Uncovering Survivorship Bias in Longitudinal Mental Health Surveys During the COVID-19 Pandemic

Czeisler, Mark É; Wiley, Joshua F.; Czeisler, Charles A.; Rajaratnam, Shanthakumar M W

doi:10.1101/2021.01.28.21250694

Cited by 18 publications

(30 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Third, we observed significant, nonrandom, attrition from the cohort. Although the degree of attrition was comparable to that reported in other cohorts established early in the pandemic [ 37 , 38 ], we saw greater dropout among those who reported poorer psychological well-being at baseline and keyworkers, who were arguably the most risk exposed to COVID-19 infection. This may mean that the relationships observed between psychological distress and COVID-19 infection outcomes underestimate the true effects in the population, given infection outcomes for those with the greatest distress were less likely to be represented in the models.…”

Section: Discussionsupporting

confidence: 76%

Psychological Predictors of Self-reported COVID-19 Outcomes: Results From a Prospective Cohort Study

Ayling

Coupland

Chalder

et al. 2022

Annals of Behavioral Medicine

View full text Add to dashboard Cite

Background Previous research has shown that psychological factors, such as stress and social support, are associated with greater susceptibility to viral respiratory illnesses and more severe symptoms. During the COVID-19 pandemic there has been a well-documented deterioration in psychological well-being and increased social isolation. This raises questions as to whether those experiencing psychological adversity during the pandemic are more at risk of contracting and/or experiencing COVID-19 symptoms. Purpose To examine the relationship between psychological factors and the risk of COVID-19 self-reported infection and the symptomatic experience of SARS-CoV-2 (indicated by the number and severity of symptoms). Methods As part of a longitudinal prospective observational cohort study, 1,087 adults completed validated measures of psychological well-being during April 2020 and self-reported incidence of COVID-19 infection and symptom experience across the pandemic through to December 2020. Regression models were used to explore these relationships controlling for demographic and occupational factors. Results Greater psychological distress during the early phase of the pandemic was significantly associated with subsequent self-reported SARS-CoV-2 infection as well as the experience of a greater number and more severe symptoms. Conclusions COVID-19 infection and symptoms may be more common among those experiencing elevated psychological distress. Further research to elucidate the mechanisms underlying these associations is needed.

show abstract

Section: Discussionsupporting

confidence: 76%

Psychological Predictors of Self-reported COVID-19 Outcomes: Results From a Prospective Cohort Study

Ayling

Coupland

Chalder

et al. 2022

Annals of Behavioral Medicine

View full text Add to dashboard Cite

show abstract

“…Second, although retention was relatively high (75.6%), veterans with a pre-pandemic diagnosis of MDD, GAD, and/or PTSD were less likely to complete the follow-up survey than veterans without these diagnoses (67.4% v. 76.7%). Therefore, it is possible that the prevalence of these disorders may have been underestimated due to retention or survivorship bias (Czeisler, Wiley, Czeisler, Rajaratnam, & Howard, 2021 ). Third, although we assessed peri-pandemic stress later in the pandemic than previous studies, mental health effects of the pandemic may be delayed and manifest over time, as economic, health, and social consequences of the pandemic compound.…”

Section: Discussionmentioning

confidence: 99%

Mental health impact of the COVID-19 pandemic in U.S. military veterans: a population-based, prospective cohort study

Hill

Nichter

et al. 2021

Psychol. Med.

View full text Add to dashboard Cite

Background: The coronavirus disease-2019 (COVID-19) pandemic has caused myriad health, social, and economic stressors. To date, however, no known study has examined changes in mental health during the pandemic in the U.S. military veteran population. Methods: Data were analyzed from the 2019-2020 National Health and Resilience in Veterans Study, a nationally representative, prospective cohort survey of 3,078 veterans. Pre-to-peri-pandemic changes in psychiatric symptoms were evaluated, as well as pre-pandemic risk and protective factors and pandemic-related correlates of increased psychiatric distress. Results: The prevalence of generalized anxiety disorder (GAD) positive screens increased from pre-to peri-pandemic (7.1% to 9.4%; p<0.001) and was driven by an increase among veterans aged 45-64 years (8.2% to 13.5%; p<0.001), but the prevalence of major depressive disorder and posttraumatic stress disorder positive screens remained stable. Using a continuous measure of psychiatric distress, an estimated 13.2% of veterans reported a clinically meaningful pre-to-peri-pandemic increase in distress (mean=1.1 standard deviation). Veterans with a larger pre-pandemic social network size and secure attachment style were less likely to experience increased distress, whereas veterans reporting more pre-pandemic loneliness were more likely to experience increased distress.Concerns about pandemic-related social losses, mental health COVID-19 effects, and housing stability during the pandemic were associated with increased distress, over-and-above prepandemic factors. Conclusions: Although most U.S. veterans showed resilience to mental health problems nearly one year into the pandemic, the prevalence of GAD positive screens increased, particularly among middle-aged veterans, and 1-of-7 veterans experienced increased distress.Clinical implications of these findings are discussed.

