Table 1.
Association between PPI use and severe outcomes of COVID-19 in the subgroup meta-analysis by various factors
| Factors | No. of studies | Summary HR (95% CI) | Heterogeneity, I2 (%) | Meta-regression |
|---|---|---|---|---|
| All7,9,13,19,20,23–27,29,30 | 10 | 1.48 (1.12–1.97) | 72.7 | |
| Mean age | 0.57 | |||
| < 60 years20,24–26,30 | 5 | 1.66 (1.08–2.57) | 83.9 | |
| > 60 years7,9,13,19,23,29 | 6 | 1.42 (1.08–1.87) | 63.2 | |
| Research location | 0.60 | |||
| Hospital7,9,19,23–27,29,30 | 10 | 1.58 (1.20–2.08) | 77.2 | |
| Community13,20 | 2 | 1.32 (0.72–2.44) | 71.7 | |
| Administration time of PPIs | 0.41 | |||
| At-home7,9,13,19,20,23,24,29,30 | 9 | 1.42 (1.13–1.77) | 64.6 | |
| In-hospital25–27 | 3 | 2.03 (1.17–3.54) | 50.7 | |
| Active use of PPIs | 0.13 | |||
| Active use9,13,20,23–27,29 | 9 | 1.54 (1.15–2.06) | 76.4 | |
| Non-active use7,19,30 | 3 | 1.51 (0.90–2.53) | 75.5 | |
| Geographical region | 0.06 | |||
| North America9,19,23,24,26 | 5 | 1.27 (0.96–1.68) | 67.2 | |
| Europe7,13,29,30 | 4 | 1.57 (1.11–2.23) | 47.3 | |
| Asia20,25,27 | 3 | 2.21 (1.48–3.29) | 43.8 | |
| Number of participants | 0.62 | |||
| Small (≤ 1000)7,19,23,27,29,30 | 6 | 1.64 (1.16–2.32) | 61.9 | |
| Large (> 1000)9,13,20,24,25,26 | 6 | 1.44 (1.01–2.07) | 83.5 | |
| Confounding adjustment | ||||
| BMI7,13,19,26 | 4 | 1.13 (0.98–1.29) | 11.5 | |
| Smoking status13,19,23,30 | 4 | 1.40 (0.91–2.14) | 76.2 | |
| NSAIDs use19,23,29 | 3 | 1.31 (0.96–1.78) | 75.6 | |
| Steroid use20,27,29 | 3 | 1.73 (1.25–2.38) | 0.0 | |
| Comorbidity7,13,20,23,24,26,29,30 | 8 | 1.45 (1.16–1.81) | 57.7 | |
| Gastrointestinal disease13,21,29 | 3 | 1.73 (1.07–2.81) | 74.8 |
ARDS acute respiratory distress syndrome, BMI body mass index, CI confidence interval, COVID-19 coronavirus disease 2019, HR hazard ratio, NA not applicable, NSAID non-steroidal anti-inflammatory drug, PPI proton pump inhibitor