中国全科医学 ›› 2026, Vol. 29 ›› Issue (31): 4635-4640.DOI: 10.12114/j.issn.1007-9572.2026.0123

• 论著·数智医疗研究 • 上一篇    

智能语音外呼辅助传统干预对社区老年"三高"共病患者健康管理的效果研究:一项随机对照试验

魏学娟1, 王利华1, 王丽1, 任静林2,*(), 闫海青1, 于海洋3, 刘景峰3   

  1. 1.100071 北京市丰台区方庄社区卫生服务中心
    2.100083 北京市,国家卫生健康委能力建设和继续教育中心(国家卫生健康委党校)
    3.100077 北京市丰台区西罗园社区卫生服务中心
  • 收稿日期:2026-06-23 修回日期:2026-09-02 出版日期:2026-11-05 发布日期:2026-10-10
  • 通讯作者: 任静林

  • 作者贡献:

    魏学娟、任静林提出主要研究目标,负责研究的构思与设计、研究的实施、统计分析,撰写论文;王利华、闫海青、于海洋、刘景峰进行数据的收集与整理;王丽负责文章的质量控制与审查,对文章整体负责,监督管理。

  • 基金资助:
    首都卫生发展全科医学与社区卫生科研专项(首发2023-3S-001)

Effectiveness of an Intelligent Voice Outbound Call System as an Adjunct to Usual Care for Health Management in Older Community-dwelling Patients with "Three Highs" Multimorbidity: a Randomized Controlled Trial

WEI Xuejuan1, WANG Lihua1, WANG Li1, REN Jinglin2,*(), YAN Haiqing1, YU Haiyang3, LIU Jingfeng3   

  1. 1. Fangzhuang Community Health Service Center of Fengtai District, Beijing 100071, China
    2. National Health Commission Capacity Building and Continuing Education Center (National Health Commission Party School), Beijing 100083, China
    3. Xiluoyuan Community Health Service Center of Fengtai District, Beijing 100077, China
  • Received:2026-06-23 Revised:2026-09-02 Published:2026-11-05 Online:2026-10-10
  • Contact: REN Jinglin

摘要: 背景 我国社区老年"三高"(高血压、2型糖尿病、高脂血症)共病患者逐年增多,多重用药与行为管理难度大,传统慢性病随访模式难以满足其整合管理需求。智能语音外呼已在单病种管理中初步证实可行性,但针对"三高"共病的整合型干预,目前相关随机对照研究仍较少,其效果尚不明确。 目的 探讨智能语音外呼辅助传统干预对社区老年"三高"共病患者健康认知、行为依从性及临床指标的影响。 方法 选取2025年1—3月在北京市丰台区方庄社区卫生服务中心和西罗园社区卫生服务中心门诊就诊的"三高"患者200例作为研究对象,采用区组随机化法,通过计算机生成的随机序列,按1∶1比例分为干预组和对照组,每组100例。对照组接受社区常规慢性病管理,干预组在此基础上增加智能语音外呼辅助管理,每季度进行1次语音随访,内容包括健康知识问答、自我监测提醒、生活方式指导及用药确认,通话后对回答错误者即时推送包含个人控制目标的短信,未接通者由家庭医生团队人工补呼,干预12个月。观察指标包括健康认知(血压、血糖、血脂控制目标知晓率)、行为依从性(血压监测、血糖监测、饮食管理、运动管理、规律用药)及临床指标[BMI、腰臀比(WHR)、收缩压(SBP)、舒张压(DBP)、空腹血糖(FPG)、糖化血红蛋白(HbA1c)、总胆固醇(TC)、甘油三酯(TG)、低密度脂蛋白胆固醇(LDL-C)]。 结果 干预后,干预组血压控制目标、血糖控制目标、血脂控制目标知晓率及血压监测、饮食管理、运动管理依从性均高于对照组(P<0.05);干预组BMI、WHR、DBP、HbA1c及TC均低于对照组(P<0.05)。 结论 智能语音外呼辅助传统干预可改善社区老年"三高"共病患者的健康认知、行为依从性及BMI、WHR、DBP、HbA1c等临床指标,该模式有望成为社区慢性病高效管理的辅助工具,但其长期效果仍需进一步验证。

关键词: 慢性病共病, 代谢性疾病, 远程管理, 人工智能语音系统, 随机对照试验

Abstract:

Background

The number of community-dwelling elderly patients with multimorbidity of the "three highs" (hypertension, type 2 diabetes, and hyperlipidemia) is increasing annually in China. These patients face substantial challenges, including polypharmacy and difficulties in behavioral self-management, and traditional chronic disease follow-up models are insufficient to meet their needs for integrated care. Although intelligent voice outbound call systems have demonstrated preliminary feasibility in single-disease management, randomized controlled trials specifically targeting integrated interventions for patients with these three concurrent conditions remain scarce, and their effectiveness is still unclear.

Objective

To investigate the effectiveness of an intelligent voice outbound call system as an adjunct to traditional intervention on health cognition, behavioral adherence, and clinical indicators in community-dwelling elderly patients with "three highs" comorbidity.

Methods

A total of 200 patients with "three highs" comorbidity who attended outpatient clinics at the Fangzhuang Community Health Service Center and Xiluoyuan Community Health Service Center in Fengtai District, Beijing, from January to March 2025 were enrolled. Using a block randomization method with a computer-generated random sequence, they were assigned at a 1∶1 ratio to an intervention group and a control group, with 100 patients in each group. The control group received routine community-based chronic disease management, while the intervention group received additional intelligent voice outbound call-assisted management on top of the usual care. Voice follow-ups were conducted once per quarter over the 12-month intervention period, covering health knowledge quizzes, self-monitoring reminders, lifestyle guidance, and medication confirmation. After each call, patients who answered incorrectly received instant text messages containing their personalized control targets; those who could not be reached were manually re-contacted by the family doctor team. Outcome measures included health cognition (awareness rates of blood pressure, blood glucose, and lipid control targets), behavioral adherence (adherence to blood pressure monitoring, blood glucose monitoring, dietary management, physical activity, and regular medication), and clinical indicators (BMI, WHR, SBP, DBP, FPG, HbA1c, TC, TG, and LDL-C).

Results

After the intervention, the intervention group demonstrated significantly higher awareness rates of blood pressure, blood glucose, and lipid control targets, as well as better adherence to blood pressure monitoring, blood glucose monitoring, dietary management, and physical activity compared with the control group (all P<0.05). Furthermore, the intervention group had significantly lower BMI, WHR, DBP, HbA1c, and TC levels than the control group (all P<0.05).

Conclusion

The intelligent voice outbound call system, when used as an adjunct to usual care, can improve health cognition, behavioral adherence, and several clinical indicators, including BMI, WHR, DBP, and HbA1c, in community-dwelling elderly patients with "three highs" comorbidity. This model shows promise as an efficient auxiliary tool for community-based chronic disease management, although its long-term effects require further validation.

Key words: Multiple chronic conditions, Metabolic diseases, Remote management, Artificial intelligence voice system, Randomized controlled trial