目標：COVID-19疫情對醫療體系造成衝擊，對癌症患者的診斷及預後造成影響。本研究旨在評估三級警戒對於台灣癌症新診斷人數的影響，並分析對不同癌症次族群是否造成不同程度的影響。方法：自癌症登記檔收集2018至2021年各縣市每週癌症新發人數。採用斷點迴歸設計，估計三級警戒造成的新診斷癌症人數的影響。以中斷時間序列分析評估三級警戒對不同臨床期別、年齡以及性別的立即影響與干預後趨勢。結果：三級警戒期間，平均每個縣市每週減少66.1(p<0.001)個新診斷癌症患者。干預後的立即效果，以臨床期別第0期減少75.1%最多，第4期減少45.0%最少;女性下降69.2%相較男性52.1%減少的比例更大。干預後各次族群均出現回復趨勢，其中第0期每周每十萬人回復0.020(p<0.001)最慢;第4期0.121(p<0.001)最快。結論：三級警戒使癌症新診斷人數銳減。中晚期癌症患者可能延誤治療造成過早的死亡，早期癌症患者可能增加癌症併發症造成較差的預後，需儘快找出延遲診斷的癌症患者。本研究結果有助於未來防疫政策與癌症臨床照護決策參考。

Objectives: The COVID-19 pandemic adversely affected healthcare systems, including the availability of cancer diagnosis and monitoring services. This study evaluated the effects of the COVID-19 lockdown on newly diagnosed cancer cases in Taiwan. Methods: Weekly data on new cancer diagnoses from 2018 to 2021 were collected from Taiwan's National Health Insurance Research Database. A regression discontinuity design was used to estimate the impact of the lockdown on the number of new cancer diagnoses, using city/county-level data. An interrupted time series analysis was conducted to evaluate the immediate impact and post-intervention trends of the lockdown on new cancer diagnoses by clinical stage, age, and sex. Results: The lockdown resulted in an average weekly reduction of 66.1 new cancer diagnoses per city or county (p < 0.001). This decrease was largest for stage 0 cancer (75.1%), smallest for stage 4 cancer (45.0%), and larger among women (69.2%) than men (52.1%). After the lockdown, the rates of diagnoses for all cancer stages increased. This recovery was slowest for stage 0 cancer (0.020 per week per 100,000 people, p < 0.001) and fastest for stage 4 cancer (0.121 per week per 100,000 people, p < 0.001). Conclusions: The COVID-19 lockdown caused a sharp drop in new cancer diagnoses. Delays in diagnosis may lead to postponed treatment, resulting in premature death for advanced-stage cancers and increased complications with worse prognoses for early-stage cancers. Prompt identification of patients affected by these delays is crucial. These findings can inform future pandemic policies and cancer care decisions.

新冠肺炎；疫情；癌症；斷點迴歸分析；中斷時間序列分析

COVID-19; pandemic; cancer; regression discontinuity design; interrupted time series analysis