(P-205) Use and Performance of Brighton Collaboration Case Definitions in Low-middle Income Settings

Background: Emerging pathogens in low- and middle-income countries (LMIC) have been the focus of recent global initiatives to develop new vaccines. Many challenging trial conditions in LMIC remain, however, including limited resources for safety monitoring. The Brighton Collaboration standard case definitions for adverse events following immunizations are widely recognized and recommended for use by multiple normative bodies; as of May 2019, 58 such case definitions have been published. Variability, however, may exist in the capacity of health facilities in LMIC to implement these case definitions.

Objectives: To review the use and performance of Brighton Collaboration case definitions in LMIC.

Methods: In July 2019, we searched for citation of the 58 case definition documents in Web of Science, Scopus, and Google Scholar. Eligible studies included case reports, clinical trials, observational studies, reviews, and protocols/guidance/commentaries reporting use or evaluation of case definitions in LMIC. An author screened the titles and abstracts of search output, reviewed full texts of potentially eligible articles, and extracted data. A sample of 20 full texts were independently reviewed and extracted by a second author, resolving discrepancies by consensus. We conducted descriptive analyses, and summarized performance of these case definitions in evaluation studies.

Results: We included 123 studies (11 case reports, 32 clinical trials, 67 observational studies, 8 reviews, and 5 protocols/guidance/commentaries). Thirty-five studies were conducted in India, 16 in Brazil, 14 in Colombia, while there were 9 or fewer studies each from other countries. Case definitions of intussusception (n = 37) and Guillain-Barré syndrome (GBS) (n = 34) were most commonly used, followed by fever (n = 16) and generalized convulsive seizure (n = 10). Twenty-nine studies assessed the performance of 22 case definitions (9 GBS, 7 intussusception, 5 viscerotropic disease, and 4 or less each for others). Studies were able to classify 47%–100% of GBS and 83%–100% of intussusception cases with levels 1–3 diagnostic certainty, and the stillbirth definition required modification to allow for the classification of facility-based deliveries in LMIC.

Conclusions: Our review found that the Brighton Collaboration case definitions have not been broadly used or assessed for performance in LMIC. We recommend involving more LMIC scientists in the case definition working and reference groups; disseminating free webinars, publicly available recordings, and case definition documents; formally assessing the barriers in LMIC; and developing tools to support implementation.