Don't Hyperventilate Over Triage Respiratory Rates

Abstract

SEE RELATED ARTICLE, P. 37. [Ann Emerg Med. 2013;61:44-45.] Are triage respiratory rates inaccurate? Absolutely. In this month's Annals, Bianchi et al1Bianchi W. Dugas A.F. Hsieh Y.H. et al.Revitalizing a vital sign: improving detection of tachypnea at primary triage.Ann Emerg Med. 2013; 61: 37-43Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar compared standard nurse-recorded triage respiratory rates with those measured by a research assistant during 60 seconds and those obtained by a commercial respiratory rate sensor (BioHarness; Zephyr Technology, Corp, Annapolis, MD). Tachypnea (>20 breaths/min) was missed 77% of the time. This statistic is alarming because emergency physicians fundamentally rely on triage vital signs to help assess their patients. With tachypnea being missed three fourths of the time, we must be receiving false reassurance for substantial numbers of patients. Tachypnea is an integral diagnostic element of systemic inflammatory response syndrome and a key component of pneumonia scores (eg, Pneumonia Severity Index, CURB 65); accordingly, erroneous triage respiratory rates should result in misclassifications of patients. Is there any evidence this is really a problem? During your last shift, how many patients were harmed because their triage respiratory rate was inaccurate? Bianchi et al1Bianchi W. Dugas A.F. Hsieh Y.H. et al.Revitalizing a vital sign: improving detection of tachypnea at primary triage.Ann Emerg Med. 2013; 61: 37-43Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar concluded that the solution is to change how respiratory rate is measured, perhaps by using the BioHarness device routinely at triage. This would be a time-consuming and expensive change for every emergency department (ED). But perhaps it must be done for the sake of accuracy. Let's critically examine what we already know about respiratory rates, examine the implications of these new data, and then try to make sense of this conundrum. Taking vital signs is a standard prelude to every ED visit and essentially all health care encounters. There is little dispute that respiratory rate is the least objective vital sign. Bianchi et al1Bianchi W. Dugas A.F. Hsieh Y.H. et al.Revitalizing a vital sign: improving detection of tachypnea at primary triage.Ann Emerg Med. 2013; 61: 37-43Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar demonstrated that routine triage assessments do not agree with measurements taken during a full minute (as recommended by the World Health Organization). Previous ED studies reported similar results.2Lovett P.B. Buchwald J.M. Stürmann K. et al.The vexatious vital: neither clinical measurements by nurses nor an electronic monitor provides accurate measurements of respiratory rate in triage.Ann Emerg Med. 2005; 45: 68-76Abstract Full Text Full Text PDF PubMed Scopus (167) Google Scholar, 3Hooker E.A. O'Brien D.J. Danzl D.F. et al.Respiratory rates in emergency department patients.J Emerg Med. 1989; 7: 129-132Abstract Full Text PDF PubMed Scopus (52) Google Scholar Indeed, the interrater reliability of measuring respiratory rates during a full minute is also poor, although this criterion standard has not been questioned.4Edmonds Z.V. Mower W.R. Lovato L.M. et al.The reliability of vital sign measurements.Ann Emerg Med. 2002; 39: 233-237Abstract Full Text Full Text PDF PubMed Scopus (112) Google Scholar It is widely assumed that triage nurses often estimate respiratory rates rather than formally counting respirations for 15 seconds and multiplying by 4. Recorded values such as 14 or 18 breaths/min, which are indivisible by 4, support this contention. Bianchi et al1Bianchi W. Dugas A.F. Hsieh Y.H. et al.Revitalizing a vital sign: improving detection of tachypnea at primary triage.Ann Emerg Med. 2013; 61: 37-43Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar reported that triage nurses most commonly record rates of 16 to 18 breaths/min, which seems consistent with anecdotal observations elsewhere. Given the inherent unreliability currently observed, how important is the accurate quantification of respiratory rate? Is there a clinically important difference between 12 versus 16 breaths/min? Depending on the source, there is a large range of “normal,” ie, 8 to 24 breaths/min.3Hooker E.A. O'Brien D.J. Danzl D.F. et al.Respiratory rates in emergency department patients.J Emerg Med. 1989; 7: 129-132Abstract Full Text PDF PubMed Scopus (52) Google Scholar, 5Tulaimat A. Patel A. Shah B. et al.Is the content of textbooks on the evaluation of a patient in respiratory distress adequate?.Respir Care. 2012; 57: 404-412Crossref PubMed Scopus (5) Google Scholar Do we need better accuracy in respiratory rate measurement all of the time, or even some of the time? Does the specific recorded value independently contribute to either triage nurses' or emergency physicians' assessments, or instead is the observation that matters a subjective judgment about the quality of respiratory effort? In a 24-center study of 10,837 adult ED patients triaged by 625 experienced nurses, we noted tachypnea (>20 breaths/min) in just 12% of subjects,6Cooper R.J. Schriger D.L. Flaherty H.L. et al.Effect of vital signs on triage decisions.Ann Emerg Med. 2002; 39: 223-232Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar less than the 23% observed by Bianchi et al.1Bianchi W. Dugas A.F. Hsieh Y.H. et al.Revitalizing a vital sign: improving detection of tachypnea at primary triage.Ann Emerg Med. 2013; 61: 37-43Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar These were not measurements during 60 seconds and so may simply represent the same measurement error Bianchi et al1Bianchi W. Dugas A.F. Hsieh Y.H. et al.Revitalizing a vital sign: improving detection of tachypnea at primary triage.Ann Emerg Med. 2013; 61: 37-43Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar reported. In this study,6Cooper R.J. Schriger D.L. Flaherty H.L. et al.Effect of vital signs on triage decisions.Ann Emerg Med. 2002; 39: 223-232Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar