Health policy implications of a cost-effectiveness analysis for admitting patients to the pediatric intensive care unit

Objective: To perform a cost-effectiveness analysis of admitting patients to the pediatric or intermediate intensive care unit and to evaluate the ethical implications of results from this analysis for health policy. Design: A decision-analytic model with outcomes focused on short-term survival and long-term life expectancy. A societal perspective is taken using a base-case discount rate of 3% per year. Cost data are based upon differences in average costs between the pediatric intensive care unit (PICU) and the intermediate intensive care unit (IICU). Three-way sensitivity analyses varied the patients' probability of acute survival, life expectancy after successful discharge, and differences in average costs between the two levels of care (PICU and IICU). Based on an arbitrary 'societal willingness to pay' of $50,000 or $100,000 per life year, the threshold incremental probability where society would be indifferent between admission of a patient to the IICU or the PICU (i.e. the advantage of being in the PICU as compared to the IICU), was determined. Results: Using a survival rate of 1% as a clinically significant difference between the PICU and IICU, there is a marked and progressive decrease in cost-effectiveness of providing care for patients in the PICU as their long term life expectancy decreases. The three- way sensitivity analysis showed that these results remained robust over a wide range of values. Conclusions: Using a societal perspective, it is a more cost-effective use of resources to provide care for patients in the PICU even with short-term mortality risks of less than 1% provided their long-term life expectancy is high, as compared with patients who have high short-term mortality risks but with relatively short, long-term life expectancy.