RUNNING HEADER: COVID-19 ANXIETY NETWORK ANALYSIS

1
Psychological Networks can Identify Potential Pathways to Specific Intervention Targets for

Anxiety in Response to Coronavirus Disease 2019 (COVID-19)

Santiago Papini M.A.1*, Justin Dainer-Best Ph.D.2, Mikael Rubin M.A.1, Eric D. Zaizar B.A.1,

Michael J. Telch Ph.D.1 & Jasper A. J. Smits Ph.D.1

1
Department of Psychology and Institute for Mental Health Research, The University of Texas at

Austin
2
Department of Psychology, Bard College

*Correspondence: Santiago Papini, Institute for Mental Health Research and Department of

Psychology, The University of Texas at Austin, 305 E. 23rd Street, Ste., 4.106, Austin, TX

78712, USA. Ph: 512-475-8095; Email: [email protected]

Word count: 2763

Tables/Figures: 2

References: 34
COVID-19 ANXIETY NETWORK ANALYSIS 2

Abstract

Addressing mental health challenges related to the COVID-19 outbreak can be facilitated

through research that characterizes the needs of subpopulations and identifies specific pathways

to targeted intervention. Toward this aim, we examined the impact of the COVID-19 outbreak on

anxiety symptoms among college students (N = 487) and explored the relative impact of coping

strategies using a psychological network approach, which models complex interactions to

identify potential pathways to symptom-level intervention. Although students showed several

significant fluctuations in pre- to post-outbreak anxiety symptom levels measured with the GAD-

7, anxiety network connectivity was not significantly different across timepoints. Consistent with

hypotheses, the post-outbreak symptom+coping network revealed that increased use of the

adaptive coping strategies of acceptance, behavioral activation, and values-based action was

associated with lower levels of fear, restlessness, and trouble relaxing. The symptom+coping

network also revealed that increased use of the maladaptive strategies of excessive cleaning,

reassurance seeking, and excessive checking was associated with higher levels of irritability and

fear. Surprisingly, the use of reappraisal and avoidance, two strategies with putatively opposing

adaptive value, highly overlapped and showed positive associations with fear and irritability.

These symptom+coping associations can guide the assessment and treatment of anxiety in the

face of COVID-19.
COVID-19 ANXIETY NETWORK ANALYSIS 3

1. Introduction

The outbreak of COVID-19 poses a variety of mental health challenges, including

barriers to treatment access as a result of overextended resources, quarantine regulations, and

financial burden (L. Yang et al., 2020). Research can help optimize intervention strategies by

identifying key targets such as dysfunctional coronavirus anxiety (Lee, Jobe, et al., 2020),

suicide (Reger et al., 2020), or loneliness (Jeste et al., 2020) and by characterizing the unique

needs of subpopulations, such as patients recovering from COVID-19 (Bo et al., 2020),

healthcare workers (Lai et al., 2020), older adults (Y. Yang et al., 2020), and individuals with

severe mental illness (Druss, 2020). Toward this broad goal, we investigated the impact of

COVID-19 on anxiety symptoms among college students and explored the relative impact of

coping strategies, which represent potential pathways to symptom-level intervention.

The structural impact of COVID-19 on millions of college students worldwide (~6% of

the U.S. population; National Center for Education Statistics) includes campus closures, sudden

transition to online instruction, reduced social interaction, loss of career/internship opportunities,

and unprecedented uncertainty about the future. Evidence from past disasters suggests

psychological resilience is the norm (Pfefferbaum & North, 2020) and currently available

evidence from the COVID-19 outbreak in China indicates low rates of moderate (2.7%) and

severe (0.9%) anxiety levels among pre-medical students in that country (Cao et al., 2020).

Nevertheless, addressing the psychological impact of these stressors in this sensitive

developmental stage is critical (Golberstein et al., 2020). Although the availability of assessment

tools and non-pharmacological interventions with relatively low risk profiles can facilitate

treatment dissemination, the heterogeneity of psychological responses to environmental stressors

underscores the need for precision interventions.

