With the continuous improvement of diabetes prevention and treatment strategies, clinical guidelines recommend first-line hypoglycemic agents while emphasizing a "patient-centered", and the selection of appropriate second-line hypoglycemic agents based on patients' clinical characteristics and preferences.

To quantitatively analyze second-line hypoglycemic agents selection preferences in patients with type 2 diabetes mellitus (T2DM) by using a discrete choice experiment, so as to provide reference for clinical diagnosis and treatment for T2DM patients.

This study conducted a questionnaire survey on T2DM patients selected from Hainan and Shanxi provinces from October 2021 to January 2022 by using a combination of multi-stage random cluster sampling. The questionnaire included basic personal information of the respondents, information related to disease treatment and discrete choice experimental choice set (seven attributes of glycemic control, risk of hypoglycemic events, risk of gastrointestinal adverse events, body mass change within six months, cardiovascular protection, mode of medication administration, and out-of-pocket/monthly costs, each of which included a various levels were included through the literature review, focus group discussion, and pre-tests). A mixed Logit regression model was used to quantify the preference of T2DM patients for second-line hypoglycemic agents, and the regression coefficients reflected the direction and magnitude of the preference for second-line hypoglycemic agents among T2DM patients. Willingness to pay (WTP) reflected the monetary value of patients' willingness to pay or receive compensation after the changes in different attribute levels.

A total of 1 443 questionnaires were distributed and 1 388 valid questionnaires were recovered, with an effective recovery rate of 96.2%. All seven attributes had an effect on medication preference of T2DM patients (P<0.05), the top three influences on medication preference were the effect of glycemic control, the risk of gastrointestinal adverse reactions, and the risk of hypoglycemic events, respectively. The change in body mass within six months had the lowest impact. Patients with T2DM were willing to pay 411.16 yuan per month when the blood glucose reduction increased from 0.5% to 2.5%, while patients with T2DM were willing to pay only 96.78 yuan per month when the change in body mass switched from an increase of 3 000 g to a decrease of 2 000 g over six months.

T2DM patients preferred second-line hypoglycemic drugs with good glycemic control, no risk of gastrointestinal adverse reactions, no risk of hypoglycemic events, cardiovascular protection, oral mode of administration, and a body mass reduction of 2 000 g within six months.

随着糖尿病防治策略的不断完善，临床指南推荐一线降糖药物的同时强调"以患者为中心"，根据患者的临床特征及其偏好选择适当的二线降糖药物进行治疗。

利用离散选择实验定量分析2型糖尿病（T2DM）患者二线降糖药物选择偏好，为T2DM患者的临床诊治提供参考。

本研究于2021年10月—2022年1月，采取多阶段分层随机抽样的方式，选取海南省和山西省T2DM患者进行问卷调查。问卷包括被调查者的个人基本信息、疾病治疗相关信息和离散选择实验选择集（通过文献研究、焦点小组讨论和预试验，纳入血糖控制效果、发生低血糖事件的风险、发生胃肠道不良反应的风险、半年内体重变化、能否保护心血管、服药方式和自付费用/月7个属性，每个属性又包括若干水平）。采用混合Logit回归模型量化T2DM患者对二线降糖药物相关属性偏好程度，回归系数反映T2DM患者对二线降糖药物各属性水平偏好的方向和大小。采用支付意愿（WTP）反映不同属性水平变动后患者愿意支付或获得补偿的货币值。

本次调查共发放问卷1 443份，回收有效问卷1 388份，有效回收率为96.2%。7个属性对T2DM患者用药偏好均有影响（P<0.05），对用药偏好影响排在前三的分别为血糖控制效果、发生胃肠道不良反应的风险和发生低血糖事件的风险，排在最后的是半年内体重变化。当血糖降低幅度由0.5%升高为2.5%时，T2DM患者每月愿意支付411.16元；而当半年内体重变化由增加3 000 g转换为降低2 000 g时，T2DM患者每月仅愿意支付96.78元。

T2DM患者选择二线降糖药物更偏好血糖控制效果好、无胃肠道不良反应风险、无低血糖事件的风险、保护心血管、服药方式为口服和半年内体重降低2 000 g的二线降糖药物。