To the Editor
We read with interest the study by Gabel et al1 regarding the reduction of postoperative nausea and vomiting using post hoc provider feedback and real-time clinical decision support implemented within their electronic medical record system. Because side effects of drugs used for postoperative nausea and vomiting prophylaxis can produce clinical complications (eg, confusion, extrapyramidal reactions, and headache), and affect operating room processes (eg, delayed emergence or prolonged postanesthesia care unit stays), overtreatment should be assessed as an unintended consequence of clinical decision support.2 In a study using individual provider feedback, Frenzel et al3 reported that overtreatment (ie, administration of more than the indicated number of antiemetics based on risk factors) increased from 13.1% to 25.4%. In contrast, Kooij et al4 demonstrated less use of antiemetics in low-risk patients with clinical decision support, as did Kappen et al.5
We contacted the authors for their summary count data, sufficient to assess the extent of overtreatment of low-risk patients (0–2 risk factors), which were generously provided (Eilon Gabel, University of California Los Angeles, personal communication, January 2, 2019).1 Because their algorithm did not include the expected use of postoperative opioids as a risk factor, we considered overtreatment as the administration of ≥2 drugs in excess of the number of risk factors. Overtreatment increased from 2.5% to 4.5% after clinical decision support implementation (relative risk, 1.79; 95% CI, 1.53–2.10; P < .0001). This relative risk was indistinguishable from the 1.94 relative risk reported by Frenzel et al.3 However, the absolute increase in risk was only 2.0%, compared to 22.3% in the study by Frenzel et al,3 mitigating the clinical importance among patients of Gabel et al.1
The results from the cited studies3–5 and our secondary analysis of the data sent1 underscore the variability in the presence and magnitude of unintended consequences of clinical decision support related to postoperative nausea and vomiting prophylaxis.2 Further research into factors explaining why some studies demonstrate greater overtreatment1,3 after clinical decision support implementation and other studies show less overtreatment4,5 is needed. Although the authors should be reassured that the extent of overtreatment of their patients resulting from their clinical decision support system is small, other hospitals should not rely on this result. Rather, each hospital needs to determine its baseline percentage of prophylactic postoperative nausea and vomiting overtreatment among low-risk patients, then monitor the change after the implementation of clinical decision support.2
Richard H. Epstein, MD, FASA
Department of Anesthesiology
Pain Management and Perioperative Medicine
University of Miami
Franklin Dexter, MD, PhD, FASA
Division of Management Consulting Department of Anesthesia
University of Iowa
Iowa City, Iowa
1. Gabel E, Shin J, Hofer I, et al. Digital quality improvement approach reduces the need for rescue antiemetics in high-risk patients: a comparative effectiveness study using interrupted time series and propensity score matching analysis. Anesth Analg. 2019;128:867–876.
2. Epstein RH. Postoperative nausea and vomiting, decision support, and regulatory oversight. Anesth Analg. 2010;111:270–271.
3. Frenzel JC, Kee SS, Ensor JE, Riedel BJ, Ruiz JR. Ongoing provision of individual clinician performance data improves practice behavior. Anesth Analg. 2010;111:515–519.
4. Kooij FO, Vos N, Siebenga P, Klok T, Hollmann MW, Kal JE. Automated reminders decrease postoperative nausea and vomiting incidence in a general surgical population. Br J Anaesth. 2012;108:961–965.
5. Kappen TH, Moons KG, van Wolfswinkel L, Kalkman CJ, Vergouwe Y, van Klei WA. Impact of risk assessments on prophylactic antiemetic prescription and the incidence of postoperative nausea and vomiting: a cluster-randomized trial. Anesthesiology. 2014;120:343–354.