Determinants of Long COVID illness duration and service use-an Irish retrospective cohort study

Abstract The Follow-up After Disease Acquisition (FADA) study is a retrospective cohort study of COVID-19 recovery, aimed to characterise the nature and severity of Long COVID (LC) in an Irish community-based sample. Duration analysis is conducted applying Cox’s proportional hazards model to those self-reported recovered from LC (n = 577) and a Gompertz distribution parametric model, determined best fit by AIC, for those still ill with LC (n = 928) at time of survey completion. A negative binomial model is estimated to understand the variation of services accessed for LC. The Hazard in this instance is recovery(<1 = contributing factor to duration). Factors significantly contributing to duration of illness are never married (Hazard Ratio (HR):0.689;p:0.01) higher physical activity levels prior to first C-19 infection (HR:0.964; p:0.06) and number of COVID-19 infections (H:0.926;p:0.07). Those that contract more recent variants are likely to recover quicker from their LC illness (baseline Wildtype. Alpha HR:1.7;p:0.0; Delta HR:3.36;p:0.0; Omicron HR:3.8; p:0.0). In variation of services used for LC, there are some major differences between those self-reporting as recovered from LC at time of survey completion and those still ill. Of those recovered, men, those sickest with LC and sick longest use a wider variation of services. Single people use a significantly narrower range of health services to combat their illness. For those still ill higher education, higher levels of physical activity pre COVID-19, healthcare workers, higher rates of comorbidity, those sick longest and those whose health has been most impacted by LC use a wider variation of services. A poorer health rating negatively impacts variation of services accessed for LC illness. More recent strains of COVID-19 are significantly less likely to lead to LC, this offers hope for the future, albeit more research is required to investigate factors to improve recovery rates for those already experiencing LC. Key messages • Those with Long COVID that recover quickly are fundamentally different to those that do not. • The number of COVID-19 infections is a key indicator of length of Long COVID illness, we need to protect our most vulnerable and put appropriate safeguards in place for healthcare provision.