Our purpose was to assess the cost-effectiveness of substituting low-osmolality contrast media for high-osmolality media in noncardiac parenteral use in radiology.
Estimates of effectiveness, taken as the reduction in risks of reactions affecting safety and comfort, were based on a recent meta-analysis of data published since 1980. Costs were those of the contrast medium and of treating reactions.
One hundred thousand uses of low-osmolality media instead of high-osmolality media would cost an additional $7.7 million. No deaths would be prevented, but there would be 80% fewer life-threatening events and 75% fewer minor reactions. Considering only life-threatening events, complete conversion to low-osmolality media costs $62,000 per event prevented. If only minor reactions are considered, this figure is $800 per event prevented. The cost-effectiveness of partial substitution of low-osmolality media in "high-risk" patients only cannot be estimated with confidence. With one set of assumptions, a 33% reduction in life-threatening reactions at a cost of $22,000 per event prevented is estimated. These estimates are insensitive to all inputs except the cost differential between high- and low-osmolality media: for every dollar decrease in cost differential per use, the cost per life-threatening event prevented drops by $790.
The allocation of limited resources requires the best available information on the relevant costs and benefits. A policy of substitution of low- for high-osmolality media will not influence mortality and will bring fairly small benefits at a considerable cost.