Novel Examination of the Impact of Time and Reliability on Minimal Detectable Differences Using the Coma Recovery...

To empirically evaluate the influence of (1) varying reliability coefficients and (2) varying time intervals between baseline and follow-up assessments on the minimal detectable difference (MDD) of the Coma Recovery Scale-Revised.

Secondary analysis of longitudinal data.

Two postacute rehabilitation centers.

Seven hundred four patients (N=704) with disorders of consciousness.

Not applicable.

Coma Recovery Scale-Revised scores were transformed to a 0-100 equal-interval scale. The MDDs were calculated using varying reliability coefficients (test-retest and Rasch-based indices) and pooled SDs across 1-, 2-, 3-, and 4-week intervals.

The MDD values ranged from 7.5-20.4, depending on the reliability coefficient. Rasch-based coefficients produced smaller MDDs than test-retest coefficients. Time intervals between assessments had minimal impact, with MDDs varying by <2 units across intervals.

Variability in MDD because of reliability coefficients underscores the need for careful consideration of the MDDs used in clinical and research settings. Time intervals had a negligible impact, suggesting consistent MDD values can be applied across subacute rehabilitation periods. These findings highlight the importance of selecting appropriate metrics to ensure accurate monitoring of neurobehavioral changes.