| ObjectivesThe aim of this study is to describe the distribution level and potential growth trajectory of mobile phone dependency(MPD)and depressive symptoms among college students,and to assess the direction between MPD and depressive symptoms.To detect the relationship between MPD and brain structure and reward circuit related functions,and to explore the moderating effects of brain structures and reward-related functional changes in the association between MPD and depressive symptoms.MethodsThis study was based on the cohort study of“Behaviors and Health Among College Students”from April 2019 to May 2020.The school-based survey was used among freshmen from 4 schools in Anhui province and Jiangxi province based purposive sampling method.The study was taken two follow-up surveys and the follow-up interval was 6 months.The baseline survey mainly collected general demographic data,mobile phone dependence,mental sub-health and depressive symptoms,while the follow-up survey mainly collected the mobile phone dependence,mental sub-health and depressive symptoms.In addition,the students of two schools in Anhui province were taken MRI scan and followed up scan(resting-state f MRI,structural MRI and DTI)in 12 months later according to the principle of informed consent and voluntariness.This study was approved by the ethics committee of Anhui Medical University(NO:20170291),and participants could withdraw from the study at any time.A total of 1 092 subjects with three complete data were selected from baseline and two follow-up surveys.Descriptive statistical analysis was used to describe the distribution and detection rate of MPD and depressive symptoms in baseline and follow-up surveys.Multivariate logistic regression was used to analyze the association between general demographic characteristics,smoking,alcohol consumption,sleep quality and other confounding factors,as well as MPD with depressive symptoms.Latent growth curve model(LGCM)was used to analyze the growth trajectories of MPD and depressive symptoms with 3 repeated measurements in Mplus software.Autoregressive cross-lagged model(ARCLM)was used to analyze the relationship between MPD and depressive symptoms.The MRI data analyzed 268 students who completed MRI examination at baseline and 151 students who participated in the follow-up examination.For brain structures(n=268),we used VBM and TBSS methods to explore the relationship between brain gray matter volume and white matter integrity with MPD based on baseline data.For reward circuit related functions mainly used ROI-wise FC method,we took two-sample t-test to explore the difference between MPD and Non-MPD group at baseline,and then extracted reward circuit related FC both in baseline and 12 months follow-up(n=151)based the two-sample t-test results.At last,we defined the functional changes of reward circuit by subtracting FC after follow-up from FC at baseline.The moderating effect of reward circuit related structure and the functional changes on the association between MPD and depressive symptoms was analyzed by PROCESS plug-in of SPSS software.We calculated the explanatory rate of the interaction effect between moderating variable and independent variable on dependent variables through the change of model determination coefficient(R~2).ResultsIn the baseline survey,the detection rate of MPD and depressive symptoms was 24.5%and 44.2%,and the detection rate of MPD and depressive symptoms was 27.7%and 32.1%after the 6 months follow-up,and the detection rate of MPD and depressive symptoms was 35.6%and 24.3%after the 12 months follow-up,the detection rate of MPD showed an increasing trend.The fitting results of the latent growth trajectory showed that there were individual differences in the initial state of MPD among college students(intercept factor I=34.42,P<0.01),but there was no statistical difference in the rate of change.The depressive symptoms not only had significant individual differences in the initial state and rate of change(intercept factor I=13.44,slope factor S=1.93,P<0.01),but also had a correlation between I and S(r=-0.61,P<0.01),suggesting that there was a significant negative correlation between individual growth rate and initial state,which was suggest that individuals with high initial depression score had a slower decline.In the baseline,multivariate logistic regression analysis found that there was a positive correlation between MPD and depressive symptoms(?