心理社会及社会经济因素最能预测间歇性跛行患者的健康状况
在国际多中心PORTRAIT注册中,1073例有症状外周动脉疾病(PAD)患者于基线和1年时用外周动脉问卷(PAQ)测量健康状况,采用随机森林算法评估47项患者特征与1年PAQ评分的关联,并用分层多变量线性回归建模。预测1年PAQ总结评分最重要的变量为基线PAQ评分、抑郁与焦虑评分、新发症状表现、保险状态、当前或既往抑郁诊断、低社会支持、初始侵入性治疗、症状持续时间和种族。加入19项临床变量后模型解释方差仅从R² 0.312 marginally升至0.335。
要点
- 1年PAD特异性健康状况可由10项以心理社会及社会经济为主的患者特征预测。
- 预测1年PAQ总结评分最重要的变量包括基线PAQ评分、抑郁评分和焦虑评分。
- 加入19项临床变量后R²仅从0.312升至0.335。
核心结论
- 患者1年PAD特异性健康状况(以PAQ测量)可由10项以心理社会及社会经济为主的患者特征预测,包括抑郁、焦虑、保险状态、社会支持和症状。
- 这些特征应被验证并在其他PAD队列中测试,以使该模型可为比较效果研究和基于绩效的医疗中的风险调整及PAD健康状况预测提供信息。
研究设计与样本
- 国际多中心PORTRAIT注册中连续纳入1073例有症状PAD患者,于基线和1年时用外周动脉问卷(PAQ)测量健康状况。
- 采用随机森林算法评估47项患者特征与1年PAQ评分的关联,保留有临床意义的变量并纳入分层多变量线性回归模型预测1年PAQ总结评分。
主要发现
- 患者平均年龄67.7 ± 9.3岁,37%为女性。
- 预测1年PAQ总结评分重要性排名最高的变量为基线PAQ总结评分、患者健康问卷-8抑郁评分、广泛性焦虑障碍-2焦虑评分、新发症状表现、保险状态、当前或既往抑郁诊断、低社会支持、初始侵入性治疗、症状持续时间和种族。
- 扩展模型加入19项临床变量后,1年健康状况的解释方差仅轻微改善,R²从0.312升至0.335。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: As a key treatment goal for patients with symptomatic peripheral artery disease (PAD), improving health status has also become an important end point for clinical trials and performance-based care. An understanding of patient factors associated with 1-year PAD health status is lacking in patients with PAD. METHODS: The health status of 1073 consecutive patients with symptomatic PAD in the international multicenter PORTRAIT (Patient-Centered Outcomes Related to Treatment Practices in Peripheral Arterial Disease: Investigating Trajectories) registry was measured at baseline and 1 year with the Peripheral Artery Questionnaire (PAQ). The association of 47 patient characteristics with 1-year PAQ scores was assessed using a random forest algorithm. Variables of clinical significance were retained and included in a hierarchical multivariable linear regression model predicting 1-year PAQ summary scores. RESULTS: The mean age of patients was 67.7 ± 9.3 years, and 37% were female. Variables with the highest importance ranking in predicting 1-year PAQ summary score were baseline PAQ summary score, Patient Health Questionnaire-8 depression score, Generalized Anxiety Disorder-2 anxiety score, new onset symptom presentation, insurance status, current or prior diagnosis of depression, low social support, initial invasive treatment, duration of symptoms, and race. The addition of 19 clinical variables in an extended model marginally improved the explained variance in 1-year health status (from R2 0.312 to 0.335). CONCLUSIONS: Patients' 1-year PAD-specific health status, as measured by the PAQ, can be predicted from 10 mostly psychosocial and socioeconomic patient characteristics including depression, anxiety, insurance status, social support, and symptoms. These characteristics should be validated and tested in other PAD cohorts so that this model can inform risk adjustment and prediction of PAD health status in comparative effectiveness research and performance-based care.
引用
- Scierka LE, Peri-Okonny PA, Romain G, et al. Psychosocial and socioeconomic factors are most predictive of health status in patients with claudication. J Vasc Surg. 2024. PMID: 38266885. doi:10.1016/j.jvs.2024.01.021
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。