Date of Graduation

2017

Document Type

Dissertation

Degree Type

PhD

College

School of Pharmacy

Department

Pharmaceutical Systems and Policy

Committee Chair

Suresh Madhavan

Committee Co-Chair

Xi Tan

Committee Member

Traci J LeMasters

Committee Member

Charles D Ponte

Committee Member

Usha Sambamoorthi

Abstract

Diabetes has been linked to lower rates of breast cancer (BC) screening, late stage of BC at diagnosis, and high mortality of incident BC. Up to date, no study has investigated the influence of diabetes complications and their severity on this linkage. The aims of the study were to explore the association between severity of diabetes-related complications and persistence with screening mammography in elderly women with diabetes; to check the association of diabetes complications severity with stage of BC at diagnosis in elderly women with incident BC and pre-existing diabetes; and to assess the effect of diabetes-related complications severity on all-cause mortality within 3 years of a BC diagnosis in elderly women with pre-existing diabetes. Aim 1 was conducted using the 5% random sample of linked Surveillance, Epidemiology, and End Results (SEER)-Medicare data among 16,526 elderly women with diabetes who were free of cancer during the years 2002 to 2008. Aim 2 was conducted using SEER-Medicare data among 7,729 elderly women with incident BC and pre-existing diabetes during the years 2004--2011. Aim 3 was conducted among a cohort of women age ≥ 67 years diagnosed with BC in 2007 to 2011 and pre-existing diabetes (N = 4,307). Chi-square tests were used to examine the significant differences in characteristics of the study cohorts by persistence with screening mammogram, stage of BC at diagnosis, and three-year mortality, respectively. Multinomial logistic regressions were used to check the association of diabetes complications severity with persistence with screening mammography and stage of BC at diagnosis. Hazards ratios (HR) of all-cause mortality within 3 years of BC diagnosis was estimated using unadjusted and adjusted Cox proportional hazards models to compare time to death based on diabetes complications severity index (DCSI). Overall, having high severity of diabetes complications was significantly associated with a decrease in the likelihood of receiving breast cancer screening as compared to those without diabetes complications. Among elderly women with diabetes, those with a DCSI ≥ 5 were significantly less likely to use screening mammogram (either persistent use (odds ratio (OR) = 0.08; 95% confidence intervals (CI) = 0.07-0.10) or non- persistent use (OR = 0.32; 95% CI = 0.28-0.37)), as compared to those without diabetes complications. Also, the severity of diabetes complications was no longer an independent predictor of BC stage II or advanced stage (III/IV) at diagnosis. However, women with DCSI =2 were significantly 26% more likely to be diagnosed at stage I (versus stage 0) of BC at diagnosis as compared to those without diabetes complications (OR = 1.26; 95% CI = 1.03-1.53). In addition, severity of diabetes complication was significantly associated with all-cause mortality within three years of BC diagnosis. Women with a DCSI =1, DCSI =2, and DCSI ≥3 had 34% (hazard ratios (HR) = 1.34; 95% CI = 1.02-1.75), 69% (HR = 1.69; 95% CI = 1.39-2.05), and 124% (HR = 2.24; 95% CI = 1.86-2.70) increased risk of death within 3 years after BC diagnosis, as compared to those without diabetes complications. The association between diabetes and worse BC outcomes could be predicted by severity of diabetes complications since this severity has negative consequences on screening mammography, diagnosis, and prognosis.

Download

Share

COinS