Influencing factors and predictive model construction of comorbidity of myopia and depression among middle school students
10.3980/j.issn.1672-5123.2026.5.25
- VernacularTitle:中学生近视与抑郁共患的影响因素及预测模型构建
- Author:
Hao SUN
1
;
Dongyang WANG
1
;
Wangcheng ZHENG
1
;
Jiaxiang ZHANG
1
Author Information
1. School of Public Health, Anhui Medical University, Hefei 230032, Anhui Province, China; Department of Public Health, Feidong Center for Disease Control and Prevention, Feidong County 231600, Anhui Province, China
- Publication Type:Journal Article
- Keywords:
myopia;
depression;
comorbidity;
middle school students;
predictive model
- From:
International Eye Science
2026;26(5):879-887
- CountryChina
- Language:Chinese
-
Abstract:
AIM: To investigate the comorbidity status of myopia and depressive symptoms among middle school students, identify key influencing factors, and establish a prediction model, thereby providing empirical evidence for the comprehensive intervention of these two conditions.METHODS: Students from 3 middle schools in Feidong county were recruited between 2022 and 2024. Myopia was defined as uncorrected visual acuity ≤5.0 with spherical equivalent refraction <-0.50 diopters(D). Depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale(CES-D), with a score ≥16 indicating the presence of depressive symptoms. A database was established and data were entered using EpiData software. Pearson's Chi-square test and multivariate Logistic regression analysis were performed to identify influencing factors and screen variables with R statistical software(version 4.5.2). Finally, a Stacking ensemble prediction model was constructed using Python3.13 software. RESULTS: The study included 2 476 students, consisting of 1 380 males and 1 096 females. The overall detection rate of myopia-depressive symptom comorbidity among the studied students was 14.54%. Univariate analysis showed that variables were significantly associated with the comorbidity, including family structure, grade level, sugar-sweetened beverage intake, exercise frequency, school bullying, and parental physical or verbal abuse(all P<0.05). Multivariate Logistic regression analysis identified the following risk factors: higher grade levels(8th grade: OR=1.9143, 95%CI: 1.1096-3.3024; 9th grade: OR=1.7884, 95%CI: 1.0506-3.0444; 11th grade: OR=2.1847, 95%CI: 1.1980-3.9840; 12th grade: OR=3.4606, 95%CI: 1.8250-6.5621), daily consumption of sugar-sweetened beverages more than once(OR=3.1383, 95%CI: 1.7112-5.7560), low frequency of moderate-to-vigorous exercise on weekends and holidays(mostly achievable: OR=3.3115, 95%CI: 1.009-10.8685), alcohol consumption(OR=4.4021, 95%CI: 2.7383-7.0766), daily sedentary time exceeding 10 h(OR=1.8594, 95%CI: 1.2141-2.8476), lack of puberty education(OR=3.0098, 95%CI: 2.0659-4.3848), and exposure to parental physical or verbal abuse(OR=2.4050, 95%CI: 1.1484-5.0364). Protective factors included no experience of school bullying(OR=0.0055, 95%CI: 0.0002-0.1602), no history of severe injury(OR=0.3118, 95%CI: 0.1823-0.5332), outdoor activities during class breaks(OR=0.1672, 95%CI: 0.0752-0.3719), and moderate after-school homework duration(2-3 h per day: OR=0.4802, 95%CI: 0.2620-0.8801). The constructed Stacking prediction model demonstrated good discriminative ability, with an area under the receiver operating characteristic curve(AUC)of 0.855, a sensitivity of 81.5%, and a specificity of 74.0%. Key predictive factors included alcohol consumption status, location of recess activities, unhealthy lifestyle composite index(interaction term between sedentary duration and sugar-sweetened beverage intake frequency), academic stress index(interaction term between sedentary duration and homework duration), and after-school homework duration.CONCLUSION: The comorbidity of myopia and depression among middle school students is jointly influenced by multiple factors such as lifestyle, academic pressure, and family/campus environment. It is advocated to implement a three-level intervention system that includes restricting the sale of sugar-sweetened beverages, conducting psychological screening for sedentary students, and carrying out family-school-medical collaborative management of drinking behaviors. This model can be applied to school health screening and the early identification of high-risk groups in community adolescent health management. It is suitable for middle school students in regions with similar economic levels, but not applicable to students receiving special education or those with severe organic diseases.
