Los Angeles, CA, USA: UCLA (doctoral dissertation).
University of California, Los Angeles, CA, USA
The demand for more accurate and equitable professional medical specialty payments from commercial insurers served as the catalyst for this study. The principal approaches to developing prospective per-member medical specialty payment rates often yielded little correlation to the actual financial risks associated with the members for whom the specialists assumed responsibility. The general risk adjustment methods adopted by insurers to develop population-wide premium rates were often poor predictors of resource consumption at the specialty stratification levels. This study reveals a more robust approach to developing prospective risk-adjusted medical specialty payments and subscriber group premium rates for members enrolled in a commercial Health Maintenance Organization.
This study includes the use of diagnoses-based risk scores instead of the traditional historical cost-based predictive approach . The Johns Hopkins School of Public Health Adjusted Clinical Group System provided individual member-level weights based upon age, gender, and medical diagnoses data found in historical member claims data. The expected resource weights, assigned ACG classification, enrolled group Standard Industry Code (SIC), benefit plan design, geographic area, enrolled group site, enrolled group tenure, and medical management enrollment were used as predictor variables in a regression model to estimate the subsequent year rating period costs. The study period composite healthcare costs by individual member was used as the dependent variable. The beta coefficients are used to predict costs at the individual member level, whether existing or prospective. The predicted costs are used to pay the specialists according to the composition of enrolled members. However, subscribing group premium rates are estimated using the same member-level aggregated predictive costs.