1547 Background: Despite advances in cancer outcomes, costs of cancer care continue to rise. Acute care utilization contributes significantly to these costs. The Cancer Support Program’s (CSP) care management services focus on providing education on symptom management, palliative care and advance care planning, addressing psychosocial needs and closing gaps in care in members with cancer who are at high risk for unplanned hospital admissions (UHA). Methods: Beginning 1/1/22, commercially insured members with an active cancer diagnosis were clinically stratified as high or low risk for UHA. All high-risk members were invited to participate in the CSP program. Individuals who agreed to participate received individualized telephonic care management support. Medical claims from 1/1/22 through 9/30/22 were evaluated, and rates of UHA were compared between participants and non-participants within the identified high-risk population for an average period of 4 months after identification (the period equivalent to the average enrollment period of participants). Results: The study population included 3,113 members who stratified as high risk. Of those, 36% were enrolled in the program. Reasons for non-enrollment include inability to reach (72%), member refused (27%) and bad contact information (1%). There were no statistically significant differences between the two groups’ demographic composition except for age (the enrolled group had slightly more in the 51-64 age group [63% vs. 59%; p < 0.001] and slightly fewer in the > 65 age group (9% vs. 15%; p < 0.001]). There were 358 UHA in the participating group and 737 UHA in the non-participating group. When normalized, there were 1,179 UHA per thousand members per year (PTMPY), 99% CI [1,171 UHA, 1,184 UHA] (bootstrap) in the participating group and 1,526 UHA PTMPY, 99% CI [1,523 UHA, 1,535 UHA] (bootstrap) in the non-participating group. Conclusions: Early results show lower UHA rates for members who participated in the care management program, suggesting the program is effective in reducing acute care utilization within the high-risk population. Limitations of this study include potential enrollment bias, changes in risk status from clinical or social factors, and the potential for human error or inconsistency in the risk stratification process. Further study is underway to validate these results.
Read full abstract