Impact of sociodemographic factors on outcomes in patients with peritoneal malignancies following cytoreduction and chemoperfusion

Abstract Background and Objectives Sociodemographic factors have been shown to impact surgical outcomes. However, the effects of these factors on patients undergoing cytoreductive surgery (CRS) and heated intraperitoneal chemotherapy (HIPEC) are not well known. This study aims to evaluate the impact of sociodemographic factors on patients undergoing CRS/HIPEC. Methods Adult patients at a tertiary center who underwent CRS/HIPEC were evaluated. Perioperative variables were collected and analyzed. A national database was also used to evaluate patients undergoing CRS/HIPEC. Results There were 90 patients who underwent CRS/HIPEC (32% non‐White). There was no statistically significant difference in postoperative complications, length of stay, or discharge disposition based upon race (white vs. non‐White patients), socioeconomic status (SES), or insurance type. Nationally, we found that Black and Hispanic patients were less likely to undergo CRS/HIPEC than Non‐Hispanic white patients (Black: odds ratio [OR]: 0.60, [confidence interval {CI}: 0.39–0.94]; Hispanic: OR: 0.52, [CI: 0.28–0.98]). However, there were no significant differences in postoperative complications based upon race/ethnicity. Conclusion Sociodemographic factors including race, SES, and insurance status did not impact postoperative outcomes in patients undergoing CRS/HIPEC at our single institution. On a national level, Black and Hispanic patients underwent CRS/HIPEC at lower rates compared to white patients.

Using the United States Census Bureau data from the American Community Survey (ACS) 5-year estimates (2013-2017), median household income levels for each patient were derived from their zip code. As a measure of a patient's SES, Area Deprivation Index (ADI) values were also derived from each patient's zip code. The ADI is a novel tool based on the United States Census and ACS Data. This is a standardized, multidimensional evaluation of an area's socioeconomic condition based on 17 variables, including housing, income, and education. 18 High ADI values indicate high levels of socioeconomic deprivation, where an ADI of 100 indicates the most disadvantaged.
For this study, ADI values > 25 were considered socioeconomically disadvantaged.
Analyses were also performed using the US Agency for Healthcare Research and Quality Healthcare Cost and Utilization Project's (HCUP) National Inpatient Sample (NIS). HCUP NIS is the largest allpayer (Medicare, Medicaid, Private, and Uninsured) inpatient care database in the United States, is designed to be representative of all community hospitals, and is intended to be used for national estimates. Community hospitals are defined as short-term, nonfederal, and nonrehabilitation hospitals. The NIS is drawn from a sampling frame that contains hospitals that have more than 95% of all discharges from statewide data organizations that contribute to HCUP.

| Outcomes
Primary outcomes of interest focused on the length of stay, intensive care unit (ICU) days, disposition, and readmissions. Secondary outcomes of interest included postoperative complications such as delayed gastric emptying, infections, organ insufficiency, fistulas, and postoperative leak. We also evaluated patients who were underinsured and insured (only private insurance) and compared patient demographics and postoperative outcomes. The underinsured group included those patients with Medicare, Medicaid, and uninsured.

| Statistical analysis
Our unadjusted comparison of two or more proportions were performed using a chi-squared test and continuous variables were compared using a t test or Wilcoxon rank sum test as appropriate.
Statistical significance was established at α = 0.05. In the HIPEC cohort, baseline characteristics of hospitalizations were presented using survey-adjusted counts and means to provide national estimates with standard errors. Logistic regression analysis was also used to compare the odds of undergoing HIPEC by race/ethnicity. All regression analyses controlled for the following variables: age, sex, and Elixhauser mortality score. All analyses were performed using STATA 14 software. This study was evaluated and approved by the Loyola University Chicago IRB (#212524).

| RESULTS
A total of 90 patients underwent CRS/HIPEC for all peritoneal malignancies during the study period. Patient demographics are presented in Table 1. Of the 90 patients, 68% were classified as Non-Hispanic white, while 32% were classified as non-White, which included Non-Hispanic Black, Hispanic, Non-Hispanic Asian, and Non-Hispanic other. In the group of Non-Hispanic white patients, 98% spoke English as their primary language, compared to 62% of patients in the non-White group. An interpreter was used in 28% of patient encounters in the minority group. There were no significant differences between the two groups in age (p = 0.14), sex (p = 0.64), BMI (p = 1.00), ASA class (p = 0.71) or insurance type (p = 0.80).
Postoperative complications were examined based upon race, ADI scores, and insurance status. There were no significant differences in postoperative complications between Non-Hispanic white and non-White groups (Table 2A), low ADI patients compared to the high ADI patients (Table 2B), or insured versus underinsured (Table 2C). Additionally, the mean ICU days, length of stay (LOS), 30day readmission rates, and disposition were similar between racial, ADI, and insurance groups (Table 3A-C). Furthermore, there were no statistically significant differences in postoperative complications when language proficiency was evaluated. When complications were combined into one variable to increase the number, there was again no difference when stratified by race/ethnicity, ADI, or insurance.
The NIS database was used to identify 900 patients nationally who underwent CRS/HIPEC in 2014 ( | 1289