Table 3.
Risk of any fracture | Risk of fracture at osteoporotic sites | |||
---|---|---|---|---|
Number of fractures | Fully adjusted HR (95 % CI)a | Number of fractures | Fully adjusted HR (95 % CI)a | |
MG by use of oral glucocorticoids by cumulative dose in grams prednisolone equivalents in the previous year | ||||
No oral glucocorticoid use | 47 | 1.00 | 27 | 1.00 |
Any oral glucocorticoid use | 28 | 0.88 (0.52–1.47) | 16 | 0.75 (0.38–1.50) |
<2.5 g prednisolone eq | 13 | 0.80 (0.42–1.53) | 7 | 0.63 (0.26–1.53) |
2.5–5.0 g prednisolone eq | 10 | 1.11 (0.54–2.26) | 5 | 0.83 (0.31–2.25) |
> = 5.0 g prednisolone eq | 5 | 0.73 (0.27–1.94) | 4 | 0.99 (0.31–3.14) |
MG by history of drug use in previous 6 months | ||||
No oral glucocorticoid use | 48 | 1.00 | 28 | 1.00 |
Oral glucocorticoid use | 27 | 0.97 (0.58–1.63) | 15 | 0.81 (0.40–1.61) |
<7.5 mg prednisolone eq/day | 10 | 0.99 (0.49–2.03) | 5 | 0.70 (0.26–1.92) |
7.5–15 mg prednisolone eq/day | 8 | 1.00 (0.46–2.16) | 3 | 0.57 (0.17–1.93) |
> = 15 mg prednisolone eq/day | 9 | 0.93 (0.44–1.99) | 7 | 1.17 (0.47–2.89) |
No antidepressant use | 59 | 1.00 | 31 | 1.00 |
Antidepressant use | 16 | 2.15 (1.22–3.79) | 12 | 3.27 (1.63–6.55) |
<20 mg fluoxetine eq/day | 9 | 1.88 (0.92–3.86) | 7 | 2.77 (1.18–6.50) |
> = 20 mg fluoxetine eq/day | 7 | 2.61 (1.18–5.80) | 5 | 4.32 (1.64–11.38) |
No anxiolytic use | 61 | 1.00 | 32 | 1.00 |
Anxiolytic use | 14 | 1.80 (0.97–3.34) | 11 | 2.18 (1.04–4.57) |
<10 mg diazepam eq/day | 10 | 1.72 (0.85–3.47) | 8 | 2.10 (0.90–4.86) |
> = 10 mg diazepam eq/day | 4 | 2.07 (0.73–5.82) | 3 | 2.41 (0.71–8.12) |
No anticonvulsant use | 64 | 1.00 | 36 | 1.00 |
Anticonvulsant use | 11 | 5.36 (2.76–10.39) | 7 | 6.88 (2.91–16.27) |
<1.0 g carbamazepine eq/day | 8 | 4.88 (2.27–10.50) | 5 | 5.45 (2.03–14.62) |
> = 1.0 g carbamazepine eq/day | 3 | 7.10 (2.13–23.62) | 2 | 18.18 (3.88–85.15) |
No antipsychotic use | 74 | 1.00 | 42 | 1.00 |
Antipsychotic use | 1 | 1.30 (0.17–9.76) | 1 | 1.41 (0.17–11.65) |
eq equivalents
aAdjusted for the same confounders as described below Table 2 for any and osteoporotic fracture, but the confounder is not added to the model if it is similar to the drug being investigated