Effects of managed care on service use and access for publicly insured children with chronic health conditions.

OBJECTIVE

Our goal was to estimate the effects of managed care program type on service use and access for publicly insured children with chronic health conditions.

METHODS

Data on Medicaid and State Children's Health Insurance Program managed care programs were linked by county and year to pooled data from the 1997-2002 National Health Interview Survey. We used multivariate techniques to examine the effects of managed care program type, relative to fee-for-service, on a broad array of service use and access outcomes.

RESULTS

Relative to fee-for-service, managed care program assignment was associated with selected reductions in service use but not with deterioration in reported access. Capitated managed care plans with mental health or specialty carve-outs were associated with a 7.4-percentage-point reduction in the probability of a specialist visit, a 6.3-percentage-point reduction in the probability of a mental health specialty visit, and a 5.9-percentage-point decrease in the probability of regular prescription drug use. Reductions in use associated with primary care case management and integrated capitated programs (without carve-outs) were more limited, and integrated capitated plans were associated with a reduction in unmet medical care need. We failed to find significant effects of special managed care programs for children with chronic health conditions.

CONCLUSIONS

Managed care is associated with reduced service use, particularly when capitated programs carve out services. This finding is of key policy importance, as the proportion of children enrolled in plans with carve-out arrangements has been increasing over time. It is not possible to determine whether reductions in services represent better care management or skimping. However, despite the reductions in use, we did not observe a corresponding increase in perceived unmet need; thus, the net change may represent improved care management.