Table 2. Proton pump inhibitor use and hospitalization with hypomagnesemia.
| Characteristic | Number (%) of Patients with Exposure | Odds Ratio (95% Confidence Interval) | ||
| Case Patients (n = 366) | Control Patients (n = 1,464) | Unadjusted | Adjusteda | |
| ≤90 days (current)b | 143 (39.1) | 305 (20.8) | 2.70 (2.08–3.51) | 1.43 (1.06–1.93) |
| 91–180 days (recent)b | 12 (3.3) | 42 (2.9) | 1.59 (0.82–3.07) | 1.03 (0.51–2.06) |
| 181–365 (remote)b | 9 (2.5) | 35 (2.4) | 1.37 (0.65–2.89) | 0.94 (0.42–2.11) |
| Number of distinct drugs used in previous year | — | — | 1.13 (1.11–1.16) | 1.12 (1.09–1.14) |
| Diabetes | 26 (7.1) | 65 (4.4) | 1.66 (1.03–2.67) | 1.45 (0.85–2.47) |
| Congestive heart failure | 45 (12.3) | 84 (5.7) | 2.64 (1.73–4.03) | 1.70 (1.06–2.73) |
| Malignancy | 49 (13.4) | 47 (3.2) | 4.71 (3.07–7.22) | 3.52 (2.20–5.65) |
| Steroid use | 54 (14.8) | 116 (7.9) | 2.01 (1.42–2.85) | 0.97 (0.65–1.46) |
a
Adjusted for diabetes, congestive heart failure, malignancy, steroid use, and number of distinct drugs used in the past year (excluding PPI).
b
Includes omeprazole, pantoprazole, lansoprazole, esomeprazole, and lansoprazole.