Network analysis of comorbid insomnia and depressive symptoms among psychiatric practitioners during the COVID-19 pandemic

A network analysis of the interrelationships between depression, anxiety, insomnia and quality of life among fire service recruits

Background

Research on the mental health and quality of life (hereafter QOL) among fire service recruits after the end of the COVID-19 restrictions is lacking. This study explored the network structure of depression, anxiety and insomnia, and their interconnections with QOL among fire service recruits in the post-COVID-19 era.

Methods

This cross-sectional study used a consecutive sampling of fire service recruits across China. We measured the severity of depression, anxiety and insomnia symptoms, and overall QOL using the nine-item Patient Health Questionnaire (PHQ-9), seven-item Generalized Anxiety Disorder scale (GAD-7), Insomnia Severity Index (ISI) questionnaire, and World Health Organization Quality of Life-brief version (WHOQOL-BREF), respectively. We estimated the most central symptoms using the centrality index of expected influence (EI), and the symptoms connecting depression, anxiety and insomnia symptoms using bridge EI.

Results

In total, 1,560 fire service recruits participated in the study. The prevalence of depression (PHQ-9 ≥ 5) was 15.2% (95% CI: 13.5-17.1%), while the prevalence of anxiety (GAD-7 ≥ 5) was 11.2% (95% CI: 9.6-12.8%). GAD4 ("Trouble relaxing") had the highest EI in the whole network model, followed by ISI5 ("Interference with daytime functioning") and GAD6 ("Irritability"). In contrast, PHQ4 ("Fatigue") had the highest bridge EI values in the network, followed by GAD4 ("Trouble relaxing") and ISI5 ("Interference with daytime functioning"). Additionally, ISI4 "Sleep dissatisfaction" (average edge weight = -1.335), which was the central symptom with the highest intensity value, had the strongest negative correlation with QOL.

Conclusion

Depression and anxiety were important mental health issues to address among fire service recruits in the post-COVID-19 era in China. Targeting central and bridge symptoms identified in network analysis could help address depression and anxiety among fire service recruits in the post-COVID-19 era.

The network structures of depressive and insomnia symptoms among cancer patients using propensity score matching: Findings from the Health and Retirement Study (HRS)

OBJECTIVE

Both depression and insomnia are found to be more prevalent in cancer patients compared to the general population. This study compared the network structures of depression and insomnia among cancer patients versus cancer-free participants (controls hereafter).

METHOD

The 8-item Center for Epidemiological Studies Depression Scale (CESD-8) and the 4-item Jenkins Sleep Scale (JSS-4) were used to measure depressive and insomnia symptoms, respectively. Propensity score matching (PSM) was used to construct the control group using data from the Health and Retirement Study (HRS). In total, a sample consisting of 2216 cancer patients and 2216 controls was constructed. Central (influential) and bridge symptoms were estimated using the expected influence (EI) and bridge expected influence (bridge EI), respectively. Network stability was assessed using the case-dropping bootstrap method.

RESULT

The prevalence of depression (CESD-8 total score ≥ 4) in cancer patients was significantly higher compared to the control group (28.56 % vs. 24.73 %; P = 0.004). Cancer patients also had more severe depressive symptoms relative to controls, but there was no significant group difference for insomnia symptoms. The network structures of depressive and insomnia symptoms were comparable between cancer patients and controls. "Felt sadness" (EI: 6.866 in cancer patients; EI: 5.861 in controls), "Felt unhappy" (EI: 6.371 in cancer patients; EI: 5.720 in controls) and "Felt depressed" (EI: 6.003 in cancer patients; EI: 5.880 in controls) emerged as the key central symptoms, and "Felt tired in morning" (bridge EI: 1.870 in cancer patients; EI: 1.266 in controls) and "Everything was an effort" (bridge EI: 1.046 in cancer patients; EI: 0.921 in controls) were the key bridge symptoms across both groups.

