Background: Alcohol withdrawal is a common disorder faced by hospitalists. The pharmacologic standard of care involves the use of benzodiazepines, administered either as fixed doses or with doses adjusted per patient symptoms. At our institution, the alcohol withdrawal protocol previously in place used an institution-specific scale to evaluate patient symptoms, and nurses had latitude within the protocol to select medication doses. Anecdotally, this prior protocol resulted in excessive variability of medication doses and inappropriately high rates of transition to lorazepam infusion, use of physical restraints, and transfer to the intensive care unit.
Purpose: In an attempt to reduce complications and standardize medication doses, an innovative new alcohol withdrawal protocol was proposed.
Description: A new revised alcohol withdrawal protocol was developed with a committee of nurses, physicians, and pharmacists following a literature review. The new protocol provided specific guidance regarding medication doses, reassessment intervals, and, as an innovative feature, instructions regarding the transition from oral medications to IV lorazepam infusion based on alcohol withdrawal severity determined using the validated CIWA-Ar scale for awake patients and the Riker Sedation-Agitation scale for sedated patients. A retrospective analysis of aggregate patient outcomes for adult inpatients that were placed on an alcohol withdrawal order set from 2010 through 2012 was performed to monitor effectiveness and safety. Comparisons of length of stay, restraint use, transfer to intensive care, and total dosage of chlordiazepoxide and lorazepam were made between patients who received the old and new protocols. A total of 3840 adult inpatients had orders for an alcohol withdrawal order set. Of these, 1598 patients (41.6%) were on the older protocol, with the remaining 2242 (57.4%) on the newer protocol. Length of stay decreased from a median of 2.9 days on the old protocol to 2.2 days on the new protocol (p < 0.001). The percentage of patients requiring transfer to an intensive care unit decreased from 7.7% with the old protocol to 5.5% with the new protocol (p = 0.009). The percentage of patients requiring restraints decreased from 28.8% on the old protocol to 21.5% on the new protocol (p < 0.001).
Conclusions: In patients with alcohol withdrawal, the implementation of an innovative new protocol resulted in a statistically significant reductions in length of stay, transfer to intensive care, and use of restraints. This quality improvement effort demonstrated the ability of a multi-disciplinary team to apply literature-based best practices to reduce complications of care. Future efforts involve the development of computer-decision aids to improve compliance and continued monitoring of outcomes for effectiveness and safety.