This research was conducted in partnership between Goldfarb School of Nursing and Missouri Baptist Medical Center. Academic nurses with terminal degrees serve as Research/Evidence-based Practice Mentors and work with hospital nursing staff. The co-primary investigator was a mastered prepared nurse from the unit where this study was performed.
The theories of caring and vigilance were used to guide the study. Joan C. Tronto’s model of Caring proposes four phases of caring which are necessary to demonstrate caring. These phases of caring involve cognitive, emotional, and action strategies: (1) caring about; (2) taking care of; (3) caregiving; and (4) care receiving. Vigilance is defined as giving attention to and identification of clinically significant observations. The nurse must care enough about the patient to identify clinical changes, be ready to act when clinical changes are observed, realize that these clinical changes also affect family members, and receive care because they can potentially be traumatized through adverse patient situations.
Six champion nurses were identified by the unit leaders and agreed to participant in the research study. They completed a three-hour training which included an educational packet, background and purpose of the study, explanation of the research procedures, and simulation training. We created two scenarios for the training. The Champion nurses task was to assist the other nurses with using the RDOS and consenting for the study.
All registered nurses on the medical/oncology unit were asked to participate in this study. Since a respiratory assessment of all patients are an essential requirement for registered nurses, we felt that the time it took to complete study materials during their shift were minimal. The consent process for each participating RN included a demographic form and a confidence scale pre-study and post-study. Each nurse was asked to complete the RDOS on five different patients with no more than three RDOS' on the same patients; the RDOS could only be completed once per day on a patient. A locked mailbox receptacle was placed on the unit in the conference room for the confidential receipt of study materials. The Primary investigator had the only key to the mailbox. Study material were retrieved biweekly through the duration of the study. The study began in September 2017 and data collection was completed in June 2018.
The results of champion nurse training showed Average RDOS Total scores decreased from Scenario 1 to Scenario 2 (10.17 vs 6.71, Z = -2.023, p = 0.043). This was expected because Scenario 2 was intended to have a less severe case of respiratory distress. All of the champion nurses rated Scenario 1 as Severe; however, they did not agree on how to rate Scenario 2. Mild/Moderate respiratory distress responses were 33% and severe responses were 66.7%. However, in Scenario 2, all the scores were > 4 and both groups performed an intervention.
A total of 27 nurses completed the consent process. The RDOS tool has 8 individual items which were summed to form the RDOS Total score (range: 0 – 16). The parameters are heart rate, respiratory rate, accessory muscle use, paradoxical breathing pattern, restlessness, grunting at end-expiration, nasal flaring, and a fearful facial display. Clinical Observation study protocol included an Initial Assessment and a Follow-up Assessment for RDOS scores > 4. In aggregate, 46 clinical Observations were performed on 35 individual patients for a total of 66 unique encounters.
Nurses demonstrated the ability to assess respiratory distress in patients unable to self-report. Among 46 encounters (69.7%) were Initial Assessments and 20 (30.3%) were Follow-up Assessments. Results established that Initial Assessments of respiratory distress (RDOS score 4–16) were associated with the nurse providing an intervention 96.2% of the time. The nurses documented that the other 3.8% of the times an invention had been performed < 2 hours before the RDOS was performed.
Non-parametric Wilcoxon signed rank tests were used to determine ordinal changes in Confidence Scale. Items 2-5 (Item 2: perform skill without hesitation; Item 3: performance shows confidence to observer; Item 4: Feel sure of myself performing skill; and Item 5: Satisfied with my performance) improved statistically from Baseline to Follow-up (p < 0.05). Using McNemar tests, the proportion of nurses moving from non-confident to confident response increased across Item 2-5 which were statistically significant during both Evaluation Periods.
A major limitation for the Confidence scale was a large proportion of participants missed the first question because it was position higher on the form and the remaining four questions were grouped together (34.8% in Pre-Period and 56.5% in Post-Period. If we use this scale again, we recommend a revision so all the questions appear in the same table.
In conclusion, RDOS cut-point identification (> 4) assisted the nurses to respond to respiratory distress symptoms with an appropriate intervention. The positive patient outcomes were noted by a decrease RDOS reassessment scores after administration of an intervention. This study demonstrated that the RDOS is a reliable tool for assessing respiratory distress in patients unable to elicit a report on regular hospital unit.