Psychosocial and Situational Risk Factors of Suicidality among Young Adults amid and after the COVID-19 Pandemic: A Three-Wave Cross-Sectional Network Analysis

doi:10.21203/rs.3.rs-5264990/v1

Download PDF

Research Article

Psychosocial and Situational Risk Factors of Suicidality among Young Adults amid and after the COVID-19 Pandemic: A Three-Wave Cross-Sectional Network Analysis

https://doi.org/10.21203/rs.3.rs-5264990/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Purpose: The COVID-19 pandemic has brought situational impacts and mental distress among young people. No studies have investigated and compared the risk factors of suicidality among young adults during and after the COVID-19 pandemic. The present study aimed to examine the risk factors of suicidality in this population via a network approach amid and after the COVID-19 pandemic.

Methods: Repeated, cross-sectional telephone surveys recruited three samples of young adults in Hong Kongvia random sampling in autumns of 2021 (N=1472), 2022 (N=1238), and 2023 (N=1226). The respondents completed measures on situational impact, meaning in life, societal pessimism, Patient Health Questionnaire-4 (PHQ-4), social withdrawal, and suicidality. Mixed graphical modeling was conducted to identify the bridge linkages between suicidality and risk factors across the three waves.

Results: The 2023 wave showed a substantial decline in COVD-19 distress but higher rates of suicidality than the 2021 and 2022 waves. The network models showed good stability. Financial impact, PHQ-4, and social withdrawal showed positive linkages with suicidality in the network. PHQ-4 consistently showed the greatest bridge strength across the three waves. Network comparison found no significant differences in the network structure and global strength across gender, age groups, and waves. Societal pessimism showed significantly greater bridge strength in the network model from 2021 to 2023.

Conclusion: The results elucidated the network structure of risk factors of suicidality in population-based samples of young adults after the pandemic, highlighting meaning in life and societal pessimism as potential targets of interventions.

COVID-19 pandemic

Hikikomori

Meaning in life

Network approach

Societal pessimism

Suicide

Suicide is a major public health issue and a leading cause of death globally [1]. The COVID-19 pandemic has brought situational impacts across various domains and social distancing measures have substantially interfered with functioning and induced psychological responses among citizens [2]. Young adults have shown greater vulnerabilities to stressors in the education, work, social domains during the pandemic [3]. Financial strain, loneliness, and prolonged social isolation have been found to contribute to depressive symptoms and suicidal ideation during the pandemic [4,5]. Existing studies have demonstrated worse mental health and increased suicidality during the COVID-19 pandemic among young adults [6,7].

Meaning in life measures the purpose and existential mattering of one’s life [8]. A recent meta-analysis found meaning in life as a protective factor against suicidal ideation in young people [9]. Societal pessimism refers to a sentiment among citizens that their society is in decline. Socioeconomic vulnerabilities such as economic decline, unemployment, and political instability have contributed to societal pessimism [10]. Although pessimistic future expectations have been found to predict depressive symptoms and suicidality in young adults [11], no existing studies have examined the potential role of societal pessimism in shaping suicidality among young adults. Hong Kong experienced the most severe and fifth wave of the COVID-19 due to the Omicron variant in early 2022, which accounted for three-quarters of the total COVID-19 deaths throughout the six waves [12]. Although the Hong Kong government removed all social distancing measures as the COVID-19 pandemic subsided in March 2023, there remain substantial challenges to mental health in the post-pandemic context [13]. Further research is needed to evaluate the psychosocial aftermath of the pandemic, or the “new normal” so as to prepare for upcoming public health crises [14].

Suicidality is the outcome of complex interaction among a multitude of risk factors in clinical, psychological, and social domains [15]. The network approach provides a new perspective on comorbidity between the mental disorders at a symptom level. de Beurs [16] introduced the network approach to conceptualize suicidality as the consequence of a network of multidimensional risk factors. This novel approach could not only elucidate the dynamics that contribute to suicidality through visualization of the comorbidity among risk factors [17], but allow enable identification of central and bridge symptoms in the network structure. Though recent studies have applied network analysis to examine the psychosocial risk factors of suicidality in various samples of young adults, the majority of them are based on cross-sectional or college student samples [18-22].

We searched PubMed and Web of Science on July 8, 2024, for publications that analyzed the impact of the COVID-19 pandemic on suicidality using the keywords “COVID-19 pandemic, mental health or suicidality, after or post, young adults or youths”. We found only one repeated, cross-sectional study that evaluated mental health problems among national population-based samples of Finnish youths before, during, and after the pandemic [23]. No studies have investigated the risk factors of suicidality among young adults after the COVID-19 pandemic. It is of importance to better understand the long-term effects of the pandemic on suicidality of young adults in a post-pandemic context. The present study has two objectives. Firstly, we estimate the network structure of situational impact, meaning in life, societal pessimism, social withdrawal, and psychological distress as psychosocial and situational risk factors of suicidality in population-based samples of young adults during and after the COVID-19 pandemic. Secondly, we aim to identify the central nodes and bridge linkages between the risk factors and suicidality and compare the network model across waves to examine differences in network characteristics across pandemic stages.