=0.27;95%CI:0.25,0.30;P<0.01).Combined with the follow-up data,using the ARCLM,we found that there was a positive correlation between the previous MPD and the subsequent depressive symptoms(P<0.01),and also found that there was a positive correlation between the previous depressive symptoms and the subsequent MPD(P<0.01),indicating that the correlation between MPD and depressive symptoms may be bidirectional relation.At baseline,VBM results showed that MPD was negatively correlated with GMV of anterior cingulate gyrus(ACC)and right fusiform gyrus(P<0.05,FDR corrected).TBSS results showed that there was a significant negative correlation between MPD and FA value of corpus callosum(P<0.05,FWE corrected).No correlation was found between MPD and other indexes(MD,AD,RD).In addition,there were differences in FC between three nodes of reward circuit and specific brain regions between MPD and Non-MPD group(P<0.05,FWE corrected).The first was FC between left inferior frontal gyrus(IFG,No.29 in the template)and left inferior occipital gyrus(IOG),the second was FC between left IFG(No.39 in the template)and left occipital polar cortex(OPC)and right middle frontal gyrus(MFG),and the third was FC between left parahippocampal gyrus(Ph G,No.115 in the template)and right middle temporal gyrus(MTG),left cuneus(Cun),right precuneus(Pcun),left superior temporal gyrus(STG),right supramarginal gyrus(SMG)and right precentral gyrus(Pr G).In the analysis of moderating effect,GMV of ACC could moderate the association between MPD and depressive symptoms at 12 months later(?=0.08,P<0.01),the explanatory rate of moderating effect was 3.30%.FC between the left IFG and the right IOG moderated the association between MPD and depressive symptoms at 12 months later(?=0.36,P<0.01),and the explanatory rate of the moderating effect was 5.96%;FC between the left IFG and the left OPC moderated the association between MPD and depressive symptoms at 12 months later(?=0.48,P<0.01),and the explanatory rate of the moderating effect was 6.17%;FC between the left IFG and the right MFG moderated the association between MPD and depressive symptoms at 12 months later(?=0.48,P<0.05),and the explanatory rate of the moderating effect was 3.30%;FC between the left Ph G and the right MTG moderated the association between MPD and depressive symptoms at 12 months later(?=0.53,P<0.01),and the explanatory rate of the moderating effect was 5.88%;FC between the left Ph G and the left Cun moderated the association between MPD and depressive symptoms at 12 months later(?=0.57,P<0.01),and the explanatory rate of the moderating effect was 8.55%;FC between the left Ph G and the right Pcun moderated the association between MPD and depressive symptoms at 12 months later(?=0.53,P<0.01),and the explanatory rate of the moderating effect was 7.31%;FC between the left Ph G and the left STG moderated the association between MPD and depressive symptoms at 12 months later(?=0.48,P<0.01),and the explanatory rate of the moderating effect was 5.09%;FC between the left Ph G and the right SMG moderated the association between MPD and depressive symptoms at 12 months later(?=0.57,P<0.05),and the explanatory rate of the moderating effect was 4.21%;FC between the left Ph G and the right Pr G moderated the association between MPD and depressive symptoms at 12 months later(?=0.59,P<0.01),and the explanatory rate of the moderating effect was 5.26%.Conclusions1.The results suggest common MPD and high depressive symptoms in college students,and MPD increases with two follow-up survey.2.There is non-linear correlation between MPD and depressive symptoms,and there is a bidirectional association between MPD and depressive symptoms in college students.3.There is a negative correlation between MPD and GMV of ACC,it means that the more severe MPD and the smaller GMV of the ACC.There is a negative correlation between MPD and the FA value of corpus callosum,it means that the more severe MPD and the worse integrity of the white matter.There is a correlation between MPD and the reward circuit related function,it means that the FC of the left IFG and left Ph G with specific brain regions were positively correlated with MPD change.These results suggest that brain structure(gray matter volume and white matter integrity)and reward circuit related function are associated with MPD.4.GMV of the ACC is negatively moderated the association between MPD and depressive symptoms at 12 months later.The changing FC of the left IFG and the left Ph G with specific brain regions are positively moderated the association between MPD and depressive symptoms at 12 months later in college students.These results suggest that brain structure and reward circuit related functional changes play a moderating role in the association between MPD and depressive symptoms,the moderating effect was from 3.30%to 8.55%.5.There is an association between MPD and depressive symptoms in college students,and this association is moderated by changes in brain structure and reward circuitry related functions.It provides evidence for the possible neural mechanism of the association between behavior and mental health of college students. |
PDF Full Text Request