The results of the study, as illustrated in Table 2, included 810 patients available for analysis. The episodic payment group saw a decrease in hospitalizations and therapeutic radiology use that contributed to the $33 million in savings for medical costs—a 34% decrease. On the other hand, chemotherapy costs were significantly underestimated in the episodic payment group, which resulted in a $13.5 million increase in cost. The study results show that the total medical costs were lower, even with increased chemotherapy drug costs. These results seem to demonstrate value for using this episodic payment model, independent of drug choice. Only the lung cancer group had a large enough sample size to compare survival for patients with episodic billing versus no episodic billing; however, analysis indicated no difference in overall survival for this study population. Overall, this study suggests that some element of cost control can be attained by using episodic payments, without sacrificing the quality and efficiency of care. Of note, one possible weakness in the study design included the potential existence of the Hawthorne effect—that is, participants were aware that they were being studied and may have behaved and acted differently as a result.
Health economics and outcomes research (HEOR) studies, both retrospective and prospective, can be utilized to support the use of bundled-payment policies. Retrospective cost-of-care studies, which focus on identifying the cost drivers and determining cost of care for patients at different stages of disease, can be utilized to determine the components of total cost of care. The usual components of total medical costs for cancer patients include costs of hospitalizations, emergency room visits, clinic visits, laboratory use, radiation, supportive care agents, adverse event management, drug administration, and drug costs. Understanding the differences in costs and cost components based on selection of chemotherapy agents and by line of treatment would be valuable for determining the preferred agents for patients with different stages of a specific cancer. These insights regarding cost drivers can also be used for determining costs that should or should not be included in a bundled-care model for cancer patients. Retrospective and prospective studies whose objectives focus on treatment patterns and subsequent resource utilization for adverse events and supportive care can be utilized to determine patient clinical and demographic factors and comorbidities that may predict increased resource utilization and therefore increased overall costs. These findings can be used in creating predictive risk models that categorize patient mix into high cost versus low cost groups and determine the stage-specific episodes or categories that need to be created to set up an episodic payment system.
Comparative effectiveness analysis studies also have a role in creating and determining bundled-payment policies and alternative payment methods. Studies that correlate adherence of chemotherapy agents to total medical costs and to clinical outcomes, such as overall survival and progression-free survival, can be conducted to determine the most effective treatment options. These studies are especially valuable when the agents being compared have the same mechanism of action and mode of administration and are indicated for the same cancer, by stage, and line of therapy. Multivariate analyses comparing different treatment options adjusting for confounding variables can be used to determine the most effective treatment options using real-world data as well as better defining outcomes, comorbidities, and other clinical factors. Once the most effective agent is identified using randomized clinical trial data and observational comparative effectiveness trial data, the use of therapies in appropriate patient populations may help decrease overall costs and lead to utilization of the most cost-effective options to consider including in a bundled-payment program.
Pharmaceutical companies, payers, and providers could benefit from using HEOR study experience when creating policies that affect treatment decisions and payment mechanisms, while still maintaining a high quality of care.