Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

This session is one of three sessions on econometric methodology issues in health economics and health policy. The papers deal with issues that are often acknowledged, but rarely addressed directly. Each of the issues addressed can have major implications for either bias in estimation or in substantial loss of precision. The first paper addresses the issues of censoring of observations before the episode of illness or lifetime of the condition is reached. By applying survival methods, it addresses issues that annual cost-analysis or per-member – per-month approaches cannot address consistently.

The second paper addresses issues of the heavy tails in health care distribution, thus going beyond the usual concerns about heteroscedasticity and skewneess in health care cost analysis.

The third paper deals with valuations of morbidity and mortality risk revealed in the labor market. Although the value of a statistical life has been used widely in some areas where there concerns about excess morbidity and mortality, much of that literature does not address the potential self-selection biases that arise in some individuals selecting into riskier jobs. The third paper uses selection models to understand the resulting biases in the estimates of the value of a statistical life.