Stopping the revolving door: An exploratory analysis of health care super-utilization in gynecologic oncology

The objective of this study was to determine the prevalence of and risk factors for health care super-utilization among gynecologic oncology patients at a single academic hospital. A retrospective cohort study of gynecologic oncology patients with an index unplanned encounter between January and December 2018 was performed. Super-utilizers were defined as patients with 3 or more unplanned hospital encounters during a 12-month period starting at the time of the index unplanned encounter. We identified 553 patients with gynecologic cancer. Of those, 37(7%) met inclusion criteria for super-utilizers accounting for 193/310(62%) of unplanned visits. The median number of unplanned visits was 4 (range 3-24). The most common cancers were uterine (N = 15 (41%)) and ovarian (N = 11 (30%)). Nineteen (51%) super-utilizers had advanced stage disease. Phases of oncologic care at index unplanned encounter included primary diagnosis (N = 24 (65%)), recurrence (N = 10 (27%)), and surveillance (N = 2 (5%)). Twelve super-utilizers (32%) had new diagnoses of cancer without prior therapy, 19(51%) had prior chemotherapy, 17(46%) had prior surgery, and 10(27%) had prior radiation therapy at the time of initial encounter. Fifteen super-utilizers (41%) were in the last year of life. The most common reasons for unplanned encounters were pain (66%) and gastrointestinal symptoms (61%). Multivariable analysis adjusting for key variables demonstrated that Medicaid insurance, ASA classification, and disease status are risk factors for health care super-utilization. The majority of health care utilization occurred during the first year of diagnosis. This exploratory analysis suggests an opportunity to decrease health care utilization, particularly during upfront treatment.