Does managed care affect the supply and use of ICU services?

This paper analyzes use of hospital intensive care units (ICUs) by adult patients who are under age 65 and not covered by Medicaid; it allows for variation in indicators of the patient's condition, severity of illness, type of admission, emergency status, and degree of constraint on the total hospital ICU supply. We use data for Massachusetts and Florida in 1992. In neither state is there a significant difference in ICU admission rates between managed care patients and other privately insured patients. In Massachusetts, we find that the length of stay in the ICU is somewhat less for managed care and uninsured patients than for other privately insured patients. In both states, a hospital's ratio of total annual supply of ICU services to expected demand has a strong effect. In Massachusetts, the differences across payer groups in length of stay disappear for hospitals where the total ICU supply is relatively constrained.