Table 8 Examples of vague instructions for Cochrane authors regarding judgments of attrition bias

Missing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups

There is no quantitative measure of “balanced”

For dichotomous outcome data, the proportion of missing outcomes compared with observed event risk not enough to have a clinically relevant impact on the intervention effect estimate

There is no quantitative measure of “not enough”

Missing data have been imputed using appropriate methods

Not specified what is considered by Cochrane to be “appropriate methods”

Reason for missing outcome data likely to be related to true outcome, with either imbalance in numbers or reasons for missing data across intervention groups

There is no quantitative measure of “imbalance”

Potentially inappropriate application of simple imputation.

Not specified what is considered to be “inappropriate application”