乳癌篩檢經濟效益評估

Abstract

背景與目的 乳癌為臺灣女性發生率最高之癌症，且好發年齡較歐美國家提早，對公共衛生與醫療資源造成重大負擔。自2004年起，國民健康署推動乳房X光攝影篩檢政策，並隨流行病學趨勢與醫療量能逐步擴大服務年齡層。2025年起，篩檢年齡正式延伸至40～74歲，然其長期經濟效益與資源配置合理性，仍需以在地實證資料加以評估。本研究目的在比較不同乳癌篩檢起始與終止年齡策略之成本效益，評估擴大篩檢年齡範圍在臺灣情境下之健康經濟可行性，作為政策優化之依據。 研究方法 本研究採社會觀點，建構決策樹結合馬可夫模型（Decision Tree combined with Markov Model），模擬乳癌自然病史與篩檢介入後之長期健康結果。比較四種策略，包括：無篩檢、50～69歲每兩年一次乳房攝影、45～69歲每兩年一次乳房攝影，以及40～74歲每兩年一次乳房攝影。模型參數主要來自臺灣癌症登記資料、全民健保資料庫及大規模篩檢實證數據，並納入直接醫療成本與間接時間成本。健康效益以品質調整生命年（QALYs）衡量，成本與效益均以3%折現率計算。經濟評估指標包括增量成本效益比（ICER）、淨貨幣效益（NMB）、成本效果可接受度曲線（CEAC），並進行確定性與機率性敏感度分析（Monte Carlo Simulation）。 研究結果 在基準分析中，相較於無篩檢策略，40～74歲每兩年一次乳房攝影篩檢可額外增加0.042 QALYs，對應之ICER為新台幣505,945元／QALY，顯著低於臺灣常用之一倍人均GDP支付意願門檻，顯示其具成本效益。45～69歲與50～69歲篩檢策略則因成本效益相對較差，呈現延伸性支配。敏感度分析結果顯示，在不同篩檢參與率與確診順從率情境下，40～74歲策略皆維持為非支配策略，且順從率提升可同時增加健康效益並降低長期醫療成本。機率敏感度分析顯示，在一倍人均GDP之支付意願門檻下，40～74歲策略成為最具成本效益方案之機率接近100%，結果具高度穩健性。 結論 在臺灣乳癌流行病學特性與醫療體系情境下，將乳癌篩檢年齡擴大至40～74歲，每兩年一次乳房X光攝影，為最具成本效益且具政策穩健性之策略。研究結果支持現行40～74歲篩檢政策方向，未來政策除持續推動擴大篩檢外，亦應同步強化確診順從率與個案管理，以進一步提升乳癌篩檢計畫之整體公共衛生效益與資源配置效率。

Background and Objectives Breast cancer is the most commonly diagnosed cancer among women in Taiwan, with an earlier peak age of onset compared with Western countries, posing a substantial burden on public health and healthcare resources. Since 2004, the Health Promotion Administration has implemented a nationwide mammography screening program, gradually expanding the eligible age groups in response to epidemiological trends and healthcare capacity. Beginning in 2025, the screening age range was formally extended to women aged 40~74 years. However, the long-term economic benefits and resource allocation efficiency of this expanded policy require evaluation based on local empirical evidence. This study aimed to compare the cost-effectiveness of different breast cancer screening initiation and cessation age strategies and to assess the economic feasibility of expanding the screening age range within the Taiwanese context to inform policy optimization. Methods A societal perspective was adopted, and a decision tree combined with a Markov model was constructed to simulate the natural history of breast cancer and long-term health outcomes following screening interventions. Four strategies were compared: no screening; biennial mammography for women aged 50~69 years; biennial mammography for women aged 45~69 years; and biennial mammography for women aged 40~74 years. Model parameters were primarily derived from the Taiwan Cancer Registry, the National Health Insurance Research Database, and large-scale population-based screening data. Both direct medical costs and indirect time costs were included. Health outcomes were measured in quality-adjusted life years (QALYs), and both costs and outcomes were discounted at an annual rate of 3%. Economic evaluation metrics included the incremental cost-effectiveness ratio (ICER), net monetary benefit (NMB), and cost-effectiveness acceptability curves (CEAC). Deterministic and probabilistic sensitivity analyses using Monte Carlo simulation were performed to assess model uncertainty. Results In the base-case analysis, biennial mammography screening for women aged 40~74 years yielded an additional 0.042 QALYs compared with no screening, with an ICER of NTD 505,945 per QALY, which is substantially below the commonly used willingness-to-pay threshold of one times Taiwan’s per capita gross domestic product, indicating that this strategy is cost-effective. Screening strategies targeting women aged 45~69 years and 50~69 years were extendedly dominated due to inferior cost-effectiveness. Sensitivity analyses demonstrated that the 40~74-year screening strategy consistently remained non-dominated across varying screening participation rates and compliance levels. Improvements in compliance were associated with both increased health benefits and reduced long-term healthcare costs. Probabilistic sensitivity analysis further showed that, at a willingness-to-pay threshold of one times per capita GDP, the probability of the 40~74-year strategy being the most cost-effective option approached 100%, indicating high robustness of the findings. Conclusions Considering Taiwan’s breast cancer epidemiology and healthcare system context, extending biennial mammography screening to women aged 40–74 years represents the most cost-effective and policy-robust strategy. The findings support the current screening policy targeting this age group. In addition to continuing the expanded screening program, future policy efforts should place greater emphasis on improving compliance and case management to further enhance the overall public health impact and resource allocation efficiency of breast.