Table 2.
Risk of bias summary
| Study | Random sequence generation | Allocation concealment | Blinding for participants and personnel | Blinding for outcome assessors | Incomplete outcome data | Selective outcome reporting | Other sources of bias |
|---|---|---|---|---|---|---|---|
| a. Risk of bias summary of hospital mortality | |||||||
| Bernard 1987 [7] | Low risk | Low risk | Low risk | Low risk | Low risk | Unclear risk | Low risk |
| Meduri 1998 [8] | Low risk | Low risk a | Low risk a | Low risk a | Low risk | Unclear risk | Low risk |
| Steinberg 2006 [9] | Low risk | Low risk | Unclear risk | Low risk | Low risk | Low risk | Low risk |
| Meduri 2007 [10] | Low risk | Low risk | Low risk a | Low risk a | Low risk | High risk a | Low risk |
| Liu 2012 [30] | Unclear risk | Unclear risk | Unclear risk | Low risk | Low risk | Unclear risk | Low risk |
| Tongyoo 2016 [11] | Low risk | Low risk | Low risk | Low risk | Low risk | Unclear risk | Low risk |
| Tomazini 2020 [12] | Low risk | Low risk | High riskb | Low risk | Low risk | Low risk | Low risk |
| Villar 2020a [13] | Low risk | Low risk | High riskb | Low risk | Low risk | High riskc | Low risk |
| Villar 2020b [31] | Low risk | Low risk | High riskb | Low risk | Low risk | Low risk | Low risk |
| b. Risk of bias summary of incidence of infection | |||||||
| Bernard 1987 [7] | Low risk | Low risk | Low risk | Low risk | Low risk | Unclear risk | Low risk |
| Meduri 1998 [8] | Low risk | Low riska | Low riska | Low riska | Low risk | Unclear risk | Low risk |
| Steinberg 2006 [9] | Low risk | Low risk | Unclear risk | Unclear risk | Low risk | Low risk | Low risk |
| Meduri 2007 [10] | Low risk | Low risk | Low riska | Low riska | Low risk | High riska | Low risk |
| Liu 2012 [30] | Unclear risk | Unclear risk | Unclear risk | Unclear risk | Low risk | Unclear risk | Low risk |
| Tongyoo 2016 [11] | Low risk | Low risk | Low risk | Low risk | Low risk | Unclear risk | Low risk |
| Tomazini 2020 [12] | Low risk | Low risk | High riskb | Unclear risk | Low risk | High riskc | Low risk |
| Villar 2020a [13] | Low risk | Low risk | High riskb | Low risk | Low risk | High riskc | Low risk |
| Villar 2020b [31] | Low risk | Low risk | High riskb | Unclear risk | Low risk | Low risk | Low risk |
| c. Risk of bias summary of ventilator-free days | |||||||
| Meduri 1998 [8] | Low risk | Low riska | Low riska | Low risk | Low risk | Unclear risk | Low risk |
| Steinberg 2006 [9] | Low risk | Low risk | Unclear risk | Low risk | Low risk | Low risk | Low risk |
| Meduri 2007 [10] | Low risk | Low risk | Low riska | Low risk | Low risk | Low riska | Low risk |
| Liu 2012 [30] | Unclear risk | Unclear risk | Unclear risk | Low risk | Low risk | Unclear risk | Low risk |
| Tongyoo 2016 [11] | Low risk | Low risk | Low risk | Low risk | Low risk | Unclear risk | Low risk |
| Tomazini 2020 [12] | Low risk | Low risk | High riskb | Low risk | Low risk | Low risk | High riskd |
| Villar 2020a [13] | Low risk | Low risk | High riskb | Low risk | Low risk | Low risk | High riskd |
aWe got information by author contact in Meduri [8] and Meduri [10]. Meduri [8] REPLY: Central randomization by a third party in order to conceal the allocation; Physicians and all medical personnel were blinded. Meduri [10] REPLY: The randomization is double blind and will remain blind throughout therapy to physicians, nursing care teams, research investigators, outcome assessor, participants and their family members. The primary objective of this prospective double blind, randomized clinical trial is to assess the effects of prolonged methylprednisolone therapy on the following response by days 7 and 28 of therapy: improvement in lung injury score, number of ventilator-free days, mortality
bThese studies did not use placebo in control group
cNot planned outcomes in pre-published protocol
dThe study was stopped earlier than planned