Model for regional collaboration: Successful strategy to implement a pediatric early warning system in 36 pediatric oncology centers in Latin America

Abstract

Background: Pediatric early warning systems (PEWS) aid in the early identification of deterioration in hospitalized children with cancer; however, they are under-used in resource-limited settings. The authors use the knowledge-to-action framework to describe the implementation strategy for Proyecto Escala de Valoracion de Alerta Temprana (EVAT), a multicenter quality-improvement collaborative, to scale-up PEWS in pediatric oncology centers in Latin America.

Methods: Proyecto EVAT mentored participating centers through an adaptable implementation strategy to: (1) monitor clinical deterioration in children with cancer, (2) contextually adapt PEWS, (3) assess barriers to using PEWS, (4) pilot and implement PEWS, (5) monitor the use of PEWS, (6) evaluate outcomes, and (7) sustain PEWS. The implementation outcomes assessed included the quality of PEWS use, the time required for implementation, and global program impact.

Results: From April 2017 to October 2021, 36 diverse Proyecto EVAT hospitals from 13 countries in Latin America collectively managing more than 4100 annual new pediatric cancer diagnoses successfully implemented PEWS. The time to complete all program phases varied among centers, averaging 7 months (range, 3-13 months) from PEWS pilot to implementation completion. All centers ultimately implemented PEWS and maintained high-quality PEWS use for up to 18 months after implementation. Across the 36 centers, more than 11,100 clinicians were trained in PEWS, and more than 41,000 pediatric hospital admissions had PEWS used in their care.

Conclusions: Evidence-based interventions like PEWS can be successfully scaled-up regionally basis using a systematic approach that includes a collaborative network, an adaptable implementation strategy, and regional mentorship. Lessons learned can guide future programs to promote the widespread adoption of effective interventions and reduce global disparities in childhood cancer outcomes.

Lay summary: Pediatric early warning systems (PEWS) are clinical tools used to identify deterioration in hospitalized children with cancer; however, implementation challenges limit their use in resource-limited settings. Proyecto EVAT is a multicenter quality-improvement collaborative to implement PEWS in 36 pediatric oncology centers in Latin America. This is the first multicenter, multinational study reporting a successful implementation strategy (Proyecto EVAT) to regionally scale-up PEWS. The lessons learned from Proyecto EVAT can inform future programs to promote the adoption of clinical interventions to globally improve childhood cancer outcomes.

Keywords: Latin America; global health; implementation science; pediatric early warning system (PEWS); pediatric oncology; quality-improvement collaborative; resource-limited.

FIGURE 1

The Proyecto EVAT implementation strategy. Modified KTA action cycle describing the Proyecto EVAT PEWS implementation strategy. CDE indicates clinical deterioration event; KTA, knowledge‐to‐action; PEWS, pediatric warning systems; Proyecto EVAT, the Early Warning Assessment Scale Project; SWOT, Strengths, Weaknesses, Opportunities, Threats assessment.

FIGURE 2

The Proyecto EVAT PEWS implementation phases. This graph describes the time required for each collaborating center to move through the PEWS implementation phases, with the x‐axis representing time and the y‐axis indicating the 36 Proyecto EVAT centers. Centers that completed PEWS implementation after March 2020 (during the COVID pandemic) are marked with the blue COVID symbol. The phases described are as follows: phase 1 (red), time from the start of prospective tracking of clinical deterioration events to completing all necessary adaptation to implement PEWS; phase 2a (orange), time from the start of PEWS training to the start of the PEWS pilot; phase 2b (yellow), pilot start to implementation completion; and phase 3 (green), implementation completion to October 2021 (maximum, 18 months of postimplementation data collection). COVID indicates coronavirus disease; KTA, knowledge‐to‐action; PEWS, pediatric warning systems; Proyecto EVAT, the Early Warning Assessment Scale Project.

FIGURE 3

The quality of PEWS use over time. These graphs describe the results of monthly monitoring of three types of PEWS errors used to assess the quality of PEWS use at all centers: (A) omissions (documented vital signs without using PEWS), (B) PEWS score calculation errors, and (C) PEWS algorithm nonadherence (not following the PEWS algorithm correctly for high scores). At each center, data for PEWS errors were collected from the start of the PEWS pilot through October 2021 (or until 18 months after implementation). PEWS errors were calculated two or three times each week through a review of nursing vital signs and PEWS documentation for all hospitalized patients by the local PEWS implementation leaders and were aggregated monthly. In each graph, the x‐axis is the implementation month or the month since the start of the PEWS pilot at each center, and the y‐axis is the percentage errors measured that month. Dots represent data for each of 36 Proyecto EVAT centers; blue dots indicate the months before the COVID‐19 pandemic (before March 2020), and red dots indicate the months after the start of the COVID‐19 pandemic (after March 2020). The solid red line represents the median percentage of errors across the 36 centers during each implementation month. The black dotted line represents the goal threshold (GOAL) used to define high‐quality PEWS use (<15% errors in each error type). Centers with monthly error results above this threshold (>15% errors) more than 6 months after the start of the PEWS pilot are marked with their center number (Center ID Code). COVID‐19 indicates coronavirus disease 2019; PEWS, pediatric warning systems; Proyecto EVAT, the Early Warning Assessment Scale Project.

FIGURE 4

The frequency of red PEWS scores (≥5). Graphs describe the total number of documented red PEWS at Proyecto EVAT centers from the start of the PEWS pilot through October 2021 (or until 18 months after implementation), with 10–30 months of data per center. Red PEWS scores, defined as scores ≥5, were documented by local PEWS implementation leadership teams through a prospective quality‐improvement registry from the start of the PEWS pilot. Monthly numbers of in‐patient hospital days and clinical deterioration events, defined as an unplanned ICU transfer, the use of ICU interventions on the wards, or nonpalliative ward death, were also documented by all centers for the same period. One center was not able to share patient‐level data because of national regulations but collected these data locally for quality improvement (resulting in n = 35 centers for some measures, as labeled in the illustration). Each dot represents data from one center with: (A) the average number of red PEWS per month (red) and the rate of red PEWS normalized to 1000 in‐patient days at each center (blue), with the black line representing the median among all centers; and (B) a comparison of the total number of red PEWS (x‐axis) and CDEs (y‐axis) at each center during the same period, with a solid red line indicating regression and the dotted red lines indicating the 95% confidence interval. CDEs indicates clinical deterioration events; ICU, intensive care unit; PEWS, pediatric early warning systems; Proyecto EVAT, the Early Warning Assessment Scale Project.