The impact of pediatric early warning score and rapid response algorithm training and implementation on interprofessional collaboration in a resource-limited setting

Introduction Improved teamwork and communication have been associated with improved quality of care. Early Warning Scores (EWS) and rapid response algorithms are a way of identifying deteriorating patients and providing a common framework for communication and response between physicians and nurses. The impact of EWS implementation on interprofessional collaboration (IPC) has been minimally studied, especially in resource-limited settings. Methods The study took place in the Pediatric Department of the main academic referral hospital in Rwanda between April 2019 and January 2020. Pediatric nurses and residents were trained on the use of the Pediatric Warning Score for Resource-Limited Settings (PEWS-RL) and a rapid response algorithm. Training included vital sign collection, PEWS-RL calculation, IPC and rapid response algorithm implementation. Prior to training, participants completed surveys on IPC with Likert scale responses (from “strongly disagree” to “strongly agree”). Follow-up surveys were then administered nine months later and also included an open-response question on the impact of the PEWS-RL implementation on IPC. Results Sixty-five (96%) nurses were trained and completed the pre-survey and thirty-seven (54%) of the trained nurses completed the post-survey. Twenty-two (59%) pediatric residents were trained in the workshop and completed the pre-survey and twenty-four physicians (4 pediatricians (40%) and 20 pediatric residents (53%)) completed the post-implementation survey. There was a statistically significant increase in the percent of nurses indicating strong agreement across all domains of communication and collaboration from the pre- to the post-survey. Although the percent of physicians indicating strong agreement increased in the post-survey for all items, only the “share information” item was statistically significant. Conclusion Training and implementation of a PEWS-RL and a rapid response algorithm at a tertiary hospital in Rwanda resulted in significant improvement of nurse and physician ratings of IPC nine months later.

Introduction Improved teamwork and communication have been associated with improved quality of care. Early Warning Scores (EWS) and Rapid Response Teams are a way of identifying deteriorating patients and providing a common framework for communication and response between physicians and nurses. The impact of EWS implementation on interprofessional collaboration (IPC) has been minimally studied, especially in resource-limited settings.

Methods
The study took place in the Pediatric Department of the main academic referral hospital in Rwanda between April 2019 and January 2020. Pediatric nurses and residents were trained on the use of the Pediatric Warning Score for Resource-Limited Settings (PEWS-RL) and a rapid response algorithm. Training included vital sign collection, PEWS-RL calculation, IPC and a rapid response algorithm implementation. Prior to training, participants completed surveys on IPC with Likert scale responses (from "strongly disagree" to "strongly agree"). Follow-up surveys were then administered 9 months later and also included an open-response question on the impact of the PEWS-RL implementation on IPC. Results Sixty-five (96%) nurses were trained and completed the pre-survey and thirty-seven (54%) of trained nurses completed the post-survey. Twenty-two (59%) pediatric residents were trained in the workshop and completed the pre-survey and twenty-four physicians (4 pediatricians (40%) and 20 pediatric residents (53%)) completed the post-implementation survey. There was a statistically significant increase in the percent of nurses indicating strong agreement across all domains of communication and collaboration from the pre-to the post-survey. Although the percent of physicians indicating strong agreement increased in the post-survey for all items, only the "share information" item was statistically significant. Conclusion Training and implementation of a PEWS-RL and a rapid response algorithm at a tertiary hospital in Rwanda resulted in significant improvement of nurse and physician ratings of IPC 9 months later. physicians and nurses. The impact of EWS implementation on interprofessional collaboration 28 (IPC) has been minimally studied, especially in resource-limited settings. 29 30 Methods 31 The study took place in the Pediatric Department of the main academic referral hospital in 32 Rwanda between April 2019 and January 2020. Pediatric nurses and residents were trained on 33 the use of the Pediatric Warning Score for Resource-Limited Settings (PEWS-RL) and a rapid 34 response algorithm. Training included vital sign collection, PEWS-RL calculation, IPC and a 35 rapid response algorithm implementation. Prior to training, participants completed surveys on 36 IPC with Likert scale responses (from "strongly disagree" to "strongly agree"). Teams of nurses rotated through five simulation scenarios to practice using the PEWS-RL and 146 rapid response algorithm. After calculating the PEWS-RL score, nurses practiced communicating 147 their concern and then escalating their concerns utilizing the rapid response algorithm. In simulations, they faced a resident stating they were too busy in the emergency department to 149 respond at that time, a resident who refused to come because he was in a lecture who then did not 150 show up within the expected time for response, a situation in which the resident assigned to the 151 ward was post-call and the covering resident did not respond to their calls, and a case in which 152 the covering resident was not responding and they had to escalate their concerns to the PICU 153 resident. (Fig 1)  This time, instead of the facilitator playing the role of the residents, the residents themselves 159 gave a scripted response to the nurses' calls. These responses included residents who said they 160 could not respond due to other emergencies or teaching conference, residents who were post-call 161 or not responding, forcing escalation to second and third call providers on the algorithm, and a 162 resident who responded that they had assessed the patient that morning and did not think they 163 needed to come back to reassess the patient. Nurses worked through the steps of conveying their 164 concern for serious illness, reiterating the elevated PEWS-RL score, informing the resident of the 165 requirement of bedside assessment in algorithm, and offering to call next person in algorithm if 166 the resident was unable to come. Once the urgency of the evaluation was adequately conveyed, 167 the simulation progressed to the resident responding to the bedside. On arrival of the resident "in 168 person" to the bedside of the simulated case, they performed a patient assessment and simulated 169 initial clinical interventions such as dextrose or fluid resuscitation, medication administration, 170 respiratory support or further laboratory or imaging studies. Simulation scenarios were each 15 171 minutes followed by 5 minutes of debriefing. 172 Likert scale survey items were dichotomized as "strongly agree" versus all other responses, 220 based on the skewed distribution of many items and the desire to quantify clinically meaningful 221 outcomes. We calculated the proportion of "strongly agree" responses for each item, at both the 222 pre-and post-assessments. We also calculated the proportion difference between the pre-and 223 post-assessments, with 95% confidence intervals calculated with a nonparametric bootstrap 224 estimation (with 100 repetitions). Analyses were conducted separately by responder type (i.e., 225 physicians and nurses). 226