COVID-19 ANXIETY NETWORK ANALYSIS 4

Psychological network approaches can model the complexity of diverse responses to

environmental stressors. This approach estimates edges, or connections among psychological

features included in the network, as well as a variety of metrics derived from edge estimates,

such as the global network connectivity, or the sum of the strength of all network edges. In the

context of psychological disorders, the network approach has been framed as a theoretical

alternative to the latent disease model (i.e., symptom networks are the disorder vs. symptoms in

the latent disease model are indicators of an underlying disorder). Readers interested in a

theoretical, methodological, and empirical overview of the network approach to psychopathology

may consult a recent review (Robinaugh et al., 2020). In the current work, we focused on

research that examined changes in network structure in the context of environmental stressors

and on the use of network metrics as prospective and prognostic markers of the emergence of (or

recovery from) stress-related psychopathology. For example, posttraumatic stress disorder

(PTSD) symptom networks demonstrated stronger connectivity 12-months after emergency room

hospitalization (Bryant et al., 2017), and symptoms with the strongest connections were better

predictors of future diagnosis in a distinct sample of assault survivors (Haag et al., 2017). We

previously found that change in PTSD symptoms that showed the strongest connections in the

pretreatment network was positively associated with overall treatment response (Papini et al.,

2020). A similar relation between node- and network-change was observed in the emergence of

prolonged grief after the death of a spouse (Robinaugh et al., 2016). Altogether, these studies

provide evidence of prospective and prognostic markers in psychological networks.

Although network analysis has been primarily used to investigate symptom connectivity

within or across disorders, more recent work has used the approach to examine symptom-level

treatment effects by conducting secondary analyses of clinical trials (i.e., network intervention
COVID-19 ANXIETY NETWORK ANALYSIS 5

analysis; Blanken et al., 2019). A third application involves the inclusion of psychological

features that can influence symptoms, including risk and resilience factors (Hoorelbeke et al.,

2016), positive and negative emotion regulation strategies (Everaert & Joormann, 2019),

rumination and executive control (Bernstein et al., 2017), and attentional biases (Heeren &

McNally, 2016). This approach represents a potential bridge between the exploration of

symptom relations (i.e, symptom network analysis) and the impact of interventions on symptoms

(i.e., network intervention analysis) because it can elucidate links between symptoms and the

psychological mechanisms through which they may be optimally targeted by treatment. For

example, positive reappraisal showed connectivity to only two of the 42 depression and anxiety

symptoms in the estimated network, suggesting that this emotion regulation strategy may not

equally impact symptoms of these disorders (Everaert & Joormann, 2019). On the basis of this

finding, Everaert and Joormann (2019) provided an example of a testable prediction about which

symptoms to target with which strategy based on their network analysis: “If patients display

symptoms related to negative expectations about the future (e.g., pessimism, fear of the worst

happening), then treatments may adopt a symptom-focused strategy and increase positive

reappraisal use (e.g., through cognitive restructuring) to obtain relief in these specific symptoms”

(p. 10). As such, data-derived hypotheses generated via network analysis of symptom-

mechanism interactions can function as an intermediate low-burden step that can guide the

development of precision clinical trials, which require focalized targets, adaptive approaches,

and precise measurement of mechanisms and outcomes (Lenze et al., 2020).

In line with this prior work, the present study applied psychological network methods to e

the potential impact of the COVID-19 outbreak on anxiety symptoms and coping strategies

among college students as they adapted to campus closure and the transition to online
COVID-19 ANXIETY NETWORK ANALYSIS 6

coursework. Our aims were (1) to examine longitudinal changes (pre- to post-outbreak) in

anxiety symptom levels and overall network connectivity, and (2) to examine the relative impact

of adaptive and maladaptive coping strategies on the post-outbreak anxiety symptom network.

Together, these aims can provide insight into the specific features of anxiety that may have been

impacted by the outbreak and identify potential pathways for optimizing symptom-level

intervention strategies.

