# Predictors and Drivers of End-of-Life Medicare Spending Among Older Adults with Solid Tumors: A Population-Based Study

**Authors:** Courtney E. Baird, Elizabeth Wulff-Burchfield, Pamela C. Egan, Lee A. Hugar, Ami Vyas, Nikolaos A. Trikalinos, Michael A. Liu, Adam J. Olszewski, Leonidas E. Bantis, Orestis A. Panagiotou, Emmanuelle Bélanger

PMC · DOI: 10.3390/cancers17061016 · Cancers · 2025-03-18

## TL;DR

This study identifies factors that influence high healthcare spending in the last month of life for older cancer patients, suggesting ways to improve end-of-life care.

## Contribution

The study identifies specific demographic, clinical, and geographic predictors of high end-of-life Medicare spending among older cancer patients.

## Key findings

- Higher end-of-life spending is associated with more health conditions, being female, being Black or non-White, advanced-stage cancer, and living in populated areas.
- Lower spending is observed in older patients, those in rural areas, and those with poor health.
- Targeted interventions like improved palliative care access could reduce unnecessary treatments and improve patient experiences.

## Abstract

Medicare patients with cancer account for a significant portion of healthcare spending, particularly in the last month of life. However, the reasons behind this high cost remain unclear. To better understand the factors contributing to these expenses, we analyzed data from older adults who died of breast, prostate, lung, or colorectal cancer. Our study found that higher end-of-life healthcare costs were linked to having more existing health conditions, being female, being Black or of another non-White race, having advanced-stage cancer, living in more populated areas, and receiving state assistance for Medicare premiums. In contrast, lower spending was observed among older patients, those living in rural areas, and those with poorer overall health. These findings suggest that targeted interventions, such as improving access to palliative care for high-risk patients, could help reduce unnecessary medical treatments and improve end-of-life experiences for patients. Understanding these patterns can help policymakers and healthcare providers ensure that resources are used efficiently while prioritizing patient comfort and quality of life.

Purpose: Medicare patients who die from cancer are responsible for about 30% of annual Medicare spending, most of which occurs during the last 30 days of life. Yet, there are significant and persisting knowledge gaps regarding which factors contribute to this high-intensity EoL spending. To that end, we conducted a retrospective analysis of SEER-Medicare data to identify risk factors associated with high-intensity EoL spending among older adults who died of breast, prostate, lung, or colorectal cancer. Methods: We used multivariable linear regression to identify clinical, demographic, socioeconomic, and geographic characteristics associated with the total inflation-adjusted Medicare spending in the last 30 days of life, including reimbursements for services provided in inpatient and outpatient settings. The study subjects included U.S. Medicare beneficiaries aged 65 and older who died of breast, prostate, lung, or colorectal cancer between 2011 and 2015. Results: Among 59,355 decedents (49.2% female; 21.2% of a non-White race/ethnicity), the factors associated with greater EoL spending were an increased comorbidity burden, the female sex, the Black race, other races/ethnicities, stage III or IV disease, living in a more populated county, and state subsidization of patient Medicare premiums. The EoL spending was lower among older patients; patients living in the Midwest, South, or West; patients living in more rural areas; and patients with a poor performance status. The results were largely consistent across cancer types. Conclusions: Our findings can inform targeted intervention development for patients with cancer who are at a higher risk of high-intensity EoL spending, such as decision support tools that facilitate referrals to palliative care for high-risk patients.

## Linked entities

- **Diseases:** breast cancer (MONDO:0004989), prostate cancer (MONDO:0005159), lung cancer (MONDO:0005138), colorectal cancer (MONDO:0005575)

## Full-text entities

- **Diseases:** Solid Tumors (MESH:D009369)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC11941710/full.md

## Figures

2 figures with captions in the complete paper: https://tomesphere.com/paper/PMC11941710/full.md

## References

51 references — full list in the complete paper: https://tomesphere.com/paper/PMC11941710/full.md

---
Source: https://tomesphere.com/paper/PMC11941710