key: cord-0259297-qcc6jdji authors: Michael, T.; Schäfer, S. K.; Schanz, C. G.; Equit, M. title: Longitudinal association between depressive symptoms and self-directed passive aggression: A random intercept cross-lagged panel analysis date: 2022-04-16 journal: nan DOI: 10.1101/2022.04.12.22273775 sha: 3f309383833399ca449c5ab329f13f5fa6b78eba doc_id: 259297 cord_uid: qcc6jdji Background: Self-directed passive aggression (SD-PAB) is defined as any behaviour harming one-self by inactivity and omission of own needs. Depressive disorders are a severe mental disorder that results from the interaction between stress exposure, coping strategies, and vulnerability. Previous cross-sectional studies found SD-PAB to be associated with depressive symptoms and to represent a mediator of the relationship between cognitive risk factors and depressive symptoms. Therefore, SD-PAB may be a potential target of prevention or treatment in the context of depressive disorders. However, prospective studies on the relationship between depressive symptoms and SD-PAB are lacking. The current study aimed at closing this gap by examining the associations of subjective stress, SD-PAB, and depressive symptoms cross-sectionally and over time. Method: In two assessment cohorts students participated three times [M1: start of the semester ( n = 352); M2: start of the exam period ( n = 293); M3 = end of the exam period ( n = 276)] in an online survey (depressive symptoms; self-perceived stress; SD-PAB). Cross-sectional data was analysed using regression models. Longitudinal data was analysed using Random Intercept Cross-lagged Panel Models. Results: Across all time points, SD-PAB demonstrated a unique cross-sectional association with depressive symptoms when controlled for self-perceived stress ( {beta} = .27 - .33; all p s < .001). Furthermore, at M2 [ {beta} = .14, t (289) = 3.71, p < .001] and M3 [ {beta} = .15, t (272) = 3.51, p < .001] the relationship between depressive symptoms and self-perceived stress was stronger for individuals reporting higher levels of SD-PAB. Depressive symptoms at M1 are a marginal significant predictor of SD-PAB at M2 ( {beta} = .31; p = .067) and depressive symptoms at M2 are a marginal significant predictor for SD-PAB at M3 ( {beta} = .17; p = .074). However, there was no evidence for SD-PAB predicting the course of depressive symptoms. Conclusion: SD-PAB may represent a symptom of depressive disorders and a moderator of unsuccessful stress coping but does not predict the course of depressive symptoms over time. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 16, 2022. ; https://doi.org/10.1101/2022.04.12.22273775 doi: medRxiv preprint 4 49 Based on Kanfer's (36) self-control theory, the self-control model of depression supposes that 50 decreased self-reinforcement and increased self-punishment derive from dysfunctional self-51 monitoring and self-evaluation processes [i.e. due to dysfunctional attitudes and negative 52 attributional style;(35)]. Self-punishment reflects active SD-AB and is seen in patients with DD 53 in form of self-blame, non-suicidal self-harm, and suicidal behaviour (37,38). Reduction of self-54 reinforcement is a form of passive SD-AB and contributes to the core symptoms of DD 55 (depressed mood and loss of interest or pleasure) by the loss of reinforcement (26,27). 56 Therefore, active and passive SD-AB could constitute risk factors for the development of DD. 57 However, one may also assume different associations between SD-AB and DD: Some studies 58 revealed active SD-AB as predictor for DD (39,40), others found DD to be a predictor for active 59 SD-AB (41-44), or DD and active SD-AB to interact (45). However, other studies found cross-60 sectional but not longitudinal associations between active SD-AB and DD (46-48). In sum, 61 there is tentative evidence that DD predicts active SD-AB, which, if confirmed, would indicate 62 the need of prevention strategies for active SD-AB in patients suffering from DD. To the best 63 of our knowledge, to date, no prospective study has investigated the relationship between 64 passive SD-AB and DD. Because insight in the longitudinal association of passive SD-AB and DD 65 could be relevant for prevention and treatment of both passive SD-AB and DD, the current 66 prospective study examined their association over the course of one semester in a student 67 sample. 68 Study aims 69 The current study has three aims: First, to replicate previous findings (15,16) on the 70 association between passive SD-AB and depressive symptoms. Second, to analyse if passive 71 SD-AB moderates the association between subjective stress and depressive symptoms, which 72 would support the notion of SD-AB being a dysfunctional self-regulation strategy. Third, to . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 16, 2022. ; https://doi.org/10.1101/2022.04.12.22273775 doi: medRxiv preprint 5 73 examine if baseline SD-AB and depressive symptoms severity predict changes in one-another 74 in the course of a moderate stressful event (an exam phase). Thereby, the current study will 75 help to clarify the question whether passive SD-AB is a risk-factor for DD or a consequence of 76 DD symptoms. Table 1 . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 16, 2022. ; https://doi.org/10.1101/2022.04.12.22273775 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 16, 2022. ; https://doi.org/10.1101/2022.04.12.22273775 doi: medRxiv preprint 7 114 Psychopathological symptoms. General symptom severity was assessed using the BSCL(52). 115 The BSCL consists of 53 items and its Global Severity Index (GSI) is a reliable and valid measure 116 for general symptom intensity (57,58). The internal consistency of GSI was excellent across all 117 assessments (α = .95 -.96). 118 Self-directed passive-aggressive behaviour. Passive SD-AB was measured using the TPA (15). . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 16, 2022. ; https://doi.org/10.1101/2022.04.12.22273775 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted April 16, 2022. ; https://doi.org/10.1101/2022.04.12.22273775 doi: medRxiv preprint Assessment periods of both waves comprised three time points: At the start of the semester 92 (measure point 1, M1), two weeks before the first exam of the semester Participants 94 received an e-mail as reminder for study participation and completed the online survey. Due 95 to the rise of COVID-19 in Germany all pending exams at the Saarland University were 96 cancelled in March 2020. Therefore, all participants which had not completed their last exam 97 at that date (57.51%) were invited for M3 For all assessments, the online survey platform SoSci Survey (49) was used for data collection After giving written informed consent in accordance with the Declaration of Helsinki (50), the 100 participants completed the short form of the the Trier Inventory for Chronic Stress the Beck Depression Inventory II Perceived subjective stress levels of the last month were assessed using the 105 short form of the Trier Inventory for Chronic Stress The TICS is a valid and reliable 106 measure for perceived stress (54). The TICS-SF comprises 12 items selected from the one 107 factor solution of the long form The 21 items 110 of the BDI-II represent depressive symptoms based on the fourth version of the Diagnostic 111 and Statistical Manual of Mental Disorders The BDI-II is a valid and reliable 112 measure of depressive symptoms (56) Cross-sectional association between self-directed passive-aggressive behaviour and 183 depressive symptoms In line with previous studies (16,59), passive SD-AB was 186 moderate to strongly associated with depressive symptoms, even when general symptom 187 severity was controlled for. Passive SD-AB also accounted for a significant amount of variance 188 in depressive symptoms when controlled for perceived stress. Moreover, individuals with 189 higher levels of passive SD-AB reacted with more severe depressive symptoms to stressful 190 events. Thus The current longitudinal study is the first investigation using state-of-the-art methods for the 195 analysis of prospective data examining whether passive SD-AB and depressive symptoms 196 represent risk factors for each other. The RI-CLPM identified depressive symptoms as a 197 marginally significant predictor of passive SD-AB, but not vice versa. This finding is in line with 198 previous results Therefore, passive SD-AB could be a relevant treatment target in DD We investigated all associations 202 in a student sample, which may have impeded the generalisability of our results, e.g. gender 203 and age are distributed in a non-representative way. Moreover, the variance of clinical 204 measures could have been restricted. However, previous studies identified female gender These findings demonstrate the relevance of SD-AB and DD 207 research in those populations. Furthermore, depressive symptoms at M2 (but not at M1 and 208 M3) matched the clinical cut-off for mild depression demonstrating the comparability of results from student and patient 213 samples. Depressive symptoms were assessed using the BDI-II, a self-report measure for 214 severity of depressive symptoms. The