2. Materials and Methods

2.1. Participants and measures

Participants were undergraduate students who participated in the online surveys for

research credit as part of an introductory course (N = 487; female: 288 [59.14%]; mean age =

19.65, SD = 2.14; self identified as White = 186 [38.19%], Hispanic or Latino origin = 111

[22.79%], Asian = 105 [21.56%], Black or African American = 25 [5.13%]). All participants

provided written, informed consent, and the Institutional Review Board of The University of

Texas at Austin approved all procedures. Data collection began at the onset of the University’s

campus closure in response to COVID-19, and continued through the end of the semester (post-

outbreak; March–May 2020). Anxiety symptoms (i.e., anxiety, uncontrollable worry, generalized

worry, trouble relaxing, restlessness, irritability, and fear) were assessed with the GAD-7

(Spitzer et al., 2006), which measures frequency on a Likert scale from “not at all sure” to

“nearly every day” (range: 0-3). For a majority of these participants (n = 443; 90.97%), GAD-7

data from the beginning of the semester (pre-outbreak; January 2020) was available.

At the post-outbreak time point, we also measured use of adaptive and maladaptive

coping strategies that represent common techniques or targets of two evidence-based

psychotherapy approaches for treating anxiety: Cognitive Behavioral Therapy (Barlow et al.,
COVID-19 ANXIETY NETWORK ANALYSIS 7

2017) and Acceptance and Commitment Therapy (Hayes et al., 2006). Adaptive coping

strategies were: reappraisal, behavioral activation, values-based action, and acceptance.

Maladaptive coping strategies were avoidance, excessive checking, excessive cleaning, and

reassurance seeking. Table 1 includes the questions used to measure the frequency of use of

these strategies, on a Likert scale from “none of the time” to “all of the time” (range: 0-4).

2.2. Analyses

2.2.1. Pre- to post-outbreak changes in anxiety symptom levels and network connectivity

All analyses were conducted in R version 3.6.1 (R Core Team, 2017); package versions

used are provided for each analysis. To test the hypothesis that the COVID-19 outbreak would be

associated with increases in symptom severity we used paired t-tests comparing pre- to post-

outbreak levels of anxiety symptoms and total severity, calculated by summing all symptoms

(stats v3.6.1; R Core Team, 2017). For these analyses we report uncorrected p-values from the

two-tailed tests but note that the Bonferroni-corrected threshold for the eight comparisons is

.006. To test the hypothesis that the COVID-19 outbreak would be associated with increased

connectivity in the anxiety symptom network, we used the network comparison test which is a

two-tailed permutation test that compares the global strength of networks estimated on randomly

regrouped participants across the two time points (NetworkComparisonTest v2.2.1; van Borkulo

et al., 2015). The procedure is repeated 1,000 times, yielding a distribution allowing a test of the

null hypothesis. Global strength connectivity was considered significantly different at p < .05.

2.2.2. Association of coping strategies with anxiety symptoms

Data from the post-outbreak phase were used to construct a symptom+coping network.

This allowed us to test the hypothesis that greater use of the adaptive coping strategies of

reappraisal, behavioral activation, values-based action, and acceptance would be associated with
COVID-19 ANXIETY NETWORK ANALYSIS 8

lower severity of anxiety symptoms, and that greater use of the maladaptive coping strategies of

avoidance, excessive checking, excessive cleaning, and reassurance seeking would be associated

with higher severity of anxiety symptoms. To increase interpretability and parsimony of the

symptom+coping network, we took a two-step approach. First, redundant nodes were identified

using the weighted topological overlap approach and combined using latent variable scoring

(EGAnet v0.9.3; Golino et al., 2020). Next, partial correlation networks were estimated in a

nonparametric bootstrapped procedure with 1,000 replications (bootnet v1.3; Epskamp & Fried,

2018). Following recent recommendations for reducing bias and increasing the stability and

replicability of edge weight estimates, networks were estimated without regularization using

Spearman correlations (Fried et al., 2020; Williams et al., 2019). Edge weights were extracted

with 95% confidence intervals to assess relations between coping strategies and anxiety

symptoms after controlling for all other nodes in the network. Edges with confidence intervals

that did not cross zero were significant at the p < .05 level; nonsignificant edges were pruned

from the network.

