Document Type

Capstone Experience

Graduation Date

12-2018

Degree Name

Master of Public Health

Department

Epidemiology

First Committee Member

Evi Farazi, PhD

Second Committee Member

Matt Lunning, DO

Third Committee Member

Liz Lyden, MS

Abstract

Background: Clinical trial participation, especially among cancer clinical trials in adult populations, continues to be low despite the large number of clinical trials available across the U.S. Estimates for clinical trial participation are as low as 5% in some adult cancer populations, however in some lymphoma populations this figure may be as high as 13.9%. This figure can be confusing, given that as much as 70% of Americans are estimated to be willing to participate in a clinical trial. Previous research shows that as much as 95% of surveyed respondents who said they had previously participated in a clinical trial stated they would consider future participation in another clinical trial. Goals and Objectives: This study contains four primary objectives. The first is to provide descriptive statistics for the Nebraska Lymphoma Study Group (NLSG) study population at the University of Nebraska Medical Center (UNMC). The second is to determine associations between demographic and clinical variables and two separate outcome variables: participation in the NLSG study and participate in a clinical trial as front-line therapy. The third objective is to determine the association between participation in the NLSG study and participation in a clinical trial as front-line therapy. The fourth objective is to develop a logistic regression model for participation in a clinical trial as front-line therapy. We hypothesize that patients who agree to participate in the NLSG study will be more likely to select a clinical trial as front-line therapy. Methods: Demographic and clinical information were provided from the NLSG research team at UNMC. Tests for association between categorical variables and two separate outcome variables, participation in the NLSG study and participation in a clinical trial as front-line therapy, were examined using Chi-square analyses. An odds ratio was calculated for participation in the NLSG study and participation in a clinical trial as front-line therapy. Logistic regression analysis among bivariate predictor variables was modeled for the outcome of participation in a clinical trial as front-line therapy. Results: The sample population consisted of N=2,343 patients. Ethnicity as a multivariate categorical variable and ethnicity as a bivariate categorical variable (White, non-Hispanic and all other ethnicities) were significantly associated with participation in the NLSG study (p=0.007 and p<0.001, respectively). N=125 patients selected a clinical trial as front-line therapy. Age at diagnosis was found to be negatively associated with participation in a clinical trial as front-line therapy (p=0.006). Those who participated in a clinical trial as front-line therapy had 3.48 times the odds of previously participating in the NLSG study, however this was not statistically significant (95% CI=0.47, 25.35). A logistic regression model for selection of a clinical trial as front-line therapy was developed which included two significant covariates, age at diagnosis and diagnosis type. The logistic regression model showed that those who were younger (below the median age at diagnosis) and those with Hodgkin lymphoma diagnosis were at greater odds of selecting a clinical trial as front-line therapy. Conclusions: We found that ethnicity, in particular those who were White, non-Hispanic were more likely to have been enrolled in the NLSG study. Additionally, we found that those who were younger (below the median age at diagnosis of 57 years), and those with a diagnosis of some subtype of Hodgkin Lymphoma, were more likely to have selected a clinical trial as front-line therapy. We found that those who chose a clinical trial as front-line therapy were at 3.48 times the odds of having previously agreed to participate in the NLSG study, however this association was not statistically significant. We recommend that further research be done to investigate the factors associated with registry and clinical trial participation among non-White ethnic groups as well as among older populations. We also recommend that future similar research be done on lymphoma populations in other geographic areas in the U.S. as well as in other disease states or health conditions to examine predictive factors for participation in clinical trials. Understanding factors that positively influence clinical trial participation may improve clinical trial participation rates.

Key Words & Phrases: Lymphoma, Predictive Factors, Registry, Clinical Trial, Logistic Regression, Nebraska Lymphoma Study Group, Clinical Trial Participation

Download

Included in

Public Health Commons

Share

COinS