A Cluster Analysis to Identify Predictors and Profiles of Financial Hardship Among Cancer Patients in the United States
Author(s)
Chia-Yun Hsu, BPharm, MS, Eberechukwu Onukwugha, MSc, PhD;
University of Maryland School of Pharmacy, Baltimore, MD, USA
University of Maryland School of Pharmacy, Baltimore, MD, USA
Presentation Documents
OBJECTIVES: Information about groups of individuals diagnosed with cancer who are more likely to experience financial hardship (FH) is limited. Information to characterize vulnerable groups can streamline the delivery of financial support services. This study aims to identify clusters and determine the prevalence of FH among individuals diagnosed with cancer.
METHODS: A cross-sectional study was conducted, using the 2011, 2016, and 2017 Medical Expenditures Panel Survey to identify adults with a cancer diagnosis who were treated or last treated within one year of completing the Cancer Self-Administered Questionnaire. We collected three types of FH variables: material, psychological, and behavioral. We calculated catastrophic health expenditure (CHE) as out-of-pocket costs > 10% of annual household income. Logistic regression was used to identify predictors of FH. Agglomerative hierarchical cluster analysis was used to determine clusters using Ward's method. The number of clusters were determined from the inspection of the dendrogram. We calculated the prevalence of FH for each cluster using the individual measures and a composite of the three types.
RESULTS: We identified 847 individuals for analysis. The mean (SD) age, in years, was 65 (13) and 51% of patients were female. Age, race, ethnicity, and household income were selected for cluster analysis as these variables were statistically significant in the regression. The cluster analysis yielded 3 groups: (1) “high income and white”; (2) “young, female, and low comorbidity”; (3) “racially diverse, low income, and publicly insured”. The prevalence of the composite FH (76%) was highest in group 2. The prevalence of CHE (21%) was highest in group 3.
CONCLUSIONS: Financial hardship was highest among the cluster distinguished by youth, female gender, and low comorbidity status while CHE was highest in a racially diverse, publicly insured, and low-income group. This study can inform the development and delivery of tailored strategies to support patients experiencing financial hardship.
METHODS: A cross-sectional study was conducted, using the 2011, 2016, and 2017 Medical Expenditures Panel Survey to identify adults with a cancer diagnosis who were treated or last treated within one year of completing the Cancer Self-Administered Questionnaire. We collected three types of FH variables: material, psychological, and behavioral. We calculated catastrophic health expenditure (CHE) as out-of-pocket costs > 10% of annual household income. Logistic regression was used to identify predictors of FH. Agglomerative hierarchical cluster analysis was used to determine clusters using Ward's method. The number of clusters were determined from the inspection of the dendrogram. We calculated the prevalence of FH for each cluster using the individual measures and a composite of the three types.
RESULTS: We identified 847 individuals for analysis. The mean (SD) age, in years, was 65 (13) and 51% of patients were female. Age, race, ethnicity, and household income were selected for cluster analysis as these variables were statistically significant in the regression. The cluster analysis yielded 3 groups: (1) “high income and white”; (2) “young, female, and low comorbidity”; (3) “racially diverse, low income, and publicly insured”. The prevalence of the composite FH (76%) was highest in group 2. The prevalence of CHE (21%) was highest in group 3.
CONCLUSIONS: Financial hardship was highest among the cluster distinguished by youth, female gender, and low comorbidity status while CHE was highest in a racially diverse, publicly insured, and low-income group. This study can inform the development and delivery of tailored strategies to support patients experiencing financial hardship.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
EE17
Topic
Economic Evaluation
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, SDC: Oncology