Participants and procedures

The present study was a three-wave, repeated cross-sectional telephone survey among young adults in Hong Kong between 2021 and 2023. The three assessment waves were held between August and November 2021 annually in 2021 (before the fifth COVID-19 wave), 2022 (after the fifth COVID-19 wave), 2023 (after the end of the pandemic), respectively [12]. The surveys were conducted by an independent survey agency in undertaking various social surveys with expertise in survey implementation and data security. Simple random sampling was used to retrieve 6500 mobile phone numbers from the Office of the Communications Authority in Hong Kong. Trained interviewers contacted potential participants in selected samples from 6:30 pm to 10:30 pm using a computer-assisted telephone interviewing system. Inclusion criteria of the participants were aged 18 - 35, residence in Hong Kong for at least 6 months, and ability to understand Cantonese. Invalid cases with age outside 18-35 years and non-working mobile phone numbers were excluded.

Eligible participants provided oral informed consent before completing a standardized questionnaire in 10-15 minutes. During the phone interviews, on-site validity checks were performed on range and consistency of the collected responses to ensure accuracy and completeness. To ensure proper adherence of the fieldwork procedures, 300 interviews were randomly drawn for quality checking and 91.7% of them confirmed that they participated in the phone interview with consistent answers for four questions in the quality back-check questionnaire as the completed questionnaire. The survey data were weighted with reference to 2021 population census data by the Census and Statistics Department of Hong Kong to adjust proportionally for the age and gender. The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2013. All procedures involving human subjects/patients were approved by the Human Research Ethics Committee of the University of Hong Kong (Reference number: EA210185). Verbal informed consent was obtained from all participants, and they understood that they could withdraw anytime.

The present study recruited samples of 1472, 1238, and 1226 eligible adults in the 2021, 2022, and 2023 telephone surveys. Response rates of the survey, which denoted the ratio of successful interview cases to the sum of successful interview cases, non-contact cases (not answering the call), and refusal cases, were 68.9%, 61.5%, and 62.2% in the three waves of surveys, respectively. Overall, half (49.7%) of the sample were females. The sample had a mean age of 27.0 years (SD = 4.74; range = 18 – 35) and was classified into two age groups of 2312 young adults (58.7%: aged 26 – 35) and 1624 youths (41.3%: aged 18 – 25). The majority of the sample attained tertiary education level (73.8%) and were working (80.7%).

Measures

The standardized questionnaire included demographic characteristics and measures on situational impact, societal pessimism, meaning in life, psychological distress, social withdrawal, and suicidality. The questionnaire was pilot-tested in 20 young adults in July 2021 to refine questionnaire wordings and operational procedures.

Situational impact

Situational impact was measured by the Mental Impact and Distress Scale: COVID-19 [2]. This scale assessed the perceived negative impact of COVID-19 pandemic in five domains over the past four months: study/job, family, social, physical, and financial, on a 5-point Likert scale (1 = “not at all” to 5 = “very severe”). This scale showed acceptable reliability (Mcdonald’s Omega ω = 0.70 – 0.85) across the three waves in the present study.

Depressive symptoms

The 4-item Patient Health Questionnaire [24] was used to assess the level of depressive symptoms of the respondents over the past two weeks. The items were scored on a 4-point Likert scale (0 = “not at all” to 3 = “nearly every day”) and the total PHQ-4 score had a theoretical range from 0 to 12. This scale showed good reliability (ω = 0.77 – 0.88) across the three waves.

Meaning in life

Meaning in life was evaluated by the 3-item Presence subscale of the Meaning in Life Questionnaire - short form [25]. The items were scored on a 7-point Likert scale (1 = “absolutely untrue” to 7 = “absolutely true”) and the total meaning in life score had a theoretical range from 1 to 7. This scale showed good reliability (ω = 0.85 – 0.89) across the three waves.

Societal pessimism

Societal pessimism was evaluated by how much the respondents agree with the following three statements: 1) I am pessimistic about the future of my society; 2) the glory days of my society have passed; and 3) the society I am living in is weakening overall. The items were developed with reference to empirical framework on societal unease by Steenvoorden [26]. The items were answered on a 7-point Likert scale (1 = “strongly disagree” to 7 = “strongly agree”) and the total score of societal pessimism had a theoretical range from 1 to 7. This scale showed acceptable reliability (ω = 0.73 – 0.81) across the three waves.

Social withdrawal

Social withdrawal were evaluated by the Hikikomori Questionnaire [27]. This questionnaire comprised four items on social withdrawal symptoms: marked social isolation in one’s home, significant functional impairment due to social isolation, distress brought by social isolation, and duration of the symptoms. Hikikomori was diagnosed with marked social isolation at home of at least 6 months and significant functional impairment or distress brought by the social isolation. This scale has shown adequate psychometric properties in Hong Kong young adults [5].

Suicidality

The respondents were asked whether they had considered suicide, attempted suicide, or injured themselves intentionally in the past 12 months. Based on these binary measures, the present study evaluated the level of suicidality on a 4-point ordinal scale: 0 = non-suicidal; 1 = suicidal or deliberate self-harm without suicide attempt; 2 = both suicidal and deliberate self-harm without suicide attempt; and 3 = had suicide attempt.

