Lloyd Andrew, McIntosh Emma, Price Martin
The MEDTAP Institute at United BioSource Corporation, London, UK.
Pharmacoeconomics. 2005;23(11):1167-81. doi: 10.2165/00019053-200523110-00008.
Antiepileptic drugs (AEDs) have been shown to reduce the severity and frequency of seizures for most patients. However, many patients experience adverse effects in order to maintain seizure control.
A stated preference discrete choice experiment (DCE) was used to explore the preferences of people with epilepsy regarding the adverse effects and seizure control of AEDs.
The main adverse effects of AEDs were identified through a literature search and expert consultation. In addition, a national epilepsy patient advocacy group helped to identify important attributes and commented on the attributes we had already identified. The DCE included five attributes related to adverse effects (alopecia, nausea, skin rash, concentration effects and weight change) plus seizure control and cost (to estimate willingness to pay [WTP]). A cost attribute was included in the DCE in order to estimate people's WTP for changes in attribute levels. Five hundred members of a national patient advocacy group with a diagnosis of epilepsy were presented with pairs of hypothetical drug profiles with varied levels of adverse effects, seizure control and cost; they were then asked to indicate which drug they preferred. Questions were also included to collect sociodemographic data (including income) and information regarding experience of adverse effects and medication. The survey was administered via the post and the Internet. Data were analysed using a random effects probit model.
A total of 148 surveys were returned. All attributes were significant and had the expected polarity, i.e. participants showed a preference for less severe adverse effects, greater seizure control and less cost. To achieve 100% seizure control and no adverse effects, participants were willing to pay 709 pounds sterlings (1105 US dollars) per month, 95% CI 451 pounds sterlings, 1278 pounds sterlings (1 pound sterling = 1.56 US dollars, 2002 exchange rate). Participants' WTP was significantly influenced by different adverse effects; for example, people with epilepsy were willing to pay only 174 pounds sterlings (271 US dollars) per month for a drug that provided seizure freedom but also caused hair loss. Segmented models showed that seizure frequency has a significant negative impact on respondents' income levels. Also, women were willing to pay twice as much as men to avoid weight gain. Participants were also willing to trade changes in seizure control for improvements in adverse effects.
Participants placed a high value on gaining total seizure control with no adverse effects. This study underlines the importance that people with epilepsy place on reducing adverse effects. The study also revealed how preferences for AEDs vary in different subgroups. Management of epilepsy is usually aimed at minimising seizures within a tolerable level of adverse effects. The present study suggests that people with epilepsy have strong preferences for reducing adverse effects as well as improving seizure control. These data may be considered useful when making medical management decisions in epilepsy.
抗癫痫药物(AEDs)已被证明可降低大多数患者癫痫发作的严重程度和频率。然而,许多患者为了维持癫痫控制会出现不良反应。
采用陈述偏好离散选择实验(DCE)来探究癫痫患者对AEDs不良反应和癫痫控制的偏好。
通过文献检索和专家咨询确定AEDs的主要不良反应。此外,一个全国性癫痫患者倡导组织帮助确定重要属性,并对我们已确定的属性发表意见。DCE包括与不良反应相关的五个属性(脱发、恶心、皮疹、注意力不集中和体重变化)以及癫痫控制和成本(以估计支付意愿[WTP])。DCE中纳入成本属性是为了估计人们对属性水平变化的支付意愿。向500名诊断为癫痫的全国患者倡导组织成员展示成对的具有不同不良反应水平、癫痫控制和成本的假设药物概况;然后要求他们指出更喜欢哪种药物。还包括一些问题以收集社会人口统计学数据(包括收入)以及关于不良反应和用药经历的信息。该调查通过邮寄和互联网进行。使用随机效应概率模型对数据进行分析。
共收回148份调查问卷。所有属性均具有显著性且具有预期的极性,即参与者表现出对较轻微不良反应、更好的癫痫控制和更低成本的偏好。为实现100%的癫痫控制且无不良反应,参与者愿意每月支付709英镑(1105美元),95%置信区间为451英镑,1278英镑(1英镑 = 1.56美元,2002年汇率)。参与者的支付意愿受到不同不良反应的显著影响;例如,癫痫患者每月仅愿意为一种能实现癫痫自由但会导致脱发的药物支付174英镑(271美元)。分段模型显示癫痫发作频率对受访者的收入水平有显著负面影响。此外,女性为避免体重增加愿意支付的金额是男性的两倍。参与者也愿意用癫痫控制的变化来换取不良反应的改善。
参与者高度重视实现完全癫痫控制且无不良反应。本研究强调了癫痫患者对减少不良反应的重视。该研究还揭示了不同亚组对AEDs的偏好如何不同。癫痫管理通常旨在将癫痫发作降至可耐受的不良反应水平内。本研究表明癫痫患者在减少不良反应以及改善癫痫控制方面有强烈的偏好。这些数据在做出癫痫医疗管理决策时可能会被认为是有用的。