Study characteristics
| All patients | RA | PMR | IBD | Low-dose | Medium-dose | Short follow-up | Long follow-up | |
| Studies (N) | 28 (3780) | 14 (1708) | 4 (167) | 10 (1905) | 17* (2184) | 13* (1596) | 14 (1762) | 14 (2018) |
| GC-using populations (N) | 39 (2382) | 17 (796) | 6 (167) | 16 (1419) | 23 (1398) | 16 (984) | 21 (1362) | 18 (1020) |
| Age (years) | 46 | 54 | 72 | 37 | 47 | 44 | 39 | 52 |
| Gender (% female) | 63 | 70 | 69 | 55 | 64 | 61 | 56 | 68 |
| Mean (SD) dose prednisone equivalent (mg) | 11.5 (14.5) | 7.5 (5.9) | 8.6 (3.2) | 13.9 (13) | 6.7 (4.8) | 18.0 (20) | 14.6 (17.4) | 7.2 (4.8) |
| Quality analysis of the (sub)populations of the included studies | ||||||||
| AE scoring per protocol (N) | 22 (2960) | 10 (1047) | 3 (142) | 9 (1771) | 15 (1817) | 9 (1143) | 11 (1470) | 11 (1490) |
| Mean time AE scoring (months) | 1.7 | 3.4 | 1.4 | 0.9 | 2.2 | 1.0 | 0.5 | 3 |
| Predefined AE (N) | 15 (1976) | 9 (1072) | 3 (142) | 3 (762) | 10 (1438) | 5 (538) | 5 (578) | 10 (1398) |
| Number of predefined AE | 7.9 | 8.8 | 5.8 | 7.0 | 8.3 | 6.8 | 6.2 | 8.6 |
| Missing data (N)† | 24 (3471) | 11 (1424) | 3 (142) | 10 (1905) | 14 (1900) | 12 (1571) | 12 (1702) | 12 (1769) |
| Missing data (%)‡ | 21.3 | 19.8 | 23.9 | 21.7 | 22.1 | 20.1 | 22 | 19.7 |
*Two studies (Sandborn, 1998; Rutgeers, 1994) were split into two subpopulations as they included both low- and medium-dose populations.
†Missing data: number of studies which noted patients who dropped out of the study.
‡Percentage of patients of the study population who dropped out of the study.
N, number of studies/populations (number of patients); IBD, inflammatory bowel disease; PMR, polymyalgia rheumatica; RA, rheumatoid arthritis.
Low dose: ⩽7.5 mg prednisolone equivalent; medium dose: >7.5 mg and ⩽30 mg prednisolone equivalent; short follow-up: study duration ⩽6 months; long follow-up: study duration >6 months.