OBJECTIVES: The league table approach to rank ordering health care programs according to the cost-effectiveness ratio is a common method to guide policy makers in setting priorities for resource allocation. The league table approach, however, has so far been described from a deterministic perspective. The objective of the present study is to propose a risk-adjusted method to ranking health care programs. METHODS: In the presence of uncertainty, ranking programs is complicated by the degree of variability associated with each program. Confidence intervals for cost-effectiveness ratios may be overlapping. Moreover, confidence intervals may include negative ratios and the interpretation of negative cost-effectiveness ratios is ambiguous. We suggest ranking health care programs according to their rate of return, which is defined as the net monetary benefit over the costs of the program. However, how does a program with a higher expected return but higher uncertainty compare to a program with a lower expected return but lower risk? RESULTS: We borrow methods used in portfolio theory. Financing a health care program is treated as an investment in a risky asset. The risky asset is combined with a risk-free asset in order to construct a combined portfolio. The weights attributed to the risk-free and risky assets are chosen in such a manner that all programs under consideration exhibit the same degree of uncertainty. CONCLUSION: The proposed method can be used to compare the performance of the individual programs by constructing a risk-adjusted league table of expected returns.