Back to: Assess for Risk of Bias
How did you fare? For reference, here’s a chart about missing participant data that indicate a “low risk of bias” and those that indicate a“high risk of bias.”
| Low risk of bias | High risk of bias |
|---|---|
| No missing participant data | Considerable missing participant data |
| It’s unlikely participants dropped out because of an effect of the treatment | It’s likely participants dropped out because of an effect of the treatment |
| Number of missing participants and reasons for leaving are similar in both the treatment and control group | Number of missing participants and reasons for leaving are very different in the treatment vs. control group |
| Sensitivity analysis does not change the study conclusions | Sensitivity analysis does change the study conclusions |
| Researchers used appropriate analytic methods to adjust for missing data | Researchers used no or inappropriate methods to adjust for missing data |
When you’re assessing a study for attrition bias, you should look at every outcome individually to see if the missing participant data about that outcome falls in the “low risk of bias” or “high risk of bias” column. If the approach isn’t described, or you’re unsure, you should label this domain as “unclear.”
Let’s move on to our last domain: incomplete outcome reporting.