Dyspnea (breathing discomfort) is commonly experienced by critically ill patients and at this time is not routinely assessed and documented. Intensive care unit nurses at the study institution recently instituted routine assessment and documentation of dyspnea in all patients able to report using a numeric scale ranging from 0 to 10.
To assess nurses’ perceptions of the utility of routine dyspnea measurement, patients’ comprehension of assessment questions, and the impact on nursing practice and to gather nurses’ suggestions for improvement.
Data were obtained from interviews with intensive care unit nurses in small focus groups and an anonymous online survey randomly distributed to nurses representing all intensive care units.
Intensive care unit nurses affirmed the importance of routine dyspnea assessment and documentation. Before implementing the measurement tool, nurses often assessed for breathing discomfort in patients by using observed signs. Most nurses agreed that routine assessment can be used to predict patients’ outcomes and improve patient-centered care. Nurses found the assessment tool easy to use and reported that it did not interfere with workflow. Nurses felt that patients were able to provide meaningful ratings of dyspnea, similar to ratings of pain, and often used patients’ ratings in conjunction with observed physical signs to optimize patient care.
Our study shows that nurses understand the importance of routine dyspnea assessment and that the addition of a simple patient report scale can improve care delivery and does not add to the burden of work-flow.
Notice to CE enrollees:
This article has been designated for CE contact hour(s). The evaluation demonstrates your knowledge of the following objectives:
Identify why patients who report dyspnea are at increased risk of adverse outcomes, including higher mortality rates in the hospital.
Describe how routine dyspnea assessment can lead to early recognition of a patient’s clinical decline and improve symptom management.
Recognize that nurses and physicians evaluating dyspnea on the basis of clinical signs routinely underestimate patients’ reports of dyspnea.
To complete the evaluation for CE contact hour(s) for this article #A2029022, visit www.ajcconline.org and click the “CE Articles” button. No CE evaluation fee for AACN members. This expires on March 1, 2023.
The American Association of Critical-Care Nurses is accredited as a provider of nursing continuing professional development by the American Nurses Credentialing Center’s Commission on Accreditation, ANCC Provider Number 0012. AACN has been approved as a provider of continuing education in nursing by the California Board of Registered Nursing (CA BRN), CA Provider Number CEP1036, for 1.0 contact hour.
Patients receiving mechanical ventilation in intensive care units (ICUs) have described the dyspnea experience thus: “I often thought about death while I was attacked by dyspnea”; “I wondered what’s going on with my breathing, I asked myself ‘will I die here?’”1 Despite evidence that dyspnea causes suffering2,3 and is a predictor of adverse patient outcomes,4,5 little information regarding routine measurement and documentation of dyspnea in the ICU has been published.
Recent studies show the harmful impact and high prevalence of dyspnea in ICU patients. In one recent study of 30 ICU patients receiving mechanical ventilation, 100% reported breathing discomfort (median score, 4 on a scale of 0 to 10).6 In another study, 47% of 96 patients receiving mechanical ventilation reported dyspnea (median score, 5 on a scale of 0 to 10).7 Dyspnea was reported to be the most distressing of 10 symptoms in a survey of 171 ICU patients.8 Dyspnea in ICU patients is associated with indicators of posttraumatic stress disorder and correlates with duration of mechanical ventilation.3,9-11
Dyspnea is an independent predictor of morbidity and mortality in specific patient populations and is not limited to diseases of pulmonary origin. Dyspnea is more closely associated with cardiac mortality than is angina12,13 and is a more reliable predictor of mortality in patients with esophageal and gastric cancers than are symptoms such as pain, nausea, loss of appetite, and fatigue.14,15 Patients on medical-surgical units experiencing any dyspnea on admission have greater than a 3-fold odds of death in the hospital.16 Assessing and documenting dyspnea are the first steps in managing this distressing symptom.
