中風相關的個人化高血壓預防之成本效益分析

Cost-effectiveness analysis of personalized prevention for hypertension associated with stroke

郭書帆(Shu-Fan Kuo)

指導教授 ： 陳秀熙

國立臺灣大學/公共衛生學院/流行病學與預防醫學研究所/碩士(2019年)

https://doi.org/10.6342/NTU201903838

摘要

<研究背景> 雖然控制高血壓已廣被接受為預防中風之必要措施，但不同階段的血壓控制對於發生中風所造成之影響不同，使得對於不同高血壓防治方法之實證效益難以進行評估。此外，近年來對於運用初段預防進行高血壓防治受到重視，然而此一介入措施對於減少後續發生中風之效益卻甚少被提及。如何於民眾運用前述以族群為基礎之初段預防措施，結合以個人高血壓進展風險分數為導向之篩檢與預防性投藥之次段預防措施，在現有之研究仍付之闕如。本論文因此希望能針對前述對結合初段與次段預防之中風防治措施所能預防中風發生的比例、介入後所增加之人年，以及調整生活品質後進行成本效益分析。

<研究目的>
(1) 以具不同危險因子之社區世代為基礎，預測其多階段多因子高血壓進展及後續中風發生；
(2) 利用(1)，進一步發展個人化高血壓防治措施在中風防治的介入分析模型；
(3) 評估彰化社區世代個人化高血壓防治措施之成本效益分析。

<材料與方法> 我們採用以基隆社區世代所發展的多階段多因子高血壓預測模式，配合不同血壓階段的中風發生率，套用於彰化社區世代具不同危險因子的族群。以馬可夫決策模型，考量中風後功能狀態，進行個人化高血壓防治相對於控制組之成本效益分析，並以機率性成本效益分析處理參數不確定性問題。主要結果呈現包括中風的減少、生活品質調整人年的增加，及增加成本效用比，並以成本效益散佈圖及接受曲線呈現。

<結果> 以彰化社區世代個人的危險因子進行的模擬分析，預測其中風的年發生率約為每千人年13例，包括3.59及9.48例的出血性和缺血性中風，個人化高血壓防治策略可降低17%中風發生(95%信賴區間：15-18%)。考慮中風的減少及其後續殘障等級的變化之後，個人化高血壓防治策略對個人平均可增加0.17個生活品質調整人年，且平均每人整體花費較控制組減少新台幣161,170元。在考量參數的變異之下，個人化高血壓防治策略的成本效益優勢仍然存在。

<結論> 本研究建立評估個人化高血壓及中風防治的架構，在考量高血壓狀態的動態轉移之下，對以全民為基礎的初段預防及高危險群者之預防性投藥及次段預防措施進行評估，結果發現個人化高血壓防治就成本效益的角度而言是一項節省成本的優勢策略，除了可增加生活品質調整人年之外，累積花費成本亦較低。本研究的結果預期可以做為個人化高血壓防治策略之參考。

關鍵字

高血壓 ； 中風 ； 馬可夫決策模型 ； 成本效益分析

並列摘要

Although the control for hypertension have widely been accepted as a necessary component for stoke prevention, the heterogeneity regarding the evolution of hypertension in associated with the occurrence of stoke render the elucidation of the effectiveness of each components of prevention strategies intractable. As the primary prevention have gained great attention in current strategies of hypertension management, its contribution to stoke prevention remain not fully addressed. The effectiveness on the application of multiple prevention strategies including the population-based primary prevention and risk-oriented carotid ultrasound screening followed by a series of treatment and therapy for stoke prevention also has not been elaborated. The cost incurred by the multi-step prevention strategies and the attributes of its effectiveness including stroke averted, life-year gained, and quality adjusted life-years gained motivate our research by applying a cost-effectiveness analysis to evaluate the decision on the proposed personalized stroke prevention strategies.

This thesis aims
(1) to explore the dynamic of stoke embedded in the evolution of hypertension defined by multiple disease status including normal, prehypertension, stage I hypertension, and stage II hypertension;
(2) to develop a risk-guided individual-tailored stoke prevention strategy incorporating primary and secondary prevention based on (1); and
(3) to assess the cost-effectiveness of (2) applied to Changhua population following the principle of decision analysis.

We calculated the personalized risk score with 4-state hypertension Markov model underpinning in light of definition from JNC 7, normal, pre-hypertension, stage 1 hypertension, stage 2 hypertension. The clinical weights of the risk score were borrowed from literature with Keelung Community-based Screening (KCIS) cohort, and applied to the Changhua Community-based Screening (CHCIS) cohort to stratify the cohort into different risk groups. A Markov decision tree, incorporating the 4-state hypertension model, hemorrhagic stroke, ischemic stroke, three functional outcomes after stroke, and death was built. Finally, the probabilistic cost-effectiveness analysis of personalized prevention of hypertension compared with control group was conducted. The main outcome measures include stroke avoided and quality-adjusted life-year prolonged, incremental cost-utility ratio, and the acceptability curve for personalized prevention against control.

Among the Changhua population, the incidence of stroke is 16 per 1,000 with 4.2 and 11.8 per 1,000 for hemorrhagic and ischemic type, respectively. The personalized prevention strategy results in stroke risk reduction by 17% (95% CI: 15-18%). Taking into account the stroke events reduction and the functional status after stroke, the personalized prevention program not only results in 0.17 QALY gained per person in a 20-year time horizon, but also leads to an average NTD 161,170 less expenditure per person. Considering the distribution of parameters, the benefit of personalized prevention program over control is still dominant even under the worst case of parameters.

We developed a framework for the evaluation of individual-tailored hypertension and stroke prevention, which incorporated primary and secondary prevention. As the personalized prevention is cost-effective against control in terms of quality-adjusted life year gained, our results support the personalized prevention as a promising policy for hypertension associated stroke.

並列關鍵字

hypertension ； stroke ； Markov decision model ； cost-effectiveness analysis

參考文獻

1. Egan BM, Li J, Hutchison FN, Ferdinand KC: Hypertension in the United States, 1999 to 2012: progress toward Healthy People 2020 goals. Circulation 2014;130:1692-9.

2. 衛生福利部國民健康署: statistical yearbook of health promotion 2016. 2018.

3. Lewington S, Clarke R, Qizilbash N, Peto R, Collins R: Age-specific relevance of usual blood pressure to vascular mortality: a meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002;360:1903-13.

4. Thomopoulos C, Parati G, Zanchetti A: Effects of blood pressure lowering on outcome incidence in hypertension. 1. Overview, meta-analyses, and meta-regression analyses of randomized trials. J Hypertens 2014;32:2285-95.

5. Rapsomaniki E, Timmis A, George J, et al.: Blood pressure and incidence of twelve cardiovascular diseases: lifetime risks, healthy life-years lost, and age-specific associations in 1.25 million people. Lancet 2014;383:1899-911.