Data analysis

The present sample showed minimal missing data (0.3% - 1.2%) in study/job impact and physical impact and no missing data in other variables. We first compared the demographic characteristics and study variables across the three waves using chi-square tests and ANOVA with post-hoc comparisons using Sidak corrections. Eta-squared denotes the effect size of the comparison with cutoff values of 0.01, 0.06, and 0.14 indicating small, moderate, and large effect sizes, respectively. Cutoff values for correlation coefficients were 0.10, 0.30, and 0.50 indicating small, moderate, and large effect sizes, respectively [28]. In the present study, statistical significance was primarily set at p < 0.01 level.

To model the inter-relationships among the 12 study variables (age, six items on situational impact, PHQ-4, meaning in life, societal pessimism, hikikomori, and suicidality), partial correlation networks were estimated in the 2021 – 2023 waves using mixed graphical model via the bootnet package in R [29].Age, situational impact, PHQ-4, meaning in life, and societal pessimism showed acceptable skewness (-0.07 – 1.07) and kurtosis (-0.92 – 0.85) and were modelled as continuous variables. The network models were estimated using the graphical least absolute shrinkage and selection operator (GLASSO) from the qgraph package to limit spurious edges [30]. The optimal model was selected based on the extended Bayesian information criterion. Tuning hyperparameter γ was set to 0.25 to filter out weaker edges to improve interpretability of the networks. All network analyses were conducted in R version 4.3.3.

Network stability was evaluated using 2,500 case-dropping bootstraps via the bootnet package [31]. The correlation stability coefficient (CS-coefficient) indicated the proportion of cases that could be removed from analysis while maintaining a correlation of at least r = 0.70 with values of 0.5 indicate good network stability. To summarize the network characteristics, expected influence was used to evaluate the cumulative influence of the nodes in the network, together with other centrality indices (strength, closeness, and betweenness). Nodes with high strength or expected influence were regarded as the central nodes in the network. The network models were visualized using the springlayout via the qgraph package [32]. Positive and negative edges were displayed in blue and red, respectively.

We focused on the bridge edges between suicidality and the risk factors. Accuracy of edge weights was assessed using 95% confidence intervals. The networktools package [33] was used to compute the bridge strength as the bridging power of the nodes with suicidality. The moderate sample sizes (Ns > 1200) in each wave provided adequate statistical power to estimate a 12-node network. Finally, network comparison was conducted with 1000 iterations using the NetworkComparisonTest package [34] on the network structure and global strength across gender (males and females), age groups (youths and young adults), and waves (2021 - 2023). We conducted invariance tests on centrality of individual nodes and edge weights. To adjust for multiple testing, invariance tests of edge weights were conducted at p < 0.001 level.

Supplementary Table 1 shows the bivariate correlations among the study variables in the whole sample and the 2022 wave. The correlations for the 2021 and 2023 waves are shown in Supplementary Table 2. Situational impacts showed significant and moderate correlations (r = 0.20–0.51, p < 0.01). Societal pessimism was negatively correlated with meaning in life (r = -0.28, p < 0.01) and positively correlated with PHQ-4 (r = 0.27, p < 0.01). PHQ-4 and hikikomori were positively correlated with suicidality (r = 0.20–0.34, p < 0.01).

As Table 1 shows, the three waves of participants showed a small difference in age (η² = 0.013, p < 0.001) but not gender. There were significant but small differences (η² = 0.005–0.026, p < 0.001) in situational impact in four domains. The 2023 wave showed a sharp decline in COVD-19 distress (η² = 0.151, p < 0.001) than 2021 and 2022 waves. There was a substantial decline in levels of towards societal pessimism (η² = 0.135, p < 0.001) from 2021 to 2023. There were significant and small declines in meaning in life and PHQ-4 (η² = 0.005–0.009, p < 0.001) from 2021 wave to 2023 wave. However, the 2022 and 2023 waves showed higher rates of hikikomori, suicidal ideation, self-harm, and suicide attempt than 2021 wave.

