key: cord-0747404-6i9sddep
authors: Vedhara, Kavita; Ayling, Kieran; Jia, Ru; Fairclough, Lucy; Morling, Joanne R; Ball, Jonathan K; Knight, Holly; Blake, Holly; Corner, Jessica; Denning, Chris; Bolton, Kirsty; Jackson, Hannah; Coupland, Carol; Tighe, Patrick
title: Relationship between Anxiety, Depression and Susceptibility to SARS-CoV-2 Infection: Proof of Concept
date: 2022-01-12
journal: J Infect Dis
DOI: 10.1093/infdis/jiac006
sha: ea7e1000d846a22699728a15824301f18c65eca7
doc_id: 747404
cord_uid: 6i9sddep

Psychological factors can influence susceptibility to viral infections. We examined whether such influences are evident in SARS-CoV-2 infection. Participants (n=102) completed measures of anxiety, depression, positive mood and loneliness and provided a blood sample for the measurement of antibodies to the SARS-CoV-2 spike and nucleocapsid proteins. SARS-CoV-2 seropositivity was significantly associated with anxiety and significantly positively associated with depression. The model remained significant after adjustment for age and gender, although anxiety and depression were no longer significant independent predictors. These findings offer early support for the hypothesis that psychological factors may influence susceptibility to SARS-CoV-2 infection.

proven to be a pernicious adversary. At the time of writing, the virus has resulted in an estimated 150 million infections and over 3 million deaths worldwide. But what factors determine susceptibility to the infection? Previous research with other respiratory viruses suggests psychological and social factors may also influence susceptibility to the infection [1] . Viral challenge studies in particular have shown that the greatest risk of disease occurs in individuals contending with chronic stressors (of 1 month or longer in duration) and where the sources of stress are interpersonal or employment related [2] . In comparison, the experience of positive emotions and social support confer protection against viral infection [2] .

This previous work suggests that several of the psychological and social consequences of the SARS-CoV-2 pandemic (i.e., chronic financial and interpersonal stressors; increased negative and reduced positive emotions and social isolation) may also have implications for an individual's risk of contracting SARS-CoV-2. To examine this hypothesis, we conducted an exploratory observational study in which we investigated the relationship between COVID-19 infection (as indicated by the presence of antibodies to the SARS-CoV-2 spike and nucleocapsid proteins) and psychological factors previously shown to be associated with disease susceptibility (i.e., positive mood and loneliness) and known to have been affected by the pandemic (i.e., anxiety and depression [3] ) .

Eligibility criteria specified that participants should be aged 18 or over.

A c c e p t e d M a n u s c r i p t 5 Participants were staff and students at a higher education institution in the United Kingdom (UK) participating in the institution's COVID-19 surveillance programme. Participants were recruited through word of mouth and a campus wide campaign and participation was voluntary. All participants were recruited between 3/12/20 and 17/2/21.

As the present sample was recruited opportunistically alongside the COVID-19 surveillance programme and the research was exploratory and hypothesis generating, no sample size target was set apriori.

As part of the COVID-19 surveillance programme, participants who agreed to antibody testing received a finger-prick blood sample kit including a disposable lance, a barcoded 10µL Mitre© sampling stick (Neoteryx, USA) and sampling instructions. Dried blood samples were assayed for IgG antibodies to SARS-CoV-2 spike and nucleocapsid proteins using ELISA and all assays were performed on Opentrons OT-2 Precision liquid handling robots. Methods have been reported previously [4] . Seropositivity was indicated according to the ratio method i.e., ratio of the average sample optical density (OD)/average Negative OD.

According to this approach a ratio of >=1.3 = seropositive; a ratio between 1.1 <1.3 = indeterminate; a ratio <1.1 = seronegative.

Participants also completed an online survey (https://www.onlinesurveys.ac.uk/) which 

All analyses were performed using SPSS (version 25). Summary statistics were calculated for the total group of participants and split by seropositivity status using counts and percentages, and means (with SDs) or medians (IQR) as appropriate. Anxiety and depression scores were also dichotomised according to established thresholds for diagnosis of 'caseness' (PHQ-9 score greater or equal to 10, GAD-7 score greater or equal to 8) [8] . Seven individuals were categorised as having an 'indeterminant' antibody result and were excluded from all analyses.

