Variables	Study Participation	Study Attrition	Prognostic Factor Measurement	Outcome Measurement	Study Confounding	Statistical Analysis and Reporting
High risk of bias	Specialized population (ie, pregnant, elderly) or subset of very ill patients only	Incomplete outcome ascertainment (some patients still hospitalized)	Unclear definition for prognostic factors	Outcome not defined	No multivariate analysis performed	Selective reporting of results, no clear analytic strategy
Moderate risk of bias	Only inpatients	Complete ascertainment but >10% loss to follow-up	Post hoc selection of cut points	Unclear definition
Low risk of bias	Inpatient and outpatient	Complete ascertainment and <10% loss to follow-up	Typical cutoffs used, clearly defined	Clear and reproducible definition	Multivariate analysis reported	Full reporting, analytic strategy clearly described