Table 1

Comparison of demographic characteristics and study variables across the three waves
	2021 wave (N = 1472)	2022 wave (N = 1238)	2023 wave (N = 1226)
Categorical variables	N (%)	N (%)	N (%)	χ²	p
Male	710 (48.2)	631 (51.0)	614 (50.1)	2.13	0.34
Tertiary education	1359 (92.3) ^c	811 (66.2) ^b	727 (59.3) ^a	431.9	< 0.001
Hikikomori	42 (2.9) ^a	71 (5.7) ^b	54 (4.4) ^ab	13.9	< 0.001
Suicidal ideation	149 (10.1) ^a	282 (22.8) ^b	291 (23.7) ^b	106.5	< 0.001
Self-harm	61 (4.1) ^a	259 (20.9) ^b	245 (20.0) ^b	199.8	< 0.001
Suicide attempt	14 (1.0) ^a	85 (6.9) ^b	91 (7.4) ^b	77.3	< 0.001
Continuous variables	Mean (SD)	Mean (SD)	Mean (SD)	F	p	η²
Age, years [18–36]	26.3 (3.78) ^a	27.4 (5.20) ^b	27.4 (5.17) ^b	25.8	< 0.001	0.013
Situational impact
COVID [1–5]	2.83 (1.21) ^b	2.81 (1.18) ^b	1.79 (0.99) ^a	349.3	< 0.001	0.151
Study/Job [1–5]	2.59 (1.08) ^b	2.48 (1.07) ^ab	2.35 (1.02) ^a	16.4	< 0.001	0.008
Financial [1–5]	2.17 (1.12) ^a	2.63 (1.26) ^c	2.43 (1.12) ^b	52.4	< 0.001	0.026
Physical [1–5]	1.95 (1.03)	1.92 (1.09)	1.94 (1.04)	0.17	0.84	0.000
Family [1–5]	1.81 (1.01) ^a	1.98 (1.08) ^b	1.96 (1.07) ^b	10.6	< 0.001	0.005
Social [1–5]	1.84 (0.96) ^a	2.08 (1.08) ^b	2.01 (1.06) ^b	18.8	< 0.001	0.009
Pessimism [1–7]	5.33 (1.22) ^c	4.49 (1.27) ^b	4.21 (1.20) ^a	306.6	< 0.001	0.135
Meaning in life [1–7]	4.38 (1.35) ^b	4.26 (1.27) ^ab	4.18 (1.16) ^a	8.99	< 0.001	0.005
PHQ4 symptoms [0–12]	2.72 (2.43) ^b	2.53 (2.57) ^b	2.16 (2.31) ^a	18.3	< 0.001	0.009
χ² = chi-square; η² = eta-squared; ^abc superscripts (a < b < c) show significant post-hoc differences across waves with Sidak correction; Square brackets show theoretical ranges of continuous variables

[Table 1 about here]

Network models in 2021–2023 waves

Figures 1–3 displays the network model on suicidality and its risk factors across 2021–2023, respectively. Figures A1 – A3 plots the stability of expected influence of the networks using case-dropping bootstrap. There was excellent stability for expected influence (CS-coefficient = 0.67–0.75) and strength (CS-coefficient = 0.60–0.67). The centrality indices of the nodes are shown in Figures A4 – A6 across waves. Out of the 66 potential edges, there were 28, 23, and 27 edge weights across 2021–2023 networks, with an average edge weight of 0.083, 0.121, and 0.105, respectively. Figures A7 – A9 show the bootstrapped CIs of the edge weights and pairwise comparisons between the edge weights are shown in Figures A10 – A12 across the three waves. In 2021, Suicidality showed positive linkages with family impact, PHQ-4, and hikikomori (r_p = 0.11–0.21), and a negative linkage with meaning in life (r_p = -0.09). In 2022, Suicidality showed positive linkages with hikikomori, COVID-19 distress, PHQ-4, and pessimism (r_p = 0.15–0.35). In 2023, Suicidality showed positive linkages with financial impact, PHQ-4, and pessimism (r_p = 0.07–0.14).

[Figures 1–3 about here]

Network comparison across age groups, gender, and waves

Figure A13 – A15 displays the network model on suicidality and its risk factors in the combined sample and across gender and age groups, respectively. The network structure did not differ significantly across gender (M = 0.11, p = 0.97) and age groups (M = 0.16, p = 0.82). There were no significant differences in global strength across gender (males: 4.13; females: 3.72; S = 0.41, p = 0.32) and age groups (youths: 4.48 versus young adults: 4.17; S = 0.30, p = 0.54).

Figure 4 shows the network model on suicidality and its risk factors across 2021–2023 waves on a common layout. There were no significant differences in the network structure (M = 0.28–0.35, p = 0.08–0.51) and the global network strength (2021: 3.74; 2022: 4.03; 2023: 4.14; S = 0.11–0.40, p = 0.53–0.79) across the three waves. Invariance tests on node centrality found that the bridge strength of societal pessimism increased significantly (p = 0.001–0.004) from 0.25 in 2021 to 0.61 in 2022 to 0.85 in 2023. Figure A16 plots the bridge strength of the network nodes across the three waves. PHQ-4 showed the greatest bridge strength (0.93–1.30) and meaning in life showed the third highest bridge strength (0.57–0.73) across the three waves. Pessimism showed significantly stronger linkages (p < 0.001) with financial impact, meaning in life, and suicidality in 2023 than 2021. The linkage between pessimism and COVID-19 distress changed significantly (p < 0.001) from positive (r_p = 0.06) in 2021 wave to negative (r_p = -0.09) in 2023.

[Figures 4 about here]

The present study systematically examined the risk and protective factors of suicidality using a network approach in Hong Kong young adults during and after the COVID-19 pandemic. Compared to previous waves, the present study found a substantial decline in COVID-19 distress in the post-pandemic wave, which aligns with the decreasing health threat of the pandemic. In line with previous findings [35, 36], the post-peak wave of the pandemic showed higher rate of suicidality. The deterioration could be attributed to the severity of fifth COVID-19 wave and the associated social distancing measures [37, 38]. The lack of improvements in hikikomori and suicidality in post-pandemic wave suggests long-term mental health effects of the pandemic [39, 7].

