# Practice and impact of selection and centralization for oral cancer services: a scenario analysis

## TL;DR

This study evaluates how organizing oral cancer treatment based on cancer stage and hospital specialization can improve patient outcomes.

## Contribution

The study proposes and evaluates a scenario-based system for centralizing oral cancer treatment based on cancer stage and hospital type.

## Key findings

- No significant survival differences were found for early-stage oral cancer patients across hospital types.
- Significant survival differences were observed for stages III and IV oral cancer patients.
- Survival rates improved for advanced-stage oral cancer after implementing the proposed scenarios.

## Abstract

Centralization of cancer treatment may potentially improve cancer outcomes, enhance quality of life, and reduce complications; however, few scientific studies have demonstrated its effectiveness. We aimed to define a cancer care delivery system with the aid of selection/ centralization scenarios to enhance outcomes for oral cancer, which is recognized as a rare malignancy in Japan. The purpose of this study was to review our scenarios and discuss their effectiveness.

We confirmed and implemented the following pragmatic scenarios: (A) early-stage cancer can be treated at any hospital, (B) Stage III cases should be treated at cancer care hospitals, and (C) Stage IV and recurrent cases should be treated at the regional cancer care center. Participants included oral and maxillofacial surgeons in one prefectural cancer care hospital, eight regional cancer care hospitals (out of 11 base hospitals), and 10 other noncancer treatment hospitals. Medical records between 2017 and 2021 were examined. In addition, treatment outcomes from eight hospitals that had provided oral cancer treatment between 2000 and 2011 were examined to compare pre- and post-scenario treatment outcomes. Survival rates were analyzed via Kaplan-Meier estimation of 5-year overall survival (OS) and assessed for significant differences via the log-rank test. A P-value of less than 0.05 was considered to indicate a significant difference. All analyses were performed using JMP ver.13.2 (SAS Institute Inc., Cary, NC, USA).

Four hundred sixty-nine patients with oral malignancies were seen at participating hospitals. A total of 158 (33.7%) patients were subsequently referred to other hospitals for cancer treatment. A comparison of survival rates stratified by hospital category showed no significant differences between hospital categories for early-stage cancers. In contrast, there were significant differences between groups in stages III and IV. A pre- and post-scenario comparison revealed that survival rates had improved, especially for those with advanced cancer (stage IVB).

The results of this study suggest that the scenarios we developed are useful and may improve the outcomes of oral cancer.