Disparities in Pancreatic Adenocarcinoma Care Using the National Cancer Data Base Open Access

Flink, Benjamin James (2015)

Permanent URL: https://etd.library.emory.edu/concern/etds/rf55z849s?locale=en


Introduction: Pancreatic adenocarcinoma is a highly lethal cancer that affects over 1% of the American population. Prior literature provided evidence of care and treatment disparities among pancreatic adenocarcinoma patients with respect to race, gender, age, insurance and socioeconomic status (SES). No national studies in the US have examined the effect of rural-urban residence on pancreatic adenocarcinoma care. Our study examines whether there are social and demographic differences in the receipt of surgery, surgery at a high volume center, thirty-day postoperative readmission/mortality and overall survival.

Methods: Using the National Cancer Data Base from 2003 to 2011, pancreatic adenocarcinoma patients were identified. Further cohorts were identified for patients with potentially resectable (T1-3M0) disease as well as receiving resection. Univariate analyses evaluated the overall cohort to look at rural-urban and racial differences. In the potentially resectable and resected cohorts, univariate and multivariate logistic regression models were used to examine receipt of resection, whether it was performed at a high volume center, and who experienced readmission or died within thirty-days of surgery. We examined overall survival in all three cohorts.

Results: Rural patients presented with earlier stage disease and received similar treatment while black patients presented younger and at later stages while receiving poorer treatment.Among those potentially resectable, older, black and uninsured patients had lower odds of receiving treatment. Among resected patients, Hispanic, rural, uninsured, low SES, and patients within 10 miles had lower odds of resection at a high volume hospital. There were no social or demographic differences leading to thirty-day readmissions. Both older and government insured patients had higher odds of thirty-day mortality. Government insured patients, uninsured, older, low SES, and black patients had a survival disadvantage. High volume hospitals and other' race patients had a survival advantage. Black patients had similar survival when treated the same as whites.

Conclusions: With pancreatic adenocarcinoma, there are several opportunities for improvement to equalize treatment outcomes with respect to race, age, insurance status and SES. While further research will be needed to elicit a causal relationship will require further research, but action can be taken now to improve access to and quality of care.

Table of Contents

Abstract. 4

Acknowledgements. 6

Table of Contents. 8

Literature Review. 14

Introduction. 14

Pancreatic Adenocarcinoma Overview. 16

Epidemiology. 16

Risk Factors. 16

Precursors and cellular progression. 17

Clinical Presentation. 19

Screening and Workup. 19

Treatment. 21

Hospital Volume and Regionalization. 27

Mortality and Survival Benefits. 27

A Related Measure: Surgeon Volume. 33

Support for a High Volume Cutoff. 37

Pancreatic Cancer Disparities. 39

Racial/Ethnic. 39

Incidence. 39

Receipt of Therapy. 42

Mortality/Survival. 48

Age Disparities. 52

Incidence. 52

Receipt of Therapy 52

Mortality/Survival. 57

Socioeconomic Disparities. 59

Incidence. 59

Receipt of Therapy. 59

Mortality/Survival. 60

Gender Disparities. 63

Incidence. 63

Receipt of Therapy. 63

Mortality/Survival. 64

Rural/Urban Disparities. 65

Receipt of Therapy. 66

Mortality/Survival. 66

Insurance Disparities. 67

Receipt of Therapy. 67

Mortality/Survival. 69

Conclusions. 69

Methods. 73

Data Source. 73

Study Design. 74

Patient Selection. 77

Selection for Non-Survival Analyses. 77

Selection for Survival Analyses. 79

Variables. 80

Statistical Analysis. 86

Results. 90

Part 1: All Pancreatic Adenocarcinoma Patients. 90

Introduction. 90

Descriptive Statistics. 90

Univariate Associations with Rurality. 93

Univariate Associations with Race. 98

Univariate Survival Differences in All Pancreatic Adenocarcinoma Patients. 103

Multivariable Cox Regression Models for Overall Survival. 111

Without Treatment Variables. 111

With Treatment Variables. 113

Part 2: Patients Eligible for Surgery. 115

Introduction. 115

Descriptive Statistics. 115

Univariate Differences by Curative Resection Attempt. 118

Multivariable Logistic Model for Receipt of Curative Resection. 122

Univariate Survival Differences in Potentially Resectable Population. 125

Multivariable Cox Regression Models for Overall Survival. 133

Without Treatment Variables. 133

With Treatment Variables. 135

Part 3: Patients Who Received a Curative Resection Attempt. 136

Introduction. 136

Descriptive Statistics. 137

Who Gets Surgery at High Volume Centers. 140

Univariate Differences by Hospital Volume. 140

Multivariable Logistic Regression Model for Who Received Surgery at High Volume Centers. 145

15 Cases/Year. 145

20 Cases/Year. 147

Thirty-Day Postoperative Readmissions. 148

Univariate Differences by Thirty-Day Postoperative Readmissions. 148

Multivariable Logistic Regression Model for Thirty-Day Postoperative Readmissions. 152

Thirty-Day Postoperative Mortality. 154

Univariate Differences by Thirty-Day Postoperative Mortality. 154

Multivariable Logistic Regression Model for Thirty-Day Postoperative Mortality. 159

Overall Survival of Resected Patients. 161

Univariate Survival Differences in Patients that Received Surgery. 161

Multivariable Cox Regression for Overall Survival of Resected Patients. 169

Discussion/Conclusions. 172

Rurality and Racial Presentation/Treatment Differences. 172

Receipt of Curative Resection Attempt. 175

Receipt of Resection at a High Volume Hospital Among Resected Patients. 182

Thirty-Day Postoperative Readmissions. 187

Thirty-Day Postoperative Mortality. 191

Overall Survival. 196

Limitations. 206

Strengths. 209

Conclusions. 210

Bibliography. 214

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
  • English
Research field
Committee Chair / Thesis Advisor
Committee Members
Partnering Agencies
Last modified

Primary PDF

Supplemental Files