In line with Fong et al. [40], our present network found high degrees of inter-connectedness among situational impacts of the pandemic across various domains. Hikikomori and PHQ-4were the strongest risk factors of suicidality, corroborating the central role of psychological factors in contributing to suicidal behavior. PHQ-4 was the central node of the network with the highest strength and consistent bridge linkages with suicidality across time. A recent study [41] found the greatest bridge strength between hopelessness and suicidal ideation in a symptom network, and Lam et al. [42] identified bridge relationships between depressive symptoms and suicidality in Macau residents near the end of the pandemic in 2022.

Though meaning in life only showed weak negative linkages with suicidality in the 2021 wave but not in 2022 or 2023 waves, our results suggest indirect protective pathways from meaning in life to suicidality through PHQ-4, hikikomori, and societal pessimism. A recent study found that depressive symptoms mediated the effects of meaning in life on suicidal ideation [24]. The invariant global strength in our networks across time implies sustained comorbid relationships between the risk factors, which indicate sustained mental health needs of young adults in post-pandemic contexts and that the era of “new normal” is not quite normal for their mental well-being.

Despite a substantial mean decline, societal pessimism showed greater influences in the network models from 2021 wave to 2023 wave. The stronger edges that connected societal pessimism with meaning in life, financial impact, PHQ-4, and hikikomori in the 2023 network suggest a greater role for this node in shaping suicidality among young adults in post-pandemic context. Young adults with greater societal pessimism are more likely to realize that their life goals have become unattainable, making it difficult for them to sustain a sense of self [43]. This could lead to feelings of hopelessness and despair and predict suicidality. Interestingly, young adults with greater COVID-19 distress showed lower levels of societal pessimism in post-pandemic wave. Experience of greater distress of the pandemic could result in post-traumatic growth and a more positive outlook of the society [44].

Study implications

The present study contributes to the literature by highlighting comorbidity of various risk factors and complex pathways in their effects on suicidality. Younger people have shown psychological vulnerability to the COVID-19 pandemic as an existential threat, which led to a loss of meaning and feelings of entrapment [45]. This highlights the need to address the existential concerns of young adults in mitigating their pessimistic views towards society. Promotion of prosocial motivation could foster prosocial behavior and a more positive outlook on society [46]. Psychological flexibility could contribute to better emotional adjustments to a new normal in young adults in post-pandemic contexts [47]. Interventions such as mindfulness and narrative interventions could promote meaning in life and in turn attenuate the risk factors of suicidality [48].

The pandemic has caused fundamental shifts in the labor markets and young adults who lack work experience often face higher risks of unemployment and insecure part-time work. Preventive interventions could be implemented by health care sectors on individual and societal levels to promote the mental health of young people through universal and targeted frameworks [49]. The COVID-19 pandemic has exacerbated the disparities in access to mental health services for young people. Practitioners should make use of digital technologies to provide universal and accessible mental health care via online and remote interventions for youths and young adults with mental distress to reduce their risk profile in post-pandemic contexts [50]. Timely interventions on mental distress could ameliorate activation of the risk factors to reduce their suicidality.

Limitations and future research

Compared to previous studies on convenience samples of Chinese college students, our random samples of young adults were more representative of the general population with a balanced gender ratio. The network results contribute to a nuanced understanding of the network structure among the risk factors of suicidality and clarify precise pathways from the risk factors to suicidality. The invariance of network structure and global strength across gender and age groups provides empirical support for the robustness of the network and suggests similar vulnerabilities for young adults of all genders to the situational impacts of the pandemic.

There were several limitations in the present study. Firstly, our repeated cross-sectional study design did not prospectively follow up the same cohort of young adults from 2021 to2023. The cross-sectional networks could not separate between-person effects from within-person effects and infer directional relationships among the variables. Longitudinal studies are needed to examine the temporal cross-lagged relationships from the risk factors to suicidality in a post-pandemic context. Secondly, network analysis was conducted in this study as an exploratory analytic approach on the comorbidity among the risk factors and their bridge linkages with suicidality. Thus, the present results should be interpreted with caution before replication studies in independent samples and other age groups such as adolescents and older adults.

Thirdly, despite acceptable response rates (> 60%) of the phone surveys, the potential self-selection and non-response biases cast doubt over the generalizability of the present results to the general population. Healthy young adults could be less motivated to participate in the surveys, resulting in overestimation of mental health impact of the pandemic. Finally, the present study only focused on a set of risk factors of suicidality in the network and was subject to confounding bias as a result of omission of relevant variables. Future studies should incorporate measures on social media influences, suicide stigma, and resilience in the suicidality network.

The present study provides fine-grained results on the network structure of risk factors of suicidality in population-based samples of young adults during and after the pandemic. Our results contribute to a better understanding of psychosocial and situational risk factors of suicidality and have clinical implications in highlighting meaning in life and societal pessimism as potential targets of future interventions to enhance population mental health.

Conflict of interest: The authors declare that they have no conflict of interest.

Ethics approval: Ethical approval was obtained from the Human Research Ethics Committee of the University of Hong Kong (Reference number = EA210185). All procedures performed in the present study were in accordance with the 1964 Helsinki Declaration and its later amendments.

