Targeted Care Reverses Racial/Ethnic Health Disparities In Colon Cancer Screening, Researchers Find

A More Nuanced Understanding of Communities and Culture Can Help Close the Equity Gap

BOSTON – Colorectal cancer is a leading cancer-related cause of death in the United States, ranking third in incidence and deaths. Effective colorectal cancer screening has been shown to reduce the risk of death from the disease by almost 70 percent, however, numerous disparities in colorectal cancer continue to be discovered and defined. Demographic factors – including race, ethnicity, gender, education level, poverty, health insurance status and patient language – have been linked with a higher or lower risk of screening, while additional cultural factors, such as distrust of medical care, may also play a role. Understanding the intersection of these social determinants of health is critical to improving health equity and survival rates for underserved populations.

In a retrospective review of patients who had a recent primary care visit in a well-resourced safety-net health system serving a diverse population, a team led by researchers at Beth Israel Deaconess Medical Center (BIDMC) aimed to better define the links between patients’ socio-demographic characteristics and colorectal screening. Evaluating self-reported factors including race, ethnicity, preferred language, mental health and substance use status, the team’s more granular assessment provided findings that contradict traditional U.S. healthcare disparities, with Hispanic and Spanish-speaking patients screening at significantly higher rates than white and English-speaking patients. The counterintuitive findings, published in Preventive Medicine, demonstrate that a healthcare system designed to provide equal access to screening for underserved patients can address the disparities commonly seen in cancer screening.

“Investment into a multicultural workforce and outreach efforts to underserved patients may counteract some of the implicit or explicit biases seen on health systems that have led to traditional racial/ethnic disparities,” said senior author Heidi J. Rayala, MD, PhD, urologist at BIDMC. “Our study showed differences in odds of successful screening based on sub-sections of traditionally defined ethnicities – such as breaking down “Hispanic” into more specific cultures and backgrounds – and that suggests that future research should focus on better understanding individual cultures and communities, rather than lumping patients into overly large groups.”

Rayala, who is also an assistant professor of surgery at Harvard Medical School, and colleagues looked at de-identified records of more than 22,000 patients between 50- and 75-years old who saw a primary care physician at Cambridge Health Alliance (CHA) in 2018 to 2019. CHA is an urban, state-funded safety-net health system in the greater Boston area consisting of 13 primary care sites and two hospitals. Intended to provide health care services to low-income and medically under-resourced residents, CHA serves a 63 percent non-white patient population, with 43 percent of patients having limited English proficiency and 51 percent having Medicaid insurance. CHA is well-resourced in interpreter services and mental health services, and has focused on researching means to improve health equity. CHA is affiliated with BIDMC, and the two organizations have a long history of working together to expand access to care in local communities.

Of the 22,000 patients included in the study, 16,065 underwent colorectal screening, an overall screening rate of 73 percent. While that rate is on par with Massachusetts’ overall colorectal screening rates, the state numbers reflect national racial and ethnic disparities, in which people of color do not get screened as often as white people. Massachusetts’ numbers show a screening rate of 56 percent of Hispanic individuals and 68 percent of Black individuals compared to 76 percent for white individuals.

In contrast, at CHA, Hispanics had the highest screening rates of 78 percent. Rayala and colleagues further broke out participants by more granular demographic factors, finding the ethnicity of Portuguese/Azorean received screening at 79 percent. Spanish speakers in general had the highest screening rate of nearly 80 percent.

Among all CHA patients, there were no differences in screening rates between patients with or without obesity, nor were neighborhood income level associated with differences in screening, though patients with commercial insurance were screened at higher rates than those covered by Medicare or Medicaid. Patients with substance use disorder or severe mental health diagnosis both had lower rates of screening. Likewise, white patients had the lowest screening rates at 69 percent. The researchers say these data points are likely related, as the white population at in this safety net population had significantly higher rates of serious mental illness and substance use disorders.

“Non-Hispanic white patients had the lowest colorectal screening rates, highlighting an opportunity to assess barriers to screening within a safety-net population in an ethnic group that is generally reported to have higher rates of screening,” said first author Benjamin G. Allar, a general surgery resident at BIDMC and lead research fellow at the Center for Surgery and Public Health at Brigham and Women's Hospital. “Our white population also had a significantly higher incidence of severe mental health diagnosis and substance use disorder, which resulted in 17 percent lower odds of screening for people with substance use disorder. Future studies could examine the intersectionality of substance use disorder and race and ethnicity in safety-net populations relating to cancer screening.”

Co-authors included Evangelos Messaris of BIDMC; Gezzer Ortega of Brigham and Women’s Hospital; Rumel Mahmood, Lorky N. Libaridian and Ketan Sheth of Cambridge Health Alliance; and Taisha Joseph of Massachusetts General Hospital.

This work did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors. There are no conflicts of interest among all the co-authors.