Abstract: Background　Polypharmacy burden is a core factor that affect patients' medication beliefs，medication compliance，and medication safety.By evaluating the patient's medication burden in time，healthcare workers may find out the problems that need to be solved in the process of medicine use，by which a scientific drug intervention plan may be formulated.However，at present，there is a lack of scientific and effective tools to evaluate the patient's polypharmacy burden in China.Objective　To translate the English version of Living with Medicines Questionnaire（LMQ）into Chinese，and test its reliability and validity in community-living elderly patients with polypharmacy，to provide a tool for assessing the medication burden in this group.Methods　The Chinese version of LMQ（initial version）including 8 dimensions and 39 items was formed by using forward translation，proofreading，back translation，expert review，cognitive interview and presurvey.From April to August 2019，convenience sampling was used to select elderly chronic disease patients with polypharmacy from communities served by three community health centers〔Zhengzhou Community Health Center（East 2nd Street），Linshanzhai Community Health Center，Science Avenue Community Health Center of Zhengzhou National High & New Industries Development Zone〕in Zhengzhou as the participants.In the first stage of this study，the data were collected for exploratory factor analysis，and the sample size was 260 cases.In the second stage，the reliability of the initial version of questionnaire was retested 2-4 weeks later in a sample of 30 cases randomly selected from the above-mentioned 260 cases.In the third stage，the data were collected for confirmatory factor analysis，with a sample of 373 cases.The investigation used the General Data Questionnaire and the Chinese version of LMQ（initial version）.The validity and reliability of the final Chinese version of LMQ were evaluated overall.Results　The questionnaire results of effective responders in the first，second，and third stages of this study〔242（93.1%），30（100.0%），and 350（93.8%）〕were included for final analysis.Item analysis showed that the item-total correlation coefficients of the Chinese version of LMQ（initial version）were 0.317-0.658（P<0.01），and the critical value for each item was > 3.0.The results of exploratory factor analysis showed that the KMO value of Chinese version of LMQ（initial version）was 0.875，and the χ2 of Bartlett's test was 8 139.877（P<0.01），which indicated that it was suitable for factor analysis.Eight common factors with eigenvalue > 1.000 were extracted，explaining 76.780% of the total variance，and the loading value of each item on its common factor was > 0.400.The results of confirmatory factor analysis showed that the pre-set model fitting indicators were not ideal，so the model was modified according to the indicator prompts.After adding 8 covariance correlations，the fitting indicators are in the acceptable range.Content validity：the item content validity indices of the Chinese version of LMQ（initial version）were 0.86-1.00，and the scale content validity index was 0.949.Reliability：the scale Cronbach's α coefficient was 0.892，with dimensions' Cronbach's α coefficients ranging from 0.867-0.943； the scale half-reliability was 0.817 with dimensions' half-reliability ranging from 0.841-0.947； the scale retest reliability was 0.904 with dimensions' retest reliability ranging from 0.802-0.875.The final Chinese version of LMQ includes 8 dimensions（39 items）：medication attitude（7 items），practical difficulties（6 items），doctor-patient relationship（5 items），medication effects（5 items），interferences with daily life（6 items），side effects（4 items），medication behavior（3 items），economic burden（3 items）.Conclusion　The Chinese version of LMQ uses simple and easy-to-understand 39 items to assess the patient's medication burden from 8 aspects（medication attitude，practical difficulties，doctor-patient relationship，medication effects，interferences with daily life，side effects，medication behavior，and economic burden），demonstrating good validity and reliability.The assessment is comprehensive and easy to implement，which may effectively assess the burden of medication in community-living elderly patients with polypharmacy，and may provide a measurement tool for healthcare workers to formulate precise drug interventions.

Key words: Polypharmacy, Aged, Community, Living with Medicines Questionnaire, Questionnaires, Reliability, Validity

摘要： 背景　多重用药负担是影响患者用药信念、用药依从性和用药安全的核心要素，医务人员以患者用药负担为中心进行评估，可及时发现其用药过程中亟需解决的问题、制定科学化的药物干预方案，但目前国内尚且缺乏科学、有效的工具来评价患者多重用药负担。目的　汉化用药生活问卷（LMQ）并分析其在社区老年多重用药患者中的信效度，为明确社区老年多重用药患者用药负担现况提供测量工具。方法　采用正译、校对、回译、专家审核、认知性访谈、预调查对LMQ进行汉化，最终形成了包括8个维度39个条目的中文版LMQ（初始版）。2019年4—8月，采用便利抽样法选取郑州市3家社区卫生服务中心〔郑州市社区卫生服务中心（东二街）、林山寨社区卫生服务中心、高新区科学大道社区卫生服务中心〕下属社区中老年多重用药患者为调查对象。本研究第一阶段收集数据用于探索性因子分析，最终确定样本量为260例；第二阶段用于重测信度评估，重测间隔时间为2~4周，最终确定样本量为30例（从第一阶段患者中采用随机数字表法选取）；第三阶段用于验证性因子分析，最终确定样本量为373例。采用一般资料调查问卷、中文版LMQ（初始版）对其进行调查，评价中文版LMQ（初始版）的信效度。结果　第一阶段共发放问卷260份，回收有效问卷242份，有效回收率为93.1%；第二阶段共发放问卷30份，回收有效问卷30份，有效回收率为100.0%；第三阶段共发放问卷373份，回收有效问卷350份，有效回收率为93.8%。项目分析结果显示，中文版LMQ（初始版）各条目得分与其总分的相关系数为0.317~0.658（P<0.01）；中文版LMQ（初始版）各条目决断值（CR）均>3.0。探索性因子分析结果显示，中文版LMQ（初始版）的KMO值为0.875，Bartlett's球形检验的χ2=8 139.877，P<0.01，表明适合做因子分析；共提取出特征根>1.000的公因子有8个，累积方差贡献率为76.780%，各条目在所属公因子上的载荷量均>0.400。验证性因子分析结果显示，预设模型拟合指标不理想，根据指标提示进行模型修正，添加8条协方差相关关系后，各拟合指标处于可接受范围。内容效度：中文版LMQ（初始版）的条目内容效度指数（I-CVI）为0.86~1.00，问卷内容效度指数（S-CVI）为0.949。信度：中文版LMQ（初始版）的Cronbach's α系数为0.892，各维度的Cronbach's α系数为0.867~0.943；中文版LMQ（初始版）的折半信度为0.817，各维度的折半信度为0.841~0.947；中文版LMQ（初始版）的重测信度为0.904，各维度的重测信度为0.802~0.875。最终形成中文版LMQ，其包括8个维度39个条目，分别为用药态度（7个条目）、实践困难（6个条目）、医患关系（5个条目）、用药效果（5个条目）、干扰日常生活（6个条目）、副作用（4个条目）、用药行为（3个条目）、经济负担（3个条目）。结论　中文版LMQ利用39个条目从用药态度、实践困难、医患关系、用药效果、干扰日常生活、副作用、用药行为及经济负担8个方面了解患者的用药负担，信效度良好，且操作简单、评估较全面、条目通俗易懂，可以有效评估社区老年多重用药患者居家生活过程中各方面的用药负担，并为医务人员制定精准化药物支持干预方案提供测评工具。

关键词: 多种药物疗法, 老年人, 社区, 用药生活问卷, 问卷调查, 信度, 效度