| Risk of Bias |
Author’s Judgment
Low Risk
Unclear
High Risk
|
Support for Judgment |
| Selection bias |
|
Change of definition through time.Sick population. ED HDU patient |
| Bias in definition and measurement |
|
AMS—determined by deduction from notesSepsis was defined by the 2001 definition |
| Outcome measurement bias |
|
|
| Handling of missing data |
|
|
| Confounding |
|
31% mortality |
| Bias of statistics or presentation of result |
|
Statistics unclearDouble-counting MEWS and SOFA in modelling |
Low Risk
Unclear
High Risk