Consent to participate:Informed consent was obtained from all individual participants included in the study, and they understood that they could withdraw anytime.

Acknowledgements: The authors would like to thank the staff members at The Hong Kong Jockey Club Centre for Suicide Research and Prevention, The University of Hong Kong for various roles such as project coordination, data cleaning and fruitful discussions of the results. We would like to express our gratitude to the Social Science Research Centre for their help in data collection and all respondents for their participation in the phone survey.

Author contributions: Ted C.T. Fong: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Software, Visualization, Writing – original draft; Paul S.F. Yip: Conceptualization, Funding acquisition, Investigation, Resources, Supervision, Validation, Writing – review & editing. All authors approved the final submitted version.

Funding: This study was supported by the Hong Kong Jockey Club Charities Trust for the OpenUp online emotional support program and the General Research Fund grant (GRF/HKU 17603323) of the Research Grants Council. This project was partly supported by the Strategic Topic Grants Scheme (STG4/M-701/23-N). The funders of the study had no role in study design; data collection, data analysis, data interpretation; or writing of the report.

Data availability: The data and codes that support the findings of this study are available from the first author upon reasonable request.

Klonsky ED, May AM, Saffer BY (2016) Suicide, suicide attempts, and suicidal ideation. Ann Rev Clin Psychol 12:307–330. 10.1146/annurev-clinpsy-021815-093204
Fong TCT, Chang K, Sit HF, Ho RTH (2023) Associations between COVID-19 mental impact and distress, resilience, burnout and well-being in Hong Kong community adults: A structural equation model. Psychol Health Med 28:1803–1817. 10.1080/13548506.2023.2229235
Varma P, Junge M, Meaklim H, Jackson ML (2021) Younger people are more vulnerable to stress, anxiety and depression during COVID-19 pandemic: A global cross-sectional survey. Prog Neuro-psychopharmacol Biol Psychiatry 109:110236. 10.1016/j.pnpbp.2020.110236
So WWY, Fong TCT, Woo BPY, Yip PSF (2024) Psychosocial and financial well-being mediated the effects of COVID-19 distress on suicidality: a serial mediation model among Hong Kong young adults. Soc Psychiatry Psychiatr Epidemiol 59:165–174. 10.1007/s00127-023-02501-4
Fong TCT, Yip PSF (2023) Prevalence of hikikomori and associations with suicidal ideation, suicide stigma, and help-seeking among 2,022 young adults in Hong Kong. Int J Soc Psychiatry 69:1768–1780. 10.1177/00207640231174376
O'Connor RC, Wetherall K, Cleare S, McClelland H, Melson AJ, Niedzwiedz CL, O'Carroll RE, O'Connor DB, Platt S, Scowcroft E, Watson B, Zortea T, Ferguson E, Robb KA (2021) Mental health and well-being during the COVID-19 pandemic: longitudinal analyses of adults in the UK COVID-19 Mental Health & Wellbeing study. Br J psychiatry: J mental Sci 218:326–333. 10.1192/bjp.2020.212
Gustafsson J, Lyyra N, Jasinskaja-Lahti I, Simonsen N, Lahti H, Kulmala M, Ojala K, Paakkari L (2023) Mental health profiles of Finnish adolescents before and after the peak of the COVID-19 pandemic. Child Adolesc Psychiatry Mental Health 17. 10.1186/s13034-023-00591-1
King LA, Hicks JA (2021) The Science of Meaning in Life. Annu Rev Psychol 72:561–584. 10.1146/annurev-psych-072420-122921
Li S, Luo H, Huang F, Wang Y, Siu Fai Yip P (2024) Associations between meaning in life and suicidal ideation in young people: A systematic review and meta-analysis. Child Youth Serv Rev 158:107477. 10.1016/j.childyouth.2024.107477
Steenvoorden EH, van der Meer TW (2017) Continent of pessimism or continent of realism? A multilevel study into the impact of macro-economic outcomes and political institutions on societal pessimism, European Union 2006–2012. Int J Comp Sociol 58:192–214. 10.1177/0020715217710809
Chang EC (2024) Expectativas Oscuras: On the Added Burden of Concomitant Negative Problem Orientation and Pessimism in Predicting Depressive Symptoms and Suicidal Behaviors in Latinx Young Adults. Cogn Therapy Res. 10.1007/s10608-024-10509-6
Yang B, Lin Y, Xiong W, Liu C, Gao H, Ho F, Zhou J, Zhang R, Wong JY, Cheung JK, Lau EHY, Tsang TK, Xiao J, Wong IOL, Martín-Sánchez M, Leung GM, Cowling BJ, Wu P (2024) Comparison of control and transmission of COVID-19 across epidemic waves in Hong Kong: an observational study. Lancet Reg Health West Pac 43:100969. 10.1016/j.lanwpc.2023.100969