3. Results

3.1. Pre- to post-outbreak changes in anxiety symptom levels and network connectivity

Table 1 provides item-level descriptive statistics of study measures. Relative to the

beginning of the semester, most GAD-7 symptoms were significantly lower in the post-outbreak

assessment, as indicated by the mean difference estimates [95% CI] for anxiety = -0.21 [-0.30 to

-0.11], p < .001; uncontrollable worry -0.35 [-0.45 to -0.25], p < .001; generalized worry -0.24 [-

0.33 to -0.14], p < .001; and restlessness -0.22 [-0.31 to -0.13], p < .001. Irritability was the only

symptom that increased, 0.19 [0.09 to 0.29], p < .001, and the remaining two symptoms (fear and

trouble relaxing) showed nonsignificant fluctuations across time (both ps > .05). Overall anxiety,
COVID-19 ANXIETY NETWORK ANALYSIS 9

indexed by the GAD-7 total score, decreased, -0.95 [-1.43 to -0.47], p < .001, but network

connectivity, or the strength of symptom co-activation, was not significantly different across

assessments (2.93 vs 2.97; p = .54). Together, these results suggest that changes in symptoms

across time did not significantly alter global strength connectivity in the network.

3.2. Association of coping strategies with anxiety symptoms

In the post-outbreak assessment, high degree of overlap was detected among anxiety

symptoms (anxiety/uncontrollable worry/generalized worry and restlessness/trouble relaxing),

two adaptive strategies (behavioral activation/values-based actions), two maladaptive strategies

(reassurance seeking/excessive checking), and two strategies with putatively different adaptive

value (reappraisal/avoidance). Figure 1 shows the network graph plotted using qgraph v1.6.5

(Epskamp et al., 2012). After combining the overlapping nodes, the final symptom+coping

network contained four anxiety symptom nodes and five coping strategy nodes connected by 16

significant edges out of 36 potential edges. Anxiety nodes had strong positive connectivity with

each other, suggesting co-activation of symptoms. Connectivity was also observed across many

coping strategies, suggesting that they are not used in isolation.

Consistent with hypothesis, significant edge estimates (i.e., partial r with [95% CI] that

did not cross zero, p < .05) showed that increased use of the adaptive strategies of behavioral

activation/values-based action was associated with lower levels of fear, -.13 [-.20 to -.05], and

acceptance was associated with lower levels of restlessness/trouble relaxing, -.13 [-.21 to -.04],

after controlling for all other relations in the network. In contrast, increased use of maladaptive

strategies of reassurance seeking/excessive checking was associated with higher levels of fear,

.21 [.12 to .29] and excessive cleaning was associated with higher levels of irritability, .17 [.08 to
COVID-19 ANXIETY NETWORK ANALYSIS 10

.25]. Reappraisal/avoidance was positively associated with fear, .13 [.04 to .22], and with

irritability, .12 [.03 to .20].

4. Discussion

Several key findings can be highlighted from our analyses of anxiety symptoms and

coping strategies in a diverse sample of university students. Surprisingly, severity of most GAD-

7 anxiety symptoms was lower post-outbreak relative to the January 2020 baseline; only

irritability showed an increase. Moreover, we did not find evidence of increased connectivity

among symptoms in the early stages of the pandemic’s impact relative to a January 2020

baseline. Together, this suggests that changes in the severity of individual nodes occurred

without significant changes in the overall connectivity of the GAD-7 network. If COVID-19

outbreak and its related consequences played a role in the observed longitudinal fluctuations of

anxiety, these findings indicate a differential impact on specific features of anxiety. For example,

frustration around online coursework, interference in social life, or reduced independence among

students who had to move back home, may have contributed to increases in irritability, even if

these changes also contributed to decreases in anxiety, worry, and restlessness. Future follow-up

assessments may help clarify whether the response to this stressor changes over time.

Nevertheless, a substantial minority of students reported severity in the moderate

(18.89%) or severe (8.21%) range, which are substantially higher rates than those reported in the

cohort of undergraduates in China (Cao et al., 2020). Connections between coping strategies and

symptom levels in the symptom+coping network suggest clinically intuitive avenues for

optimizing treatment. For example, individuals experiencing fear may benefit from

psychoeducation around the paradoxical impact of reassurance seeking and excessive checking,
COVID-19 ANXIETY NETWORK ANALYSIS 11

and strategies targeting behavioral activation and valued actions, whereas individuals exhibiting

restlessness and trouble relaxing may benefit from acceptance-based strategies.

