健保放寬降血脂藥物給付範圍對心血管疾病發生、死亡及醫療費用之長期影響

Abstract

背景：心血管疾病是全球主要死因之一，並構成沉重的疾病負擔。其中，高血脂為其重要危險因子，且臺灣近年來高血脂的盛行率快速上升，20歲或以上人口約有26%罹患此症。因應國際治療共識的更新，臺灣健保署自2013年8月1日起放寬降血脂藥物給付範圍，使糖尿病患者等心血管高風險族群能更早接受藥物治療。雖然先前研究已證實該政策的初步成效，但其長期影響仍有待釐清。 目的：本研究旨在評估2013年健保放寬降血脂藥物給付範圍後，對藥物使用狀況、臨床結果、經濟影響三個層面之及短、中、長期影響。 方法：本研究採用縱貫性回溯設計，分析2010年8月1日至2022年7月31日的衛生福利資料。研究對象包括以無心血管疾病史的糖尿病患者作為政策介入組，及以無心血管疾病與糖尿病史的高血壓患者作為對照組。本研究運用差異中的差異法（difference in difference, DID），並使用廣義估計方程式（generalized estimating equations, GEE）進行多變項分析，比較給付政策實施後不同時期，介入組相對於對照組在藥物使用狀況、臨床結果及經濟影響三個層面的變化，以評估政策之淨影響。 結果：本研究共納入146,786名介入組及148,873名對照組之研究對象。結果發現給付範圍放寬政策實施後，在藥物使用狀況方面，介入組在短期內的降血脂藥物使用比例相較對照組高3.06%（p<0.0001），但在中期轉為低2.89%（p=0.0010），長期則進一步下降至低12.59%（p<0.0001），顯示政策影響隨時間遞減，可能反映對照組逐漸開始使用降血脂藥物。在臨床結果方面，介入組的心血管疾病發生風險皆相較對照組低，短期、中期與長期分別低7.15%（p=0.0070）、9.95%（p=0.0002）及16.71%（p<0.0001），顯示政策對臨床成效的影響隨時間增強。在經濟影響方面，介入組平均降血脂藥物費用未相較對照組增加，且平均心血管疾病醫療花費亦較對照組低，短期、中期與長期分別低7.69%（p=0.1637）、10.88%（p=0.0589）及22.55%（p<0.0001），顯示政策帶來的費用節省效果隨時間擴大，其經濟效益亦隨之提升。 結論：健保2013年放寬降血脂藥物給付範圍，使政策影響族群─糖尿病患者的心血管疾病發生風險及相關醫療支出均顯著降低，且此效應隨時間推移（由短期至長期）更加明顯。顯示該政策具有長期正向且累積性的影響，並進一步驗證針對心血管疾病高風險族群放寬降血脂藥物給付條件的臨床與經濟價值。

Background: Cardiovascular diseases (CVD) are among the leading causes of death worldwide and impose a substantial disease burden. Hyperlipidemia is a major risk factor for CVD, and its prevalence has been rising rapidly in Taiwan, with approximately 26% of adults over 20 years old affected. In response to updates in international treatment guidelines, Taiwan’s National Health Insurance Administration (NHIA) expanded reimbursement criteria for lipid-lowering agents on August 1, 2013. This policy allowed high-risk populations, such as patients with diabetes, to initiate lipid-lowering medication earlier. While previous studies have demonstrated the initial benefits of this policy, its long-term impact remains unclear. Objective: This study aims to evaluate the short-, medium-, and long-term effects of the expansion of reimbursement criteria for lipid-lowing agents on three key aspects: medication use, clinical outcomes, and economic impact. Methods: A longitudinal retrospective study was conducted using Health and Welfare Data Science Center (HWDC) Data from August 1, 2010, to July 31, 2022. The study population consisted of diabetes patients without a history of CVD as the intervention group and hypertension patients without a history of CVD or diabetes as the control group. The difference-in-differences (DID) approach was adopted to evaluate the net impact of the reimbursement criteria change. Generalized estimating equations (GEE) were further applied for multivariable analysis. The analysis compared changes in medication use, clinical outcomes, and economic impact between the intervention and control groups across different periods following expansion of reimbursement criteria. Results: A total of 146,786 individuals were included in the intervention group and 148,873 in the control group in this study. The findings indicate that following the implementation of the expanded reimbursement criteria, the intervention group showed a 3.06% higher rate of lipid-lowering medication use in the short term compared to the control group (p<0.0001). However, this trend reversed in the mid-term, with a 2.89% lower rate (p=0.0010), and further declined to 12.59% lower in the long term (p<0.0001). This suggests that the policy’s impact on medication use diminished over time, potentially reflecting a gradual increase in lipid-lowering medication use among the control group. In terms of clinical outcomes, the intervention group consistently demonstrated a lower risk of CVD compared to the control group—by 7.15% in the short term (p=0.0070), 9.95% in the mid-term (p=0.0002), and 16.71% in the long term (p<0.0001)—indicating that the policy's clinical effectiveness increased over time. Regarding economic impact, the average lipid-lowering medication cost in the intervention group did not increase relative to the control group. Furthermore, the average medical expenditures related to cardiovascular disease were lower in the intervention group by 7.69% (p=0.1637), 10.88% (p=0.0589), and 22.55% (p<0.0001) in the short, mid, and long term, respectively. These findings suggest that the cost-saving effect of the policy expanded over time, with its economic benefits becoming more pronounced. Conclusion: The 2013 National Health Insurance (NHI) expansion of reimbursement criteria for lipid-lowering agents led to additional reductions in both CVD risk and related healthcare expenditures among the policy-targeted population—patients with diabetes. These effects became increasingly pronounced over time, from the short term to the long term. The findings indicate that the policy has a positive and cumulative long-term impact, validating the clinical and clinical and economic value of expanding lipid-lowering agents reimbursement criteria for populations at high risk of cardiovascular disease.