Evaluating the effect of a scalable cancer-navigation program on total cost of care.

Abstract

4 Background: Cancer-patient navigation programs have been shown to reduce total cost of care while improving outcomes. Previous research has focused on navigation programs embedded within a single institution or a single network of practices. We explore whether an independent cancer-patient navigation program deployed in collaboration with a Medicare Advantage (MA) plan can demonstrate similar cost savings when deployed across a wide variety of practice settings in the state of New Jersey. Methods: We conducted a retrospective observational study to determine the per member per month (PMPM) total cost of care (TCOC) difference between cancer patients with and without a navigation intervention provided by Thyme Care, a technology-enabled virtual cancer-navigation platform. The team, staffed by both lay-health workers and nurses, was assigned members of a single MA plan’s population in NJ beginning on March 2, 2021. Patients with a cancer-specific ICD-10 code who received systemic antineoplastic therapy were eligible for the program. The navigation team coordinated care, addressed barriers to care, facilitated goals of care discussions, and conducted proactive and reactive symptom assessment. The cadence of the symptom assessment was determined by both patient and disease-specific criteria. Concerning responses were triaged to the patient’s oncology team the same day. The study period was from January 1, 2018 to June 13, 2022. The primary outcome was TCOC excluding part D drug spend. We compared TCOC for patients who received the intervention to a control group of eligible individuals that were matched using propensity scores derived from clinical and demographic covariates. We used a repeated measure generalized linear model to estimate the monthly effect of navigation on TCOC. Additionally, we conducted a monte-carlo sensitivity analysis to provide an upper and lower bound on the cost savings. Results: Of 4,172 eligible members, 662 were in the treatment group (across 25 community practices and 45 hospital based practices) and 3,510 were in the control group (across 58 community practices and 73 hospitals). The propensity matching resulted in 188 members in both the treatment and the control group (376 total). The mean TCOC declined $459 more PMPM for the navigated group compared to the non-navigated group. The sensitivity analysis suggested a plausible range of $209-708 PMPM savings in the intervention group. Conclusions: This is the first study to demonstrate that a cancer-navigation program can reduce overall costs when deployed in collaboration with a health plan across a wide geography and range of practice types.