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

Discussions of the value of medical technology often focus on average treatment effects. However, the benefits of medical technology typically vary within the population according to age, disease severity, life expectancy, and preferences over different types of outcomes. For a technology to be used efficiently, treatment decisions must be individualized, meaning that the technology is used in patients where benefits exceed costs and not used in patients where benefits are insufficiently large.

There are good reasons to believe that the individualization of medical care is sub-optimal. Fee-for-service reimbursement and concerns about malpractice provide incentives to overuse technology. "Default" bias may lead physicians to use the same treatment for every patient. In some cases, there are inadequate mechanisms in place for patients to communicate their preferences to physicians.

Each paper in this session focuses on a specific medical technology and examines the degree to which use of the technology reflects patient heterogeneity. The technologies are anti-depressants, where benefits vary by age (abstract 1), prostate cancer screening, where benefits vary by life expectancy (abstract 2), and intensive therapy for diabetes, where benefits vary by preferences (abstract 3). In each case, the authors use novel methods for examining the impact of patients' expected benefit from treatment on treatment decisions.