Contributions of prognostic factors to socioeconomic disparities in cancer survival: protocol for analysis of a cohort with linked data

Racial/ethnic disparities in the cause of death among patients with prostate cancer in the United States from 1995 to 2019: a population-based retrospective cohort study

Background

Racial/ethnic disparities in prostate cancer are reported in the United States (US). However, long-term trends and contributors of racial/ethnic disparities in all-cause and cause-specific death among patients with prostate cancer remain unclear. We analysed the trends and contributors of racial/ethnic disparities in prostate cancer survivors according to the cause of death in the US over 25 years.

Methods

In this retrospective, population-based longitudinal cohort study, we identified patients diagnosed with first primary prostate cancer between 1995 and 2019, with follow-up until Dec 31, 2019, using population-based cancer registries' data from the Surveillance, Epidemiology, and End Results (SEER) Program. We calculated the cumulative incidence of death for each racial/ethnic group (Black, white, Hispanic, Asian or Pacific Islander [API], and American Indian or Alaska Native [AI/AN] people), by diagnostic period and cause of death. We quantified absolute disparities using rate changes for the 5-year cumulative incidence of death between racial/ethnic groups and diagnostic periods. We estimated relative (Hazard ratios [HR]) racial/ethnic disparities and the percentage of potential factors contributed to racial/ethnic disparities using Cox regression models.

Findings

Despite a decreasing trend in the cumulative risk of death across five racial/ethnic groups, AI/AN and Black patients consistently had the highest rate of death between 1995 and 2019 with an adjusted HR of 1.48 (1.40-1.58) and 1.40 (1.38-1.42) respectively. The disparities in all-cause mortality between AI/AN and white patients increased over time, with adjusted HR 1.32 (1.17-1.49) in 1995-1999 and 1.95 (1.53-2.49) in 2015-2019. Adjustment of stage at diagnosis, initial treatment, tumor grade, and household income explained 33% and 24% of the AI/AN-white and Black-white disparities in all-cause death among patients with prostate cancer.

Interpretation

The enduring racial/ethnic disparities in patients with prostate cancer, call for new interventions to eliminate health disparities. Our study provides important evidence and ways to address racial/ethnic inequality.

Funding

National Key R&D Program of China, National Natural Science Foundation of China, Beijing Municipal Administration of Hospitals Clinical Medicine Development of Special Funding Support, the Open Research Fund from Beijing Advanced Innovation Center for Big Data-Based Precision Medicine, Key Projects of Philosophy and Social Sciences Research, Ministry of Education of China.

Socioeconomic disparities in colorectal cancer survival: contributions of prognostic factors in a large Australian cohort

Purpose

We quantified the contributions of prognostic factors to socioeconomic disparities in colorectal cancer survival in a large Australian cohort.

Methods

The sample comprised 45 and Up Study participants (recruited 2006–2009) who were subsequently diagnosed with colorectal cancer. Both individual (education attained) and neighbourhood socioeconomic measures were used. Questionnaire responses were linked with cancer registrations (to December 2013), records for hospital inpatient stays, emergency department presentations, death information (to December 2015), and Medicare and Pharmaceutical Benefits claims for subsidised procedures and medicines. Proportions of socioeconomic survival differences explained by prognostic factors were quantified using multiple Cox proportional hazards regression.

Results

1720 eligible participants were diagnosed with colorectal cancer after recruitment: 1174 colon and 546 rectal cancers. Significant colon cancer survival differences were only observed for neighbourhood socioeconomic measure (p = 0.033): HR = 1.55; 95% CI 1.09–2.19 for lowest versus highest quartile, and disease-related factors explained 95% of this difference. For rectal cancer, patient- and disease-related factors were the main drivers of neighbourhood survival differences (28–36%), while these factors and treatment-related factors explained 24–41% of individual socioeconomic differences. However, differences remained significant for rectal cancer after adjusting for all these factors.

Conclusion

In this large contemporary Australian cohort, we identified several drivers of socioeconomic disparities in colorectal cancer survival. Understanding of the role these contributors play remains incomplete, but these findings suggest that improving access to optimal care may significantly reduce these survival disparities.