Abstract
Prostate Cancer (PCa) is the most commonly diagnosed non-skin cancer and the second leading cause of cancer death, after lung cancer. In 2024, there were roughly 299,010 survivors and approximately 35,250 men who died from PCa in the United States (CDC, 2024). While PCa mortality rates have declined in recent years, minority PCa survivors, particularly Black men, continue to experience a disproportionate burden from the disease. They are at increased risk of developing the disease, are more likely to be diagnosed with later-stage tumors, and are more likely to die from PCa at higher rates than men of other racial and ethnic groups (Giaquinto et al., 2022). However, race and ethnicity can only partly explain disparities in PCa healthcare. Other factors, including age at diagnosis, choice of healthcare plan, as well as poverty and education levels of the patient can have significant associations. This monograph dissertation uses secondary data from the 2011-2019 Surveillance, Epidemiology, and End Results (SEER) dataset linked to the Medicare Consumer Assessment of Healthcare Providers and Systems (CAHPS) survey to examine patient care experiences in four patient subgroups with PCa: non-Hispanic (NH) White patients, NH Black patients, Hispanic patients, and patients with Other/Unknown-race/ethnicity. Hypothesis testing and pooled ordinary least squares (OLS) regression were used to explore the patient care experiences for Medicare Fee-For-Service (FFS) enrollees and Medicare Advantage (MA) beneficiaries. The study identified males aged ≥35 years who completed a CAHPS survey within 6–60 months post-PCa diagnosis. It first aimed to determine whether race/ethnicity were associated with the patient experience . Next, it aimed to examine whether the patient experience differed by plan type (i.e., FFS and MA). Finally, this study explored whether the association of race/ethnicity with the patient care experience differed by plan type.
The four study hypotheses are as follows: H1: Minority PCa patients are likely to report lower patient experience scores than NH White PCa patients. H2: PCa patients associated with the FFS plan type will report significantly higher Getting Needed Care and Getting Care Quickly scores than PCa patients associated with the MA plan type. H3: Due to potential fragmentation of care, PCa patients associated with the FFS plan type will report significantly lower Provider Communication and Care Coordination scores than PCa patients associated with the MA plan type. H4: NH White PCa patients enrolled in the FFS plan type will likely report higher patient care experience scores than minority PCa patients enrolled in the FFS plan type.
A total of 19,492 participants met the inclusion criteria for this research study, with 68% being NH White males, 10.4% being NH Black males, 9.7% being Hispanic males, and 11.5% being Other/Unknown males. The dependent variables for this study are a series of patient care experiences, measured from CAHPS questions with responses on a 4-item Likert scale (always, usually, sometimes, never), and consists of the following five composite measures: 1) getting care quickly, 2) getting needed care, 3) getting needed prescription drugs, 4) doctor communication and 5) care coordination.
By and large, the initial hypotheses were confirmed that mean values for the race/ethnicity variable are not equal. Hispanic and Other/Unknown race are associated with lower scores on getting care quickly outcome. Also, for that outcome, the MA plan type was associated with higher scores than the FFS plan type. For the getting needed care model, the coefficients for all three race variables were negative and statistically significant, confirming the first hypothesis. The first two hypotheses were confirmed with the getting needed prescription drugs model, where three races/ethnicities were statistically significant and negative. The plan type was also significant in this model. The model also confirms the first two hypotheses, with Hispanic and Other/Unknown race coefficients being negative and statistically significant. Finally, the care coordination model confirmed the first hypothesis, showing significant and negative coefficients for two of the race variables. The second hypothesis was also confirmed in that there was a significant and negative interaction of NH Black with the MA plan type.
For the getting care quickly outcome, Model 2 found that age and college education categories were statistically significant and positive predictors. For the getting needed care outcome, the categories of age and low-income subsidy were also significant and positive, confirming the fourth hypothesis. For the doctor communication outcome, the fourth hypothesis was confirmed because the low-income subsidy variable was negative and statistically significant. For the care coordination outcome, there was a negative and statistically significant association with education. These findings support the fourth hypothesis regarding covariates.
This study suggests that the overall role of plan type associated with the patient care experience can be instrumental in fostering trust and better interaction between patient and provider. Future studies should explore the significance different routes to care might play on patient experience for populations with diverse socioeconomic backgrounds at other cancer sites.