The American Thoracic Society defines dyspnea as a “subjective experience of breathing discomfort that consists of qualitatively distinct sensations that vary in intensity.”17 This statement emphasizes that dyspnea can be perceived, and thus reported, only by the person experiencing it. Clinicians, including critical care nurses, respiratory therapists, and physicians, can decrease suffering and improve quality of care by asking for dyspnea ratings from all patients who are able to respond. The Joint Commission’s current standards require that organizations establish policies regarding pain assessment and treatment and conduct educational efforts to ensure compliance,18 so clinicians in health care systems worldwide routinely assess and document patients’ pain ratings. Although less common than pain, dyspnea is prevalent in ICU patients and can be more distressing.8,19
In 2013, nurses at our institution began routinely measuring and documenting dyspnea in all medical-surgical patients using a numeric scale ranging from 0 to 10, with 10 indicating unbearable dyspnea. In 2015, this practice was extended to all ICU patients, and nurses were asked to include dyspnea in routine patient assessments every 4 hours. (Data entry in our electronic documentation system is not mandatory.) We conducted focus sessions and an anonymous online survey to obtain nurses’ feedback regarding routine assessment.
Patients on medical-surgical units who report dyspnea are more likely to suffer serious adverse events in hospital.
The study design included a 2-part data collection method with qualitative (focus group interviews) and quantitative (random anonymous survey) components. These studies were approved by the medical center’s institutional review board.
In August 2016, we invited ICU nurses to participate in focus groups designed to assess nurses’ perceptions of the utility of routine dyspnea assessment, ease of use, patients’ comprehension of questions, and impact on practice and to collect suggestions for improvement. The location, date, and time of the sessions were arranged with local nursing leaders and invitations were sent to nurses via email 1 week before each session, with a reminder sent the day before the session. Inclusion in the sessions was limited to nurses on duty the day interviews were held, and written consent was obtained. Sessions were held during nurses’ nonpaid lunch breaks in locations near the 8 ICUs. Participation was voluntary, and nurses were allowed to enter and exit throughout the session as necessary. The purposes of the focus groups were to help us tailor the online survey to ICU nurses and to provide verbatim comments to illustrate the survey results.
We held two 30-minute focus sessions. Facilitators took field notes and recorded sessions to document discussion accurately. Participants were informed that sessions were recorded and that all identifying information would be omitted when the recordings were transcribed. A staff nurse (N.S.V.) and a clinical nurse specialist (K.M.B.) led the focus sessions. An interview guide was provided to participants to facilitate conversation and ensure that all discussion points were addressed. Seven and 10 nurses representing 6 of the ICUs attended the first and second session, respectively.
On the basis of experience and information obtained from the focus interviews, we made adjustments to a survey we had previously administered to medical-surgical nurses at our institution19 to better suit the survey to ICU nurses. All of the survey questions and responses are shown in the Supplement (available online only at www.ajcconline.org). In September 2016, the modified survey was distributed via a Web-based software platform (REDCap, hosted at Harvard Catalyst–Beth Israel Deaconess Medical Center) to 48 registered nurses (14% of the ICU nursing staff) selected randomly from each of the ICUs. Nurses were given 4 weeks to complete the survey. Study researchers and local unit leaders sent reminders via email. The survey allowed participants to add comments; some of the more illustrative are included in the Results.
Importance of Dyspnea Assessment
Thirty-seven of 48 surveys were completed (77% response rate). Nurses reported that assessing dyspnea with a uniform tool and tracking dyspnea at every shift are important. They also stated that routine measurement can be useful in predicting patients’ outcomes and can lead to improved patient-centered care (Figure 1). Comments from the survey and focus sessions enhanced the survey finding that nurses believe routine assessment is important:
“I have always completed the dyspnea assessment when I assess respiratory distress.”
“Dyspnea assessment was already part of my patient assessment if the patient was able to report their level of respiratory distress.”
“Allows for patient to explain in their own words how they are feeling.”
Implementation of Dyspnea Assessment
Most nurses (92%) found the assessment tool easy or very easy to administer. Most (68%) noted the addition of dyspnea assessment did not interfere with workflow, and some (32%) said it improved workflow (Figure 2). Comments from the 8% of nurses who found the tool “difficult” to administer included the following:
“There are too many options for the different levels of distress … mild, moderate, and severe would suffice.”
“Our patients often cannot rate/score their dyspnea. They don’t understand the scale.”
“Make the scale simpler … normal, worse than normal, worse than it’s ever been before.”
