| Risk of sampling bias |
The meta-research study evaluated a random sample of primary articles. |
The meta-research study included a non- or pseudo-random sample of primary articles. |
The sampling frame for the sample of primary articles is unclear. |
| Risk of selective reporting bias |
Eligible outcomes and associations reported in the protocol for the meta-research study are fully reported in the results section of the publication. |
Not all eligible outcomes and associations reported in the protocol for the meta-research study are reported in the results section of the publication. |
It is unclear if all eligible outcomes and associations are fully reported in the results section of the publication (e.g., because a study protocol for the meta-research study is unavailable). |
| Risk of article selection bias |
Details about which studies were excluded from the study and why have been shared and match the criteria described in the methods. |
Details about which studies were excluded and why were not reported. |
Details about the eligibility criteria and study selection process is unclear. |
| Risk of errors in the accuracy of reported estimates |
All outcome data were either: manually coded by at least two people independently in parallel or coded by one person and checked in full by another. |
Outcome data was manually coded by: only one researcher, only an automated algorithm, or according to another methodology different from that outlined in the Low Risk category. |
The method used to extract data from the included primary studies is unclear. |
