Psychometric Evaluation of the Sophia Observation Withdrawal Symptoms Scale in Critically Ill Children

• DOI: 10.1097/pcc.0b013e31829f5be1
• Publication Date: 2013-08-20T12:18:47Z
• Authors: Erwin Ista; Matthijs de Hoog; Dick Tibboel; Hugo J. Duivenvoo...; Monique van Dijk
• Keywords: Male; EMC OR-02-54-06; Adolescent; Psychometrics; Critical Illness; EMC MM-03-54-04-A; Intensive Care Units, Pediatric; Sensitivity and Specificity; Severity of Illness Index; Hospitals, University; Benzodiazepines; 03 medical and health sciences; 0302 clinical medicine; Risk Factors; Humans; Prospective Studies; Child; Infant, Newborn; Infant; Substance Withdrawal Syndrome; 3. Good health; Analgesics, Opioid; Child, Preschool; Female; EMC MGC-02-53-01-A

The Sophia Observation withdrawal Symptoms scale is an instrument for screening benzodiazepine and opioid withdrawal syndrome in pediatric critical care patients. The objectives of this study were to establish cutoff scores and to test sensitivity to change. Second, risk factors for withdrawal syndrome were explored.Prospective observational study with repeated measures.Level IV ICU at a university children's hospital.A total of 154 children with median age 5 months (interquartile range, 0-42 mo) who received continuous infusion of benzodiazepines and/or opioids for 5 or more days.None.Nurses repeatedly applied the Sophia Observation withdrawal Symptoms scale and the Numeric Rating Scale withdrawal when children were weaned off benzodiazepines and opioids. The latter represents the nurse's expert opinion. We analyzed 3,754 paired assessments; the median number per child was 15 (interquartile range, 7-31) over a median of 5 days (interquartile range, 3-11 d). Sensitivity and specificity were 0.83 and 0.93, respectively, for the Sophia Observation withdrawal Symptoms scale cutoff score of 4 or higher against a Numeric Rating Scale-withdrawal score of 4 or higher. Sensitivity to change was determined by comparing 156 Sophia Observation withdrawal Symptoms scale assessments (n = 51 patients) before and after additional sedatives or opioids. Multilevel regression analysis showed a mean decline of 1.5 points (at score range 0-15) after intervention (p < 0.0001). Logistic regression analysis identified duration of preweaning of midazolam, duration of weaning of midazolam, duration of preweaning of morphine, duration of weaning of morphine, and number of additional sedatives/opioids as statistically significant risk factors for withdrawal syndrome in these children.The Sophia Observation withdrawal Symptoms scale is a valid tool with good psychometric properties to assess withdrawal symptoms in PICU patients.