Practicalities of Assessment
Most nurses reported using more than 1 phrase to explain to the patient what they were supposed to rate. Many (65%) included the phrase “breathing discomfort,” as contained in the American Thoracic Society’s definition of dyspnea. Other phrases commonly used were “difficulty breathing/trouble breathing” (92%), “short of breath” (73%), and “can’t catch your breath” (57%).
Most nurses (76%) reported first asking a yes/ no question (eg, “Are you having any breathing difficulty?”) when assessing for dyspnea. For patients who responded no, 42% of nurses said they recorded a score of 0 without asking the patient to provide a number rating. The following comments explain this process:
“I typically ask if they are having difficulty breathing or feeling short of breath. If the answer is no, I presume that the number rating is 0/10 as I would presume with the pain scale… .”
“I document ‘0’ if the patient responds no if their diagnosis and my assessment support this. [I] ask more questions or rephrase … if objective findings suggest otherwise.”
“If the patient does not appear to be in any distress, oxygen saturations are fine on room air, etc, and they report no, I would document a ‘0.’ If their breathing is off baseline, I will investigate further.”
Nurses in focus sessions and the survey recalled patients who responded no to the initial question but then gave a nonzero rating. Nurses felt that the initial response of no was often because patients with chronic disorders become accustomed to living with breathing discomfort:
“I have noticed that patients often give themselves a 1-3 rating even when they answer no.”
“A patient with COPD [chronic obstructive pulmonary disease] may say no, but their baseline dyspnea score could be 4, so it is important to obtain the baseline report.”
“Every patient has a different baseline level … such as with COPD, asthma patients. It is important to ask the number as their baseline may be 3.”
Patient’s Report Versus Observed Signs to Assess Dyspnea
Dyspnea training for nurses in our hospital includes the concept that dyspnea, like pain, is defined as the patient’s experience, so the patient’s report is the best indicator of severity. However, ICU nurses reported that they often supplement or even supplant patients’ reports with their own estimates from observed signs. Nurses used physical signs to modify patients’ reports most often because the patient was unresponsive or was unable to use the number scale for some other reason. Comments included the following:
“I don’t complete the dyspnea score if the patient cannot give the report.”
“Often patients are intubated, confused, delirious, or have dementia and cannot answer.”
Nurses reported using physical signs rather than patients’ reports twice as often for patients who seemed more uncomfortable than they claimed as for patients who seemed less uncomfortable than they claimed (Figure 3). Nurses reported assessing respiratory distress with the following physical signs listed in the survey (in descending order of frequency of selection): accessory muscle use, tachypnea, difficulty speaking, restless movements, heart rate, nasal flaring, ventilator dyssynchrony, and fearful facial expression.
The online survey confirmed that ICU nurses often used some combination of patient’s report and observed signs to obtain a rating for dyspnea; 15 of 36 respondents said they always use a combination. Six nurses report they always used the patient’s report alone, and only 3 nurses said they always used physical signs alone.
Nurses in the focus groups and in the survey suggested that a solution to this conflict might be to provide a separate scale allowing nurses to report their respiratory distress estimates according to observed signs. Comments regarding the use of observed physical signs rather than the patient’s report included the following:
“In the ICU, most patients cannot speak due to the ventilator or altered mental status … it is important to use nonverbal cues from the patient to assess.”
“Patients who are clearly labored sometimes say their breathing is ‘not bad, 3/10.’ This … is sometimes because at baseline their breathing is terrible or they just don’t realize how hard they are working to breathe, whether because of delirium or they just don’t realize.”
Patients’ Ability to Rate Dyspnea
Seventy percent of ICU nurses reported that patients give a meaningful rating for dyspnea at least half of the time, and 81% reported that patients give a meaningful rating for pain at least half of the time. In other words, nurses had slightly less confidence in patients’ dyspnea ratings than in their pain ratings.
