Understanding the relative impact of ethnicity and sex on patient experience in cancer care.

247 Background: Patient experience is a key indicator of healthcare quality and outcomes. Research suggests that patient ethnicity and sex have a significant impact on patient experience scores. Therefore, our hospital created a process for continuously monitoring patient experience scores stratified by patient demographic data. In 2021, this process caught what appeared to be a significant drop in some areas of patient experience for non-white female inpatients. This drop prompted an investigation that attempted to better understand the relative impact of ethnicity and sex on patient experience at our hospital. Methods: We conducted a cross-sectional study using 13,560 Press Ganey inpatient experience surveys collected from 11,435 unique patients between 2021 and 2023 at Memorial Sloan Kettering Cancer Center. This investigation analyzed the responses to “likelihood of recommending this hospital to others” using a ordinal logistic regression model including patient ethnicity, sex, language, age, expected mortality on admission, admitting department, insurance, marital status, and length of stay observed to expected ratio. Results: Asian (odds ratio [OR] 0.51, 95% confidence interval [CI] (-0.86,-0.47), p=1.98e-11), female (odds ratio [OR] 0.76, 95% confidence interval [CI] (-0.38,-0.15), p=2.76e-06), and patients admitted to emergency departments female (odds ratio [OR] 0.59, 95% confidence interval [CI] (-0.73,-0.30), p=1.65e-06) reported lower scores than white, male, and patients admitted to medicine. Differences associated with marital status, patient insurance, and patient primary language were insignificant. The largest impact on patient experience was the patient’s expected mortality (odds ratio [OR] 0.30, 95% confidence interval [CI] (-2.20,-0.19), p=1.98e-02). Patients with higher expected mortality were much more likely to give low scores. Non-white patients were also more likely to have high expected mortality when compared to their white counterparts (p = 1.90e-24). Conclusions: Our study suggests that while a patient’s expected mortality on admission has the largest impact on patient experience scores. There are also significant differences in patient experience scores based on ethnicity and sex, but they are relatively small in magnitude. The data suggests that focusing on the intersection of high expected mortality and female patients could be an effective method for improving healthcare quality and outcomes. We recommend additional exploration of end of life care and its interactions with patient experience to better understand optimal methods and times for process improvement projects.