《一个关于药物经济学的分析模型的Case study.ppt》由会员分享，可在线阅读，更多相关《一个关于药物经济学的分析模型的Case study.ppt（14页珍藏版）》请在三一办公上搜索。
1、Development of an Economic Model to Assess the Cost Effectiveness of Asthma Management Strategies,Martin J.Price 1 and Andrew H.Briggs2 1 Global health outcomes,GlaxoSmithKline Research Development,Greenford,Middlesex,United Kingdom 2 Health Economics Research Centre,University of Oxford,Oxford,Unit
2、ed Kingdom,Pharmaceconomics 2002;20(3):183-194,Disease,Disease:Asthma 慢性发作性疾病 急性症候（不同严重程度）性发作：wheeze,chest tightness,breathlessness,cough,nacturnal wakingTreatment:A:salmeterol/flucticasone propionate combination(SFC,沙美特罗/丙酸氟替卡松联合给药)50/100g B:flucticasone propionate(FP,丙酸氟替卡松)100g 实验条件：一天两次，吸入式给药，持续
3、12周，随机，双盲的临床实验Treatment success：Asthma control（defined in world-wide asthma management guidelines and in terms of QALYs）Results：1,Effectiveness:Proportion of successfully controlled weeks per patient:SFC(66%)FP(47%)2,Cost:Mean weekly direct asthma management cost:SFC(15.77)FP(11.83)3,Average increme
4、ntal cost per successfully controlled week with SFC:20.83 4,PSA(probabilistic sensitivity analysis):ICER(95%uncertainty intervals):-64.94 to 112.66 if decision maker WTP 45 for an additional successful controlled week,SFC more cost-effective in 80%of the time,Methods：Markov model,Health states：1,Suc
5、cessful control:table I2,Hospital-managed exacerbation:hospital3,Primary-care-managed exacerbation:outpatient4,Sub-optimal control:between“successful control“and“exacerbation”5,Treatment failure:discontinue the study treatmentCycle length:1 weekTransition parameters:figure I,Health states：1,Successf
6、ul control:table I,well-controlled 代替total control,原因：1，对于一部分病人，不清楚total control 是否是现实的和可以达到的。2，对于某些决策者来说，2,Hospital-managed exacerbation:哮喘恶化病人在医院接受治疗，一些病人：急救室后回家，另一些病人：住院治疗3,Primary-care-managed exacerbation:门诊病人接受专业治疗。英国：general practice clinic.美国：urgent care centres4,Sub-optimal control：成功控制和恶化之
7、间，无法确诊，但又不值得干预5,Treatment failure：1，由于时间短，哮喘致死率低，而无法使用death作为markov model的absorbing state，而是用treatment failure 代替。2，而与一般的markov model 不同，处在treatment failure 的病人还会增加成本和效果，因为他们可能会选择另外的治疗方式。3，故使用B:flucticasone propionate(FP,丙酸氟替卡松)100g的成本效果来代替treatment failure 的成本效果Cycle length：对于哮喘的周期长度来说，1周是比较合适的。Tra
8、nsition parameters：定义了健康状态和转变方向，其中任意两个状态是可以相互转化的，只有treatment failure 是absorbing state，是单向的。,Data source,Population:1,patients with asthma aged 12-70,2,FEV1(forced expiratory volume in 1 second)40%-85%,3,receiving treatment corticosteroids(beclomethasone dipropionate(二丙酸倍氯米松）252-420g/day or equivalen
9、t)or salmeterol（沙美特罗）42 g/dayGrouping：1,salmeterol/flucticasone propionate combination(SFC,沙美特罗/丙酸氟替 卡松联合给药)50/100g 2,B:flucticasone propionate(FP,丙酸氟替卡松)100g 3,salmeterol(沙美特罗)50g 4,placebo Recording:1,Patient diary card(symptoms,lung function,nocturnal waking,use of rescue salbutamol(舒喘宁）2,case re
10、port form,Data coding,Assignment to health statesDetermining transition probabilitiesModel costHealth outcomesFitting distributions,Assignment to health states,Successful control(1):well-controlledHospital-managed exacerbation(2):weekly data examinedPrimary-care-managed exacerbation(3):weekly data e
11、xaminedSub-optimal control:not 2 and 3,any of criteria of 1 not metTreatment failure:withdraw from the study for prematurely clinical reasons,Determining transition probabilities(1),1，存在的问题：有些transition 的数量为0，这意味着这些transition很小，而不是不可能发生。2，采用bayesian methods,Determining transition probabilities(2),Mo
12、del cost,Medication cost(medication usage,physician time,hospital costs):2000 UK pharmacy prices(taking from MIMS(Monthly Index of Medical Specialities)Cost associated with Primary and secondary exacerbation health states(from a paper published in 1998)Inflation(HCHS(hospital and community health se
13、rvices)inflation index),Hospital-managed exacerbation中，cost采用急救室和住院的平均费用计算，Hoskins et al.,Health outcomes,Outcomes:Proportion of successfully controlled weeks All other states except treatment failure are assigned 0 benefitCosts and Outcomes of Treatment failure states:use FP treatment arm（lacking“u
14、sual care”sub-model),Fitting distributions,Transition probabilities:dirichlet distribution(multinomial equivalent of the-distribution)Costs:1,drug:no distribution 2,unit costs of primary care consultations and the cost of an inpatient day:normal distribution 3,the length of stay for a hospital-manag
15、ed exacerbation:log normal distribution 4,costs of primary and secondary care-managed exacerbations:normal distribution Treatment failure:probabilistically(FP arm),Data analysis,Cost-effectivenessPSA(Probabilistic sensitivity analysis),1，控制周数增加比例：0.10.26.2，成本增加：-10.9717.51每周3，ICER：-64.94112.664，25%S
16、FC处于dominant，75%，SFC效果更好，成本更高5，WTP=45时，80%SFC的NB0,Discussion,Limitations management exacerbation cost factors usual care cost data QALYsFurther improvement usual care prior distribution asthma control QALYs,限制：，恶化成本因素（临床实验时间短），一般治疗使用了FP arm代替，成本数据来源于不同的病人群体，的效用值说服力不够（个哮喘病人，轻度中度，平均效用值医院.，门诊.，次级健康.进一步研究：，组可以用来作为一般治疗，先验性的数据可以代替根据方法得出饿先验分布，提供一个更广泛的评价,THANK YOU！,