CONCLUSION

Although cancer patients had more frequent and severe depressive symptoms compared to controls, no significant difference was observed in the network structure or strength of the depressive and insomnia symptoms. Consequently, psychosocial interventions for treating depression and insomnia in the general population could be equally applicable for cancer patients who experience depression and insomnia.

Inter-relationships of depression and insomnia symptoms with life satisfaction in stroke and stroke-free older adults: Findings from the Health and Retirement Study based on network analysis and propensity score matching

BACKGROUND

Depression and insomnia are common co-occurring psychiatric problems among older adults who have had strokes. Nevertheless, symptom-level relationships between these disorders remain unclear.

OBJECTIVES

In this study, we compared inter-relationships of depression and insomnia symptoms with life satisfaction among older stroke patients and stroke-free peers in the United States.

METHODS

The study included 1026 older adults with a history of stroke and 3074 matched controls. Data were derived from the US Health and Retirement Study. Depression, insomnia and life satisfaction were assessed. Propensity score matching was employed to identify demographically-similar groups of stroke patients and controls. Central and bridge symptoms were assessed using Expected influence (EI) and bridge EI, respectively.

RESULTS

The prevalence of depression in the stroke group (25.0 %) was higher than that of controls (14.3 %, P < 0.001). In stroke group, "Feeling depressed" (CESD1; EI: 5.80), "Feeling sad" (CESD7; EI: 4.67) and "Not enjoying life" (CESD6; EI: 4.51) were the most central symptoms, while "Feeling tired in the morning" (JSS4; BEI: 1.60), "Everything was an effort" (CESD2; BEI: 1.21) and "Waking up during the night" (JSS2; BEI: 0.98) were key bridge symptoms. In controls, the most central symptoms were "Lack of happiness" (CESD4; EI: 6.45), "Feeling depressed" (CESD1; EI: 6.17), and "Feeling sad" (CESD7; EI: 6.12). Furthermore, "Feeling tired in the morning" (JSS4; BEI: 1.93), "Everything was an effort" (CESD2; BEI: 1.30), and "Waking up too early" (JSS3; BEI: 1.12) were key bridge symptoms. Life satisfaction had the most direct associations with "Not enjoying life" (CESD6) and "Feeling lonely" (CESD5) in the two groups, respectively.

CONCLUSION

Older adults with stroke exhibited more severe depression and insomnia symptoms. Interventions targeting central and bridge symptoms may help to mitigate the co-occurrence of these symptoms.

Prevalence and network structure of depression, insomnia and suicidality among mental health professionals who recovered from COVID-19: a national survey in China

Psychiatric syndromes are common following recovery from Coronavirus Disease 2019 (COVID-19) infection. This study investigated the prevalence and the network structure of depression, insomnia, and suicidality among mental health professionals (MHPs) who recovered from COVID-19. Depression and insomnia were assessed with the Patient Health Questionnaire (PHQ-9) and Insomnia Severity Index questionnaire (ISI7) respectively. Suicidality items comprising suicidal ideation, suicidal plan and suicidal attempt were evaluated with binary response (no/yes) items. Network analyses with Ising model were conducted to identify the central symptoms of the network and their links to suicidality. A total of 9858 COVID-19 survivors were enrolled in a survey of MHPs. The prevalence of depression and insomnia were 47.10% (95% confidence interval (CI) = 46.09-48.06%) and 36.2% (95%CI = 35.35-37.21%), respectively, while the overall prevalence of suicidality was 7.8% (95%CI = 7.31-8.37%). The key central nodes included "Distress caused by the sleep difficulties" (ISI7) (EI = 1.34), "Interference with daytime functioning" (ISI5) (EI = 1.08), and "Sleep dissatisfaction" (ISI4) (EI = 0.74). "Fatigue" (PHQ4) (Bridge EI = 1.98), "Distress caused by sleep difficulties" (ISI7) (Bridge EI = 1.71), and "Motor Disturbances" (PHQ8) (Bridge EI = 1.67) were important bridge symptoms. The flow network indicated that the edge between the nodes of "Suicidality" (SU) and "Guilt" (PHQ6) showed the strongest connection (Edge Weight= 1.17, followed by "Suicidality" (SU) - "Sad mood" (PHQ2) (Edge Weight = 0.68)). The network analysis results suggest that insomnia symptoms play a critical role in the activation of the insomnia-depression-suicidality network model of COVID-19 survivors, while suicidality is more susceptible to the influence of depressive symptoms. These findings may have implications for developing prevention and intervention strategies for mental health conditions following recovery from COVID-19.

