Background: Little is known about how morbidity levels progress over time and the implications of these morbidity trajectories for healthcare utilization.
Objectives: To identify and compare characteristics of people in different morbidity trajectories and to evaluate how morbidity trajectories impact the performance of diagnostic risk-adjustment models.
Research design: Morbidity trajectories were derived from 3-year (2002 to 2004) of claims from a national insurance system. These trajectories, with or without 2004 claims-based risk adjusters developed from the Adjusted Clinical Group case-mix system, were used to explain medical utilization in 2005.
Subjects: A random sample of Taiwanese National Health Insurance beneficiaries continuously enrolled from 2002 to 2005 (n=147,892).
Measures: Adjusted R of 5 types of healthcare expenditures.
Results: On the basis of naturally occurring patterns, we identified 6 morbidity trajectory groups. People assigned to different trajectory groups have distinct demographics and medical utilization. The effect of adding morbidity trajectory indicators differed substantially by the comprehensiveness of baseline risk-adjustment models: the increase in adjusted R ranged from 0.3% in the most comprehensive model to 5.7% in the demographics model.
Conclusions: A simple morbidity trajectory classification over a 3-year period is almost as powerful a predictor of prospective medical utilization as more comprehensive baseline risk adjusters. It may be unnecessary to construct longitudinal morbidity trajectories if a comprehensive baseline model was adopted, especially for healthcare systems without the stability of continuous enrollment.