show abstract

“…Although significant improvements in depression and anxiety were evident after easing of restrictions compared with baseline during the lockdown among completers, such results might only be representative of those who participated in all surveys but not those who dropped out (or survivorship bias). (22) In fact, the levels of depression and anxiety may be underestimated in our cohort, given those with highest levels of depression and anxiety at baseline were less likely to complete further questionnaires. To address this issue, we estimated means and prevalence of depression and anxiety at both Time 2 and 3 using MI data.…”

Section: Discussionmentioning

confidence: 85%

“…This is comparable to other cohorts established early in the pandemic. (22, 32) Those who dropped out were also significantly different compared with those who completed all three surveys, both in demographic characteristics and baseline mental health. The high proportion of drop-outs is likely to have led to an under-estimation of depression and anxiety in our cohort.…”

Section: Discussionmentioning

confidence: 92%

“…Indeed, reaching and retaining individuals most in need of mental health support is not uncommon in such research. (22) The high proportion of keyworkers in our cohort (50% at baseline) may have also contributed to the high drop-out rate. These individuals were, by definition, providing essential services throughout the pandemic and will, therefore, have had less capacity to remain engaged in the research.…”

Section: Discussionmentioning

confidence: 94%

See 1 more Smart Citation

The prevalence, incidence, prognosis and risk factors for depression and anxiety in a UK cohort during the COVID-19 pandemic

Ayling

Chalder

Massey

et al. 2021

Preprint

View full text Add to dashboard Cite

Background: The COVID-19 pandemic had profound immediate impacts on population mental health. However, in whom the effects may be prolonged is less clear. Aims: To investigate the prevalence, incidence, prognosis, and risk factors for depression and anxiety reported in a UK cohort over three distinct periods in the pandemic in 2020. Method: An online survey was distributed to a UK community cohort (n=3097) at three points: April (baseline), July-September (T2) and November-December (T3). Participants completed validated measures of depression and anxiety on each occasion and we prospectively explored the role of socio-demographic factors and psychological factors (loneliness, positive mood, perceived risk of and worry about COVID-19) as risk factors. Results: Depression (PHQ-9 means - baseline: 7.69, T2: 5.53, T3: 6.06) and anxiety scores (GAD-7 means -baseline: 6.59, T2: 4.60, T3: 4.98) were considerably greater than pre-pandemic population norms. Women reported greater depression and anxiety than men. Being younger, having prior mental health disorders, more negative life events due to COVID-19, as well as greater loneliness and lower positive mood at baseline were significant predictors of poorer mental health outcomes. Conclusion: The negative impact of the COVID-19 pandemic on mental health has persisted to some degree. Younger people and individuals with prior mental health disorders were at greatest risk. Easing of restrictions might bring the opportunity for a return to social interaction, which could mitigate the risk factors of loneliness and positive mood.

show abstract

Uncovering Survivorship Bias in Longitudinal Mental Health Surveys During the COVID-19 Pandemic

Cited by 18 publications

References 36 publications

Psychological Predictors of Self-reported COVID-19 Outcomes: Results From a Prospective Cohort Study

Psychological Predictors of Self-reported COVID-19 Outcomes: Results From a Prospective Cohort Study

Mental health impact of the COVID-19 pandemic in U.S. military veterans: a population-based, prospective cohort study

The prevalence, incidence, prognosis and risk factors for depression and anxiety in a UK cohort during the COVID-19 pandemic

Contact Info

Product

Resources

About