BDI-II is a well validated and widely used measure of 215 depressive symptoms with good sensitivity and specificity (56,67), however it is a screening 216 tool and does not allow for categorical diagnostics of DD. Thus, the current study design allows 217 for the examination of the course of depressive symptoms, but not for the onset of depressive 218 episodes. Therefore, future studies should additionally use structured clinical interviews to 219 enable the reliable and valid diagnosis of DD. Another possible limitation is the relatively small 220 amount of perceived stress induced by the exam phase (M2 and M3) Psychometric Properties of the Test of Passive Aggression Self-directed passive-aggressive behaviour as 277 an essential component of depression: findings from two cross-sectional observational 278 studies Diagnostic and statistical manual of mental disorders : 280 DSM-5. 5 The size and 282 burden of mental disorders and other disorders of the brain in Europe The increasing burden of depression Lifetime prevalence 288 and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey 289 Replication The Rise and Fall of Depressive 291 Symptoms and Academic Stress in Two Samples of University Students The Unique 294 Associations of Academic Experiences With Depressive Symptoms in Emerging 295 Prevalence, correlates, and prospective predictors of non-suicidal 345 self-injury among New Zealand adolescents: cross-sectional and longitudinal survey data Nonsuicidal self-injury in adolescence: Prospective rates and risk 348 factors in a 2 ½ year longitudinal study. Psychiatry Res. 30 Differences in risk factors for 350 self-harm with and without suicidal intent: findings from the ALSPAC cohort Non-suicidal self-injury and depressive 353 symptoms during middle adolescence: a longitudinal analysis Deliberate self-harm and psychological 356 problems in young adolescents: evidence of a bidirectional relationship in girls Prospective prediction of nonsuicidal self-injury: a 1-year 359 longitudinal study in young adults Alcohol use and psychiatric 361 comorbid disorders predict deliberate self-harm behaviour and other suicidality among 362 depressed adolescent outpatients in 1-year follow-up Depressed 365 adolescents as young adults -predictors of suicide attempt and non-suicidal self-injury 366 during an 8-year follow-up Verfügbar unter: www. soscisurvey.de . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity World Medical Association Declaration of Helsinki: ethical 369 principles for medical research involving human subjects Manual for the Beck Depression Inventory-II. Psychological 372 Corporation Franke GH. Brief Symptom Checklist (BSCL). Hogrefe The Trier Inventory for the Assessment of Chronic Stress (TICS): scale 375 construction, statistical testing, and validation of the scale work overload Factor structure and psychometric properties of 378 the trier inventory for chronic stress (TICS) in a representative german sample American Psychiatric Association. Diagnostic and statistical manual of mental disorders Psychometric properties of the Beck Depression Inventory-II: a 383 comprehensive review. Rev Bras Psiquiatr Sao Paulo Braz Reliability and validity of the Brief Symptom Inventory The validity, reliability and factor structure of the Brief Symptom Inventory Convergent and Discriminant Validity of the 391 Test of Passive Aggression in a Psychotherapy Outpatient Sample IBM Cooperation. IBM SPSS Statistics for Windows Introduction to mediation, moderation, and conditional process analysis: A 395 regression-based approach A critique of the cross-lagged panel model lavaan: An R Package for Structural Equation Modeling Gender differences in the prevalence of nonsuicidal self-injury: 401 A meta-analysis Self-harm in young people Prediction of 405 major depressive disorder onset in college students Sensitivity and specificity of depression questionnaires in a college-408 age sample the author/funder, who has granted medRxiv a license to display the preprint in perpetuity . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 16, 2022. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The copyright holder for this preprint this version posted April 16, 2022. ; https://doi.org/10.1101/2022.04.12.22273775 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)The copyright holder for this preprint this version posted April 16, 2022. ; https://doi.org/10.1101/2022.04.12.22273775 doi: medRxiv preprintThe copyright holder for this preprint this version posted April 16, 2022. ; https://doi.org/10.1101/2022.04.12.22273775 doi: medRxiv preprint