Background and Aim: Cancer imposes a significant economic burden on patients and their families, which can cause or exacerbate food insecurity, with an associated impact on cancer disease management. Additionally, several studies indicate that, among cancer patients, high diet quality is associated with a lower risk of cancer mortality. There is, however, limited research investigating food insecurity and diet quality as well as characteristics associated with their onset among cancer patients. The purpose of this study, therefore, was to determine the prevalence and factors associated with food insecurity and diet quality among cancer patients, as well as the relationship between food insecurity and diet quality. Identifying predictors of increased food insecurity and low diet quality risk among this vulnerable population allows care givers to target those in need of food resources to help patients better manage their disease.
Methods: Recruitment of 242 cancer patients who consented for this study took place in central Ohio cancer treatment facilities. Data were gathered on sociodemographic, disease, and treatment related characteristics. Food insecurity was calculated using the 10-item Food Security Survey Module. A Chi-squared test was used to determine sociodemographic and disease related characteristics that were significantly associated with food insecurity. A logistic regression model was run and forwards selection was used to determine the final model. Diet quality was scored using Healthy Eating Index (HEI) 2010. Independent sample t-tests and one-way analysis of variance (ANOVA) were used to compare mean HEI scores across characteristics. A multiple linear regression analysis was used to predict the HEI score of patients with significant socio-demographic and disease related characteristics. One-way ANOVA was used to determine if the HEI scores and HEI subcategory scores were significantly different across groups with different food security levels.
Results: The prevalence of food insecurity in this sample was 10.3% which is below the national average. Characteristics significantly associated with food insecurity included race (p = 0.04), marital status (p = 0.02), education (p = 0.006), employment (p = 0.02), monthly income (p = 0.001), insurance type (p < 0.001), borrowing money to pay medical bills (p < 0.001), paying bills late due to medical expenses (p < 0.001), use of federal (p < 0.001) and private (p < 0.001) food assistance, and use of alcohol (p = 0.002). The regression model was statistically significant Χ² (5, N=231) = 41.452 p<0.005. A person who is divorced or widowed (B = -1.28, p=0.02), pays bills late (B = -2.26, p < 0.0005), or uses public food assistance (B = -2.50, p < 0.0005) is less likely to be food secure than individuals who do not exhibit these characteristics. Interestingly, those who consume alcohol are more likely to be food secure than individuals who do not (B = 1.89, p = 0.01). Individuals without a high school degree or only a high school degree or GED had significantly lower HEI scores (ß = -3.94, p = 0.04; ß = -8.31, p < 0.005, respectively) compared to those with college degrees. Additionally, those who were homemakers had significantly higher HEI scores (ß = 8.84, p = 0.003) compared to those who work over 40 hours per week. Current smokers had significantly lower HEI scores (ß = -4.59, p = 0.05). Also, individuals with endometrial or uterine cancer had significantly higher HEI scores (ß = 8.65, p = 0.01) than individuals with breast cancer. There was no significant difference in total HEI score by food security status [F (1, 240) = 0.513, p = 0.474] or in HEI subcategories by food security status, except for the “fatty acids” category (p = 0.022).
Conclusion: It is confirmed by our findings that many cancer sufferers continue to face challenges that inhibit their access to healthy food. Oncologists and healthcare professionals can use the characteristics identified here to identify cancer patients at risk for food insecurity and direct them to supplemental food resources.