Implementation of a Standardized Data Collection Tool for Febrile Episodes in Childhood Cancer Patients in Central America and the Caribbean

Abstract

Abstract Fever is a common sign of infection and is associated with morbidity and treatment delay in childhood cancer patients. A multinational group of physicians identified the need to develop standardized measures to assess and intervene in the quality of care delivery as measured by adherence to the fever management guidelines published by this group in 2020. Our objective was to create an implement a febrile neutropenia registry that included data for quality assessment. Six sites in Central America and the Caribbean implemented a prototype of a Fever Registry for standardized data collection. Variables collected in the registry included demographic information and quality measures like the performance of selected laboratory tests (neutrophil count and blood culture) and time to antibiotic administration. Sites also collected clinical outcomes, including infection diagnosis, microbiologic diagnosis, critical care support requirements, length of hospital stay, and mortality. The initial implementation cohort included one dedicated pediatric oncology center and five oncology services situated within general pediatric hospitals. From September 9, 2021 to December 5, 2022, the six sites contributed 188 fever episodes from 148 unique patients. The maximum number of fever episodes entered for a single patient was 6 episodes. The distribution of cases entered in the database by site ranged between 5 to 54 cases (average patients enrolled per site=25). 84.9% of iCRFs(Case Report forms) had complete data. The most common incomplete data fields were: Initial antimicrobial administered (9/175=5.1%), nutritional status (6/181=3.3%), and was urinalysis performed at onset of fever (6/181=3.3%). Implementation strategies were developed incorporating responses to a survey based on the Consolidated Framework for Implementation Research. Infectious diseases specialists, who were the site leads, directed the implementation efforts at each site. The coordinating team encouraged sites to incorporate hematologists/oncologists and data managers as part of the team. However, team composition and internal meeting frequency were at the discretion of the site leads. Every site elected to use a paper data collection tool to facilitate centralized data entry, which was conducted by a data manager in all but one site. Staff at local sites received two initial trainings and a retraining by St. Jude staff to identify eligible patients and complete data fields appropriately. Group meetings with all site leads were conducted monthly to review progress and challenges and for quality assurance. The challenges most cited were personnel limitations for data collection and data entry. Surges exacerbated these challenges in caseloads related to respiratory illnesses, including SARS-CoV-2, influenza, and RSV. Conclusion: This group has created and implemented a tool to facilitate standardized data collection for a common clinical problem. Implementation challenges highlight the role of human resources and simplifying data collection to minimize data entry time. Current efforts include developing interim reports so clinicians can use data to improve the quality of care delivery in real-time. The group is also revising data fields to remove data points that are often unavailable or unnecessary to measure care quality. Project implementation in these six sites exposed opportunities to streamline the data collection system and optimize the implementation process.