The Association between Diabetes and Social Determinants of Health, and Cancer in the United States: Results from the Behavioral Risk Factor Surveillance System (BRFSS)

Abstract

Several studies have found an association between diabetes and cancer, wherein the incidence of cancer was higher among diabetic patients. However, these studies did not account for the effect of social determinants of health (SDOH), despite that they were shown to play a significant role in the increase in the prevalence of non-communicable diseases. Accordingly, the literature is still lacking publications that focus on triangulating the relationship between SDOH, diabetes, and cancer. Therefore, in this study, we aimed to examine the association between SDOH, and diabetes as predictors, and cancer in general and specific types of cancer as outcomes of interest, while controlling for age, sex, body-mass index (BMI), smoking, and alcohol consumption.

We conducted a secondary data analysis on a dataset adopted from the Behavioral Risk Factor Surveillance System (BRFSS), a cross-sectional study performed by the Centers for Disease Control and Prevention (CDC) in the year 2020. BRFSS encompasses a nationally representative sample of randomly selected participants from all the states of the United States of America (USA) using a multistage cluster design. In our study, we only considered the 22 states that had questionnaires in both cancer and diabetes amounting to a total sample size of 155,377 participants. Outcome variables include cancer, and type of cancer. Exposure variables include diabetes and SDOH: homeownership, marital status, healthcare coverage, employment status, urban/rural county, education level, income level, and race. Our analyses were adjusted for age, sex, BMI, smoking status, and alcohol consumptions. The survey required weighted analyses using sampling and cluster weights. Crude and adjusted analyses were carried out using weighted simple and multiple logistic regression to generate weighted crude and adjusted odds ratios (OR) and their respective 95% confidence intervals.

Participants with diabetes had higher odds of having any type of cancer compared to participants with no diabetes (OR 1.10; 95% CI: 1.02-1.19). They also showed higher odds of having endometrial cancer (OR 1.61; 95% CI: 1.06-2.43), head and neck cancer (OR 2.67; 95% CI: 1.31-5.44), esophageal cancer (OR 4.39; 95% CI: 1.24-15.49), pancreatic cancer (OR 3.03; 95% CI: 1.03-8.91), and bone cancer (OR 1.91; 95% CI: 1.01-3.61). By contrast, diabetics showed lower odds of having non-melanoma skin cancer compared to non-diabetics (OR 0.83; 95% CI: 0.71-0.97). A further sub-analysis showed that this could be due to insulin use. Among diabetics, those using insulin had significantly lower odds of having non-melanoma skin cancer compared to those not using insulin (OR 0.51; 95% CI: 0.33-0.80). Furthermore, several SDOH were found to be associated with having cancer and having specific types of cancer. Home ownership was associated with having breast, cervical, endometrial, oral, colon, stomach, prostate, bone, and brain cancer. Marital status was associated with having any type of cancer, endometrial, thyroid, pancreatic, testicular, and bladder cancer. Healthcare coverage was associated with having any type of cancer, colon, pancreatic, and stomach cancer. Employment Status was associated with having any type of cancer, cervical, thyroid, liver, pancreatic, Hodgkin’s lymphoma, prostate, lung, and brain cancer. People living in urban counties showed lower odds of having liver cancer compared to people living in rural counties. Education level was associated with having any type of cancer, endometrial, head/neck, pharyngeal, colon, leukemia, non-melanoma skin, lung, and renal cancer. Income level was associated with cervical, endometrial, esophageal, pancreatic, rectal, stomach, and non-melanoma skin cancer. Race was associated with having any type of cancer, breast, ovarian, head/neck, colon, stomach, leukemia, prostate, skin, and renal cancer. The other variables also showed association with cancer and several types of cancer. Age was associated with having any type of cancer, breast, cervical, ovarian, thyroid, stomach, Hodgkin’s lymphoma, leukemia, prostate, testicular, non-melanoma skin cancer, bladder, and brain cancer. Sex was associated with having any type of cancer, oral, pharyngeal, thyroid, colon, esophageal, liver, Hodgkin’s lymphoma, non-Hodgkin’s lymphoma, melanoma, bladder, renal, and brain cancer. BMI was associated with having any type of cancer, head/neck, pharyngeal, esophageal, pancreatic, stomach, Hodgkin’s lymphoma, leukemia, lung, renal, and bone cancer. Smoking was associated with having any type of cancer, breast, cervical, pharyngeal, thyroid, pancreatic, prostate, non-melanoma skin cancer, lung, bladder, and brain cancer. Finally, heavy alcohol consumers showed greater odds of having oral cancer compared to no or light alcohol consumers.

Understanding more the relationship between diabetes, SDOH, and cancer is of importance to a better setting of priorities for screening and early detection of cancer. Furthermore, the identification of the specific types of cancer that are more prevalent in patients with diabetes is essential to detect high-risk patients for cancer and to determine the right preventative measures for this pool of patients. Our findings offer a novel social dimension for the study of the association between cancer and diabetes. This research offers implications for addressing social inequities that affect people’s health.