Interventions After Dyspnea Assessment
For patients who report increasing shortness of breath, nearly all nurses (95%) said that they employ nonpharmacologic interventions such as repositioning techniques and coaching the patient to take slow, deep breaths. Many (60%) reported assessing the need for narcotics. Nearly all nurses (86%) said they notify a physician when shortness of breath increases, and 65% summon a respiratory therapist. Fifty-seven percent of nurses surveyed said it would be useful to have an algorithm with specific options for the treatment of dyspnea (eg, repositioning, facial fan, morphine).
Most nurses found the dyspnea assessment tool easy to administer and noted that using it did not interfere with workflow.
Nurses reported that physicians would always or usually take the following actions when their patients reported increased respiratory distress: request vital signs and/or oxygen saturation (95%), order laboratory or imaging studies (42%), order a pharmacologic or nonpharmacologic intervention to relieve dyspnea (47%), evaluate the patient (75%), or request that nurses reassess the patient later (81%). Rarely (3% of the time), nurses reported that the physician responsible would take none of these actions (see Table).
Our data show that most ICU nurses feel it is important to routinely assess and document Dyspnea. Most nurses in our study found that the single-dimension scale was easy to use and did not impede workflow. We found 3 areas for possible improvement. (1) Descriptive terms used to ask patients to rate their dyspnea are not uniform. (2) Nurses reported frequently asking a yes/no question (eg, “Are you having shortness of breath?”) and recording a response of no as a 0 rating rather than asking the patient for a rating. (3) Nurses reported frequently using observed signs to modify or supplant the patient’s rating.
Importance of Dyspnea Assessment
Consistent with the results of our earlier survey of medical-surgical nurses,20 most ICU nurses stated that measuring dyspnea with a standard tool and periodically documenting dyspnea are important. Nurses also agreed that routine assessment can improve care delivery and patients’ outcomes. Dyspnea in ICU patients is clearly correlated with suffering that may lead to long-term harmful sequelae, including post-traumatic stress disorder. Patients experiencing moderate to severe dyspnea during hospitalization may be at increased risk for negative outcomes.4
Many ICU nurses reported that they had always included dyspnea in their patient assessments but had not used a standard scale or routinely documented findings. Nurses reported that incorporating routine measurement is easy and that providing a standard entry on the electronic form could even improve workflow.
Areas for Improvement
Although a widely used plain-language definition of dyspnea (breathing discomfort) exists, many nurses used other terms (eg, shortness of breath or breathing difficulty) when describing to patients what they should rate. This difference undoubtedly leads to more variance in the data, but no direct evidence shows whether this is an important problem in clinical measurements.
Response of No Versus Rating of 0
Many nurses reported asking a yes/no question about dyspnea and recording responses of no as 0. It takes only a few seconds more to collect a rating from the patient, and pertinent information may be lost by using this shortcut.
Patient’s Report Versus Nurse’s Observation
By definition, dyspnea is a patient’s own experience of breathing discomfort, so measuring dyspnea requires asking the patient. Only patients know how much discomfort they are experiencing, so a patient’s report of symptoms is often considered the gold standard. Several studies have shown that clinicians’ estimates of dyspnea are poorly correlated with patients’ reports (R2 values ranging from 0.2 to 0.4).6,8,21,22 Although these correlations may be statistically significant in populations, one can have little confidence in estimating the discomfort of a given patient. One study showed that in patients who reported a dyspnea rating of 4 or greater (on a scale of 0 to 10), 47% of nurses underestimated the patient’s dyspnea rating by at least 4 points (Figure 4).21
A reliable patient report is unavailable in some instances (eg, cognitively impaired or unresponsive patients). In these cases, nurses use their best alternative to fill the information gap and properly care for the patient. Although well intended, this method creates difficulty in interpreting the record. A comparison of the results of this study and those of our earlier study20 indicates that this problem is much more common in ICU patients, who are more likely to be impaired because of heavier sedation and more severe illness. Nurses who used physical signs to substitute for or modify the patient’s report did so 92% of the time because the patient was unresponsive (eg, sedated). We suggest that this information gap would be best documented by providing nurses with a separate scale to note their own assessment of respiratory distress, rather than by having nurses act as proxies for a patient’s report. Less commonly, nurses modified patients’ ratings because patients appeared more uncomfortable than their reports suggested (the nurses thereby erred on the side of providing improved symptom relief) or patients appeared less uncomfortable than their reports suggested (uncommon in our survey). A separate scale to note the clinician’s assessment of respiratory distress would provide a better way to document this disagreement.
