Multivariable logistic regression for post-COVID-19 breathlessness
| Post-COVID-19 breathlessness | OR (95% CI) | ||||
| No | Yes | Univariable | Multivariable | Multiple imputation | |
| Sex at birth | |||||
| Male | 416 (54.2) | 352 (45.8) | 1 (ref) | 1 (ref) | 1 (ref) |
| Female | 195 (42.6) | 263 (57.4) | 1.59 (1.26–2.01) (p<0.001) | 1.44 (1.10–1.90) (p=0.009) | 1.56 (1.21–2.00) (p=0.001) |
| Age at admission, years | |||||
| <30 | 10 (40.0) | 15 (60.0) | 1.12 (0.49–2.63) (p=0.795) | 1.20 (0.47–3.12) (p=0.706) | 1.35 (0.57–3.23) (p=0.498) |
| 30–39 | 38 (48.1) | 41 (51.9) | 0.80 (0.49–1.31) (p=0.378) | 0.83 (0.48–1.44) (p=0.511) | 0.86 (0.51–1.44) (p=0.568) |
| 40–49 | 87 (45.8) | 103 (54.2) | 0.88 (0.62–1.26) (p=0.482) | 0.96 (0.64–1.44) (p=0.832) | 0.96 (0.66–1.40) (p=0.832) |
| 50–59 | 151 (42.7) | 203 (57.3) | 1 (ref) | 1 (ref) | 1 (ref) |
| 60–69 | 183 (51.3) | 174 (48.7) | 0.71 (0.53–0.95) (p=0.022) | 0.62 (0.44–0.88) (p=0.007) | 0.70 (0.51–0.96) (p=0.025) |
| 70–79 | 107 (63.7) | 61 (36.3) | 0.42 (0.29–0.62) (p<0.001) | 0.41 (0.26–0.64) (p<0.001) | 0.43 (0.28–0.64) (p<0.001) |
| ≥80 | 29 (72.5) | 11 (27.5) | 0.28 (0.13–0.57) (p=0.001) | 0.27 (0.11–0.60) (p=0.002) | 0.31 (0.14–0.66) (p=0.003) |
| Index of Multiple Deprivation | |||||
| 5, least deprived | 127 (55.2) | 103 (44.8) | 1 (ref) | 1 (ref) | 1 (ref) |
| 4 | 111 (51.9) | 103 (48.1) | 1.14 (0.79–1.66) (p=0.480) | 1.30 (0.85–1.99) (p=0.220) | 1.22 (0.82–1.81) (p=0.328) |
| 3 | 116 (50.9) | 112 (49.1) | 1.19 (0.82–1.72) (p=0.352) | 1.21 (0.80–1.84) (p=0.365) | 1.22 (0.82–1.79) (p=0.327) |
| 2 | 135 (50.8) | 131 (49.2) | 1.20 (0.84–1.71) (p=0.321) | 1.31 (0.87–1.96) (p=0.195) | 1.20 (0.82–1.76) (p=0.338) |
| 1, most deprived | 112 (42.1) | 154 (57.9) | 1.70 (1.19–2.42) (p=0.004) | 1.87 (1.24–2.84) (p=0.003) | 1.67 (1.14–2.44) (p=0.009) |
| Ethnicity | |||||
| White | 422 (48.3) | 451 (51.7) | 1 (ref) | 1 (ref) | 1 (ref) |
| South Asian | 82 (53.6) | 71 (46.4) | 0.81 (0.57–1.14) (p=0.231) | 0.83 (0.55–1.26) (p=0.378) | 0.80 (0.55–1.17) (p=0.244) |
| Black | 52 (56.5) | 40 (43.5) | 0.72 (0.46–1.11) (p=0.137) | 0.57 (0.34–0.95) (p=0.031) | 0.56 (0.35–0.89) (p=0.015) |
| Mixed | 17 (51.5) | 16 (48.5) | 0.88 (0.44–1.77) (p=0.720) | 0.98 (0.44–2.20) (p=0.956) | 0.85 (0.41–1.75) (p=0.656) |
| Other | 27 (51.9) | 25 (48.1) | 0.87 (0.49–1.52) (p=0.616) | 0.84 (0.44–1.62) (p=0.606) | 0.80 (0.44–1.44) (p=0.448) |
| BMI, kg·m−2 | 31.4±7.1 | 32.7±7.1 | 1.03 (1.01–1.04) (p=0.002) | 1.08 (0.97–1.21) (p=0.164) | 1.08 (0.98–1.19) (p=0.107) |
| Number of comorbidities | 2.0±2.0 | 2.4±2.3 | 1.09 (1.03–1.15) (p=0.002) | 1.09 (1.00–1.18) (p=0.049) | 1.08 (1.00–1.17) (p=0.049) |
| Pre-existing respiratory condition | |||||
| No | 449 (50.3) | 444 (49.7) | 1 (ref) | 1 (ref) | 1 (ref) |
| Yes | 162 (48.6) | 171 (51.4) | 1.07 (0.83–1.37) (p=0.611) | 0.85 (0.62–1.17) (p=0.312) | 0.82 (0.61–1.11) (p=0.195) |
| Pre-existing depression or anxiety | |||||
| No | 538 (53.3) | 471 (46.7) | 1 (ref) | 1 (ref) | 1 (ref) |
| Yes | 66 (33.3) | 132 (66.7) | 2.28 (1.66–3.16) (p<0.001) | 1.54 (1.00–2.38) (p=0.050) | 1.58 (1.06–2.35) (p=0.026) |
| Admission duration, days | 13.2±17.2 | 17.2±22.2 | 1.01 (1.00–1.02) (p=0.001) | 1.01 (1.00–1.02) (p=0.064) | 1.01 (1.00–1.02) (p=0.002) |
| WHO clinical progression scale | |||||
| WHO class 3–4 | 110 (49.3) | 113 (50.7) | 1 (ref) | 1 (ref) | 1 (ref) |
| WHO class 5 | 252 (52.8) | 225 (47.2) | 0.87 (0.63–1.19) (p=0.388) | 0.88 (0.61–1.29) (p=0.522) | 0.84 (0.60–1.18) (p=0.314) |
| WHO class 6 | 136 (53.1) | 120 (46.9) | 0.86 (0.60–1.23) (p=0.407) | 0.90 (0.58–1.38) (p=0.619) | 0.80 (0.54–1.18) (p=0.260) |
| WHO class 7–9 | 113 (41.9) | 157 (58.1) | 1.35 (0.95–1.93) (p=0.097) | 1.17 (0.70–1.98) (p=0.548) | 0.92 (0.57–1.47) (p=0.715) |
Data are presented as n (%) or mean±sd, unless otherwise stated. The dependent variable was post-COVID-19 breathlessness. World Health Organization (WHO) clinical progression scale: not requiring continuous supplemental oxygen (levels 3–4); continuous supplemental oxygen only (level 5); continuous positive airway pressure ventilation, bilevel positive airway pressure or high-flow nasal oxygen (level 6); invasive mechanical ventilation, extra-corporeal membrane oxygenation and acute renal replacement therapy (levels 7–9). The logistic regression model also included body mass index (BMI)2.