On the other hand, the strong overlap between reappraisal and avoidance is more

challenging to interpret. In-depth measurement may be necessary to tease apart these two

strategies (see 4.1. Limitations). Within our analysis, the combined avoidance and reappraisal

node’s positive connectivity with both fear and irritability indicates its influence in the network

may have been maladaptive. This may have resulted from a blurring of the boundary between the

way these two strategies were executed in this sample; perhaps reappraisal ultimately functioned

as an avoidance strategy or had avoidance as its goal (e.g., trying to think about the situation

differently vs trying to think about something different when confronted with the situation).

Alternatively, it could indicate that when attempts to avoid the situation failed, reappraisal was

the fallback strategy (or vice versa). Moreover, reappraisal may be a challenging strategy in the

context of COVID-19 and successful reappraisal may require clinician-guided cognitive

restructuring in order to confer therapeutic benefit.

4.1. Limitations

Our findings should take several limitations into account. Results from the longitudinal

and cross-sectional analyses in our study should not be interpreted as evidence of causality. The

longitudinal analyses provide empirical evidence of pre- to post-outbreak changes in anxiety that

are likely to have multiple causes. The post-outbreak connections between symptoms and coping

strategies may be unidirectional in either direction (e.g., fear activates reassurance seeking or

reassurance seeking activates fear), bi-directional (e.g., fear and reassurance seeking form a

feedback loop), or purely associational (e.g., fear and reassurance seeking are activated by a

variable that is not part of the network). Accordingly, the symptom+coping network should be
COVID-19 ANXIETY NETWORK ANALYSIS 12

considered a step toward generating testable causal hypotheses in randomized controlled

frameworks (e.g., targeting reassurance seeking will lead to greater reductions of fear vs an

alternative treatment strategy). Finally, the pace of research is such that several relevant COVID-

19 measures have been published after the initiation of our data collection (e.g., Ahorsu et al.,

2020; Arpaci et al., 2020; Lee, Mathis, et al., 2020; Taylor et al., 2020). These can be

incorporated into future studies to examine their links to coping strategies and other

psychological mechanisms, as well as replicability and generalizability across samples.

5. Conclusions

Despite the abovementioned limitations, our findings highlight the utility of applying

network analysis to increase precision in our understanding of the interactions between anxiety

symptoms and coping strategies. Although we did not find evidence of post-outbreak increases in

anxiety symptom severity (with the exception of irritability) or overall network connectivity, a

meaningful proportion of students had moderate to high levels of anxiety post-outbreak, and

individual symptoms showed differential relationships to several adaptive and maladaptive

coping strategies. Beyond these results, our study highlights the utility of a network approach

that combines symptoms and mechanisms to identify potential cognitive and behavioral

pathways for optimizing treatment.

COVID-19 ANXIETY NETWORK ANALYSIS 13

Declaration of financial interests

J. Dainer-Best, E. Zaizar, and M. J. Telch report no relevant financial interests. S. Papini receives

support from the National Institutes of Health and the Donald D. Harrington Foundation. M.

Rubin receives support from the National Institutes of Health. J. Smits receives support from the

National Institutes of Health, compensation for his work as a consultant to Big Health, as editor

for Elsevier and the American Psychological Association, and royalties from various book

publishers. The authors declared that these financial interests had no influence on the design and

conduct of the study; collection, management, analysis, and interpretation of the data; and

preparation, review, or approval of the manuscript.

COVID-19 ANXIETY NETWORK ANALYSIS 14

Figure 1. Symptom+Coping Psychological Network in Response to COVID-19

Edge thickness and labels correspond to the strength of the positive (solid edges) or negative

(dashed edges) partial correlations between anxiety symptoms (red nodes), adaptive coping

(green nodes), maladaptive coping (blue nodes), or a combined (mal)adaptive node (yellow).

Edge estimates with CIs that crossed zero were pruned from the network.
COVID-19 ANXIETY NETWORK ANALYSIS 15

Table 1. Descriptive Statistics of Measures.

