Harrison and colleagues conducted a separate regression analyses on 42 infants who received GHT to examine the level of change from baseline to touch. The purpose of the analyses was to identify potential factors related to physiologic and behavioral responses to individual GHT session that could help predict when infants might have problems tolerating GHT. The findings from these regression analyses suggest that baseline behavioral state and baseline levels of motor activity, no movement, and behavioral distress were the most significant predictors of change from B to T phases on the 9 dependent variables that were examined. However, the amount of variation in the dependent variables that was explained by the independent variables in the analyses was relatively small, ranging from 1.2% to 8.6%. For instance, very preterm infants who were more active or distressed at baseline had increased levels of rapid eye movement (REM) sleep and a decreased heart rate, suggesting that GHT had a relaxing or calming effect. On the contrary, very preterm infants with higher baseline levels of quiet sleep were associated with increased levels of REM sleep, active sleep, motor activity, and heart rate, and with decreased levels of modified behavioral distress, suggesting that GHT might have aroused infants who were already in a quiet state. Infants with higher morbidity scores were associated with decreased levels of REM sleep and increased levels of motor activity, suggesting that sicker infants became more agitated during the touch intervention.