Depression, anxiety and suicidality among Chinese mental health professionals immediately after China's dynamic zero-COVID policy: A network perspective

BACKGROUND

Using network analysis, the interactions between mental health problems at the symptom level can be explored in depth. This study examined the network structure of depressive and anxiety symptoms and suicidality among mental health professionals after the end of China's Dynamic Zero-COVID Policy.

METHODS

A total of 10,647 mental health professionals were recruited nationwide from January to February 2023. Depression and anxiety were assessed using the 9-item Patient Health Questionnaire (PHQ-9) and 7-item Generalized Anxiety Disorder Scale (GAD-7), respectively, while suicidality was defined by a 'yes' response to any of the standard questions regarding suicidal ideation (SI), suicide plan (SP) and suicide attempt (SA). Expected Influence (EI) and Bridge Expected Influence (bEI) were used as centrality indices in the symptom network to characterize the structure of the symptoms.

RESULTS

The prevalence of depression, anxiety, and suicidality were 45.99 %, 28.40 %, and 7.71 %, respectively. The network analysis identified GAD5 ("Restlessness") as the most central symptom, followed by PHQ4 ("Fatigue") and GAD7 ("Feeling afraid"). Additionally, PHQ6 ("Guilt"), GAD5 ("Restlessness"), and PHQ8 ("Motor disturbance") were bridge nodes linking depressive and anxiety symptoms with suicidality. The flow network indicated that the strongest connections of S ("Suicidality") was with PHQ6 ("Guilt"), GAD7 ("Feeling afraid"), and PHQ2 ("Sad mood").

CONCLUSIONS

Depression, anxiety, and suicidality among mental health professionals were highly prevalent after China's Dynamic Zero-COVID Policy ended. Effective measures should target central and bridge symptoms identified in this network model to address the mental health problems in those at-risk.

Machine Learning and Bayesian Network Analyses Identifies Psychiatric Disorders and Symptom Associations with Insomnia in a national sample of 31,285 Treatment-Seeking College Students

Background

A better understanding of the structure of relations among insomnia and anxiety, mood, eating, and alcohol-use disorders is needed, given its prevalence among young adults. Supervised machine learning provides the ability to evaluate the discriminative accuracy of psychiatric disorders associated with insomnia. Combined with Bayesian network analysis, the directionality between symptoms and their associations may be illuminated.

Methods

The current exploratory analyses utilized a national sample of college students across 26 U.S. colleges and universities collected during population-level screening before entering a randomized controlled trial. Firstly, an elastic net regularization model was trained to predict, via repeated 10-fold cross-validation, which psychiatric disorders were associated with insomnia severity. Seven disorders were included: major depressive disorder, generalized anxiety disorder, social anxiety disorder, panic disorder, post-traumatic stress disorder, anorexia nervosa, and alcohol use disorder. Secondly, using a Bayesian network approach, completed partially directed acyclic graphs (CPDAG) built on training and holdout samples were computed via a Bayesian hill-climbing algorithm to determine symptom-level interactions of disorders most associated with insomnia [based on SHAP (SHapley Additive exPlanations) values)] and were evaluated for stability across networks.