Pre- Post-
outbreak outbreak
Network node name: survey item text M (SD) M (SD)

GAD-7 (range: total 0 - 21; item 0 - 3) n = 443 N = 487

Total score (sum of items) 7.50 (5.61) 6.48 (5.28)

Anxiety: Feeling nervous, anxious, or on edge 1.32 (1.03) 1.10 (0.95)

Uncontrollable worry: Not being able to stop or control worrying 1.14 (1.08) 0.79 (0.94)

Generalized worry: Worrying too much about different things 1.30 (1.05) 1.07 (0.98)

Trouble relaxing: Trouble relaxing 1.07 (1.02) 0.96 (0.94)

Restlessness: Being so restless that it's hard to sit still 0.85 (0.97) 0.61 (0.87)

Irritability: Becoming easily annoyed or irritable 1.01 (0.94) 1.18 (0.99)

Fear: Feeling afraid as if something awful might happen 0.81 (0.99) 0.77 (0.96)

Coping with COVID-19 Anxiety (range 0 - 4)

Adaptive strategies

Acceptance: I accepted my anxiety about COVID-19 instead of - 1.37 (1.18)

trying to reduce, avoid, or control it.

Reappraisal: I tried reducing my anxiety by changing the way I - 1.24 (1.07)

think about the COVID-19 situation.

Valued actions: I tried reducing my anxiety about COVID-19 by - 2.01 (1.27)

focusing on engaging in things that are important to me, including
hobbies, relationships, organizations, religion, or work.

Behavioral activation: I tried reducing my anxiety about COVID-19 - 1.56 (1.36)

by establishing a daily routine.

Maladaptive strategies

Excessive cleaning: I tried reducing my anxiety by washing my - 2.09 (1.4)

hands, using hand sanitizers, or taking extra showers, even when I
had little reason to believe that I may have been exposed to
COVID-19.
COVID-19 ANXIETY NETWORK ANALYSIS 16

Excessive checking: I tried reducing my anxiety by checking myself - 0.95 (1.07)

for symptoms of COVID-19 or mentally reviewing past situations to
check whether I may have been exposed to the virus.

Avoidance & suppression: I tried avoiding anxiety by not thinking, - 1.18 (1.21)
reading, or talking about anything related to COVID-19.

Reassurance seeking: I tried reducing my anxiety about COVID-19 - 1.18 (1.12)

by seeking to be reassured by others that everything was going to
be OK.
Note. The pre- to post-outbreak analysis included 443 participants with data at both time points,

whereas the symptom+coping network included the full sample (N = 487) at the post-outbreak

timepoint.
COVID-19 ANXIETY NETWORK ANALYSIS 17

References

Ahorsu, D. K., Lin, C.-Y., Imani, V., Saffari, M., Griffiths, M. D., & Pakpour, A. H. (2020). The

Fear of COVID-19 Scale: Development and Initial Validation. International Journal of

Mental Health and Addiction, 1–9.

Arpaci, I., Karataş, K., & Baloğlu, M. (2020). The development and initial tests for the

psychometric properties of the COVID-19 Phobia Scale (C19P-S). Personality and

Individual Differences, 110108.

Bernstein, E. E., Heeren, A., & McNally, R. J. (2017). Unpacking Rumination and Executive

Control: A Network Perspective: Clinical Psychological Science.

https://doi.org/10.1177/2167702617702717

Blanken, T. F., Van Der Zweerde, T., Van Straten, A., Van Someren, E. J. W., Borsboom, D., &

Lancee, J. (2019). Introducing Network Intervention Analysis to Investigate Sequential,

Symptom-Specific Treatment Effects: A Demonstration in Co-Occurring Insomnia and

Depression. Psychotherapy and Psychosomatics, 88(1), 52–54.

https://doi.org/10.1159/000495045

Bo, H.-X., Li, W., Yang, Y., Wang, Y., Zhang, Q., Cheung, T., Wu, X., & Xiang, Y.-T. (2020).

