THE RISK PREDICTION MODEL FOR OUT-OF-POCKET PAYMENTS AMONG PATIENTS UNDERGOING CHEMOTHERAPY FOR COLORECTAL CANCER IN UKRAINE DURING THE WAR

Abstract

Introduction. In Ukraine colorectal cancer (CRC) is top 3 by incidence and top 2 by mortality rate, 9.9% and 14.4% respectively. Despite recently introduced healthcare reform and resulting improvements in accessibility of health services, out-of-pocket payments (OOP payments) remain to be a serious challenge in Ukraine.

Aim. To develop a risk prediction model for OOP payments based on organizational and regional features of CRC chemotherapy, patient sociodemographic and clinical characteristics.

Materials and methods. A cross-sectional study surveyed 459 respondents undergoing chemotherapy for colorectal cancer in Ukraine. The participants were divided into two groups depending on the presence of OOP payments (121 – yes, 338 – no). Statistical data analysis was performed using the R programming language version 4.3.1 with the EZR graphical user interface version 1.64. Logistic regression was used to evaluate the impact of various factors on the risk of OOP payments.

Results. A five-variable logistic regression model was developed to predict the odds ratio (OR) of incurring OOP payments. The model demonstrated very good concordance with an area under the curve (AUC) of 0.798 (0.744 – 0.851). The model included the following statistically significant indicators that increased the risk of OOP payments: age 60–69 years (p = 0.007), treatment in the Northern and Central subregions (p < 0.001 and p = 0.008, respectively), underweight based on patient body mass index (BMI) < 18.5 (p = 0.006), and waiting time for treatment initiation for more than 2 months (p = 0.017). At the same time, being overweight with a BMI of 25.0 – 29.9 decreased the risk of OOP payments (p = 0.03). The average monthly income per household member was present in the final model, but did not demonstrate an independent statistically significant effect (p > 0.05).

Conclusions. The risk prediction model reveals that the risk of incurring OOP payments is primarily driven by clinical parameters (underweight, BMI < 18.5; and overweight, BMI 25.0 – 29.9), access to care (waiting time of more than 2 months), socio-demographic factors (age 60–69 years), and regional disparities in healthcare delivery. Actionable strategies for rapid implementation to mitigate OOP payments include optimizing healthcare organization to reduce treatment waiting times, increasing patient awareness regarding free treatment.