Results

Of 31,285 participants, 20,597 were women (65.8%); mean (standard deviation) age was 22.96 (4.52) years. The elastic net model demonstrated clinical significance in predicting insomnia severity in the training sample [R2 = .449 (.016); RMSE = 5.00 [.081]), with comparable performance in accounting for variance explained in the holdout sample [R2 = .33; RMSE = 5.47). SHAP indicated the presence of any psychiatric disorder was associated with higher insomnia severity, with major depressive disorder demonstrated to be the most associated disorder. CPDAGs showed excellent fit in the holdout sample and suggested that depressed mood, fatigue, and self-esteem were the most important depression symptoms that presupposed insomnia.

Conclusion

These findings offer insights into associations between psychiatric disorders and insomnia among college students and encourage future investigation into the potential direction of causality between insomnia and major depressive disorder.

Trial registration

Trial may be found on the National Institute of Health RePORTER website: Project Number: R01MH115128-05.

A comparison of psychiatric symptoms between mental health professionals with and without post-infection sequelae of COVID-19

Post-infection sequelae of COVID-19 (PISC) have raised public health concerns. However, it is not clear whether infected mental health professionals (MHPs) with PISC have experienced more psychiatric symptoms than MHPs without PISC do. This study examined differences in the prevalence of self-reported depression, anxiety, insomnia and suicidality as well as the network structures of these symptoms between these two groups. Participants completed questionnaire measures of psychiatric symptoms and demographics. Expected influence was used to measure centrality of symptoms and network comparison tests were adopted to compare differences in the two network models. The sample comprised 2,596 participants without PISC and 2,573 matched participants with PISC. MHPs with PISC had comparatively higher symptom levels related to depression (55.2% vs. 23.5 %), anxiety (32.0% vs. 14.9 %), insomnia (43.3% vs. 17.3 %), and suicidality (9.6% vs. 5.3 %). PHQ4 ("Fatigue"), PHQ6 ("Guilt"), and GAD2 ("Uncontrollable Worrying") were the most central symptoms in the "without PISC" network model. Conversely, GAD3 ("Worry too much"), GAD5 ("Restlessness"), and GAD4 ("Trouble relaxing") were more central in the "with PISC" network model. In sum, MHPs with PISC experienced comparatively more psychiatric symptoms and related disturbances. Network results provide foundations for the expectation that MHPs with PISC may benefit from interventions that address anxiety-related symptoms, while those without PISC may benefit from interventions targeting depression-related symptoms.

Psychosocial factors of insomnia in depression: a network approach

BACKGROUND

Insomnia symptoms in patients with major depressive disorder (MDD) are common and deleterious. Childhood trauma, personality traits, interpersonal distress, and social support contribute to insomnia, but how they interact to affect insomnia remains uncertain.

METHODS

A total of 791 patients with MDD completed the Insomnia Severity Index, Eysenck Personality Questionnaire, Interpersonal Relationship Comprehensive Diagnostic Scale, Childhood Trauma Questionnaire, Social Support Rating Scale and Hamilton Depression Scale-17. This study utilized network analyses to identify the central symptoms of insomnia and their associations with psychosocial factors.

RESULTS

Worrying about sleep was identified as the central symptom in the insomnia network, insomnia and associated personality network, insomnia and associated interpersonal disturbance network, insomnia and associated childhood trauma network, insomnia and associated social support network, and the integrated network of insomnia symptoms and associated psychosocial factors. In the networks of insomnia symptoms and individual psychosocial factors, most psychosocial factors (other than childhood trauma) were directly or indirectly related to insomnia symptoms; however, neuroticism was the only factor directly associated with insomnia symptoms before and after controlling for covariates. In the final integrated network of insomnia symptoms and psychosocial factors, neuroticism was a bridge node and mediated the relationships of social support and interpersonal disturbances with insomnia symptoms, which is clearly presented in the shortest pathways.

CONCLUSIONS

Worrying about sleep and neuroticism were prominent in the integrated network of insomnia symptoms and associated psychosocial factors, and the edge between them connected psychosocial factors and insomnia symptoms in MDD patients.