This was an IRB-approved, single-institutional prospective review of patients at OSU-CCC. Only inpatient data was used in this study.
The primary objectives of this study were to evaluate the prevalence of food insecurity among patients hospitalized at OSU-CCC utilizing the HVS and to validate it in our adult patients with cancer population with the standardized 18-item HFSS. The secondary objective was to assess disparities in food access based on race, zip code and insurance status.
The two HVS questions are: " Within the past 12 months we were worried whether our food would run out before we got money to buy more,” and “Within the past 12 months the food we bought just didn’t last and we didn’t have money to get more.” The possible answers were: “often true,” “sometimes true,” or “never true.”
The 18-item HFSS is divided into different subsections: 10 questions that elicit the experiences of adults and the household in general and 8 questions that elicit experiences of providing food to any children in the household. Not all questions are answered if the patient does not meet criteria for food security.
Patients included were 18 years of age or older who were hospitalized at OSU-CCC. Patients were enrolled from September 2019 through February 2020. Patients were excluded if they were non-English speaking.
Data Collection Process
Every patient that is hospitalized at OSU-CCC is evaluated by a social worker on admission as part of standard of care. Standardized basic assessments are performed and if a patient is found to have a social need, further investigations are performed by the social worker. Through collaboration with the Department of Social Work, patients were asked if they were interested in participating in a food security study. The social workers asked patients for their permission to be contacted by a member of the study team. Once that permission was obtained, the social worker contacted the primary investigator. Then members of the research team discussed the study with the patient and if they agreed to participate, written consent was obtained. The patients were then asked the 2-question HVS screening tool along with the 18-item HFSS. Further data were then obtained via electronic medical record queries: race, zip code and insurance status. Given that patients came from several different geographical locations, zip codes were categorized as following: within the city of Columbus, within the greater Columbus area, within Ohio and out of state.
Statistical Analysis
The study was originally powered to achieve a 5% margin of error for a 95% confidence interval (CI) when estimating the prevalence of food insecurity where the prevalence of food insecurity was assumed to be 17% based on Franklin County Ohio data.21 To achieve this goal it was estimated that a minimum of 217 patients were needed. However, during the middle of data collection, the novel coronavirus, SARS-CoV-2 pandemic developed. This led to the temporary halting of clinical trials and all non-essential patient encounters. Therefore, the study was forced to halt recruitment after 112 patients had been enrolled into the study.
An interim analysis was performed on the data that was collected prior to recruitment being halted. Characteristics of the population were summarized with frequencies and percentages or means and standard deviations, as appropriate. For food insecurity measures which grouped patients into high/moderate/low/very low food security, patients with low or very low food security were considered to be food insecure. The prevalence of food insecurity as measured by the HVS as well as the standard 18-item HFSS, was estimated and 95% Wilson score confidence intervals (CI) were calculated. Further, the agreement between the prevalence of food insecurity as measured by the two different methods was compared through the use of McNemar’s test. Disparities in food access based on race/ethnicity, insurance status and zip code were assessed with Fisher exact tests. SAS 9.4 was used for all analyses and a p-value < 0.05 was considered to be statistically significant.