Oregon's Medicaid ranking and cost-effectiveness: is there any relationship?

Abstract

The authors examine whether Oregon's 1990, 1991, 1992, and 1993 prioritized lists were ranked in a manner consistent with cost-effectiveness. Two sets of cost-effectiveness data are used: data from economic analyses and Oregon's own cost-effectiveness data. Comparing the ranks of Oregon's lists with the ranks of cost-effectiveness estimates from the literature reveals Spearman correlations of -0.08 for the 1990 list, +0.39 for the 1991 list, +0.25 for the 1992 list, and +0.24 for the 1993 list. Comparing Oregon's lists with Oregon's own cost-effectiveness data reveals rank correlations of +0.99 for the 1990 list, +0.06 for the 1991 list, -0.05 for the 1992 list, and -0.03 for the 1993 list. Thus, there appear to be essentially no relationship between the 1990 list and cost-effectiveness estimates from the economic literature and modest positive relationships between the 1991-93 lists and the literature. In addition, there is virtually no relationship between the 1991-93 lists and Oregon's own cost-effectiveness data. Further, the correlations are very different from +1.0, suggesting that other factors are at play. For example, the 1993 list that is currently being implemented was ranked primarily by improvement in five-year survival and human judgment, not cost-effectiveness.