Validation of the Withdrawal Assessment Tool–1 in Adult Intensive Care Patients

Anissa Capilnean, PharmD, MSc; Amanda Martone, PharmD, MSc; Vlad A. Rosu, MSc, BPharm; Patricia R. Sandu, PharmD, MSc; Paul Murgoi, MSc, BPharm, BCPS; Anne Julie Frenette, MSc, BPharm; David Williamson, PhD; Annie Lecavalier, MD; Dev Jayaraman, MD; Philippe Rico, MD; Patrick Bellemare, MD; Céline Gélinas, PhD, RN; Marc M. Perreault, PharmD, MSc


Am J Crit Care. 2019;28(5):361-369. 

In This Article

Abstract and Introduction


Background: The Withdrawal Assessment Tool–1 (WAT-1) has been validated for assessing iatrogenic withdrawal syndrome in critically ill children receiving mechanical ventilation, but little is known about this syndrome in critically ill adults.

Objective: To evaluate the validity and reliability of the WAT-1 in critically ill adults.

Methods: A prospective, observational, open-cohort pilot study of critically ill adults receiving mechanical ventilation and regular administration of opioids for at least 72 hours. Patients were assessed for withdrawal twice daily on weekdays and once daily on weekends using the WAT-1 after an opioid weaning episode. The presence of iatrogenic withdrawal syndrome was evaluated once daily using modified Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) (DSM-5) criteria. All evaluations were blinded and performed independently. The criterion validity of the WAT-1 and the interrater reliability for WAT-1 and DSM-5 evaluations were determined.

Results: During 8 months, 52 adults (median age, 51.5 years) were enrolled. Eight patients (15%) had at least 1 positive assessment during their intensive care unit stay using the DSM-5, compared with 19 patients (37%) using the WAT-1. The overall sensitivity of the WAT-1 was 50%, and its specificity was 65.9%. Agreement between WAT-1 and DSM-5 assessments was poor (κ = 0.102). The interrater reliability for the WAT-1 was 89.1% and for the DSM-5 was 90.1%.

Conclusion: Despite showing reliability, the WAT-1 is not a valid tool for assessing the presence of iatrogenic withdrawal syndrome in adults.


Critically ill patients receiving mechanical ventilation often require analgesics and sedatives to help them tolerate invasive procedures and to relieve discomfort, pain, and anxiety.[1] In a prospective cohort study conducted in 51 Canadian intensive care units (ICUs), 92% of patients received opioids and sedatives at least once during mechanical ventilation, with opioids being used more frequently than sedatives and administered primarily as a continuous infusion.[1,2]

Repeated and prolonged exposure to opioids in the ICU puts patients at risk of a physiological tolerance developing as a result of prolonged stimulation of opioid receptors in the central nervous system. This process leads to alteration of intracellular second-messenger signaling, requiring an increase in opioid dose to maintain the same level of analgesia.[3] When regularly administered opioids are tapered or stopped abruptly, with sudden removal of the inhibitory stimulus, an iatrogenic withdrawal syndrome (IWS) can occur.[4,5] Signs and symptoms of IWS include central nervous stimulation (eg, agitation, irritability, tremors, increased wakefulness), sympathetic nervous system hyperactivation (eg, fever, hypertension, tachycardia, tachypnea, sweating), and gastrointestinal disturbance (eg, vomiting, diarrhea).[6,7]

Opioid tolerance develops more quickly and with greater intensity in patients who have undergone major trauma, burn injury, and prolonged mechanical ventilation, as well as in pediatric patients.[8] The latter population has been studied extensively and has been reported to have a high incidence of IWS, with the main predictors being duration of opioid infusion and cumulative dose.[7] Presentation is complex and variable across individuals.[7] The most common symptoms are sweating, tremors, irritability, soft or liquid stool, low-grade fever, and nausea and/or vomiting during opioid withdrawal.[4] In the pediatric population, IWS has been associated with negative clinical outcomes, such as prolonged mechanical ventilation, psychological distress, longer hospital stay, higher morbidity, and increased health care costs.[9] In critically ill adult patients, IWS is a diagnosis of exclusion, and although its reported incidence is high, no outcome data are available. However, it is strongly suspected that IWS may prolong ICU stay, and symptoms associated with withdrawal may mimic symptoms of other medical conditions, leading to misdiagnosis and unwarranted interventions. In a recent pediatric prospective study,[5] researchers reported an incidence of 37% (22 of 60 patients) of IWS after 3 or more days of opioid and benzodiazepine administration. A retrospective study of adult ICU patients showed an incidence of opioid and benzodiazepine IWS of 32% (9 of 28 patients) after 7 or more days of analgesic and sedative administration.[10] A multicenter prospective study indicated an incidence of 17% (9 of 54 patients) after more than 72 hours of opioids.[11] Identification of IWS is challenging; the scientific literature on IWS in the adult critically ill population, although scarce, continues to recommend tapering opioid doses over a few days to avoid opioid-induced withdrawal symptoms.[8]

To date, no standardized screening tool has been tested or validated in critically ill adults to help clinicians rapidly predict IWS. In critically ill children, various tools for identifying the presence of opioid and benzodiazepine withdrawal syndrome have been developed and validated for bedside use by nurses. Among these, the Withdrawal Assessment Tool–1 (WAT-1),[4] an 11-item (12-point) scale to be administered twice daily, is more logistically feasible than other validated tools (see Supplements 1 and 2, available online only at The WAT-1 consists of 4 main sections: a review of the patient's signs of withdrawal in the past 12 hours, direct observation of the patient's state for 2 minutes, and finally observation during a progressive stimulus and during poststimulus recovery.[4] The WAT-1 seems to be more specific for opioid than for benzodiazepine withdrawal.[7] The tool was validated on the basis of nurses' opinions on a numeric rating scale ranging from 0 to 10. It demonstrated the highest sensitivity and specificity—87% and 88%, respectively—with a cutoff score of 3 or higher, detecting IWS in 45% of 26 pediatric patients who received opioids for at least 5 days.[12] This tool has recently been recommended by the European Society of Paediatric Neonatal Intensive Care for assessment of IWS as part of standard care.[13]

Providing optimal comfort and sedation during mechanical ventilation in the ICU is a key role of intensive care nurses, and preventing withdrawal symptoms should be part of this role. Using a validated and reliable assessment tool for IWS in adults could facilitate recognition of this condition and prompt early therapeutic management. Given that most commonly described IWS symptoms in the adult population are nonspecific (eg, agitation, diarrhea, vomiting, tachycardia, tachypnea) and that these symptoms are accounted for in the WAT-1 scale, we hypothesized that this tool could be applicable in the adult as well as the pediatric population.

The primary objective of this study was to determine the reliability and validity of the WAT-1 for detecting IWS in critically ill adult patients receiving mechanical ventilation who are regularly administered opioids for 72 hours or more.