Karimi M, Brazier J, Paisley S
Health Economics and Decision Science, School of Health and Related Research, University of Sheffield, Sheffield, UK.
Health Economics and Evidence Synthesis Research Unit, Luxembourg Institute of Health, Strassen, Luxembourg.
Health Qual Life Outcomes. 2017 May 18;15(1):105. doi: 10.1186/s12955-017-0678-9.
The use of preference-elicitation tasks for valuing health states is well established, but little is known about whether these preferences are informed. Preferences may not be informed because individuals with little experience of ill health are asked to value health states. The use of uninformed preferences in cost-effectiveness can result in sub-optimal resource allocation. The aim of this study was to pilot a novel method to assess whether members of the public are informed about health states they value in preference-elicitation tasks.
The general public was said to be informed if the expectations of the public about the effect of ill health on people's lives were in agreement with the experience of patients. Sixty-two members of the public provided their expectations of the consequences of ill health on five life domains (activities, enjoyment, independence, relationships, and avoiding being a burden). A secondary dataset was used to measure patient experience on those five consequences.
There were differences between the expectations of the public and the experience of patients. For example, for all five life consequences the public underestimated the effects of problems in usual activities compared to problems in mobility. They also underestimated the effect of 'anxiety or depression' compared to physical problems on enjoyment of life and on the quality of personal relationships.
This proof-of-concept study showed that it is possible to test whether preferences are informed. This study should be replicated using a larger sample. The findings suggest that preferences over health states in this sample are not fully informed because the participants do not have accurate expectations about the consequences of ill health. These uninformed preferences may not be adequate for allocation of public resources, and research is needed into methods to make them better informed.
使用偏好诱导任务来评估健康状态的价值已得到广泛认可，但对于这些偏好是否基于充分信息却知之甚少。由于让几乎没有健康不佳经历的个体去评估健康状态，偏好可能并非基于充分信息。在成本效益分析中使用未基于充分信息的偏好可能导致资源分配欠佳。本研究的目的是试行一种新方法，以评估公众在偏好诱导任务中对他们所评估的健康状态是否具备充分信息。
如果公众对健康不佳对人们生活的影响的期望与患者的经历一致，那么就认为公众具备充分信息。62名公众提供了他们对健康不佳在五个生活领域（活动、享受、独立性、人际关系以及避免成为负担）产生后果的期望。使用一个辅助数据集来衡量患者在这五个方面的经历。
公众的期望与患者的经历存在差异。例如，对于所有五个生活方面的后果，与行动不便问题相比，公众低估了日常活动问题的影响。与身体问题相比，他们还低估了“焦虑或抑郁”对生活享受和个人人际关系质量的影响。
这项概念验证研究表明，有可能测试偏好是否基于充分信息。本研究应用更大样本进行重复。研究结果表明，该样本中对健康状态的偏好并非完全基于充分信息，因为参与者对健康不佳的后果没有准确的期望。这些未基于充分信息的偏好可能不足以用于公共资源的分配，需要对使偏好更具充分信息的方法进行研究。
