The moderating effects of reported pre-pandemic social anxiety, symptom impairment, and current stressors on mental health and affiliative adjustment during the first wave of the COVID-19 pandemic

ABSTRACT

Background and objectives

Individuals with social anxiety (SA) have well-established fears of being negatively evaluated and exposing self-perceived flaws to others. However, the unique impacts of pre-existing SA on well-being and interpersonal outcomes within the stressful context of the pandemic are currently unknown.

Design

In a study that took place in May 2020, we surveyed 488 North American community participants online.

Methods

We used multiple linear regression to analyze whether retrospective reports of pre-pandemic SA symptoms predicted current coronavirus anxiety, loneliness, fears of negative evaluation, use of preventive measures, and affiliative outcomes, and whether pre-pandemic functional impairment and recent COVID-related stressors moderated these relations.

Results

Results highlighted the negative effects of pre-pandemic SA on current mental health functioning, especially for participants with higher pre-pandemic functional impairment and greater exposure to COVID-related stressors. Although participants with higher pre-pandemic SA reported currently feeling lonelier and more fearful of negative evaluation, they also endorsed greater efforts to affiliate with others.

Conclusions

High SA individuals may have heightened desire for social support within the isolating context of the pandemic, in which COVID-related social restrictions enable greater avoidance of social evaluation but may also mask the enduring impairment associated with pre-pandemic SA.

KEYWORDS:

• Social anxiety
• pandemic
• COVID-19
• impairment
• stressors
• social support

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.

Notes

1 Because the four reverse-scored items of the scale have been shown to compromise validity, while the remaining straightforward items have demonstrated excellent internal consistency across a variety of samples (for example, α = .95 in an MTurk sample; Rodebaugh et al., 2007; Rodebaugh et al., 2011), only straightforward items were included. We replicated the results of our models using the SIAS with reverse-coded items retained. Thus, results are reported with the reverse-scored items removed, consistent with Rodebaugh et al.’s (2011) recommendations.

Additional information

Funding

This work was supported by Social Sciences and Humanities Research Council of Canada: [Grant Number Insight Grant 435-2018-0959].