nurses performed routine triage and then assigned an acuity level twice, before and after knowing the vital signs. Nurses altered triage category assessments according to vital signs in just 720 patients (7%) overall; however, in only 16 cases was respiratory rate the only abnormal vital sign (2% of altered triage categories, 0.1% of all adults). Given this rarity of influence, even if we could measure respiratory rate more accurately would it really be worth the cost and effort? Triage is a complex process, involving history, vital signs, and important visual cues. These visual cues include physical signs that may or may not be explicitly measured that give the provider an indication of respiratory compromise, signs including depth of respirations, retractions, tripoding, pursed lips, and nasal flaring. Visual cues, separate from the vital signs, are an important part of triage decisions.7Salk E.D. Schriger D.L. Hubbell K.A. et al.Effect of visual cues, vital signs, and protocols on triage: a prospective randomized crossover trial.Ann Emerg Med. 1998; 32: 655-664Abstract Full Text Full Text PDF PubMed Scopus (54) Google Scholar Bianchi et al1Bianchi W. Dugas A.F. Hsieh Y.H. et al.Revitalizing a vital sign: improving detection of tachypnea at primary triage.Ann Emerg Med. 2013; 61: 37-43Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar pointed out that respiratory rate is the only triage vital sign not captured electronically and that the errors noted with manual measurement or estimation can be obviated by switching to the BioHarness. This device is wrapped around the patient's chest over a single layer of clothing and contains a pressure sensor in the fabric that measures chest wall movement. After a short period of calibration, the device counts respirations for 60 seconds while the patient remains (if possible) still and quiet. Suggesting we use the BioHarness is certainly in keeping with the current paradigm in medicine: our belief that clinical assessment is not sufficiently objective and therefore must be inferior to a machine, decision instrument, biomarker, or other reproducible technology.8Schriger D.L. Newman D.H. Medical decisionmaking: let's not forget the physician.Ann Emerg Med. 2012; 59: 219-220Abstract Full Text Full Text PDF PubMed Scopus (22) Google Scholar On the plus side, using the BioHarness would provide undeniably more accurate measures of respiratory rate. Maybe we should buy stock in the company and then have our medical directors order devices for all our EDs. Would our patients be receiving better care if we did this? Yes, because we would be richer, happier, and therefore in a better mood to provide great care! However, there are disadvantages. First, this proprietary device must be purchased. Second, the device must be physically placed on each patient (after he or she has undressed to a single clothing layer) and then its 60-second measurement supervised. This seems unlikely to save the triage nurse any time because the patient cannot be interviewed or examined during this period. Indeed, patients with respiratory distress would appear least likely to be able to tolerate the device. Finally, the challenge of keeping the BioHarness clean and fomite free between serial patients may be large. Worried about respiratory rate accuracy but not anxious to buy a BioHarness? A simple solution you could implement in your ED today would be to require triage nurses to measure respirations manually during a full minute for every patient. Although this may take longer than current practice, the duration is shorter than the time it takes to apply the BioHarness, let the machine calibrate, and record for a minute. An ironic new problem arises once you've implemented the BioHarness or somehow persuaded your triage nurses to measure respirations for a full minute. Yes, you now have much more accurate values. Unfortunately, everything we know is based on the old system. How would emergency physicians respond to suddenly having 4 times more patients identified as tachypneic? Would they simply order more low-yield tests? And what about all of the scores and instruments that include respiratory rate, eg, systemic inflammatory response syndrome, Pneumonia Severity Scores? These scores were derived from inaccurate measurements made with traditional triage, not any “better” method during a full minute. Paradoxically, these scoring systems may be invalid with more accurate respiratory rates, and they would require recalibration (ie, repeated derivation and validation) should we collectively decide to change our standard for measuring this vital sign. Should we implement a method to record respiratory rates during a full minute? The merits of such a change appear dubious, given that respiratory rate only rarely alters triage decisions.6Cooper R.J. Schriger D.L. Flaherty H.L. et al.Effect of vital signs on triage decisions.Ann Emerg Med. 2002; 39: 223-232Abstract Full Text Full Text PDF PubMed Scopus (72) Google Scholar There is no evidence that having more accurate respiratory rates improves patient outcomes. Indeed, perhaps clinicians only crudely interpret the quantitative nature of this vital sign, mentally simplifying it to either normal or abnormal. Should this be true, then we could save triage nurses further time by abandoning its quantitative assessment altogether and having them simply classify each patient's respiratory rate as one of 3 categories: abnormally slow, generally normal, or abnormally fast. To suggest a major change in triage practice, we need data that such change would benefit enough patients to warrant the additional time and expense. Although new technology can allow us to be more and more precise, there is not always a theoretical or empiric reason to assume that this will translate to improved patient-centered outcomes. Just because you can measure something more accurately does not mean you need to, and just because one can buy a BioHarness or other device to do this doesn't mean one should. Are triage respiratory rates inaccurate? Absolutely. Do I care? No.