Abstract: Objective  To analyze the epidemiological characteristics of stroke among the elderly in Jing’an District, Shanghai, to explore the independent and interactive effects of sleep quality and smoking on stroke in the elderly, and to calculate the population attributable fraction (PAF), so as to provide a scientific basis for formulating primary prevention strategies for stroke in the elderly.
Methods  A cross-sectional survey was conducted among 2380 permanent residents (residing continuously in Jing’an District for more than 6 months) aged≥60 years in Jing’an District, Shanghai, from April to July 2025 using a multistage stratified random sampling method. Information such as demographic characteristics, stroke prevalence, smoking status, and chronic disease history was collected. The sleep quality of the study participants was assessed using the Pittsburgh sleep quality index (PSQI). Multivariate logistic regression was used to analyze influencing factors of stroke in the elderly and to test the interaction between poor sleep quality (PSQI total score>7 points) and smoking. The PAF of poor sleep quality and smoking, as well as their combined PAF for stroke in the elderly, was calculated. Two sensitivity analyses were conducted to test the robustness of the results, including adjusting the cutoff for poor sleep quality (PSQI total score>5 points) and excluding patients with atrial fibrillation.
Results  A total of 224 stroke patients were included, with a prevalence rate of 9.41%. There were 751 participants (31.55%) with poor sleep quality and 587 smokers (24.66%). After adjusting for confounding factors including age and hypertension history, multivariate logistic regression showed that poor sleep quality (OR 2.782, 95%CI 2.082-3.717, P<0.001) and smoking (OR 1.538, 95%CI 1.116-2.120, P=0.009) were independent risk factors for stroke in the elderly, with no significant interaction between the two factors (Pfor interaction=0.791). PAF analysis showed that the PAF of poor sleep quality was 35.99%, the PAF of smoking was 11.71%, and the combined PAF was 43.49%. Sensitivity analyses confirmed robust associations between poor sleep quality and stroke (OR 2.29-3.01), with a combined PAF of 44.02%-45.57%, further confirming the robustness of the results.
Conclusions  The prevalence of stroke is relatively high among the elderly in Jing’an District, Shanghai. Poor sleep quality and smoking are important risk factors for stroke in this population. This provides a scientific basis for integrating sleep health management into the routine health management system for the elderly and promoting it jointly with tobacco control measures to reduce the stroke burden in this population.

Key words: Stroke; Smoking; Sleep quality; Population attributable fraction

摘要： 目的 分析上海市静安区老年卒中的流行特征，探讨睡眠质量、吸烟对老年卒中的独立及交互作用，并计算其人群归因分数（population attributable fraction，PAF），为老年卒中一级预防策略的制订提供科学依据。
方法 采用多阶段分层随机抽样方法，于2025年4—7月对上海市静安区2380名≥60岁的常住居民（在静安区连续居住6个月以上）开展横断面调查，收集研究对象的人口学特征、卒中患病情况、吸烟状态、慢性病病史等信息。采用匹兹堡睡眠质量指数量表（Pittsburgh sleep quality index，PSQI）评估研究对象的睡眠质量。运用多因素logistic回归分析探讨老年卒中的影响因素，检验睡眠质量差（PSQI总分＞7分）与吸烟的交互作用。计算睡眠质量差、吸烟对老年卒中的PAF及联合PAF。通过调整睡眠质量差的判定标准（PSQI总分>5分）及排除心房颤动患者等2项敏感性分析进一步检验结果的稳健性。
结果 共纳入卒中患者224例，患病率为9.41%，睡眠质量差者751名（31.55%），吸烟者587名（24.66%）。多因素logistic回归分析结果显示，在校正年龄、高血压病史等混杂因素后，睡眠质量差（OR 2.782，95%CI 2.082～3.717，P<0.001）与吸烟（OR 1.538，95%CI 1.116～2.120，P=0.009）是老年卒中的独立危险因素，且二者无显著交互作用（P交互=0.791）。PAF分析结果显示，睡眠质量差的PAF为35.99%，吸烟的PAF为11.71%，联合PAF为43.49%。敏感性分析结果显示，睡眠质量差与老年卒中的关联结果稳健（OR 2.29～3.01），其与吸烟的联合PAF为44.02%～45.57%，进一步验证了结果的稳健性。
结论 上海市静安区老年卒中的患病率较高，睡眠质量差与吸烟是老年卒中的重要危险因素。这为将睡眠健康管理整合到老年人群常规健康管理体系，并与控烟措施联合推进，以降低该人群的卒中疾病负担提供了科学依据。

关键词: 卒中; 吸烟; 睡眠质量; 人群归因分数