Posttraumatic stress symptoms and attitude toward crisis mental health services among

clinically stable patients with COVID-19 in China. Psychological Medicine.

https://doi.org/10.1017/S0033291720000999

Bryant, R. A., Creamer, M., O’Donnell, M., Forbes, D., McFarlane, A. C., Silove, D., & Hadzi-

Pavlovic, D. (2017). Acute and Chronic Posttraumatic Stress Symptoms in the

Emergence of Posttraumatic Stress Disorder: A Network Analysis. JAMA Psychiatry,

74(2), 135–142. https://doi.org/10.1001/jamapsychiatry.2016.3470

COVID-19 ANXIETY NETWORK ANALYSIS 18

Cao, W., Fang, Z., Hou, G., Han, M., Xu, X., Dong, J., & Zheng, J. (2020). The psychological

impact of the COVID-19 epidemic on college students in China. Psychiatry Research,

287, 112934. https://doi.org/10.1016/j.psychres.2020.112934

Druss, B. G. (2020). Addressing the COVID-19 Pandemic in Populations With Serious Mental

Illness. JAMA Psychiatry. https://doi.org/10.1001/jamapsychiatry.2020.0894

Epskamp, S., Cramer, A. O., Waldorp, L. J., Schmittmann, V. D., & Borsboom, D. (2012).

qgraph: Network visualizations of relationships in psychometric data. Journal of

Statistical Software, 48(4), 1–18.

Epskamp, S., & Fried, E. I. (2018). A tutorial on regularized partial correlation networks.

Psychological Methods.

Everaert, J., & Joormann, J. (2019). Emotion Regulation Difficulties Related to Depression and

Anxiety: A Network Approach to Model Relations Among Symptoms, Positive

Reappraisal, and Repetitive Negative Thinking. Clinical Psychological Science, 7(6),

1304–1318. https://doi.org/10.1177/2167702619859342

Fried, E. I., van Borkulo, C. D., & Epskamp, S. (2020). On the Importance of Estimating

Parameter Uncertainty in Network Psychometrics: A Response to Forbes et al.(2019).

Multivariate Behavioral Research, 1–6.

Golberstein, E., Wen, H., & Miller, B. F. (2020). Coronavirus Disease 2019 (COVID-19) and

Mental Health for Children and Adolescents. JAMA Pediatrics.

https://doi.org/10.1001/jamapediatrics.2020.1456

Golino, H., Shi, D., Christensen, A. P., Garrido, L. E., Nieto, M. D., Sadana, R., Thiyagarajan, J.

A., & Martinez-Molina, A. (2020). Investigating the performance of exploratory graph

analysis and traditional techniques to identify the number of latent factors: A simulation
COVID-19 ANXIETY NETWORK ANALYSIS 19

and tutorial. Psychological Methods.

Haag, C., Robinaugh, D. J., Ehlers, A., & Kleim, B. (2017). Understanding the Emergence of

Chronic Posttraumatic Stress Disorder Through Acute Stress Symptom Networks. JAMA

Psychiatry, 74(6), 649–650. https://doi.org/10.1001/jamapsychiatry.2017.0788

Heeren, A., & McNally, R. J. (2016). An integrative network approach to social anxiety disorder:

The complex dynamic interplay among attentional bias for threat, attentional control, and

symptoms. Journal of Anxiety Disorders, 42, 95–104.

https://doi.org/10.1016/j.janxdis.2016.06.009

Hoorelbeke, K., Marchetti, I., De Schryver, M., & Koster, E. H. W. (2016). The interplay

between cognitive risk and resilience factors in remitted depression: A network analysis.

Journal of Affective Disorders, 195, 96–104. https://doi.org/10.1016/j.jad.2016.02.001

Jeste, D. V., Lee, E. E., & Cacioppo, S. (2020). Battling the Modern Behavioral Epidemic of

Loneliness: Suggestions for Research and Interventions. JAMA Psychiatry.

https://doi.org/10.1001/jamapsychiatry.2020.0027

Lai, J., Ma, S., Wang, Y., Cai, Z., Hu, J., Wei, N., Wu, J., Du, H., Chen, T., Li, R., Tan, H.,

Kang, L., Yao, L., Huang, M., Wang, H., Wang, G., Liu, Z., & Hu, S. (2020). Factors

