By Megan Mueller

Schulich School of Business Professor Sylvia Hsu led a six-person team of researchers from Yale University as they followed the healthcare pathways of over 100,000 cancer patients. Hsu’s team examined regional end-of-life expenditures (as a proxy for regional practice patterns) and found racial/ethnic differences in intensity of end-of-life care.

They discovered that most white patients (70 per cent) used hospices, while racial minorities were significantly more likely to receive aggressive end-of-life care, such as being admitted into an intensive care unit or an ER visit, which is associated with high health care expenditures. This research, which looked at high- and low-expenditure regions, led Hsu’s team to conclude that regional practice patterns may influence racial/ethnic differences in end-of-life care.

This work could inform future policy. “Programs to promote less aggressive care, such as hospices, could help to reduce racial/ethnic differences and improve the quality of end-of-life care,” Hsu says. Such programs, in other words, could be better for the patients and for expenditures.

The results, published in the Wiley journal Health Services Research (2018), will be of interest to physicians; healthcare practitioners; administrators of hospitals, hospices and other healthcare facilities; and patients and their families.

Researchers look at expenditures to explain underlying disparities

End-of-life care is increasingly important given the aging population, but it also pertains to those younger individuals with terminal, progressive or incurable diseases or conditions.

Existing research has revealed that end-of-life care differs substantially by race/ethnicity. Black, Hispanic and Asian patients with advanced cancer are more likely to opt for more aggressive end-of-life care and less likely to use hospice services than white patients.

Hsu’s original work adds to this literature because she takes a different approach: She considers racial/ethnic factors with a focus on regional end-of-life expenditures (as a proxy for regional practice patterns) and end-of-life care.

Why? In low-expenditure regions where hospice care is common and aggressive care is not, Hsu suspected that doctors might be telling patients about the benefits of hospice care. Here, patients would be more likely to receive hospice care regardless of their racial/ethnic background. Conversely, Hsu suspected that in high-expenditure areas, where aggressive care is more common than hospice use, diversity in patient preferences may result in greater variation in hospice use and aggressive care.

Researchers followed 102,759 cancer patients’ path through health care system

To conduct this research, Hsu turned to the SEER (Surveillance, Epidemiology, and End Results)-Medicare database through which it was possible to follow a large sample of 102,759 patients and their path through the healthcare system at the end of their lives.

The sample consisted of patients with breast, prostate, lung, colorectal, pancreas, liver, kidney, melanoma or hematological cancer, diagnosed from 2004 to 2011. All patients were between 66 and 94 years of age, had died within three years of diagnosis, and were continuously enrolled in Medicare during the last 18 months of life.

The researchers turned to Medicare claims to see how end-of-life health care services were used. They looked at corresponding end-of-life expenditures for the complete picture.

Results speak volumes about disparity

This research showed that racial minority patients entered end-of-life care in dire straits, when compared to white patients:

• Compared to white patients, Hispanic and black patients were younger.
• Compared to white patients, racial minority patients were significantly more likely to have multiple comorbidities – that is, two or more chronic diseases or conditions.

What kinds of health care services did they receive – hospice versus aggressive end-of-life care?

• Hospice use was highest among white patients (70 per cent), next most common in Hispanic (65 per cent) and black patients (62 per cent), and lowest in Asian patients (53 per cent).
• Racial minorities were significantly more likely to receive aggressive end-of-life care, compared to white patients.

What can high- and low-expenditures regions this tell us?

Overall end-of-life expenditures per cancer decedent in the last six months of life by race/ethnicity ranged from $10,800 to $15,400. Isolating this analysis by low- and high-expenditure regions (again, as proxy for regional practice patterns), reinforces the racial differences.

Hsu sums up: “Hispanic and Asian patients had end-of-life care patterns that were similar to white patients in low-expenditure areas, but received more aggressive end-of-life care than whites in high-expenditure areas. Black patients were more likely than white patients to receive aggressive end-of-life care in all expenditure regions.”

Policy implications

This research tests and confirms Hsu’s hypothesis that regional practice patterns may be influencing people’s choices around end-of-life care. She believes that this has two key policy implications.

1. Since high heath care expenditures are associated with aggressive end-of-life care, and regional practice patterns (as determined via expenses) may play a role in this, then reducing regional expenditures might decrease racial differences in health care utilization.
2. Earlier research has told us that an integrated palliative care model and reduced fragmentation of care may decrease aggressive end-of-life care. (Fragmentation refers to the systemic misalignment of incentives, or lack of coordination, that spawns inefficient allocation of resources or harm to patients. Itadversely impacts quality, cost, and outcomes.) Given this, health care administrators in high-expenditure areas could adopt the approaches used in the low-expenditure areas to cultivate regional practice norms promoting hospice services over aggressive end-of-life care.

Physicians could be trained to talk about treatment plans to patients of diverse racial/ethnic groups

Elaborating on the second policy implication, Hsu suggests that physicians and hospital administrators could modify their approach and practice patterns to reduce the expenditure of end-of-life care and address the cultural/ethnic differences.

“Physicians could get communication skills training to efficiently deliver information about treatment plans to patients of diverse racial/ethnic groups. These systemic changes could help minority patients accept hospice care and avoid aggressive end-of-life care,” Hsu explains.

Megan Mueller is the senior manager, Research Communications, Office of the Vice-President Research & Innovation, York University.