Table 1.
Characteristics of randomised trials included in systematic review and network meta-analysis
| Characteristics | Number (%) |
|---|---|
| Drugs | |
| Oral NSAIDs | 288 (55.5) |
| Topical NSAIDs | 32 (6.2) |
| Opioids | 53 (10.2) |
| Paracetamol | 16 (3.1) |
| Placebo | 130 (25) |
| Type of control | |
| Placebo | 130 (67.7) |
| Active intervention | 62 (32.3) |
| Average age (years) | |
| <55 | 10 (5.2) |
| 55-59.9 | 31 (16.1) |
| 60-64.9 | 123 (64.1) |
| 65-69.9 | 17 (8.9) |
| ≥70 | 6 (3.1) |
| Not reported | 5 (2.6) |
| Average body mass index | |
| <25 | 2 (4.2) |
| 25-30 | 14 (29.2) |
| >30 | 32 (66.7) |
| Average baseline pain on 10 cm scale, median (IQR) | 6.5 (5.7-7.0) |
| Average disease duration (years) | |
| <5 | 17 (20.2) |
| 5-10 | 61 (72.6) |
| >10 | 6 (7.1) |
| Average % of women | |
| <50 | 9 (4.7) |
| 50-59 | 31 (16.1) |
| 60-69 | 91 (47.4) |
| ≥70 | 58 (30.2) |
| Not reported | 3 (1.6) |
| Publication status | |
| Published | 177 (92.2) |
| Unpublished | 15 (7.8) |
| Year of publication | |
| 1982-89 | 12 (6.3) |
| 1990-99 | 33 (17.2) |
| 2000-09 | 89 (46.4) |
| 2010-21 | 58 (30.2) |
| Sample size (No of participants) | |
| 200-250 | 31 (16.1) |
| 251-500 | 73 (38) |
| 501-1000 | 63 (32.8) |
| >1000 | 25 (13) |
| Follow-up duration (weeks) | |
| <12 | 96 (50) |
| 12-24 | 73 (38) |
| 24-48 | 12 (6.3) |
| >48 | 11 (5.7) |
| No of arms | |
| 2 | 86 (44.8) |
| 3 | 60 (31.3) |
| 4 | 28 (14.6) |
| ≥5 | 18 (9.4) |
| Industry funding | |
| Yes | 153 (79.7) |
| Unclear | 37 (19.3) |
| No | 2 (1) |
| Deterministic imputation of missing data | |
| Yes | 103 (53.6) |
| Unclear | 79 (41.1) |
| No | 10 (5.2) |
| Centre | |
| Multicentre | 181 (94.3) |
| Single centre | 2 (1) |
| Unclear | 9 (4.7) |
NSAID=non-steroidal anti-inflammatory drug.
All variables are presented at trial level, except for drugs, which are presented at level of randomised treatment arms. Information on body mass index and disease duration was reported in <50% of the trials. For unpublished reports, publication date was date study data were made available online or date reported in official files or regulatory documents. Examples of deterministic methods of imputation include last observation carried forward, baseline observation carried forward, mean imputation.