Skip to main content

Table 3 Multivariable-adjusted hazard-ratios of COVID-19-Like Symptoms (CLS) according to covariate valuesa

From: Incidence and risk factors of COVID-19-like symptoms in the French general population during the lockdown period: a multi-cohort study

 

Hazard-Ratio

P-Value

Age group

  < 40

Reference

 

 [40–50[

0.80 (0.72; 0.90)

< 0.0001

 [50–60[

0.68 (0.60;0.76)

< 0.0001

 [60–70[

0.52 (0.44;0.62)

< 0.0001

  > =70

0.34 (0.27; 0.44)

< 0.0001

Gender

 Female

reference

 

 Male

0.99 (0.91; 1.07)

0.7267

Regions

 Ile-de-France

1.31 (1.19; 1.44)

< 0.0001

 GrandEst

1.29 (1.14; 1.47)

< 0.0001

 Other French metropolitan regions

reference

 

Living Area

 Rural

Reference

 

  < 20,000 inhab.

1.07 (0.93; 1.22)

0.3629

 20–000-100,000 inhab.

1.14 (1.00; 1.30)

0.0471

  > 100,000 inhab.

1.15 (1.02; 1.29)

0.0241

Household size and composition

 Nb persons (incl. participant)

  

  1

Reference

 

  2

0.95 (0.85; 1.06)

0.3405

  3 or +

1.00 (0.87; 1.16)

0.9603

 Nb children (< 18 yrs)

  

  0

Reference

 

  1 or +

1.28 (1.13; 1.45)

0.0001

BMI (kg/m2)

  < 18.5

1.04 (0.85; 1.27)

0.6916

 [18.5; 25[

Reference

 

 [25; 30[(overweight)

1.17 (1.07; 1.27)

0.0005

  > =30 (obese)

1.41 (1.25; 1.58)

< 0.0001

Professional activity before lockdown

 Student

1.01 (0.71; 1.44)

0.9620

 Working

Reference

 

 Looking for a job

1.05 (0.86; 1.27)

0.6618

 Retired

0.86 (0.72; 1.03)

0.1057

 Not working due to health conditions

1.30 (0.96; 1.77)

0.0894

 No professional activity (house wife or husband)

0.89 (0.68; 1.16)

0.3773

Essential job position

 Health care worker (Y vs N)

1.18 (1.00; 1.40)

0.0535

 Other essential job (Y vs N)

1.01 (0.90; 1.14)

0.8922

Chronic diseases (Y vs N)

 Asthma, COPD, other resp. diseases

1.41 (1.24; 1.60)

< 0.0001

 Anxiety, depression

1.31 (1.08; 1.58)

0.0065

 Other

1.22 (1.08; 1.37)

0.0013

  1. a with stratification on the cohort. 6567 (6%) participants excluded from the multivariable model due to missing values including 190 with CLS; educational level was not significantly associated with CLS when included in the multivariable model (high school degree or undergraduate versus <high-school degree, P = 0.7743; graduate or doctorate versus <high-school degree, P = 0.6299), and was not kept in the final model due to missing information on this covariate in the E4NG2 cohort. To avoid overfitting Positive RT-PCR in another household member was not entered in the model. However, results were unchanged when this variable was entered (not shown)