General discussion 9 275 9.2.1 Spotlight on the occurrence, clinical features, immunopathology and determinants of long COVID How often does long COVID occur? Establishing the frequency at which long COVID occurs is of vital importance for several reasons. For example, clinicians may wish to be able to predict the proportion of individuals who develop persistent symptoms according to the severity of their initial infection, so that tailored interventions soon after initial infection can be implemented. Meanwhile, public health officials and politicians may wish to know the total population of individuals long COVID. In evaluating the occurrence of long COVID, it is therefore important to delineate between the occurrence of the condition in specific sub-groups versus the prevalence of persistent symptoms in a specific study or wider population. Below, I discuss different approaches towards estimating the magnitude of long COVID as a health issue and touch upon the factors that make comparisons between studies difficult. Data from the RECoVERED cohort were used to estimate the occurrence of long COVID according to initial disease severity. Using survival analysis of recovery from COVID-19 symptoms, we found that approximately one-third of those with initially mild COVID-19, twothirds of those with moderate illness and four-fifths of those with severe or critical disease reported at least one ongoing symptom (of a total of 18 symptoms) 12 weeks after illness onset (i.e., using the National Institute for Health and Care Excellence [NICE] definition of long COVID[50]) (this thesis, Chapter 4). These severity-specific estimates provide clinicians with a detailed view of prognosis within each band of COVID-19 severity. However, as we did not incorporate multiple events of symptom onset and recovery in our survival analysis, our estimates do not take into account that symptoms may recur. As such, they do not reflect the proportion of study participants with ongoing symptoms at different time-points. We therefore additionally evaluated the proportion of participants who reported at least one symptom in each monthly symptom questionnaire (using the number of participants who had completed the questionnaire as the denominator in a cross-sectional analysis). We also hypothesized that the number of symptoms deemed relevant to long COVID may impact the proportion of participants reporting at least one symptom. In this analysis, long COVID prevalence at 12 months since illness onset ranged from 39.6% (95%CI=33.4-46.2) when using a limited selection of symptoms to 80.6% (95%CI=74.8-85.4) when considering any reported symptom (Figure 9.1)[51]. The selection of symptoms leads to high variability of prevalence estimates even within the same study population. Hence, harmonised data collection tools are required to achieve greater reproducibility and comparability of results. The effect of study design and outcome measurement on prevalence estimates was also demonstrated
RkJQdWJsaXNoZXIy MTk4NDMw