Agreement, conflict and its causes, then GRADE certainty for each outcome: the NU504 Unit 8 literature synthesis argues the ICU sleep studies together rather than one at a time. Searches like "nu 504 unit 8 assignment example", "nu504 unit 8 sample" and "nu504 unit 8 example" land here.
Consistent in Direction, Weak in Certainty: A Literature Synthesis on Nighttime Sleep Aids and Bundles in Adult Intensive Care
[Student Name]
Purdue University Global
NU504: Scientific and Analytic Approaches to Advanced Evidence-Based Practice
Unit 8 Assignment
[Instructor Name]
[Date]
Synthesis of the nine studies in the Unit 7 evidence table; certainty ratings are the writer's, following GRADE.
Question and Approach
Among adults in intensive care, do earplugs and eye masks at night, used on their own or as one part of a bundle, reduce delirium and improve sleep compared with usual care? The nine studies carried forward from the evidence table include three trials with random allocation, four uncontrolled comparisons of a before period with an after period, and two reviews. No new pooled estimate is attempted here. The interventions differ, from earplugs alone to bundles that change lighting, noise and daytime routines; the delirium tools differ, from the CAM-ICU to a confusion scale; and follow-up windows run from one night to the whole ICU stay. Two published meta-analyses have already pooled subsets of these studies, and the synthesis uses their estimates while explaining what they combine.
Delirium
The direction of effect is consistent: every study that measured delirium or confusion found less of it with the aids or the bundle. The size and the strength of the evidence vary widely.
The randomized evidence is thin. Only one trial measured a cognitive outcome, and its reported benefit, a hazard ratio of 0.47 for delirium or mild confusion, came almost entirely from mild confusion; delirium itself occurred in 19 percent of the earplug group and 20 percent of controls (Van Rompaey et al., 2012). The Cochrane review found two trials of earplugs or eye masks with delirium data and pooled a relative risk of 0.55, 95% CI 0.38 to 0.80, from 177 participants, rating the evidence low quality (Hu, Jiang, Chen, et al., 2015).
The before-and-after studies show larger effects. A noise and light bundle was followed by a fall in delirium from 33 to 14 percent (Patel et al., 2014); a medical ICU sleep bundle was associated with lower odds of delirium or coma, odds ratio 0.46 (Kamdar et al., 2013); and a surgical ICU bundle that included earplugs and eye masks reduced the proportion of ICU days with delirium from 20 to 15 percent (Tonna et al., 2021). The larger effects in uncontrolled studies are what one would expect if part of the change came from other improvements made over the same years, such as lighter sedation and earlier mobility, which these designs cannot separate from the bundle. The meta-analysis that pooled randomized and nonrandomized studies together found a relative risk of 0.59 (95% CI, 0.44-0.78) with moderate heterogeneity, and its authors judged the bias risk high in all nine studies (Litton et al., 2016).
Sleep Quality
Effects on sleep are consistent in direction and small. Earplugs and eye masks with music improved perceived depth of sleep and overall quality after cardiac surgery (Hu, Jiang, Hegadoren, & Zhang, 2015), and earplugs alone improved sleep perception after the first night but not later (Van Rompaey et al., 2012). With polysomnography, the one randomized trial found fewer long awakenings but did not meet its primary outcome, deep sleep as a proportion of total sleep (Demoule et al., 2017). Two bundle studies found no significant change in overall perceived sleep quality, although noise ratings improved (Kamdar et al., 2013; Tonna et al., 2021). The Cochrane review pooled a longer total sleep time of 2.19 hours from two small trials (Hu, Jiang, Chen, et al., 2015). All self-reported sleep measures share one limitation: they can be completed only by patients awake and lucid enough to answer, who are not the patients at highest risk.
Adherence and Tolerability
Adherence is the outcome most likely to decide whether the intervention works on a real unit. In one trial, 30 percent of patients assigned to earplugs did not keep them in all night (Demoule et al., 2017). Across the studies in one meta-analysis, mean noncompliance was 13 percent (Litton et al., 2016). Patients in a before-and-after study named noise as the main barrier to sleep and named earplugs and masks among the things that helped (Jones & Dawson, 2012). No study reported serious harms.
Gaps the Studies Leave
Three gaps limit what this body of work can tell our unit. First, no included trial randomized a bundle with the CAM-ICU as its primary outcome, so the best-designed studies and the most relevant outcome never meet. Second, sedated patients were excluded from the one trial of earplugs with a cognitive outcome, and they make up a large share of delirium in a mixed unit. Third, publication bias cannot be ruled out: the search found [two] completed registered trials without published results, and small positive studies are more likely to reach print than small null ones. Each gap points to a design choice for our pilot: measure delirium with the CAM-ICU, include lightly sedated patients where the aids can be used, and report the result whichever way it falls.
Certainty of Evidence
Certainty is rated with GRADE, which starts randomized evidence at high and observational evidence at low and then moves it down for risk of bias, inconsistency, indirectness, imprecision and publication bias, or up for large effects (Guyatt et al., 2008).
1. Delirium. Starting point: high, from randomized trials. Down one level for risk of bias, since assessment was unblinded or only partly blinded. Down one level for imprecision, since the randomized evidence comes from fewer than 200 patients. Indirectness is considered but not downgraded again, although the only trial with a cognitive outcome used a combined confusion outcome. Final: low.
2. Sleep quality. Starting point: high. Down one level for risk of bias; self-report by unblinded patients. Not downgraded for imprecision, since effects are consistent. Final: moderate for perceived sleep; low for objective sleep, where the one polysomnography trial missed its primary outcome.
3. Adherence. Observational data, consistent. Final: moderate that a meaningful minority will not tolerate earplugs all night.