Vadivel R, Shoib S, El Halabi S, El Hayek S, Essam L, Gashi Bytyçi D, Karaliuniene R, Schuh Teixeira AL, Nagendrappa S, Ramalho R, Ransing R, Pereira-Sanchez V, Jatchavala C, Adiukwu FN, Kudva Kundadak G (2021) Mental health in the post-COVID-19 era: challenges and the way forward. Gen Psychiatr 34:e100424. 10.1136/gpsych-2020-100424
Boden M, Zimmerman L, Azevedo KJ, Ruzek JI, Gala S, Abdel Magid HS, Cohen N, Walser R, Mahtani ND, Hoggatt KJ, McLean CP (2021) Addressing the mental health impact of COVID-19 through population health. Clin Psychol Rev 85:102006. 10.1016/j.cpr.2021.102006
Turecki G, Brent DA (2016) Suicide and suicidal behaviour. Lancet 387:1227–1239. 10.1016/s0140-6736(15)00234-2
de Beurs D (2017) Network Analysis: A Novel Approach to Understand Suicidal Behaviour. Int J Environ Res Public Health 14:219. 10.3390/ijerph14030219
de Beurs D, Bockting C, Kerkhof A, Scheepers F, O’Connor R, Penninx B, van de Leemput I (2021) A network perspective on suicidal behavior: Understanding suicidality as a complex system. Suicide Life-Threatening Behav 51:115–126. 10.1111/sltb.12676
Calati R, Romano D, Magliocca S, Madeddu F, Zeppegno P, Gramaglia C (2022) The interpersonal-psychological theory of suicide and the role of psychological pain during the COVID-19 pandemic: A network analysis. J Affect Disord 302:435–439. 10.1016/j.jad.2022.01.078
Lu J, Jin Y, Liang S, Wang Q, Li X, Li T (2024) Risk factors and their association network for young adults’ suicidality: a cross-sectional study. BMC Public Health 24:1378. 10.1186/s12889-024-18860-9
Zhou H-y, Luo Y-H, Shi L-j, Gong J (2023) Exploring psychological and psychosocial correlates of non-suicidal self-injury and suicide in college students using network analysis. J Affect Disord 336:120–125. 10.1016/j.jad.2023.05.089
Ma S, Yang J, Xu J, Zhang N, Kang L, Wang P, Wang W, Yang B, Li R, Xiang D, Bai H, Liu Z (2022) Using network analysis to identify central symptoms of college students' mental health. J Affect Disord 311:47–54. 10.1016/j.jad.2022.05.065
Yang T, He Y, Wu L, Ren L, Lin J, Wang C, Wu S, Liu X (2023) The relationships between anxiety and suicidal ideation and between depression and suicidal ideation among Chinese college students: A network analysis. Heliyon 9:e20938. 10.1016/j.heliyon.2023.e20938
Kiviruusu O, Ranta K, Lindgren M, Haravuori H, Silén Y, Therman S, Lehtonen J, Sares-Jäske L, Aalto-Setälä T, Marttunen M, Suvisaari J (2024) Mental health after the COVID-19 pandemic among Finnish youth: a repeated, cross-sectional, population-based study. Lancet Psychiatry 11:451–460. 10.1016/S2215-0366(24)00108-1
Fong TCT, Ho RTH, Yip PSF (2023) Psychometric properties of the Patient Health Questionnaire-4 among Hong Kong young adults in 2021: Associations with meaning in life and suicidal ideation. Front Psychiatry 14:1138755. 10.3389/fpsyt.2023.1138755
Yuen M, Chan RT-H, Wong S-W, Yau J, Zhang J, Ho EY-F, Fung E, Chan S, Wu F (2022) Appropriateness of the Dispositional Hope Scale (DHS) and Meaning in Life Questionnaire: Short form (MLQ-SF) for use with regular and gifted students in Hong Kong. High Ability Stud 1–13. 10.1080/13598139.2022.2159336
Steenvoorden EH (2015) A General Discontent Disentangled: A Conceptual and Empirical Framework for Societal Unease. Soc Indic Res 124:85–110. 10.1007/s11205-014-0786-4
Kato TA, Kanba S, Teo AR (2020) Defining pathological social withdrawal: proposed diagnostic criteria for hikikomori. World Psychiatry 19:116–117. 10.1002/wps.20705
Cohen J (1988) Statistical power analysis for the behavioral sciences, 2nd edn. Lawrence Erlbaum Associates, Hillsdale, NJ
Haslbeck JMB, Waldorp LJ (2020) MGM: Estimating time-varying mixed graphical models in high-dimensional data. J Stat Softw 93:1–46. 10.18637/jss.v093.i08
Epskamp S, Fried EI (2018) A tutorial on regularized partial correlation networks. Psychol Methods 23:617–634. 10.1037/met0000167
Epskamp S, Borsboom D, Fried EI (2018) Estimating psychological networks and their accuracy: A tutorial paper. Behav Res Methods 50:195–212. 10.3758/s13428-017-0862-1
Epskamp S, Cramer AOJ, Waldorp LJ, Schmittmann VD, Borsboom D (2012) qgraph: Network Visualizations of Relationships in Psychometric Data. J Stat Softw 48:1–18. 10.18637/jss.v048.i04
Jones PJ, Ma R, McNally RJ (2021) Bridge centrality: A network approach to understanding comorbidity. Multivar Behav Res 56:353–367. 10.1080/00273171.2019.1614898