Associated With Mental Health Outcomes Among Health Care Workers Exposed to

Coronavirus Disease 2019. JAMA Network Open, 3(3), e203976–e203976.

https://doi.org/10.1001/jamanetworkopen.2020.3976

Lee, S. A., Jobe, M. C., & Mathis, A. A. (2020). Mental health characteristics associated with

dysfunctional coronavirus anxiety. Psychological Medicine, 1–2.

https://doi.org/10.1017/S003329172000121X

Lee, S. A., Mathis, A. A., Jobe, M. C., & Pappalardo, E. A. (2020). Clinically Significant Fear
COVID-19 ANXIETY NETWORK ANALYSIS 20

and Anxiety of COVID-19: A Psychometric Examination of the Coronavirus Anxiety

Scale. Psychiatry Research, 113112.

Lenze, E. J., Rodebaugh, T. L., & Nicol, G. E. (2020). A framework for advancing precision

medicine in clinical trials for mental disorders. JAMA Psychiatry.

Papini, S., Rubin, M., Telch, M. J., Smits, J. A. J., & Hien, D. A. (2020). Pretreatment

Posttraumatic Stress Disorder Symptom Network Metrics Predict the Strength of the

Association Between Node Change and Network Change During Treatment. Journal of

Traumatic Stress, 33(1), 64–71. https://doi.org/10.1002/jts.22379

Pfefferbaum, B., & North, C. S. (2020). Mental Health and the Covid-19 Pandemic. The New

England Journal of Medicine. https://doi.org/10.1056/NEJMp2008017

R Core Team. (2017). R: A language and environment for statistical computing. R Foundation

for Statistical Computing. https://www.R-project.org/

Reger, M. A., Stanley, I. H., & Joiner, T. E. (2020). Suicide Mortality and Coronavirus Disease

2019—A Perfect Storm? JAMA Psychiatry.

https://doi.org/10.1001/jamapsychiatry.2020.1060

Robinaugh, D. J., Hoekstra, R. H. A., Toner, E. R., & Borsboom, D. (2020). The network

approach to psychopathology: A review of the literature 2008–2018 and an agenda for

future research. Psychological Medicine, 50(3), 353–366.

https://doi.org/10.1017/S0033291719003404

Robinaugh, D. J., Millner, A. J., & McNally, R. J. (2016). Identifying Highly Influential Nodes

in the Complicated Grief Network. Journal of Abnormal Psychology, 125(6), 747–757.

https://doi.org/10.1037/abn0000181

Spitzer, R. L., Kroenke, K., Williams, J. B., & Löwe, B. (2006). A brief measure for assessing
COVID-19 ANXIETY NETWORK ANALYSIS 21

generalized anxiety disorder: The GAD-7. Archives of Internal Medicine, 166(10), 1092–

1097.

Taylor, S., Landry, C., Paluszek, M., Fergus, T. A., McKay, D., & Asmundson, G. J. (2020).

Development and initial validation of the COVID Stress Scales. Journal of Anxiety

Disorders, 102232.

van Borkulo, C., Boschloo, L., Borsboom, D., Penninx, B. W., Waldorp, L. J., & Schoevers, R.

A. (2015). Association of symptom network structure with the course of depression.

JAMA Psychiatry, 72(12), 1219–1226.

Williams, D. R., Rhemtulla, M., Wysocki, A. C., & Rast, P. (2019). On Nonregularized

Estimation of Psychological Networks. Multivariate Behavioral Research, 54(5), 719–

750. https://doi.org/10.1080/00273171.2019.1575716

Yang, L., Yin, J., Wang, D., Rahman, A., & Li, X. (2020). Urgent need to develop evidence-

based self-help interventions for mental health of healthcare workers in COVID-19

pandemic. Psychological Medicine, 1–2. https://doi.org/10.1017/S0033291720001385

Yang, Y., Li, W., Zhang, Q., Zhang, L., Cheung, T., & Xiang, Y.-T. (2020). Mental health

services for older adults in China during the COVID-19 outbreak. The Lancet.

Psychiatry, 7(4), e19. https://doi.org/10.1016/S2215-0366(20)30079-1