The associations between psychological factors (anxiety, depression, positive mood and loneliness) and the binary SARS-CoV-2 seropositivity outcome were examined using logistic regression controlling for age and gender. These latter indices were included in view of evidence that the risk of infection may be greater in males and older individuals [9, 10] . Study analyses were not preregistered.

We convened a virtual PPI Group to support this research. Individuals participated via Microsoft (MS) Teams in one-to-one or group discussions which informed the length and structure of the survey (e.g., order of presentation of items) and language of the information sheet (i.e., to maximise interest in the research).

SARS-CoV-2 antibody status and survey data were available for N=102 individuals. Of these, N=85 (83%) were seronegative and N=17 (17%) were seropositive (as indicated by serum levels of IgG antibody to SARS-CoV-2 spike and nucleocapsid combined). Of A c c e p t e d M a n u s c r i p t 7 participants who reported a previous positive COVID-19 test result (N=7), the average number of days since their test was 74 days (range 51-105 days). A summary of the characteristics of the whole cohort and according to seropositivity status is shown in Table 1 .

The results from the first step of the multivariable logistic regression model suggested that the likelihood of being seropositive was significantly associated with anxiety and depression:

with the likelihood being 23% lower per unit increase in anxiety (OR=0.77, 95%CI: 0.60, 0.98); and 27% higher per unit increase in depression (OR=1.27, 95%CI: 1.05, 1.54). After adjusting for age and gender, anxiety and depression were no longer significant independent predictors of seropositivity, although the overall model remained statistically significant, the direction of the relationships remained unchanged and the effect on odds ratios was modest.

To our knowledge, this is the first reported analysis of the relationship between psychological factors and SARS-CoV-2 seropositivity. Our findings provide early support for the hypothesis that psychological factors may be related to the likelihood of SARS-CoV-2 infection: with the primary analyses, prior to adjustment for age and gender, suggesting that lower levels of anxiety, and higher levels of depression, were significantly and independently associated with SARS-CoV-2 seropositivity. Several issues are worthy of further discussion.

First, our approach was predicated on antibody seropositivity being an established surrogate indicator for recent infection with SARS-CoV-2 i.e., it is unlikely that someone would be seropositive for SARS-CoV-2 and have not previously been infected with the virus. Our primary analysis, is consistent with this assumption. However, it is important to acknowledge several factors that can confound this relationship. First, antibodies are less likely to be detectable several months after natural infection [11] . This means that the absence of antibodies cannot be assumed to mean the absence of previous infection. Second, A c c e p t e d M a n u s c r i p t 8 asymptomatic SARS-CoV-2 infection is reasonably prevalent (estimated to occur in one-third of infected individuals). Such individuals may be less likely to seek SARS-CoV-2 tests (thus the absence of a previous positive COVID-19 test may not accurately represent presence/absence of previous infection). Similarly, such individuals may also be less likely to seek SARS-CoV-2 antibody testing (and may therefore be underrepresented in the present cohort). Thus, further work is needed to examine the relationship between psychological risk factors and SARS-CoV-2 infection and would ideally be examined in large cohorts undertaking asymptomatic testing for COVID-19 to mitigate the confounding effects of asymptomatic infection and antibody decay over time.

A second issue concerns the direction of the relationship between anxiety and seropositivity. This initially appears counterintuitive and contrary to previous work which has more commonly shown that negative moods are positively associated with risk of infection. This more expected relationship relates to negative emotions altering biological or behavioural pathways that lead to immune dysregulation and, in turn, increased disease risk [2] . We hypothesise that the existence of an inverse relationship may point to a behavioural pathway in which individuals who are less anxious may be more likely to engage in behaviours that increase their risk of exposure and infection. Some support for such a pathway can be seen in work which has shown that increased anxiety is related to a greater fear of contracting COVID-19 infection [12] and increased willingness to engage in preventative behaviours [13] . Thus, it is plausible that lower anxiety and reduced fear of infection, precipitate more risk-taking behaviours (e.g., increased social contact, alcohol consumption) [14] , or indeed less protective behaviours (e.g., less social distancing) [13] . In post-hoc analyses we sought to interrogate this possibility by examining whether and how the relationship between anxiety, depression and seropositivity changed when loneliness was excluded from the A c c e p t e d M a n u s c r i p t 9 models (Supplementary Table 1 ). The models remained unchanged, although it should be noted that our measure of loneliness might be a poor proxy for social contact.

