Award Date

May 2018

Degree Type

Thesis

Degree Name

Master of Public Health (MPH)

Department

Environmental and Occupational Health

First Committee Member

Paulo S. Pinheiro

Second Committee Member

Sheniz Moonie

Third Committee Member

Brian Labus

Fourth Committee Member

Daniel Young

Number of Pages

54

Abstract

Hispanics are now the youngest, largest, and fastest growing minority group in the U.S. Prostate cancer (PC) is the most commonly diagnosed cancer in men and is the second-leading cause of cancer deaths among Hispanics. For the first time, we examined PC-specific survival among distinct Hispanic groups that include Mexicans, Cubans, Dominicans, Puerto Ricans, as well as Central and South Americans. We compared these groups to the main reference population in the U.S., non-Hispanic Whites (NHW), after adjustment for prognostic factor risk categories (prostate-specific antigen (PSA) level, Gleason score, and tumor stage), as well as sociodemographic covariates (e.g., health insurance, and marital status). Surveillance, Epidemiology, and End Results (SEER) data from 2004 to 2013 were used. Cox proportional hazards regression revealed that Hispanics, overall, show an increased risk of death in comparison to NHW HR = 1.17 (95% CI: 1.12-1.22), over time, but that difference disappears after adjustment for prognostic factors HR = 0.97 (95% CI: 0.93-1.02). This result is likely due to the known overdiagnosis of initial indolent PC cancers more common in the referent group (NHW). Moreover, among the subcategory of unknown stage (as opposed to known stage), Hispanics do relatively poorly which may account for some of the increased risk of death in relation to NHW. To further examine a more meaningful disparity, we restricted our analysis to those with localized stage and PSA larger than 9.9 ng/ml. Initially, for these intermediate risk stages that are more prone to disparities, due to differential treatment or access to quality healthcare, there were no differences between Hispanics and the referent group HR = 1.03 (95% CI: 0.96 – 1.09, p-value = 0.44); however, the HR appeared to improve after adjusting for prognostic factors (HR = 0.95, 95% CI: 0.89 – 1.01, p-value = 0.09), which may indicate that Hispanics present marginally worse biological characteristics in comparison to NHW. Finally, after adjusting for all prognostic and social factors, Hispanics showed a theoretical survival advantage in comparison to NHW HR = 0.93 (95% CI: 0.87-0.99). Contrary to expected results, social factors that were included in this study did not appear to add a survival advantage for Hispanics. Among the different Hispanic groups, Puerto Rican men living in the U.S. (HR = 1.38, p-value <0.01, 95% CI: 1.17 – 1.63) showed the highest disparity in relation to NHW but this estimate may be impacted to some extent by the Not Otherwise Specified (NOS) bias. This bias may be large enough to significantly influence and artificially increase the risk of death among Puerto Ricans, as seen in our results. More should be done to improve stage at diagnosis and access to quality healthcare for all Hispanics to eliminate the persisting disparities.

Keywords

Cox Proportional Hazards Model; Hispanic; Population Based Cohort Study; Prostate Cancer; Surveillance; Epidemiology; and End Results (SEER); Survival Analysis

Disciplines

Biostatistics | Epidemiology

Language

English


Share

COinS