Nurses in our study felt that patients could provide meaningful dyspnea ratings nearly as often as meaningful pain ratings. This finding was very similar to our findings with medical-surgical nurses. When ICU nurses were concerned that a patient’s rating did not truly reflect the patient’s discomfort, they responded by asking further questions or rephrasing and by using objective findings to modify the patient’s report.
Need for a Universal Dyspnea Assessment Tool
Implementing routine assessment with a tool that is easy, fast, and inexpensive can lead to early recognition of deterioration in a patient’s status and subsequent interventions to improve symptom management. Intensive care unit nurses strongly endorsed the use of a standard tool to measure dyspnea and to record and track it at every shift. The 0-to-10 numeric scale was chosen for its brevity and ease of use, for its consistency with the common pain assessment tool, and because 10-point scales are broadly recognized and understood by the general population. Other assessment scales exist and can provide a much richer collection of data23,24 but are more cumbersome to use and thus may not gain acceptance for routine screening in the busy inpatient setting. Nurses in focus sessions indicated that a simpler tool may be easier for patients to comprehend; further studies to compare such scales may be useful.
Because nurses and physicians evaluating Dyspnea on the basis of physical signs nearly always underestimate the patient’s reported dyspnea,21 a uniform, systematic tool to evaluate signs of Dyspnea in nonverbal patients would be useful. The Respiratory Distress Observation Scale is a recently developed instrument for the behavioral assessment of respiratory distress.25 This tool is intended for use only in patients who are unable to communicate. Validation trials in patients able to provide a dyspnea rating have shown a statistically significant correlation; at best the Respiratory Distress Observation Scale explains about half the variance in patients’ ratings (ie, R2 values generally ranging from 0.4 to 0.5 but sometimes ranging as low as 0.1).25-28 A modified version has been developed for ICU patients.29 The Respiratory Distress Observation Scale and other observation scales may prove to be acceptable surrogates for dyspnea ratings in patients unable to report but require further psychometric testing in ICU patients unable to report.
Once dyspnea is recognized, management includes pharmacologic and nonpharmacologic interventions2,30 that are implemented while pursuing an accurate diagnosis so the underlying cause can be appropriately treated. Nurses readily reported using interventions such as repositioning; remaining with the patient; coaching slow, deep breathing and providing reassurance; and notifying the physician or respiratory therapist. Physicians often request closer monitoring, additional laboratory or imaging studies, and pharmacologic interventions. Most nurses surveyed (57%) reported that a treatment algorithm for acutely dyspneic patients would be useful.
The nurses at our institution may not be representative of nurses at all hospitals. Ours is an academic tertiary care hospital, and several of the attending physicians in our medical ICUs work closely with researchers in our dyspnea laboratory. These physicians may have influenced the level of dyspnea awareness among nurses and respiratory therapists. (Our sample was drawn about equally from medical and surgical ICUs.)
Implementation of dyspnea assessment and documentation in the ICU requires acceptance and support by the nurses who perform these assessments. A significant barrier to adoption of yet another nursing requirement is the perception that it will impede work-flow, burden busy nurses, and cause resentment. Our study of front-line ICU nurses shows, on the contrary, that routine dyspnea measurement is feasible in the highly stressful critical care environment, that nurses were able to readily incorporate it into their practice, and that nurses endorse it as important to patient care. The Joint Commission established standards for pain assessment and intervention decades ago in response to the national report of undertreated pain. We believe that the same standards should exist for dyspnea.
The authors are very grateful to the unit nurses for their enthusiastic participation, the Lois E. Silverman Department of Nursing for enabling and supporting this project, and Dr Robert W. Lansing for his generous time and valuable input.
This study was supported by NIH grant NR10006 to Robert B. Banzett.
To purchase electronic or print reprints, contact American Association of Critical-Care Nurses, 101 Columbia, Aliso Viejo, CA 92656. Phone, (800) 899-1712 or (949) 362-2050 (ext 532); fax, (949) 362-2049; email, firstname.lastname@example.org.