Characteristics of the 232 NMAs included in the re-analysis
| Characteristics of networks | Median | IQR |
| Median number of treatments compared | 6 | (5 to 9) |
| Median number of studies included | 19 | (12 to 34) |
| Median total sample size | 6100 | (2514 to 17264) |
| Number of NMAs | % | |
| Beneficial outcome | 97 | 41.8% |
| Dichotomous outcome | 185 | 79.7% |
| Continuous outcome | 47 | 20.3% |
| Published before 2010 | 42 | 18.1% |
| Ranking metric used in original publication (non-exclusive) | ||
| Probability of producing the best value | 83 | 35.8% |
| Rankograms | 7 | 3% |
| Median or mean rank | 3 | 1.3% |
| SUCRA | 16 | 6.9% |
| Other | 2 | 0.9% |
| None | 133 | 57.3% |
| Published in general medicine journals* | 125 | 53.9% |
| Published in health services research journals† | 3 | 1.3% |
| Published in specialty journals | 104 | 44.8% |
*Includes the categories Medicine, General and Internal, Pharmacology and Pharmacy, Research and Experimental, Primary Healthcare.
†Includes the categories Healthcare Sciences and Services, Health Policy and Services.
IQR, interquartile range; NMA, network meta-analysis; SUCRA, surface under the cumulative ranking curve.