背景：随着人口老龄化加剧，失能、半失能老年人数量持续增长，居家照护服务已成为我国养老服务体系的重要组成部分。然而，当前对于老年人居家照护服务的实际偏好了解仍有限，尤其是在测量方法方面，国际研究多采用离散选择实验（DCE），而优劣尺度法（BWS）在老年人群体中的应用和比较研究较少。

目的：本研究旨在比较DCE与BWS两种陈述性偏好测量方法在中国老年人群中对居家照护服务偏好测量的适用性，并揭示不同老年群体的偏好异质性，为居家照护服务制度优化和政策制定提供实证依据与方法学参考。

方法：首先通过文献回顾、专家咨询、老年人访谈与预调研确定偏好测量的4个核心属性（服务内容、服务质量、是否对服务过程进行监督或收集反馈意见、个人自付额）及相应水平，设计DCE与BWS问卷。随后以多阶段随机抽样在北京与成都两地选取社区老年人进行面对面调查。采用条件logit模型和混合logit模型对DCE与BWS问卷数据分别进行分析，探索偏好方向、重要性排序及个体间异质性特征；同时评估两种方法在结果一致性及受访者可接受性方面的差异，以比较其在老年人群中的适用性。

结果：调查共获得480份有效问卷。当聚焦老年人居家照护服务偏好时，DCE和BWS方法在偏好测量方面的结果相当一致。在答题完成难度方面，BWS（2.74分）总体高于DCE（2.62分）。BWS和DCE结果均显示，受访者更偏好包含以下特征的居家照护服务：提供生活照料与医疗护理，服务过程中有监督与反馈收集，服务质量较高，自付费用较低。此外，受访者对低自付费用最为重视，其次是服务内容，而后是服务质量，监督/反馈收集的影响相对最弱。BWS的异质性分析结果表明，北京受访者相较于成都受访者更偏好包含医疗护理的服务，高收入群体对费用敏感性较低，而低收入者更偏好低自付额服务，有照护经历者与60–75岁群体更倾向于选择低费用服务，低收入者则更重视服务质量。DCE的异质性分析结果显示，北京老年人更偏好包含医疗护理的整合型服务，低收入者对自付额较低的居家照护服务的关注度较高，未接受过照护服务的个体对服务质量更为敏感，成都老年人相较于北京老年人更倾向于较低的自付额。

结论：本研究发现DCE与BWS均可有效测量老年人对居家照护服务的偏好，且结果具有相当的一致性。但相对于BWS，DCE在受访者接受度方面更具优势。老年人偏好低费用、整合型服务与高质量照护，需设计阶梯化服务方案，完善长期护理保险报销机制，并推动医疗与照护资源协同。本研究为在中国情境下老年照护服务的制度设计与个性化服务包制定提供了有力的证据支持和方法学借鉴。

Background: As China’s population ages rapidly, the number of disabled and semi-disabled older adults is steadily increasing, making home care services a key part of the country’s eldercare system. While the government has introduced policies to support home-based care, there is still limited understanding of older adults’ actual preferences—especially regarding how those preferences are measured. Discrete Choice Experiments (DCE) are widely used internationally, but Best-Worst Scaling (BWS) has been less explored in older populations. Objective: This study aimed to compare the use of DCE and BWS in measuring preferences for home care services among older adults in China, and to explore differences in preferences across subgroups. Methods: Based on literature review, expert consultation, and pilot interviews with older adults, four key service attributes were selected: service content, service quality, supervision and feedback, and out-of-pocket cost. A DCE design was generated using a D-optimal algorithm, and a BWS design using an Orthogonal Main Effects Design (OMED), each including eight choice tasks. Using stratified random sampling, face-to-face surveys were conducted with older adults in Beijing and Chengdu, resulting in 480 valid responses. Data were analyzed using conditional and mixed logit models to estimate preference directions, importance rankings, and individual-level variation. The two methods were also compared in terms of consistency and respondent acceptability. Results: DCE and BWS produced largely similar results in terms of preference patterns. However, DCE was slightly easier for participants to complete (mean difficulty score: DCE = 2.62, BWS = 2.74). Both methods showed that older adults preferred services that offered a combination of medical and daily care, included supervision and feedback mechanisms, ensured high service quality, and involved lower out-of-pocket costs. Cost was the most important factor, followed by service content, quality, and supervision. BWS results revealed that Beijing respondents favored integrated medical services more than those in Chengdu; low-income individuals placed more emphasis on cost, while high-income groups were less sensitive to price. Those with care experience and people aged 60–75 were more likely to choose low-cost services, and low-income participants paid more attention to service quality. DCE results showed similar trends: Beijing respondents preferred integrated care, low-income participants cared more about cost, and those without prior care experience were more sensitive to quality. Chengdu respondents showed a stronger preference for lower costs. Conclusion: Both DCE and BWS are effective tools for measuring older adults’ preferences for home care services, with strong agreement between results. However, DCE appeared to be more acceptable to older respondents. Overall, older adults preferred services that are affordable, integrated, and high-quality. These findings suggest the need to develop tiered service packages, improve long-term care insurance reimbursement, and strengthen coordination between medical and care services. This study provides solid evidence to support policy development and personalized service planning for home-based eldercare in China.