Table 2. Predictors of persisting symptoms in COVID-19, based on a multivariate logistic regression model (mean, proportion in percent/95% CI).

	Persisting Symptoms		Multivariable
	yes (n = 305)	no (n = 591)	aOR [95% CI]	p-value
Age (years)	46.4 [44.5– 48.3]	39.5 [38– 40.9]	1.02 [1.01– 1.03]	<0.01
BMI (kg/m²)	26.3 [25.7– 26.9]	26.3 [25.9– 26.7]
Follow-up time (months)	6.6 [6.4– 6.9]	7.2 [6.8– 7.7]	1.02 [0.98– 1.07]	0.4
Female sex	198 (64.9%)	264 (44.7%)	1.92 [1.4– 2.7]	<0.01
Symptoms during acute phase
– At least one COVID-19 symptom	288 (94.4%)	480 (81.2%)	4.2 [1.7– 10.6]	<0.01
– Cough	167 (54.8%)	246 (41.6%)
– Anosmia/ageusia	160 (52.5%)	189 (32.0%)	1.8 [1.3– 2.6]	<0.01
– Fever	135 (44.3%)	185 (31.3%)
– Headache	128 (42.0%)	192 (32.5%)
– Limb pain	129 (42.3%)	183 (31.0%)
– Rhinitis	81 (26.6%)	146 (24.7%)
– Sore throat	75 (24.6%)	139 (23.5%)
– Shortness of breath	88 (28.9%)	71 (12.0%)	2.1 [1.4– 3.2]	<0.01
– Abdominal pain/diarrhoea	50 (16.4%)	50 (8.5%)
– Fatigue	21 (6.9%)	11 (1.9%)
– Total (mean/patient)	3.5 [3.3– 3.7]	2.5 [2.3– 2.6]	1.1 [1.0– 1.3]	0.048

An aOR of 1.02 for “age” translates to a 2% increased risk of persisting symptoms, an aOR of 4.2 for “at least one COVID-19 symptom” translates to a 320% increased risk of persisting symptoms in patients with symptomatic vs. those with asymptomatic COVID-19 infection.

BMI, body mass index; aOR, adjusted odds ratio; CI, confidence interval