TABLE 1.
Summary of MR results for causal relationship between substance/medication traits (exposure) and COVID-19 adverse outcomes.
| Substance/Medication category | COVID-19 outcome | LCV | Egger | IVW | WM | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| GCP | p-value | rg (SE) | p-value | OR (95%CI) | p-value | OR (95%CI) | p-value | OR (95%CI) | |||||
| Cannabis abuse | Hospitalized | −0.82 | 8.07e–38 | 0.14 (0.31) | 4.25e–03 | 0.87 (0.87-0.88) | ↓ | 1.18e–08 | 0.77 (0.77–0.78) | ↓ | 0.74 | 0.97 (0.97–0.98) | |
| Cigarettes Per Day (Without UKB) | −0.90 | 1.03e–30 | −0.27 (0.24) | 5.39e–31 | 1.02 (1.02–1.03) | ↑ | 1.16e–122 | 1.04 (1.04–1.05) | ↑ | 3.76e–12 | 1.02 (1.02–1.03) | ↑ | |
| Vasodilators used in cardiac diseases | −0.73 | 9.05e–21 | 0.17 (0.39) | 2.56e–16 | 1.01 (1.00–1.01) | ↑ | 6.60e–28 | 1.01 (1.01–1.02) | ↑ | 2.01e–16 | 1.01 (1.01–1.02) | ↑ | |
| Statin | 0.05 | 0.03 | 0.24 (0.26) | 3.46e–56 | 1.06 (1.05–1.07) | ↑ | 8.32e–199 | 1.08 (1.07–1.08) | ↑ | 2.31e–62 | 1.09 (1.08–1.10) | ↑ | |
| Opioid (medication) | 0.00 | 0.94 | 0.40 (0.38) | 9.20e–48 | 1.03 (1.03–1.04) | ↑ | 2.06e–56 | 1.02 (1.02–1.03) | ↑ | 4.98e–42 | 1.03 (1.03–1.04) | ↑ | |
| Opioid exposure | 0.00 | 0.89 | −0.98 (1.34) | 1.21e–01 | 0.50 (0.21–1.20) | 1.96e–11 | 5.23 (3.23–8.49) | ↑ | 0.40 | 1.24 (0.75–2.04) | |||
| Diuretics | Severe Respiratory | −0.53 | 1.21e–06 | −0.31 (0.20) | 0.15 | 1.22 (0.93–1.59) | 2.66e–15 | 0.47 (0.39–0.57) | ↓ | 0.18 | 0.87 (0.70–1.07) | ||
| Antithrombotic agents | 0.22 | 4.64e–4 | −0.39 (0.52) | 0.70 | 1.07 (0.75–1.54) | 0.016 | 0.74 (0.58–0.95) | 0.01 | 1.38 (1.07–1.77) | ||||
| Cannabis abuse | −0.11 | 9.80e–3 | 0.95 (0.28) | 0.84 | 1.01 (0.91–1.12) | 0.36 | 1.04 (0.95–1.14) | 0.02 | 0.87 (0.77–0.98) | ||||
| Agents acting on the renin-angiotensin system | −0.62 | 0.01 | −0.43 (0.18) | 0.20 | 0.80 (0.56–1.13) | 1.87e–17 | 0.36 (0.29–0.46) | ↓ | 0.01 | 0.70 (0.53–0.92) | |||
| Alcohol intake frequency | −0.60 | 0.03 | 0.42 (0.17) | 0.19 | 0.70 (0.41–1.19) | 8.92e–4 | 0.53 (0.37–0.77) | ↓ | 2.08e–07 | 0.30 (0.19–0.48) | ↓ | ||
| Opioid (medication) | −0.08 | 0.59 | 0.12 (0.30) | 8.01e–20 | 2.68 (2.17–3.31) | ↑ | 1.50e-16 | 1.85 (1.60–2.14) | ↑ | 6.11e–07 | 1.43 (1.24–1.65) | ↑ | |
| Opioid exposure | −0.01 | 0.78 | *1.72 (10.63) | 0.12 | 0.50 (0.21–1.20) | 1.25e–12 | 7.71 (4.39–13.55) | ↑ | 0.41 | 1.24 (0.74–2.07) | |||
Note: COVID-19 adverse outcomes include both hospitalization and severe respiratory symptoms. Significant values are highlighted in bold, and arrows indicate the direction of causality. According to O’Connor and Price (2018) the genetic causality proportion (GCP) greater than |0.6| is suggestive of a causal relationship. Negative GCP, values are indicative of a reverse causality relationship. The null hypothesis for the reported p-value in LCV analysis is GCP = 0. For each category of COVID-19 outcome, only exposure traits that reached the significance level across at least one MR method are shown. rg, The estimate of genetic correlation from the LCV method; Egger, Egger regression; IVW, inverse-variance weighted linear regression; WM, Weighted median. (* LD score regression estimate of rg is not bounded. Estimates outside of [−1, 1] are possible; however, the extremely large standard error indicates that estimates of genetic correlation, in this case, is unstable.)