南方医科大学学报 ›› 2022, Vol. 42 ›› Issue (7): 1038-1043.doi: 10.12122/j.issn.1673-4254.2022.07.11

• • 上一篇    下一篇

我国围绝经期女性抑郁症状变化及影响因素分析:基于CHARLS面板数据

李佳欣,肖 艳,廖 娟,杨春霞   

  1. 四川大学华西公共卫生学院//华西第四医院流行病与卫生统计学系,四川 成都 610041;华西-协和陈志潜卫生健康研究院慢性病研究中心,四川 成都 610041
  • 出版日期:2022-07-20 发布日期:2022-07-15

Temporal trend and contributing factors of depressive symptoms in Chinese menopausal women: analysis based on CHARLS panel data

LI Jiaxin, XIAO Yan, LIAO Juan, YANG Chunxia   

  1. Department of Epidemiology and Health Statistics, West China School of Public Health/West China Fourth Hospital, Sichuan University, Chengdu 610041, China; Non-communicable Diseases Research Center, West China-PUMC C.C. Chen Institute of Health, Sichuan University, Chengdu 610041, China
  • Online:2022-07-20 Published:2022-07-15

摘要: 目的 了解我国围绝经期女性抑郁症状变化情况及影响因素,从而为制定防治策略提供依据。方法 利用CHARLS数据选取年龄45~60岁关键变量值完整的围绝经期女性作为调查对象,去除缺失值,得出4年的截面数据,分别为4318、4200、3930和4147份。通过截面数据匹配面板数据,得到4期完整的追踪数据5040份,构成n=1260,T=4的平衡短面板数据集。利用面板数据分析围绝经期女性抑郁症状检出率及变动情况,采用面板二值选择模型的随机效应Logistic回归模型分析围绝经期女性抑郁症状的影响因素。结果 不同年份追踪样本的围绝经期女性的抑郁症状检出率分别为35.9%,33.1%,36.7%,43.7%,追踪数据检出率的变化趋势不具有统计学意义(APC=3.25%,P=0.183)。回归结果显示:城镇地区OR=0.570(0.457,0.710)、文化程度越高 OR=0.759(0.655,0.879)、有配偶 OR=0.363(0.236,0.558)是围绝经期女性抑郁症状的保护因素;自评健康差 OR=2.704(2.152,3.396)、有残疾 OR=1.457(1.087,1.954)、有慢性疾病 OR=1.407(1.179,1.680)、最近两年有摔倒 OR=2.028(1.613,2.550)、睡眠时长不正常OR=2.249(1.898,2.664)、生活较不满意OR=4.803(3.757,6.140)是围绝经期女性抑郁症状的危险因素。结论 我国围绝经期女性抑郁症状检出水平较高,应加强对农村地区,文化程度低,无配偶,自评健康差,有残疾,有慢性疾病,最近两年有摔倒,睡眠时长不正常,生活较不满意等人群的心理保健和控制措施。

关键词: 围绝经期;抑郁;影响因素;面板数据

Abstract: Objective To understand the temporal trend of and the factors affecting depressive symptoms in Chinese menopausal women to provide evidence for the development of prevention and treatment strategies. Methods CHARLS data were used to select menopausal women aged 45-60 years. Complete values of the key variables were screened and missing values were removed to obtain the cross-sectional data of the years 2011 (n=4318), 2013 (n=4200), 2015 (n=3930), and 2018 (n=4147). The panel data were matched by the cross-sectional data, and a total of 5040 cases with complete record of the follow-up data were obtained for the 4 years to constitute a balanced short panel dataset with n=1260 and T=4. The prevalence and temporal trend of depressive symptoms in the menopausal women were analyzed based on the panel data. The random-effects Logit model with a panel dichotomous choice model was used to explore the factors affecting depressive symptoms in the menopausal women. Results The prevalence of depressive symptoms in the menopausal women calculated based on the panel data was 35.9%, 33.1%, 36.7% and 43.7% in the 4 years, respectively, showing no statistically significant changes in the temporal trend (APC=3.25%, P=0.183). The results of the random-effects Logit model analysis showed that living in the urban area (OR=0.570, 95%CI: 0.457-0.710), a high education level (OR=0.759, 95%CI: 0.655-0.879), and having a spouse (OR=0.363, 95% CI: 0.236-0.558) were associated with a decreased incidence of depressive symptoms, while poor self-reported health (OR=2.704, 95% CI: 2.152-3.396), disability (OR=1.457, 95%CI: 1.087-1.954), chronic disease (OR=1.407, 95% CI: 1.179-1.680), falls in the last two years (OR=2.028, 95% CI: 1.613-2.550), abnormal sleep duration (OR=2.249, 95% CI: 1.896-2.664), and dissatisfaction with life (OR=4.803,95% CI: 3.757-6.140) were associated with an increased incidence of depressive symptoms. Conclusion The prevalence of depressive symptoms is relatively high in menopausal women in China. Measures should be taken to ensure that the menopausal women living in rural areas, with low education level, without spouse, with a poor self-reported health status, disability, chronic diseases, falls in recent two years, abnormal sleep time and dissatisfaction with life have access to psychological health care services and interventions.

Key words: menopause; depression; risk factors; panel data