Van Borkulo CD, van Bork R, Boschloo L, Kossakowski JJ, Tio P, Schoevers RA, Borsboom D, Waldorp LJ (2023) Comparing network structures on three aspects: A permutation test. Psychol Methods 28:1273–1285. 10.1037/met0000476
García-Fernández L, Romero-Ferreiro V, Izquierdo-Izquierdo M, Rodríguez V, Alvarez-Mon MA, Lahera G, Santos JL, Rodriguez-Jimenez R (2023) Dramatic increase of suicidality in children and adolescents after COVID-19 pandemic start: A two-year longitudinal study. J Psychiatr Res 163:63–67. 10.1016/j.jpsychires.2023.04.014
Bersia M, Koumantakis E, Berchialla P, Charrier L, Ricotti A, Grimaldi P, Dalmasso P, Comoretto RI (2022) Suicide spectrum among young people during the COVID-19 pandemic: A systematic review and meta-analysis. eClinicalMedicine 54. 10.1016/j.eclinm.2022.101705
Burki T (2022) Hong Kong's fifth COVID-19 wave – the worst yet. Lancet Infect Dis 22:455–456. 10.1016/S1473-3099(22)00167-0
Elbogen EB, Lanier M, Blakey SM, Wagner HR, Tsai J (2021) Suicidal ideation and thoughts of self-harm during the COVID-19 pandemic: The role of COVID-19-related stress, social isolation, and financial strain. Depress Anxiety 38:739–748. 10.1002/da.23162
Khan KS, Mamun MA, Griffiths MD, Ullah I (2022) The Mental Health Impact of the COVID-19 Pandemic Across Different Cohorts. Int J Mental Health Addict 20:380–386. 10.1007/s11469-020-00367-0
Fong TCT, Chang K, Ho RTH, Chio FHN, Yip PSF, Wen M (2024) Psychological responses to the COVID-19 pandemic and situational impact: A network analysis in Hong Kong residents. J Affect Disord 362:152–160. 10.1016/j.jad.2024.06.105
Junus A, Yip PSF (2023) Preventing comorbidity between distress and suicidality: a network analysis. npj Mental Health Res 2:2. 10.1038/s44184-023-00022-1
Lam MI, Cai H, Chen P, Lok KI, Chow IHI, Si TL, Su Z, Ng CH, An FR, Xiang YT (2024) The Inter-Relationships Between Depressive Symptoms and Suicidality Among Macau Residents After the Relatively Static Management COVID-19 Strategy: A Perspective of Network Analysis. Neuropsychiatr Dis Treat 20:195–209. 10.2147/ndt.S451031
TenHouten W (2023) The Emotions of Hope: From Optimism to Sanguinity, from Pessimism to Despair. Am Sociologist 54:76–100. 10.1007/s12108-022-09544-1
Jian Y, Hu T, Zong Y, Tang W (2023) Relationship between post-traumatic disorder and posttraumatic growth in COVID-19 home-confined adolescents: The moderating role of self-efficacy. Curr Psychol 42:17444–17453. 10.1007/s12144-021-02515-8
Vacchiano M, Politi E, Lueders A (2023) The COVID-19 pandemic as an existential threat: Evidence on young people's psychological vulnerability using a Multifaceted Threat Scale. PLoS ONE 18:e0292894. 10.1371/journal.pone.0292894
Maher A (2024) Tracing the bounds of distress: Mental health and the Lancet Commission on lessons for the future from the COVID-19 pandemic. J Glob Health 14:03022. 10.7189/jogh.14.03022
Cheng C, Wang HY, Ebrahimi OV (2021) Adjustment to a New Normal: Coping Flexibility and Mental Health Issues During the COVID-19 Pandemic. Front Psychiatry 12:626197. 10.3389/fpsyt.2021.626197
Manco N, Hamby S (2021) A Meta-Analytic Review of Interventions That Promote Meaning in Life. Am J Health Promotion 35:866–873. 10.1177/0890117121995736
Fusar-Poli P, Correll CU, Arango C, Berk M, Patel V, Ioannidis JPA (2021) Preventive psychiatry: a blueprint for improving the mental health of young people. World Psychiatry 20:200–221. 10.1002/wps.20869
Fischer-Grote L, Fössing V, Aigner M, Fehrmann E, Boeckle M (2024) Effectiveness of Online and Remote Interventions for Mental Health in Children, Adolescents, and Young Adults After the Onset of the COVID-19 Pandemic: Systematic Review and Meta-Analysis. JMIR Ment Health 11:e46637. 10.2196/46637

No competing interests reported.

Download PDF

Editor assigned by journal
18 Oct, 2024
Submission checks completed at journal
17 Oct, 2024
First submitted to journal
15 Oct, 2024

You are reading this latest preprint version

Psychosocial and Situational Risk Factors of Suicidality among Young Adults amid and after the COVID-19 Pandemic: A Three-Wave Cross-Sectional Network Analysis

Status:

Version 1

Abstract

Figures

Introduction

Methods

Results

Network models in 2021–2023 waves

Network comparison across age groups, gender, and waves

Discussion

Study implications

Limitations and future research

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1