Using a Dyspnea Assessment Tool to Improve Care at the End of Life

Lorri Birkholz, DNP, RN, NE-BC; Tina Haney, DNP, CNS, RN


Journal of Hospice and Palliative Nursing. 2018;20(3):219-227. 

In This Article


The RDOS was developed as a method for health care providers to assess dyspnea in patients who cannot self-report.[14] Since its inception, the RDOS tool has been modified to include an eighth assessment item, which is the presence/absence of paradoxical breathing. Paradoxical breathing is described as breathing movements in which the chest moves in on inspiration and out on expiration, which is the reverse of normal breathing.[20]

Project Design

This pre-experimental study was designed using a pretest/posttest format to determine improvement in the assessment and treatment of dyspnea after educational instruction on the use of the RDOS was provided. Scenarios depicting 3 degrees of dyspnea severity were developed. The process variable was the educational theory and fidelity experience on the use of the RDOS tool. The dependent variables were the outcomes of improved dyspnea assessment and dyspnea treatment when the RDOS tool was applied in the simulation setting.[21] Studies have shown that the use of standardized patients in nursing education creates a level of realism that enriches health assessment competence.[22] The standardized patients utilized for this study were highly trained actors from a national center for simulation education and used role play to realistically depict patients with varying degrees of dyspnea. The study was approved by the quality improvement/institutional review board committee at an Eastern US medical system and the author's affiliated university institutional review board committee prior to the start of the study.


Nurses who provide end-of-life care were the target population to receive the educational program on the use of the RDOS tool for this study. Nurses participating in the study worked in settings where there was currently no standardized tool for dyspnea assessment in use. There were 2 settings used for this study. The first was a not-for-profit hospice agency located in the Northeast that provides hospice care both in the community and at a dedicated hospice inpatient unit. The other was a medical/surgical unit of a rural community hospital in the Western United States where the nurses provide end-of-life and "comfort measures only" care without access to any palliative care specialty services. All nurse participants received the RDOS educational program along with the questionnaires, pretest/posttest, and evaluations. Inclusion criteria were limited to registered nurses (RNs) older than 18 years who currently were employed in settings that provide end-of-life care. Exclusion criteria included any staff who were not RNs.


The simulation scenarios were developed by the author and filmed using standardized patients of multiple ethnicities at an Eastern US medical school simulation center. The dyspnea scenarios were designed to depict mild, moderate, and severe levels of dyspnea using various combinations of the 8 variables of the RDOS. All but 1 of the scenarios was recorded without sound because the RDOS is designed to be used on patients who cannot self-report. The 1 scenario that had sound demonstrated expiratory grunting, which is an audible variable. Each scenario ranged from 45 to 60 seconds in length. The completed audiovisual program was formatted for portability so that on-site training and education could occur at both study locations.

Sociodemographic Collection Tool. A researcher-developed demographic tool assessed the variables of age, years of experience as a nurse, years of experience as a hospice nurse, highest level of education, employment status, primary shift worked, specialty certification possession, previous specialty End-of-Life Nursing Education Consortium (ELNEC) training, and a self-assessed rating of clinical skill in managing end-of-life dyspnea.

Nursing Assessment and Treatment Tool. This author-developed pretest/posttest was used to assess clinical assessment and treatment decisions for dyspnea in the noncommunicative patient prior to the participants receiving formalized education on the use of the RDOS. The pretest documented the nurse's assessment of patient comfort, presence and degree of dyspnea, differential diagnoses, need for nursing intervention, and initiation of nursing interventions as appropriate. Similarly, the posttest contained these same questions in addition to the documented RDOS score.

Respiratory Distress Observation Scale Tool. Campbell's[15,16] 8-item RDOS assessment tool evaluates the variables of heart rate (beats per minute), respiratory rate (breaths per minute), restlessness, and the presence/absence of paradoxical breathing pattern, accessory muscle use, grunting at end expiration, nasal flaring, and look of fear. Each variable is assigned a numerical value of 0, 1, or 2. The individual item scores are totaled, and sums range from 0 to 16, indicating the degree of dyspnea. The total score is used to determine the severity of dyspnea. A score of 0 to 2 signifies little or no respiratory distress, 3 signifies mild distress, scores 4 to 6 indicate moderate distress, and scores greater than 7 signify severe distress. Internal consistency for the RDOS[16] across studies has a Cronbach's α range of .64 to .86. Construct, convergent, and discriminant validity with the RDOS has been determined, but there is no established validity for use of the tool with the end-of-life hospice population.[16] The RDOS is not valid on neonates, young pediatric patients, or patients with cervical spine conditions resulting in quadriplegia or bulbar amyotrophic lateral sclerosis but is applicable to other adult patients who are unable to self-report dyspnea.[16]

Respiratory Distress Observation Scale Evaluation Tool. The author developed the RDOS Evaluation Tool using a 4-point Likert scale (1 = strongly agree, 2 = agree, 3 = do not agree, 4 = strongly do not agree) to evaluate the RDOS's ease of use and clinical application to the end-of-life setting. Prior to using, the researcher-developed tools were evaluated and approved by 2 content and tool development experts.

Theoretical Model

Humphreys'[21] Learning From Simulation Conceptual Model was used as a framework for this study. Specifically, the Learning From Simulations Model uses the concepts of preparedness, activation, and reflection to provide both individual and team-oriented learning based on the simulation of real-life experiences without causing harm to actual patients. The use of simulation in nursing education can strengthen clinical skills, support continued excellence, and refine practice.[1,21]


After obtaining participant consent and collecting demographic data, the RDOS educational program was presented to the nurses using an audiovisual format presented at their place of employment. The nurses were asked to view the 6 videotaped simulation scenarios pre- and post-RDOS education. With the initial viewing of the nonverbal scenarios, nurses completed the pretest using only experiential knowledge. After the RDOS education was presented, the nurses rewatched the video and applied the RDOS on the posttest to obtain a numerical score that indicated the patient's degree of dyspnea. Three levels of dyspnea (mild, moderate, severe) were presented in a nonsequential order utilizing different combinations of the RDOS assessment variables. Using a simulation video allowed for consistently delivered scenarios regardless of the setting.

The video was paused after each scenario to allow all participants time to complete the required assessment and treatment information. For each scenario, the nurses were asked to determine whether the patient appeared comfortable, identify the patient's level of dyspnea, and identify their treatment plan based on experiential practice. These data were collected as preintervention data. Once completed, the educational program on the development and use of the RDOS tool and its cutoff points was presented by the nurse researcher who is a palliative nurse educator. This educational session was immediately followed by a second viewing of the 6 dyspnea scenarios with the incorporation of the RDOS score for assessment and treatment; participants scored the scenarios in the same manner for the postintervention data. The presentation ended with the completion of the single-page RDOS Evaluation Tool, which was a 2-part evaluation tool: 1 portion specific to the ease of use of the RDOS tool and the other related to the educational offering/presentation.

Data Management and Analysis

Data were collected using the questionnaires, pretests/posttests, and evaluations. Nurse participant's identification was protected using a numbering system allowing for the identification of the agency only. Each nurse participant received a packet of documents that was labeled with their self-selected identifier, which allowed the researcher to correlate pretest/posttest scores and determine completeness of the study components. A total of 39 packets were analyzed to provide the results of this study using SPSS 24.0 (IBM Corp, Armonk, New York) and Intellectus Statistics[23] version 1.01 for data analysis and interpretation.