NUR 5300 Module 3 Randomized Trial Appraisal Example

Design Before Results: Appraising a Three-Arm Trial of Inhaled Isopropyl Alcohol and Oral Ondansetron for Acute Nausea

Student Name

Master of Science in Nursing (MSN), Thomas Edison State University

NUR-5300: Evidence-Based Nursing Practice

Mentor Name

Month Day, Year

Design Before Results: Appraising a Three-Arm Trial of Inhaled Isopropyl Alcohol and Oral Ondansetron for Acute Nausea

Of the studies carried forward from my search, the most relevant to my question is a randomized trial by April et al. (2018), because one of its three arms combined inhaled isopropyl alcohol with an antiemetic, the closest match in the literature to the intervention I am studying. This paper appraises that trial using the Critical Appraisal Skills Programme (2020) checklist for randomized controlled trials, which asks whether the trial addressed a focused question, whether its methods were sound, what its results were and whether they apply locally. Following the principle that design should be weighed before results are believed, the methods are appraised first and the findings only afterward.

The Question and the Design

The trial asked whether aromatherapy with inhaled isopropyl alcohol, alone or combined with oral ondansetron, reduces nausea more than oral ondansetron alone among adult emergency department patients with nausea who did not need immediate intravenous access. The question is focused, with a defined population, interventions, comparison and outcome. The design was a randomized, blinded, placebo-controlled trial with three arms: isopropyl alcohol plus 4 milligrams of oral ondansetron, isopropyl alcohol plus an oral placebo, and an inhaled saline placebo plus oral ondansetron. The double-dummy structure, in which each participant received one inhaled and one oral treatment, is a strength, because it allowed the three combinations to be compared while keeping the number and form of treatments the same.

The primary outcome was the mean reduction in nausea on a 0 to 100 millimeter visual analog scale from enrollment to thirty minutes after the intervention, with rescue antiemetic use and adverse events as secondary outcomes. A thirty-minute window and a continuous patient-reported scale are appropriate for a symptom that changes quickly and is subjective by nature.

Recruitment, Randomization and Blinding

Participants were a convenience sample of adults at an urban tertiary care emergency department, enrolled when research staff were available. Convenience sampling does not bias the comparison between randomized groups, but it limits generalizability, because patients enrolled at certain hours may differ from the full population. One hundred twenty-two participants were randomized and 120 completed the study, a completion rate above 98 percent, so attrition bias is unlikely.

Blinding is the trial's most important methodological question. The authors describe the trial as blinded and used a saline pad as the inhaled placebo, but isopropyl alcohol has a strong, familiar smell that saline lacks, so participants in the alcohol arms could probably tell which inhaled treatment they received. Because the primary outcome is a self-reported symptom score, any expectation created by recognizing the alcohol could inflate its apparent effect. The oral placebo, by contrast, could be blinded effectively. This means the comparison between the two alcohol arms, which differ only in the oral drug, is better protected from expectation than the comparison between alcohol and saline.

Allocation concealment is not described in the abstract, and readers should check the full report for how the sequence was generated and hidden. If staff enrolling patients could predict the next assignment, they might, without intending to, steer more or less nauseated patients into particular arms, which is a different bias from unblinding and would affect the comparison even if participants could not tell what they inhaled.

Results

Only after weighing the design do the results deserve attention. Mean nausea reduction at thirty minutes was 30 millimeters in the alcohol plus ondansetron arm, 32 millimeters in the alcohol plus placebo arm and 9 millimeters in the saline plus ondansetron arm, with confidence intervals that did not overlap between the alcohol arms and the ondansetron-only arm. Rescue antiemetics were given to 27.5 percent and 25.0 percent of participants in the alcohol arms and 45.0 percent of those receiving ondansetron alone. No adverse events were reported. A difference of more than twenty millimeters on a 100-millimeter scale is generally considered clinically meaningful for symptom relief, so the effect, if real, matters to patients.

Two features of the results need interpretation. First, adding ondansetron to isopropyl alcohol produced no additional benefit within thirty minutes. That may reflect the time oral ondansetron takes to act, which the thirty-minute window could be too short to capture, rather than a true lack of effect. Second, the size of the alcohol effect is plausible in light of the systematic review by Hines et al. (2018), in which pooled trials, rated moderate in quality, showed nausea halving sooner with alcohol than with the usual drugs. Consistency with a pooled analysis increases confidence, although the pooled studies share the same blinding problem.

Applicability to an Ambulatory Surgery PACU

Several differences limit how directly the findings apply to my question. The population was emergency department patients with undifferentiated nausea, not postoperative patients, whose nausea has different causes, including anesthetic agents and opioids. The comparator was oral ondansetron, whereas PACU rescue is typically intravenous and from a class not used for prophylaxis. And the setting differs in staffing and time pressure. Against those differences, the intervention is identical to what PACU nurses could offer, it is cheap and quick, and no adverse events were reported. The trial supports the plausibility of a benefit in my population but does not establish it.

Dose and technique also limit transfer. The trial's inhalation instructions and number of repeat inhalations may differ from what a busy PACU nurse would do, and fidelity to the technique was not something the trial could guarantee outside its research protocol. A local protocol would need to specify technique carefully to have any chance of reproducing the effect.

Overall Judgment

The trial addressed a focused question with a sound randomized design, low attrition and a meaningful outcome, and its results are consistent with a systematic review. Its main weakness is the likelihood that participants recognized the smell of alcohol, which threatens the blinding of the self-reported primary outcome, along with convenience sampling and a short follow-up for oral ondansetron. In the evidence hierarchy it is a single randomized trial at moderate risk of bias. For my synthesis, it is the strongest available evidence on combining isopropyl alcohol with an antiemetic, but its findings will need to be weighed alongside postoperative trials before any recommendation is made for the PACU.

References

April, M. D., Oliver, J. J., Davis, W. T., Ong, D., Simon, E. M., Ng, P. C., & Hunter, C. J. (2018). Aromatherapy versus oral ondansetron for antiemetic therapy among adult emergency department patients: A randomized controlled trial. Annals of Emergency Medicine, 72(2), 184-193. https://doi.org/10.1016/j.annemergmed.2018.01.016

Critical Appraisal Skills Programme. (2020). CASP randomised controlled trial standard checklist. https://casp-uk.net/casp-tools-checklists/

Hines, S., Steels, E., Chang, A., & Gibbons, K. (2018). Aromatherapy for treatment of postoperative nausea and vomiting. Cochrane Database of Systematic Reviews, 2018(3), Article CD007598. https://doi.org/10.1002/14651858.CD007598.pub3