Objective: To explore the direction of the association between smartphone multitasking behavior and comorbid symptoms of anxiety and depression (CAD) among college students. Methods: College students from one college located in Shanxi, Chongqing, and Shenzhen were selected between October and December 2021 using a multistage random cluster sampling method, and a follow-up visit was conducted in May 2022. The Assessment of Smartphone Multitasking for Adolescents, the Patient Health Questionnaire-9 Items, and the Generalized Anxiety Disorder Questionnaire-7 Items were used to assess the smartphone multitasking behaviors and CAD of college students. Multivariate logistic regression models were used to analyze the associations between smartphone multitasking behavior and CAD. Autoregressive cross-lagged models (ARCLM) were used to analyze the bidirectional associations between smartphone multitasking behavior and CAD. Results: A total of 953 college students were included in this study, including 323 (33.9%) males and 630 (66.1%) females, with an age of (18.89±1.33) years. The detection rates of CAD at baseline and follow-up were 25.5% and 27.5%, respectively. The mean of the total smartphone multitasking index at baseline was 2.11±0.48. The mean of smartphone and non-media activity multitasking index was 2.00±0.55. The mean of smartphone and other media activity multitasking index was 1.92±0.67. The mean of smartphone functional use multitasking index was 2.18±0.54. The multivariate logistic regression model showed that smartphone and non-media activity multitasking index (OR=1.85, 95%CI: 1.39-2.45), smartphone and other media activity multitasking index (OR=1.42, 95%CI: 1.10-1.83), and total smartphone multitasking index (OR=1.66, 95%CI: 1.21-2.28) were positively associated with the risk of CAD among college students at baseline. The smartphone and non-media activity multitasking index (OR=1.74,95%CI: 1.30-2.32), smartphone and other media activity multitasking index (OR=1.37, 95%CI: 1.04-1.79) and total smartphone multitasking index (OR=1.48, 95%CI: 1.06-2.06) were positively associated with the risk of CAD at follow-up. There was no statistical association between smartphone functional use multitasking index at baseline and CAD. The ARCLM showed that smartphone and non-media activity multitasking index (β=0.974, P=0.029), smartphone and other media multitasking index (β=0.798, P=0.003), and the total smartphone multitasking index (β=1.379, P=0.011) could positively predict the risk of CAD at follow-up. The CAD at baseline could positively predict smartphone and other media multitasking index (β=0.004, P=0.016) and mobile phone functional use multitasking index at follow-up (β=0.004, P=0.016). Conclusion: There is a bidirectional association between smartphone and other media activity multitasking and CAD among college students.
目的： 探讨大学生手机多任务行为与焦虑抑郁症状共患（CAD）的关联方向。 方法： 于2021年10—12月采用多阶段随机整群抽样方法，选取山西、重庆、深圳3个地区各1所高等院校的大学生作为调查对象，于半年后开展随访。采用《青少年手机多任务行为评定问卷》《患者健康问卷—9项》《广泛性焦虑障碍问卷—7项》评估大学生手机多任务行为和CAD。采用多因素logistic回归分析基线大学生手机多任务行为与CAD之间的关联；采用自回归交叉滞后模型分析大学生手机多任务行为与CAD的双向关联。 结果： 本研究共纳入953名大学生，男生323名（33.9%），女生630名（66.1%），年龄为（18.89±1.33）岁。基线和随访时CAD检出率分别为25.5%和27.5%；基线平均总手机多任务指数为（2.11±0.48），手机和非媒体活动多任务指数为（2.00±0.55），手机和其他媒体活动多任务指数为（1.92±0.67），手机功能使用多任务指数（2.18±0.54）。多因素logistic回归模型分析结果显示，基线手机和非媒体活动多任务指数（OR=1.85，95%CI：1.39~2.45）、手机和其他媒体活动多任务指数（OR=1.42，95%CI：1.10~1.83）以及总手机多任务指数（OR=1.66，95%CI：1.21~2.28）与大学生基线CAD发生风险呈正相关；基线手机和非媒体活动多任务指数（OR=1.74，95%CI：1.30~2.32）、手机和其他媒体活动多任务指数（OR=1.37，95%CI：1.04~1.79）以及总手机多任务指数（OR=1.48，95%CI：1.06~2.06）与随访时CAD发生风险呈正相关。基线手机功能使用多任务指数与CAD发生无统计学关联。自回归交叉滞后模型分析显示，基线手机和非媒体活动多任务指数（β=0.974，P=0.029）、手机和其他媒体多任务指数（β=0.798，P=0.003）及总手机多任务指数（β=1.379，P=0.011）均可正向预测随访时CAD；基线CAD可正向预测随访时手机和其他媒体多任务指数（β=0.004，P=0.016）和手机功能使用多任务指数（β=0.004，P=0.016）。 结论： 大学生手机和其他媒体多任务行为与CAD呈双向关联。.