Table 4.
Use of PPIs or H2RAs and risk of hip fracture, by exposure to oral corticosteroids
| Cases (n = 6,763) | % | Controls (n = 26,341) | % | Crude OR (95% CI) | Adjusteda OR (95% CI) | |
|---|---|---|---|---|---|---|
| PPI use before | ||||||
| Never | 5,810 | 85.9 | 23,430 | 88.9 | 1.00 | 1.00 |
| Current use | 305 | 4.5 | 773 | 2.9 | 1.62 (1.41–1.86) | 1.20 (1.04–1.40) |
| By oral corticosteroid use in the 6 months beforeb | ||||||
| Unexposed | 256 | 3.8 | 682 | 2.6 | 1.54 (1.33–1.79)c | 1.19 (1.02–1.40) |
| <7.5 mg/day | 21 | 0.3 | 47 | 0.2 | 1.86 (1.11–3.12) | 1.31 (0.77–2.22) |
| 7.5–15 mg/day | 12 | 0.2 | 20 | 0.1 | 2.51 (1.21–5.18) | 1.91 (0.90–4.07) |
| ≥15 mg/day | 13 | 0.2 | 14 | 0.1 | 3.67 (1.72–7.84)c | 2.35 (1.07–5.20) |
| H2RA use before | ||||||
| Never | 5,624 | 83.2 | 22,545 | 85.6 | 1.00 | 1.00 |
| Current use | 196 | 2.9 | 520 | 2.0 | 1.52 (1.28–1.80) | 1.19 (1.00–1.42) |
| By oral corticosteroid use in the 6 months beforeb | ||||||
| Unexposed | 165 | 2.4 | 468 | 1.8 | 1.42 (1.19–1.71) | 1.18 (0.98–1.43) |
| <7.5 mg/day | 16 | 0.2 | 24 | 0.1 | 2.64 (1.39–4.99) | 1.73 (0.90–3.35) |
| 7.5–15 mg/day | 9 | 0.1 | 16 | 0.1 | 2.29 (1.01–5.19) | 1.43 (0.61–3.38) |
| ≥15 mg/day | 5 | 0.1 | 6 | 0.0 | 3.59 (1.09–11.78) | 2.34 (0.68–8.06) |
OR odds ratio, CI confidence interval
aAdjusted for same confounders listed in Table 2
cCorticosteroids by prednisolone equivalents; data not shown for patients with only 1 oral steroid dispensing before the index date
dWald statistic: the risk of hip fracture is statistically significantly higher among PPI users exposed to corticosteroids ≥15 mg/day compared with PPI users unexposed to corticosteroids (P < 0.05)