227
An inductive content analysis strategy was used to analyze our qualitative responses. Open 228 response answers were recorded directly into a REDCap database. Responses were reviewed to 229 identify positive impacts or barriers to change and each identified phrase was assigned a 230 preliminary label from which a coding scheme was then developed. Data were coded according 231 to the coding scheme using Taguette software (https://app.taguette.org/) with initial coding of 232 English responses by two English-speaking coders and initial coding of French responses by two 233 bilingual English/French-speaking coders who assigned English codes to the French responses. 234 All coders then subsequently reviewed the responses and codes and any areas of differential 235 coding were resolved. Codes were categorized and then categories of codes were grouped into 236 overall themes. Finally, each comment was again reviewed by all coders to confirm that 237 consensus was reached about the comment belonging to the assigned category and theme. 238 239

241
Participants 242 Sixty-five (96%) of the nurses were trained in the workshops and completed the pre-243 implementation survey. Thirty-seven (54%) of the nurses who underwent training completed the 244 9-month follow-up survey. Twenty-two (59%) pediatric residents were trained in the workshop 245 and completed the physician pre-implementation survey. No attending pediatricians completed 246 the training workshop, though a brief training was conducted during a staff meeting. Twenty-247 four physicians completed the physician 9-month follow-up survey: 4 (40%) attending 248 pediatricians and 20 (53%) pediatric residents. 249

250
In quantitative analyses, physicians' report of "strong agreement" with survey items in the pre-251 period ranged from 4.6% ("Nurses are accurate in their assessment of patient status") to 72.7% 252 ("When a nurse calls me regarding a patient they are worried about I always go and assess that 253 patient"). The proportion reporting "strong agreement" increased in the post-period for all 254 items, but a statistically significant increase was only detected for the item "Physicians share all 255 information with the nurses when making decisions on patient care" (Table 1) with an increase 256 of 29.3%. 257 Among nurses, strong agreement with survey items in the pre-period ranged from 14.0% 261 ("Decision-making responsibilities for patients are shared among nurses and physicians") to 262 43.1% ("On my ward physicians and nurses work together as a team to monitor and assess 263 patients"). The proportion reporting "strong agreement" significantly increased in the post-264 period for all items, with proportion increases ranging from 27% ("When I feel there is an error 265 made by the physician (verbal or written order) I feel comfortable notifying that physician when 266 error is identified") to 37.5% ("Decision-making responsibilities for patients are shared among 267 nurses and physicians") ( Table 2). 268 Qualitative results

273
In our qualitative analysis, nurses and physicians commented on positive impacts of PEWS-RL 274 training and implementation in three major categories: 1) Teamwork, 2) Care Improvements, and 275 3) Respect and empowerment. They identified barriers to improvement in three major categories 276 1) Not following PEWS-RL/RRT protocol, 2) Resource limitations, and 3) Need for more 277 PEWS-RL training. 278

279
A large number of both nurses and physicians commented on improved collaboration and 280 communication leading to shared decision-making and better teamwork. Table 3 demonstrates 281 the positive impact categories and themes within those categories along with selected 282 representative comments. Many commented that PEWS-RL was important to care and resulted 283 in the earlier identification of sick patients, faster response and interventions to signs of 284 worsening illness, the ability to prevent deterioration, and the belief that PEWS-RL reduced 285 morbidity and mortality. One nurse commented on improved knowledge of vital signs. Several 286 nurses expressed that they were more respected by physicians and felt more confident. One 287 physician noticed that nurses were more proactive following PEWS-RL implementation. 288 289 Barriers to improvement 293 Some respondents felt that no significant change had taken place and several identified barriers 294 to improvement. The categories of barriers and themes expressed within these categories along 295 with representative quotes are displayed in Table 4. A few nurses expressed the opinion that 296 physician behavior had not changed significantly in response to the PEWS-RL implementation 297 or that physicians were not following the PEWS-RL/RRT protocol. One physician commented 298 that nurses were not calculating the PEWS score. Both nurses and a physicians expressed the 299 need for more training on PEWS-RL both for reinforcing skills as well as for training newly 300 rotating physicians. 301 The new doctors must be informed (of) the PEWS process in the first days of orientation"nurse "More training for nurses and residents as they are primarily (the) one(s) who are with patients everyday" -physician