Lung cancer is the leading cause of cancer-related deaths in China. Early screening in high-risk populations is the most effective strategy for detecting lung cancer and improving patient prognosis. However, there is limited research on the factors influencing the screening behaviors of first-degree relatives of lung cancer patients, who represent a concentrated high-risk group.

This study aims to analyze the factors influencing early lung cancer screening behaviors among first-degree relatives of lung cancer patients using multifactorial Logistic regression analysis and decision tree models.

From March to June 2023, a convenience sampling method was employed to select 310 first-degree relatives of lung cancer patients currently hospitalized in the respiratory and thoracic surgery departments of Shanxi Province Cancer Hospital. Data were collected through a general information questionnaire, a lung cancer knowledge questionnaire, a Chinese version of the lung cancer screening health belief scale, and a cancer worry scale. Factors affecting lung cancer screening behavior were analyzed, and both Logistic regression and decision tree models were established to compare their predictive effectiveness.

The screening rate among first-degree relatives of lung cancer patients was 23.9% (74/310) . The overall lung cancer knowledge awareness rate was 75.2% (4 662/6 200) , and 56.5% (175/310) expressed a willingness to undergo lung cancer screening. Multifactorial Logistic regression analysis identified age, type of medical insurance, receipt of lung cancer screening recommendations, willingness to screen, perceived barriers, perceived benefits, and perceived risks as significant influencing factors for screening behavior (P<0.05) . The decision tree model highlighted perceived barriers and age as key determinants of screening behavior. A statistically significant difference was observed in the predictive effectiveness between the Logistic regression and decision tree models (P<0.05) .

The overall level of lung cancer knowledge among first-degree relatives of lung cancer patients is relatively high, but awareness of screening methods is low. Health beliefs and willingness to screen are moderate, while the actual screening rate remains low. Healthcare professionals should utilize both models to facilitate the development of accurate screening awareness and health beliefs among first-degree relatives, thereby promoting their screening behaviors.

肺癌是导致中国癌症死亡的首要原因。高危人群早期筛查是发现肺癌，改善预后最有效且关键的方法。目前对于肺癌患者一级亲属这一集中高危群体的筛查行为影响因素研究少有报道。

采用多因素Logistic回归分析和决策树模型分析肺癌患者一级亲属的肺癌早期筛查行为影响因素。

采用便利抽样法，选取2023年3—6月山西省肿瘤医院呼吸内科与胸外科住医院治疗的310名肺癌患者一级亲属为调查对象。采用一般资料调查表、肺癌知识问卷、中文版肺癌筛查健康信念和癌症担忧量表进行调查。分析影响肺癌筛查行为的因素，分别建立Logistic回归模型与决策树模型，并比较2种模型的预测效果。

肺癌患者一级亲属接受过肺癌筛查率为23.9%（74/310），总体肺癌知识知晓率为75.2%（4 662/6 200），56.5%（175/310）愿意接受肺癌筛查。多因素Logistic回归分析结果显示年龄、医疗保险类型、是否获得过肺癌筛查建议、筛查的意愿、感知障碍、感知效益、感知风险是肺癌患者一级亲属肺癌筛查行为的影响因素（P<0.05），决策树模型结果显示感知障碍和年龄是一级亲属肺癌筛查行为的影响因素，Logistic回归模型与决策树模型的预测效果比较，差异有统计学意义（P<0.05）。

肺癌患者一级亲属的整体肺癌认知水平较高但对筛查手段的认知较低，健康信念及筛查意愿处于中等水平，肺癌筛查率偏低。医务人员可联合应用两种模型，采取措施使一级亲属建立正确的筛查认知与健康信念，以期促进一级亲属的筛查行为。