Preprint: Does pre-infection stress increase the risk of long COVID? Longitudinal associations between adversity worries (...), 2022, Paul, Fancourt

Full title: Does pre-infection stress increase the risk of long COVID? Longitudinal associations between adversity worries and experiences in the month prior to COVID-19 infection and the development of long COVID and specific long COVID symptoms

Abstract

Background Long COVID is increasingly recognised as public health burden. Demographic and infection-related characteristics have been identified as risk factors, but less research has focused on psychosocial predictors such as stress immediately preceding the index infection. Research on whether stressors predict the development of specific long COVID symptoms is also lacking.

Methods Data from 1,966 UK adults who had previously been infected with COVID-19 and who took part in the UCL COVID-19 Social Study were analysed. The number of adversity experiences (e.g., job loss) and the number of worries about adversity experiences within the month prior to COVID-19 infection were used to predict the development of self-reported long COVID and the presence of three specific long COVID symptoms (difficulty with mobility, cognition, and self-care).

The interaction between a three-level index of socio-economic position (SEP; with higher values indicating lower SEP) and the exposure variables in relation to long COVID status was also examined. Analyses controlled for a range of COVID-19 infection characteristics, socio-demographics, and health-related factors.

Findings Odds of self-reported long COVID increased by 1.25 (95% confidence interval [CI]: 1.04 to 1.51) for each additional worry about adversity in the month prior to COVID-19 infection. Although there was no evidence for an interaction between SEP and either exposure variable, individuals in the lowest SEP group were nearly twice as likely to have developed long COVID as those in the highest SEP group (OR: 1.95; 95% CI: 1.19 to 3.19) and worries about adversity experiences remained a predictor of long COVID (OR: 1.43; 95% CI: 1.04 to 1.98).

The number of worries about adversity experiences also corresponded with increased odds of certain long COVID symptoms such as difficulty with cognition (e.g., difficulty remembering or concentrating) by 1.46 (95% CI: 1.02 to 2.09) but not with mobility (e.g., walking or climbing steps) or self-care (e.g., washing all over or dressing).

Interpretation Results suggest a key role of stress in the time preceding the acute COVID-19 infection for the development of long COVID and for difficulty with cognition specifically. These findings point to the importance of mitigating worries and experiences of adversities during pandemics both to reduce their psychological impact but also help reduce the societal burden of longer-term illness.

Open access: https://www.medrxiv.org/content/10.1101/2022.04.06.22273444v1.full-text

 

"Further, research into other post-viral syndromes such as myologic encephalomyelitis/chronic fatigue syndrome (ME/CFS) has shown a heightened risk for individuals who have experienced stressful life events,29 and there is substantial overlap in the symptoms of long COVID and ME/CFS.30,31 Therefore, a relationship between stress and the development of long COVID is theoretically plausible."

 

A couple things jump out about this:

1. No adjustment for vaccination status (they acknowledge this as a shortcoming due to insufficient numbers). But you would think there may be a significant correlation between worries about losing your job, worries about not having enough to eat, worries about accessing medication, etc. and vaccination status. If there were then that could certainly change the results.
   
2. Their survey found 20% reporting Long Covid after adjustment, with 12% unsure. Similar survey results from ONS around the same time period, from a larger and much more representative data set with a similar survey question, found 13% reporting LC. They note as a limitation that fewer than half of those reporting LC had symptoms lasting more than four weeks, which may explain a lot of this difference. They say they did not find a significant result when they only included those with symptoms more than four weeks. Not sure that LC symptoms for less than four weeks is a very meaningful endpoint.

 

Another explanation is that it is just response bias - people who are more likely to report stress on scales are also more likely to report symptoms post-COVID.