By Linda Chlan, PhD, RN, FAAN

Dean’s Distinguished Professor of Symptom Management Research, The Ohio State University, College of Nursing, Columbus

Dr. Chlan reports no financial relationships relevant to this field of study.

SYNOPSIS: A checklist can be used to implement comprehensive sleep-promoting interventions, but it remains a challenge to maintain adherence and sustainability among ICU clinicians.

SOURCE: Kamdar BB, et al. Developing, implementing, and evaluating a multifaceted quality improvement intervention to promote sleep in an ICU. Am J Med Quality 2014;29:546-554.

It is no surprise to any clinician that ICU patients do not get much sleep while hospitalized for a critical illness or injury. Sleep disturbances emanate from many sources, including noise, lights, care procedures, staff conversations, and receipt of mechanical ventilation to name but a few. Efforts have been made over the past couple of years to test and implement interventions in the ICU to promote sleep with the goal of improved quality and quantity of sleep. Given the complexity of sleep and the numerous factors that impair sleep in the ICU, the problem requires a much more comprehensive approach for promoting sleep among critically ill patients.

The quality improvement (QI) project by Kamdar and colleagues summarized a multifaceted, comprehensive approach to improving sleep in one medical ICU (MICU) at the Johns Hopkins Hospital in Baltimore, MD. The MICU consists of 16 beds that are staffed by registered nurses, nursing technicians, respiratory therapists, pharmacists, and physical and occupational therapists. Two physician teams comprised of an intensivist, fellow, and resident physicians round out the multidisciplinary care team.

The team based their sleep improvement project on an established four-step QI model1: 1) summarizing the evidence to identify beneficial relationships, 2) identifying barriers to implementation, 3) selecting and developing performance measures, and 4) ensuring all patients receive the interventions. The QI model also applies the four Es algorithm to engage and educate staff, execute the intervention, and evaluate performance.1

The team first reviewed the literature for potential, feasible sleep-promoting interventions (step 1). A “bundled” approach was selected and included environment modifications for day (window blinds open) and nighttime (dimming overhead lighting); nonpharmacological sleep aids such as ear plugs, eye masks, or tranquil music; and development of a pharmacological sleep guideline that discouraged medications known to influence sleep, such as benzodiazepines. The interested reader should refer to the original article and reference list for additional details or suggestions to promote sleep. A key component for any QI program is to address barriers to implementation. A prominent, team-identified barrier was that the MICU staff might be overwhelmed with multiple interventions and included the physical limitations of the MICU itself, such as the inability to adjust lighting to a desired level or ability to fully minimize overhead paging to reduce noise (step 2). A checklist was developed to document whether the specific shift interventions were implemented per the sleep bundle (step 3). These checklists were completed by nurses on the day shift and night shift, as well as by the unit clerk.

The four Es model was applied to make sure all patients received the bundled sleep interventions (step 4). The team made a concerted effort to garner buy-in from the staff; education was provided throughout the project. Daily nurse completion of the checklists ranged from 84-90% for day-shift and 76-86% for night-shift. Checklist completion by clerks was 79-94%. The nonpharmacological sleep aids saw the lowest implementation rates due to patients being asleep, delirious/comatose, or refusing. Medications to appropriately promote sleep increased from 0% to 60% across the project period. The two main challenges to adherence to the sleep bundle interventions were: 1) the large number of interventions required on each shift (three on days, 14 on nights), and 2) sleep bundle intervention adherence was achieved 60-80% of the time, which was significantly below the 100% goal. The team also discussed the challenges of bundle sustainability among the clinical staff, which can be extremely difficult in any setting.


This QI project reported by Kamdar and colleagues highlights a comprehensive, multicomponent intervention to promote sleep among ICU patients that utilized day- and night-shift specific checklists to promote adherence. Unfortunately, the report also provides a stark reminder to the reader of the realities of promoting adherence to any new initiative and sustaining “buy-in” after completion of any QI project. Perhaps the most significant contribution of the article is the frank discussion of barriers to implementation and strategies used by an experienced QI team to improve adherence rates to the sleep bundle during the study period. For example, the sleep promoting interventions were implemented in stages, which demonstrates the iterative nature of any QI project. Efforts to promote adherence to the checklist interventions included staff education throughout the project.

The QI model and four Es algorithm1 can be a useful framework for conducting a QI project in the ICU to improve patient-centered outcomes. The interested reader is advised to examine the article’s reference list for suggestions on conducting and reporting a QI project.


  1. Pronovost PJ, et al. Translating evidence into practice: A model for large scale knowledge translation. BMJ 2008;337:a1714.