Study population and setting
This prospective cohort study based in Tours, France, from March 17 to June 3, 2020, followed 150 patients (aged 18-75) with non-critical COVID-19 (symptoms not requiring intensive care unit admission) who did not reside at a nursing home for 60 days from laboratory-confirmed COVID-19 diagnosis to assess COVID-19 symptom persistence. Baseline data on age, sex, and comorbidities (obesity, chronic respiratory disease, dialysis, heart failure or previous cardiovascular event, liver cirrhosis, insulin-dependent diabetes, immunosuppression, and/or pregnancy) were collected via standardized electronic health record review, whereas symptoms seven, 30, and 60 days after diagnosis were collected via telephone on standardized case report forms. This study focused on responses at day 30 and 60; persisting symptoms were defined as the presence of at least one of the following: weight loss (> -5%), shortness of breath, low energy, chest pain, palpitations, lack of smell or taste, headache, skin changes, joint pain, muscle aches, digestive disorders (i.e. diarrhea, vomiting, pain), fever (>38°C), or sick leave that was not present prior to PCR detection of SARS-CoV-2. They compared symptoms across the baseline, 30-day, and 60-day timepoints with chi-square, Fisher exact, t-tests, or Mann-Whitney tests as appropriate and then used logistic regression to calculate bivariate odds ratios (OR) and 95% confidence intervals (CIs) for baseline characteristics and the presence of any persistent symptoms at day 30 and 60, respectively.
Summary of Main Findings
Baseline characteristics revealed that slightly more than half of the included participants were female (n=84, 56%), a plurality were between 50 and 59 years old (n=37, 24.7%), a majority had one (n=52, 34.7%) or two or more (n=28, 18.7%) comorbidities, the most common symptoms at baseline were cough, sneezing, and/or rhinitis (n=135, 91.2) and/or muscle aches, headache, and/or low energy (n=129, 87.2), and 53 (35.5%) participants required hospitalization. Symptoms persisted in 103 (68%) participants at 30 days and 86 (66.1% of the 130 remaining participants) at 60 days, with loss of taste or smell, muscle aches, headache, and/or low energy remaining the most common symptoms. Of note, 26 (19.7%) and 14 (11.2%) participants reported taking sick leave due to their symptoms at 30 and 60 days respectively. Baseline oxygen therapy (OR 3.4, 95% CI 1.2-9.5), abnormal chest sounds (OR 3.3, 95% CI 1.3-8), hospitalization (OR 2.8, 95% CI 1.2-6.2), ages 40-49 (OR 13.3, 95% CI 2.8-64.1), and ages 50-59 (OR 5.2, 95% CI 1.5-18.3) were significantly associated with persistent COVID-19 symptoms at 30 days, whereas only baseline hospitalization (OR 2.9, 95% CI 1.3-6.9) and ages 40-49 (OR 15.3, 95% CI 2.8-83.9) were significantly associated with persistent symptoms at 60 days.
This study prospectively follows a group of adult patients for 60 days from SARS-COV-2 diagnosis, which limits the influence of participant recall bias. Additionally, this study includes patients with non-critical COVID-19, who represent a sizable portion of patients infected with SARS-CoV-2. Finally, the results from this study support patient experiences that, to this point, have been underreported in the scientific literature.
This study includes a relatively small sample of patients from a single city in France, limiting its generalizability to other settings and the precision of the effect estimates. While the authors write that they characterized symptom severity in their standardized case report forms, they did not report those data, which limits interpretability of persistent symptoms. It is also unclear how meaningful 30 and 60 days are in the natural history of non-critical COVID-19; and the authors did not collect potentially confounding baseline variables of interest (such as smoking status). In their analysis of the relationship between comorbidities and persistent symptoms, the authors combined all of the comorbidities together (likely because of their small sample size), which may have obscured any relationships between particular comorbidities of interest (e.g. obesity, diabetes) and persistent COVID-19 symptoms. Finally, the authors also did not adjust for multiple comparisons or account for potential confounding variables when calculating odds ratios, which makes it difficult to tell which, if any, of the statistically significant findings are due to chance alone or may be explained by other risk factors.
This study provides evidence that persistent COVID-19 symptoms are common among patients with non-critical COVID-19 disease up to 60 days from diagnosis.
This review was posted on: 5 December 2020