Association of guideline-concordant care with survival, health care utilization, and costs among older women with advanced epithelial ovarian cancer.

Objectives: To examine guideline-concordant care (GCC) for ovarian cancer, identify its predictors, and evaluate the associations between GCC and survival, health care expenditures, and utilization.

Study Design: A retrospective cohort study using Surveillance, Epidemiology, and End Results-Medicare data.

Methods: Women aged 66 to 90 years who received a diagnosis of stage II or higher epithelial ovarian cancer during 2011-2015 were included (N = 3237). The National Comprehensive Cancer Network clinical practice guidelines were used to identify GCC. Logistic regression was conducted to identify predictors of GCC, a Cox proportional hazards model was used to examine mortality, and generalized linear models were used to examine mean monthly Medicare expenditures and health care utilization.

Results: Approximately 57% of women received GCC and 11.6% of women did not receive any cancer-specific treatment. Women who were relatively older (adjusted odds ratio [AOR], 0.272; 95% CI, 0.210-0.351), had Census tract income of $50,000 or less (AOR, 0.709; 95% CI, 0.551-0.913), had a psychiatric condition (AOR, 0.655; 95% CI, 0.464-0.923), and had adenocarcinoma histology (AOR, 0.564; 95% CI, 0.441-0.721) were significantly less likely to receive GCC. Race/ethnicity was not found to be a significant predictor of GCC. Women who received surgery only or chemotherapy only had a significant higher hazard of all-cause mortality and ovarian cancer-specific mortality compared with those who received GCC (surgery only: adjusted HR [AHR], 2.307; chemotherapy only: AHR, 1.802). Receiving chemotherapy only was associated with 45% (P < .0001) higher mean monthly expenditures compared with those who received GCC.

Conclusions: Non-GCC was associated with worsened survival, higher health care utilization, and increased expenditures. It is important to highlight that women who received GCC were associated with better survival likely due to favorable prognostic clinical factors.