P J Roohan1, R E Josberger, F C Gesten. 1. NewYork State Department of Health, Office of Managed Care, Albany 12237, USA. pjr02@health.state.ny.us
Abstract
OBJECTIVE: To demonstrate the effect of risk adjustment methodologies compared to crude rates in evaluating the rate of primary cesarean deliveries in managed care plans, after accounting for known demographic and clinical factors. Risk adjustment allows for a more accurate comparison of primary cesarean delivery rates among plans, eliminating potential confounding factors that could influence rates. METHODS: Data was collected from managed care plans as part of their 1998 Quality Assurance Reporting Requirements (QARR). Medicaid and commercial populations were matched to New York State Department of Health Vital Statistics birth file to produce a crude measure of cesarean deliveries per plan. Logistic regression models were then used to adjust for maternal education, age, race/ethnicity, obstetrical history, preexisting comorbid conditions, obstetrical conditions, and pregnancy-related conditions to produce adjusted rates. RESULTS: For Medicaid, the crude analysis showed four plans that were significantly lower than the statewide Medicaid managed care rate of 9.5 per 100 live births. One plan was significantly higher. The risk-adjusted results showed one plan being significantly lower than the statewide average and none being higher. For the commercial population, seven plans were significantly lower than the average of 16.3 and four plans were higher. After risk-adjusting, three plans were significantly lower and three plans were significantly higher than the statewide average. CONCLUSIONS: Risk-adjustment of primary cesarean delivery rates allows for a more accurate comparison among managed care plans. It is hoped that the generation and publication of more accurate rates will facilitate the acceptance and use of this information by clinicians in managed care plans to focus on improving health outcomes.
OBJECTIVE: To demonstrate the effect of risk adjustment methodologies compared to crude rates in evaluating the rate of primary cesarean deliveries in managed care plans, after accounting for known demographic and clinical factors. Risk adjustment allows for a more accurate comparison of primary cesarean delivery rates among plans, eliminating potential confounding factors that could influence rates. METHODS: Data was collected from managed care plans as part of their 1998 Quality Assurance Reporting Requirements (QARR). Medicaid and commercial populations were matched to New York State Department of Health Vital Statistics birth file to produce a crude measure of cesarean deliveries per plan. Logistic regression models were then used to adjust for maternal education, age, race/ethnicity, obstetrical history, preexisting comorbid conditions, obstetrical conditions, and pregnancy-related conditions to produce adjusted rates. RESULTS: For Medicaid, the crude analysis showed four plans that were significantly lower than the statewide Medicaid managed care rate of 9.5 per 100 live births. One plan was significantly higher. The risk-adjusted results showed one plan being significantly lower than the statewide average and none being higher. For the commercial population, seven plans were significantly lower than the average of 16.3 and four plans were higher. After risk-adjusting, three plans were significantly lower and three plans were significantly higher than the statewide average. CONCLUSIONS: Risk-adjustment of primary cesarean delivery rates allows for a more accurate comparison among managed care plans. It is hoped that the generation and publication of more accurate rates will facilitate the acceptance and use of this information by clinicians in managed care plans to focus on improving health outcomes.