and that these effects are much larger having accounted for endogeneity. The failure of epidemiological analyses to account for unobserved heterogeneity can explain their low estimates of the relevance of lifestyle in the socio-economic status-health relationship. Indicators for prudent alcohol consumption and eating breakfast in 1984 are not found to be statistically significant determinants of SAH in 1991.

Choice Behavior↗