The impact of using Elsevier ClinicalPath oncology treatment pathways on survival and cost of care.

Abstract

6620 Background: Cone Health is a 921 bed hospital system in North Carolina with a comprehensive cancer center. Our aim was to determine the value of using Elsevier’s ClinicalPath oncology pathways on short-term survival and cost of care from the patient and health system perspectives. Cancer treatment has become more complex as it progresses more toward precision medicine modalities. As such, clinical decision support tools, such as ClinicalPath, have gained traction. Health systems struggle to survive after the COVID pandemic, with residual financial impacts lingering into 2023. It is more important than ever before that health systems focus on value where the patient benefits and financial sustainability are improved. Methods: We used a retrospective cohort study design with matching. Exposure was the use of the ClinicalPath decision support tool in Epic (exposed) versus no use of the tool (not exposed). We matched on cancer body site, AJCC clinical stage, the goal of treatment, Elixhauser comorbidity score, and year of a cancer diagnosis. We were able to match 509 exposed to 509 unexposed. Our primary outcomes were 12-month survival from the time of the first treatment and contribution margin. We used a log-rank calculation based on our hazard ratios (Freedman method) to determine the sample size required for each group, assuming a power of 80%. We used a Weibull parametric model to assess 12-month survival with additional covariates. An OLS model was used to evaluate the contribution margin. Results: Results demonstrate that when ClinicalPath is used to guide treatment, patients are half as likely to die within 12 months from treatment compared to the unexposed (HR 0.50; 95% CI 0.37-0.68). The OLS contribution model found, on average, that margin has an associated 74.0% (0.74; 95% CI 0.58-0.90) increase when ClinicalPath is used to guide treatment compared to the unexposed. This equates to a crude increase in contribution margin for cancer treatment of $58,362, on average per case, when ClinicalPath is used. Variable direct costs are higher when using ClinicalPath. This means, on average, it costs more to treat patients on pathway using best available evidence. Still, they have better short-term survival and payers are covering the additional costs given the increase in contribution margin for our system. Conclusions: Short-term recurrence free survival is significantly improved when providers utilize ClinicalPath compared to not using the tool consistent with the primary aim of the CDS solution. Newer, more efficacious treatments may have higher initial costs, as expected. Still, the effectiveness of these treatments on survival in the real-world setting is born out in the current study. The higher initial direct costs are counterbalanced by a higher contribution margin per patient providing evidence that improved outcomes in cancer therapy can also provide a positive ROI for provider organizations.