Table 3. Determinants associated with admission to intensive care units in 321 Covid-19 patients at Lyon University hospitals, France.
A: With age in the model | |||
Adjusted OR | 95%CI | P-value | |
Sex | |||
Female (17/147) | 1 | ||
Male (39/174) | 1.66 | 0.76–3.62 | 0.2 |
Age (years) | |||
< 70 (27/144) | 1 | ||
70–85 (21/100) | 0.75 | 0.33–1.7 | 0.49 |
85–103 (8/77) | 0.17 | 0.05–0.51 | 0.002 |
aTemperature Continuous variable | 1.56 | 1.06–2.28 | 0.02 |
Oxygen saturation (%) | |||
90–100 (26/265) | 1 | ||
<90 (30/56) | 12.45 | 5.27–29.4 | <0.0001 |
Abnormal lung auscultation | |||
No (13/141) | 1 | ||
Yes (43/180) | 3.58 | 1.58–8.11 | 0.002 |
C Reactive Protein, mg/L (Normal range <5 mg/L) | |||
≤100 (21/217) | 1 | ||
>100 (35/104) | 2.7 | 1.29–5.66 | 0.008 |
Monocyte count, x109/L (Normal range 0.2–0.9 x109/L) | |||
≥ 0.3 (36/263) | 1 | ||
< 0.3 (20/58) | 3.28 | 1.40–7.68 | 0.006 |
B: With delay between onset and hospital admission in the model | |||
Adjusted OR | 95%CI | P-value | |
Sex | |||
Female (17/147) | 1 | ||
Male (39/174) | 1.88 | 0.87–4.10 | 0.11 |
Delay between onset and hospital admission (day) | |||
0–2 (5/73) | 1 | ||
3–10 (38/200) | 5.96 | 1.65–21.5 | 0.006 |
>10 (13/48) | 6.90 | 1.68–28.4 | 0.007 |
aTemperature Continuous variable | 1.62 | 1.10–2.39 | 0.01 |
Oxygen saturation (%) | |||
90–100 (26/265) | 1 | ||
<90 (30/56) | 11.48 | 4.86–27.1 | <0.0001 |
Abnormal lung auscultation | |||
No (13/141) | 1 | ||
Yes (43/180) | 3.50 | 1.56–7.85 | 0.002 |
C Reactive Protein, mg/L (Normal range <5 mg/L) | |||
≤100 (21/217) | 1 | ||
>100 (35/104) | 2.01 | 0.95–4.23 | 0.07 |
Monocyte count, x109/L (Normal range 0.2–0.9 x109/L) | |||
≥ 0.3 (36/263) | 1 | ||
< 0.3 (20/58) | 2.86 | 1.26–6.50 | 0.02 |
OR: odds ratio, CI: confdence interval
In multivariable logistic regression: i) ICU admission was the outcome and ii) age or delay between onset and hospital admission sex, continuous temperature, oxygen saturation, abnormal lung auscultation, monocyte count and C-reactive protein were additive covariates, Akaike information criterion = 216.0, test of Hosmer and Lemeshow goodness of fit with 10 bins: P = 0.59
aodds ratio of ICU admission was multiplied by 1.62 per degree celsius increase