How Far the Evidence Reaches
The evidence supports a narrow conclusion. Earplugs and eye masks, especially as part of a bundle that also reduces noise and light, are probably associated with better perceived sleep and may be associated with less delirium; the delirium evidence is of low certainty. The aids are cheap and without serious harms, and about one patient in seven to one in three will not tolerate them all night. That combination supports a monitored pilot on our unit, with careful measurement of delirium and adherence, and does not support a claim that earplugs prevent delirium.
References
Demoule, A., Carreira, S., Lavault, S., Pallanca, O., Morawiec, E., Mayaux, J., Arnulf, I., & Similowski, T. (2017). Impact of earplugs and eye mask on sleep in critically ill patients: A prospective randomized study. Critical Care, 21(1), 284. https://doi.org/10.1186/s13054-017-1865-0
Guyatt, G. H., Oxman, A. D., Vist, G. E., Kunz, R., Falck-Ytter, Y., Alonso-Coello, P., & Schünemann, H. J. (2008). GRADE: An emerging consensus on rating quality of evidence and strength of recommendations. BMJ, 336(7650), 924-926. https://doi.org/10.1136/bmj.39489.470347.AD
Hu, R.-F., Jiang, X.-Y., Chen, J., Zeng, Z., Chen, X. Y., Li, Y., Huining, X., Evans, D. J., & Wang, S. (2015). Non-pharmacological interventions for sleep promotion in the intensive care unit. Cochrane Database of Systematic Reviews, 2015(10), CD008808. https://doi.org/10.1002/14651858.CD008808.pub2
Hu, R.-F., Jiang, X.-Y., Hegadoren, K. M., & Zhang, Y.-H. (2015). Effects of earplugs and eye masks combined with relaxing music on sleep, melatonin and cortisol levels in ICU patients: A randomized controlled trial. Critical Care, 19(1), 115. https://doi.org/10.1186/s13054-015-0855-3
Jones, C., & Dawson, D. (2012). Eye masks and earplugs improve patient's perception of sleep. Nursing in Critical Care, 17(5), 247-254. https://doi.org/10.1111/j.1478-5153.2012.00501.x
Kamdar, B. B., King, L. M., Collop, N. A., Sakamuri, S., Colantuoni, E., Neufeld, K. J., Bienvenu, O. J., Rowden, A. M., Touradji, P., Brower, R. G., & Needham, D. M. (2013). The effect of a quality improvement intervention on perceived sleep quality and cognition in a medical ICU. Critical Care Medicine, 41(3), 800-809. https://doi.org/10.1097/CCM.0b013e3182746442
Litton, E., Carnegie, V., Elliott, R., & Webb, S. A. R. (2016). The efficacy of earplugs as a sleep hygiene strategy for reducing delirium in the ICU: A systematic review and meta-analysis. Critical Care Medicine, 44(5), 992-999. https://doi.org/10.1097/CCM.0000000000001557
Patel, J., Baldwin, J., Bunting, P., & Laha, S. (2014). The effect of a multicomponent multidisciplinary bundle of interventions on sleep and delirium in medical and surgical intensive care patients. Anaesthesia, 69(6), 540-549. https://doi.org/10.1111/anae.12638
Tonna, J. E., Dalton, A., Presson, A. P., Zhang, C., Colantuoni, E., Lander, K., Howard, S., Beynon, J., & Kamdar, B. B. (2021). The effect of a quality improvement intervention on sleep and delirium in critically ill patients in a surgical ICU. Chest, 160(3), 899-908. https://doi.org/10.1016/j.chest.2021.03.030
Van Rompaey, B., Elseviers, M. M., Van Drom, W., Fromont, V., & Jorens, P. G. (2012). The effect of earplugs during the night on the onset of delirium and sleep perception: A randomized controlled trial in intensive care patients. Critical Care, 16(3), R73. https://doi.org/10.1186/cc11330
How this NU504 Unit 8 example is structured
Disagreement is treated as information. The delirium section sets the thin randomized evidence beside the larger before-and-after effects and gives a reason the uncontrolled studies look stronger, other practice changes made over the same years. Sleep quality follows, consistent in direction and small, with the caution that only lucid patients can rate their sleep. Adherence gets its own section. A section on gaps names what no included study tested and what the pilot should therefore measure. GRADE is applied openly, outcome by outcome, with each downgrade named: delirium lands at low certainty, perceived sleep at moderate. The conclusion claims only what that permits, a monitored pilot.
Get an NU504 Unit 8 example written to your instructions
Send the Unit 8 prompt, the rubric and the studies you tabled earlier, and note if GRADE or some other certainty framework is expected. The synthesis argues your findings together by outcome, explains the disagreements instead of hiding them, and sizes its conclusion to the certainty. 24-48h, and your first custom sample is free. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.
NU504 Unit 8 questions, answered
What is GRADE, and do I have to use it?
GRADE is a widely used system for rating certainty of evidence for each outcome as high, moderate, low or very low. Randomized evidence starts high and observational evidence low; ratings move down for risk of bias, inconsistency, indirectness, imprecision or publication bias, and observational evidence can move up in limited cases. Whether it is required depends on the prompt; some sections name another hierarchy.
Should I calculate a pooled estimate across the studies?
Only if the studies are similar enough and the method is sound, which usually means a formal meta-analysis rather than adding numbers by hand. When interventions, measures or populations differ, a narrative synthesis that explains the differences is more honest. If a published meta-analysis exists, its pooled estimate and heterogeneity can be reported and discussed instead.
How can two good trials reach opposite conclusions?
Differences in the intervention, the comparison group, the population, adherence or the outcome measure can all produce different results, and so can chance in smaller trials. A synthesis examines those differences before assuming either trial is wrong. Often the disagreement narrows the claim usefully, revealing conditions under which the intervention helps and others where it does not.