A third issue concerns the fact that individuals seropositive for SARS CoV-2 tended to also have higher depression scores. The presence of antibody indicates recent infection, with antibody levels usually detectable within a few weeks of infection, and persisting for between 2-8 months [11] , depending on the nature (i.e., symptomatic versus asymptomatic) and duration of infection. This raises the possibility that the increased depressive symptoms seen in our seropositive individuals may be secondary to the original infection [15] and is consistent with evidence that depression is common post COVID-19 infection [11] .

Finally, in further post-hoc analyses we included an interaction term to capture the interaction between anxiety and depression and explore its relationship to seropositivity (Supplementary   Table 2 ). However, the interaction was found to be non-significant.

In concluding, we acknowledge some potential methodological limitations. First, both the sample size and infection rate reported here are modest. This means that only relatively large effect sizes could have been reliably detected as being statistically significant. It is important therefore these results are seen as hypothesis generating and requiring replication. Second, our study design does not allow us to elucidate the temporal relationship between our psychological risk factors and SARS-CoV-2 infection. All scales asked respondents to indicate their experiences in the previous 2 weeks, and while antibody seropositivity indicates recent infection we cannot be certain about when the infection occurred. Thus, we are, at best describing cross-sectional relationships from which the causal direction cannot be determined.

Notwithstanding these considerations, these findings suggest that psychological factors may play a role in determining SARS-CoV-2 infection. Specifically we have observed an inverse relationship between anxiety and SARS-CoV-2 seropositivity which we hypothesise may be A c c e p t e d M a n u s c r i p t 10 indicative of a behavioural pathway in which the least anxious individuals increase their risk of infection through engaging in more risk-taking behaviours or indeed less engagement with protective behaviours.

The study sponsor and funders did not play a role in the study design, collection; analysis, and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. All authors were independent of the funding source and had full access to all the data and can take responsibility for the integrity of the data and accuracy of the analyses. A c c e p t e d M a n u s c r i p t 13 

Stress and Health: A Review of Psychobiological Processes

Psychosocial Vulnerabilities to Upper Respiratory Infectious Illness: Implications for Susceptibility to Coronavirus Disease 2019 (COVID-19)

Mental health in the UK during the COVID-19 pandemic: cross-sectional analyses from a community cohort study

Potent anti-SARS-CoV-2 Antibody Responses are Associated with Better Prognosis in Hospital Inpatient

The Patient Health Questionnaire Somatic, Anxiety, and Depressive Symptom Scales: a systematic review

A brief measure for assessing generalized anxiety disorder: the GAD-7

New Well-being Measures: Short Scales to Assess Flourishing and Positive and Negative Feelings

National Collaborating Centre for Mental Health. The Improving Access to Psychological Therapies Manual -Appendices and helpful resources

Seroprevalence of SARS-CoV-2 antibodies and associated factors in healthcare workers: a systematic review and metaanalysis

Population-based seroprevalence surveys of anti-SARS-CoV-2 antibody: An up-to-date review

Dynamics of SARS-CoV-2 neutralising antibody responses and duration of immunity: a longitudinal study

Psychosocial predictors of virus and social distancing fears in undergraduate students living in a US COVID-19 "hotspot

Association Between Preventive Behaviour and Anxiety at the Start of the COVID-19 Pandemic in Slovenia

Alcohol narrows physical distance between strangers

Long-COVID and Post-COVID Health Complications: An Up-to-Date Review on Clinical Conditions and Their Possible Molecular Mechanisms