Abstract

Objective: To develop an objective and reliable method to assess drug withdrawal in newborns by quantitatively estimating the amount of movement rather than scoring individual withdrawal signs.

Design: In this cross sectional study, a commercial portable motion detector with computer memory, similar to a wrist watch (the actigraph) was used to measure movement. The measurements were compared with a clinical decision based on the neonatal abstinence syndrome (NAS) score. Movement was analysed, using non-parametric tests, in three groups: a control group of 10 infants, 13 opiate exposed newborns not treated for NAS, and 30 newborns treated for NAS (17 before treatment, eight within 24 hours of treatment, five when stabilised).

Results: There were significant differences in the median activity score, expressed as counts per minute (cpm), in the pretreatment group (124 cpm) compared with the control (42 cpm, p < 0.0001), non-treated (74 cpm, p  =  0.001), and stabilised treatment (75 cpm, p  =  0.007) groups. The accuracy of the actigraph in the identification of newborns requiring treatment from those who did not was high compared with the clinical scores; sensitivity 94%; specificity 85%; positive and negative predictive values 88% and 92% respectively.

Conclusions: The measure of movement is comparable to the clinical score in the identification of newborns who require treatment and in determining the severity of withdrawal. The clear advantage of this method is its objectivity, reliability, and efficiency as a simple, non-invasive, bedside measure. Further evaluation in a randomised, controlled trial would establish comparative benefits, potential harms, safety, and acceptability.