Skip to main content
. 2017 Apr 12;357:j1415. doi: 10.1136/bmj.j1415

Table 3.

Incidence rate ratios for adverse events associated with short term use of oral corticosteroids

Adverse event No of participants Median dose (mg/day) Median No of days using steroids 5-30 days* 31-90 days*
Incidence rate ratio† (95% CI) P value Incidence rate ratio† (95% CI) P value
All doses v no corticosteroids:
 Sepsis 1556 20 6 5.30 (3.80 to 7.41) <0.001 2.91 (2.05 to 4.14) <0.001
 Venous thromboembolism 4343 17.5 6 3.33 (2.78 to 3.99) <0.001 1.44 (1.19 to 1.74) <0.001
 Fracture 20 090 19 6 1.87 (1.69 to 2.07) <0.001 1.40 (1.29 to 1.53) <0.001
Dose: <20 mg/day v 0 mg/day:
 Sepsis 708 17.5 6 4.02 (2.41 to 6.69) <0.001 2.62 (1.58 to 4.34) <0.001
 Venous thromboembolism 2139 17.5 6 3.61 (2.81 to 4.64) <0.001 1.27 (0.96 to 1.67) 0.10
 Fracture 9941 17.5 6 1.83 (1.60 to 2.10) <0.001 1.41 (1.24 to 1.59) <0.001
Dose: 20-39 mg/day v 0 mg/day:
 Sepsis 652 32 7 7.10 (4.20 to 12.01) <0.001 2.91 (1.64 to 5.18) <0.001
 Venous thromboembolism 1713 35 7 2.83 (2.09 to 3.84) <0.001 1.40 (1.03 to 1.90) 0.03
 Fracture 8009 35 7 1.95 (1.66 to 2.30) <0.001 1.33 (1.15 to 1.54) <0.001
Dose: ≥40 mg/day v 0 mg/day:
 Sepsis 196 60 5 4.98 (1.69 to 14.72) 0.004 5.20 (1.77 to 15.25) 0.003
 Venous thromboembolism 491 60 5 4.15 (2.45 to 7.03) <0.001 2.27 (1.38 to 3.74) 0.001
 Fracture 2140 60 5 1.77 (1.31 to 2.39) <0.001 1.61 (1.26 to 2.05) <0.001

*Number of days from date when corticosteroid prescription was filled. Reference period was 5-180 days before prescription date.

†Sepsis was adjusted for antibiotics, 5-HT3 receptor antagonists, antidepressants, anti-inflammatory agents, antimuscarinics, opiate agonists, and phenothiazine. Venous thromboembolism was adjusted for antibiotics, androgens, anxiolytics, anti-inflammatory agents, azoles, calcium channel blockers, coumarin, diuretics, opiate agonists, and platelet aggregation inhibitors. Fractures were adjusted for anti-inflammatory agents, COX-2 inhibitors, and opiate agonists.