An examination of factors associated with disparities in clinical trial eligibility guided by the Socioecological Model

Abstract

Background: Patients from diverse demographic groups remain underrepresented in cancer clinical trials (CCTs) because of various factors, including disparities in eligibility. In the current study, the authors investigated associations between demographic characteristics and factors associated with CCT eligibility and enrollment using real-world data.

Methods: A cross-sectional analysis was conducted among 113,030 adult patients with cancer who were treated at Moffitt Cancer Center between 2011 and 2021. Eligibility criteria from 100 CCTs were manually abstracted and conceptually mapped to variables in the Moffitt Cancer Analytics Platform. Other factors potentially affecting patient eligibility or enrollment in CCTs were identified following the Socioecological Model and were compared across patient groups by age, sex, race, and ethnicity.

Results: The most frequent eligibility criteria in the abstracted CCTs included laboratory results associated with liver function (89%) and bone marrow function (88%). Thirty-nine percent of the protocols required genetic testing. Black/African American patients experienced a higher prevalence of diabetes (Δ = 3.2%), human immunodeficiency virus infection (Δ = 1.4%), and abnormal laboratory results compared with White patients. Hispanic patients experienced a higher prevalence of human immunodeficiency virus infection (Δ = 0.2%) and a history of organ transplantation (Δ = 0.2%) compared with non-Hispanic patients. Compared with younger patients, older adults had a higher prevalence of nononcologic conditions. Female, Black, and/or Hispanic patients were more likely to experience barriers to care, including childcare and transportation, and to live in more socially deprived areas compared with male, White, or non-Hispanic patients.

Conclusions: Restrictive CCT eligibility criteria and social determinants of health may pose structural barriers that disproportionately affect patient diversity. Investigators should leverage real-world data to design appropriate trial eligibility criteria.

Keywords: cancer treatment trials; clinical trials; disparities; eligibility criteria; oncology; social determinants of health (SDOH); socioecological model.

FIGURE 1

The Socioecological Model was used as the guiding framework to identify data elements for the analysis of clinical characteristics and social determinants of health that may affect patient diversity in cancer clinical trials. This figure depicts the data elements selected for the current analysis within the context of the various levels of the model. ICD indicates the International Classification of Diseases.

FIGURE 2

Distribution of eligibility criteria in a random sample of 100 cancer clinical trials conducted at Moffitt Cancer Center between 2020 and 2021. This figure illustrates the frequency distribution of the abstracted clinical trial inclusion and exclusion criteria. These eligibility criteria were categorized into four groups, including: laboratory test results, health conditions, bacterial/viral infection, and overall health status. ECOG indicates Eastern Cooperative Oncology Group; HIV, human immunodeficiency virus.

FIGURE 3

Differences in health conditions by demographic factors among patients who presented to Moffitt Cancer Center with cancer during 2011–2021. The figure illustrates the prevalence of health conditions, including oncologic and other comorbid conditions, by race and ethnicity. The color-coded cells indicate statistically significant differences, and the color shading represents the magnitude of the observed difference. *Post-hoc comparisons were conducted between White patients and each of the racial minority groups, and p values were adjusted using the Bonferroni method. **Includes only patients with breast cancer. AA indicates African American.

FIGURE 4

Social determinants of health in association with demographic factors among patients who presented to Moffitt Cancer Center with cancer between 2011 and 2021, by race and ethnicity. This figure illustrates the distribution of social determinants of health (payment method, education level, employment status, language preference, difficulties when seeking cancer care, and the social deprivation index) by race and ethnicity. The color-coded cells indicate statistically significant differences, and the color shading represents the magnitude of the observed difference. *Post-hoc comparisons were conducted between White patients and each of the racial minority groups, and p values were adjusted using the Bonferroni method. AA indicates African American, VA, Veterans Affairs.