Group-based trajectory analysis identifies varying diabetes-related cost trajectories among type 2 diabetes patients in Texas: an empirical study using commercial insurance

Background The trend of Type 2 diabetes-related costs over 4 years could be classified into different groups. Patient demographics, clinical factors (e.g., A1C, short- and long-term complications), and rurality could be associated with different trends of cost. Study objectives are to: (1) understand the trajectories of cost in different groups; (2) investigate the relationship between cost and key factors in each cost trajectory group; and (3) assess significant factors associated with different cost trajectories. Methods Commercial claims data in Texas from 2016 to 2019 were provided by a large commercial insurer and were analyzed using group-based trajectory analysis, longitudinal analysis of cost, and logistic regression analyses of different trends of cost. Results Five groups of distinct trends of Type 2 diabetes-related cost were identified. Close to 20% of patients had an increasing cost trend over the 4 years. High A1C values, diabetes complications, and other comorbidities were significantly associated with higher Type 2 diabetes costs and higher chances of increasing trend over time. Rurality was significantly associated with higher chances of increasing trend over time. Conclusions Group-based trajectory analysis revealed distinct patient groups with increased cost and stable cost at low, medium, and high levels in the 4-year period. The significant associations found between the trend of cost and A1C, complications, and rurality have important policy and program implications for potentially improving health outcomes and constraining healthcare costs. Keywords: Type 2 Diabetes related cost, Group-based trajectory analysis, Multivariable logistic regression models, Diabetes self-management education and support