This retrospective cohort study used the SEER-Medicare linked database to evaluate geographic patterns, temporal trends, and clinical and facility predictors of one-day discharge following radical prostatectomy. The authors aimed to quantify changes over time and identify patient- and hospital-level factors associated with postoperative day 1 discharge among Medicare beneficiaries.
The study population comprised men diagnosed with prostate cancer between 2008 and 2019 in the SEER-Medicare linkage. Patients were excluded if they had metastatic disease, were enrolled in a health maintenance organization (HMO), or were younger than 66 years at diagnosis. Type of surgery and length of stay were derived from inpatient claims data. The final analytic cohort included 40,728 patients treated at 1,048 hospitals.
Comparisons were made between an early period (2008–2013) and a late period (2014–2019), across U.S. census regions, and by individual- and facility-level characteristics. The authors used generalized estimating equations to estimate adjusted odds ratios (aOR) with 95% confidence intervals (CI), accounting for clustering of patients within hospitals. Sensitivity analyses were performed that attenuated regional differences, according to the report.
Overall, 52.4% of patients in the cohort were discharged on postoperative day 1. The prevalence of one-day discharge increased over time, from 48.8% in the early period (2008–2013) to 56.8% in the late period (2014–2019). The analysis included care provided at 1,048 hospitals captured in SEER-Medicare claims.
Key independent predictors of higher odds of one-day discharge included robotic surgery and treatment at certain high-level centers. Specifically, robotic approach was associated with markedly higher odds of day 1 discharge (aOR 2.88, 95% CI 2.54–3.27). Treatment at a National Cancer Institute-designated comprehensive cancer center was also associated with increased odds (aOR 1.88, 95% CI 1.28–2.76).
Several patient-level factors were associated with lower odds of one-day discharge. Black patients had lower adjusted odds compared with White patients (aOR 0.76, 95% CI 0.70–0.82). Unmarried patients were less likely than married patients to be discharged on day 1 (aOR 0.79, 95% CI 0.73–0.85). Patients residing in the lowest neighborhood socioeconomic quintile had lower odds than those in the highest quintile (aOR 0.86, 95% CI 0.79–0.94).
The late versus early treatment era was associated with higher odds of one-day discharge (aOR 1.31, 95% CI 1.20–1.43), indicating an increasing trend over time.
The authors reported comparisons by census region and noted regional differences in one-day discharge rates. However, those regional differences were attenuated in sensitivity analyses, according to the published abstract. Specific regional effect estimates and the sensitivity analysis methods beyond that statement were not detailed in the abstract.
One-day discharge after radical prostatectomy increased across the study period and is now the predominant practice among the Medicare population in this cohort. The observed increase was driven primarily by uptake of the robotic surgical approach and by later treatment era. Persistent racial and socioeconomic differences in discharge patterns were identified, with Black race, unmarried status, and lower neighborhood socioeconomic status associated with reduced odds of one-day discharge.
The authors highlight the need for further research to determine whether early discharge is equally safe and beneficial across patient subgroups and to understand drivers of the observed disparities.
As a claims-based retrospective analysis using SEER-Medicare, clinical detail is limited to that available in linked registry and claims fields. The abstract notes exclusion criteria (metastatic disease, HMO coverage, age <66) and that generalized estimating equations accounted for clustering by hospital. Detailed limitations beyond those described in the abstract (for example, unmeasured confounding or clinical outcome measures tied to discharge timing) were not reported in the abstract text.
Ethical oversight: the study received approval or exemption details consistent with institutional review; the authors report adherence to relevant ethical guidelines and that informed consent was waived because the analysis used de-identified secondary data under an NCI data use agreement.
The underlying SEER-Medicare data are third-party data obtained from the U.S. National Cancer Institute and are not publicly shareable by the authors because they contain protected health information. The authors note that other researchers may request access to the same SEER-Medicare data through the NCI Healthcare Delivery Research Program, subject to approval and a data use agreement. The authors indicated that SAS code for cohort construction and analysis is available from the corresponding author on request.