Economic Evaluation of Healthcare Point Solutions Versus Standard Care to Supplement Benefit Plans for Behavioral Health Adults Patients Using Real-World Synthetic Data: A Doubly Robust Causal Analysis
Author(s)
Vassiki Sanogo, MSc, PhD;
Integrity Analytics, Haines City, FL, USA
Integrity Analytics, Haines City, FL, USA
Presentation Documents
OBJECTIVES: There is an increasing demand for offering healthcare services addressing specific conditions, tailored to meet individual needs. Over 30+ healthcare Point Solutions (PS) marketed encompassing condition specific solutions are used for behavioral health. Thus, supporting the adoption of PS supplement to benefit plan programs by demonstrating its real-world value using simulated synthetic data is critical. This study aimed to assess the cost-effectiveness of PS in behavioral health adult patients.
METHODS: Synthetic data were simulated using patients’ clinical and demographic characteristics, disease onset prevalence, drug and digital solutions utilization for 50,000 patients covered by public or private health plans from January 2021 to January 2023. Comprehensive data analysis was conducted, including descriptive, matching strategies for three-groups samples, difference-in-difference, and cost-effectiveness analyses. Of the three matching approaches, the generalized propensity scores matching weights method was retained to generate a matched retrospective cohort with standardized characteristics.
RESULTS: Out of the matching analysis the patient samples were N=4310 for control group, N=26206 for SoC group, and N=13029 PS group. The study determined total costs, disease prevalence yield (DPY), and incremental cost-effectiveness ratio (ICER) adjusted for selection bias and right censoring. Sensitivity analyses explored the robustness of ICER through bootstrapping. PS resulted in a DPY of 10.92% (p = 0.0022) while standard of care (SoC) was 9.07% (p = 0.0543). PS resulted in a mean saving per person of $967 (SD = 1,516, p < 0.001) and a total of $12.6 M for 13,029 patients from matched sample.
CONCLUSIONS: PS supplement for health plans in behavioral health adult patients is cost-effective, cost-saving, and positive DPY from SoC traditional approaches for health plans support.
METHODS: Synthetic data were simulated using patients’ clinical and demographic characteristics, disease onset prevalence, drug and digital solutions utilization for 50,000 patients covered by public or private health plans from January 2021 to January 2023. Comprehensive data analysis was conducted, including descriptive, matching strategies for three-groups samples, difference-in-difference, and cost-effectiveness analyses. Of the three matching approaches, the generalized propensity scores matching weights method was retained to generate a matched retrospective cohort with standardized characteristics.
RESULTS: Out of the matching analysis the patient samples were N=4310 for control group, N=26206 for SoC group, and N=13029 PS group. The study determined total costs, disease prevalence yield (DPY), and incremental cost-effectiveness ratio (ICER) adjusted for selection bias and right censoring. Sensitivity analyses explored the robustness of ICER through bootstrapping. PS resulted in a DPY of 10.92% (p = 0.0022) while standard of care (SoC) was 9.07% (p = 0.0543). PS resulted in a mean saving per person of $967 (SD = 1,516, p < 0.001) and a total of $12.6 M for 13,029 patients from matched sample.
CONCLUSIONS: PS supplement for health plans in behavioral health adult patients is cost-effective, cost-saving, and positive DPY from SoC traditional approaches for health plans support.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
EE519
Topic
Economic Evaluation
Topic Subcategory
Cost/Cost of Illness/Resource Use Studies
Disease
SDC: Mental Health (including addition)