Aaron Stinnett

Aaron Stinnett

Lutheran Pastor
Formerly Associate Professor of Public Health, University of Alabama at Birmingham

Dissertation Title: "Issues in the Economic Evaluation of Health Interventions"The objective of this thesis is to shed light on unresolved issues in the economic evaluation of health interventions.

The opening paper addresses the use of constrained optimization techniques for health resource allocation, demonstrating that a general mathematical programming framework can accommodate much more complex information regarding returns to scale, indivisibilities, and program independence than previously has been considered.

The second paper presents a new framework for evaluating uncertainty in cost-effectiveness analysis (ECA). This method expresses the results of an evaluation in terms of the Net Health Benefit (NHB) conferred, where a program's NHB measures the health gain expected to result from implementing that program, compared to implementing its comparator and investing any cost differential in a marginally cost-effective program. This framework offers theoretical and practical advantages over the statistical analysis of cost-effectiveness (C/E) ratios.

The third paper evaluates two methods that have been proposed for estimating C/E ratios under second-order uncertainty. One method estimates a mean ratio of cost to effect, and the other estimates a ratio of mean cost to mean effect. It is demonstrated that the "ratio of means" approach follows from the theoretical foundations of CEA and is consistent with an intuitively appealing vector algebra approach to the problem; in contrast, the "mean ratio" approach is internally inconsistent and can prescribe economically inefficient choices.

The final paper investigates the cost-effectiveness of diet-and drug-based clinical strategies for cholesterol reduction in men and women at various degrees of risk for coronary heart disease (CHD). A computer simulation model is used to estimate the costs and health effects of strategies employing various combinations of a low-fat diet, niacin, and lovastatin. The results indicate that cholesterol reduction for the prevention of CHD can be relatively cost-effective in a variety of risk factor groups, but that some strategies are significantly more cost-effective than others. Strategies based on diet, niacin therapy, and stepped care (in which patients begin with niacin therapy and switch to lovastatin if they cannot tolerate niacin) tend to be relatively cost-effective. In contrast, nearly all strategies that employ lovastatin as a first-line medication are dominated.


Graduation Year


Dissertation Committee Member