Skip to main content

Table 3 Impact of the different analysis approaches on the trial conclusion

From: Multiple imputation for patient reported outcome measures in randomised controlled trials: advantages and disadvantages of imputing at the item, subscale or composite score level

Analysis approach

Treatment effect (95% CI)

Number of participants included

Complete cases analysis

0.9 (−2.6, 4.4)

167

MI at the composite score level

0.7 (−2.8, 4.2)

200

MI at the subscale level

0.7 (− 2.7, 4.1)

200

MI at the item level (ologit)

0.9 (−2.5, 4.3)

200

MI at the item level (PMM)

0.6 (−2.7, 3.9)

200