Evidence in decision-making in the context of COVID-19 in Latin America.

ABSTRACT

Background: The pace of the COVID-19 pandemic poses an unprecedented challenge to the evidence-to-decision process. Latin American countries have responded to COVID-19 by introducing interventions to both mitigate the risk of infection and to treat cases. Understanding how evidence is used to inform government-level decision-making at a national scale is crucial for informing country and regional actors in ongoing response efforts. Objectives: This study was undertaken between February-May 2021 and aims to characterise the best available evidence (BAE) and assess the extent to which it was used to inform decision-making in 21 Latin American countries, in relation to pharmaceutical (PI) and non-pharmaceutical interventions (NPI) related to COVID-19, including the use of therapeutics (corticosteroids, hydroxychloroquine/chloroquine and ivermectin), facemask use in the community setting and the use of diagnostic tests as a requirement for international travel. Method: A three-phase methodology was used to; (i) characterise the BAE for each intervention using an umbrella review, (ii) identify government-level decisions for each intervention through a document review and (iii) assess the use of evidence to inform decisions using a novel adapted framework analysis. Findings: The BAE is characterized by 17 living and non-living systematic reviews as evolving, and particularly uncertain for NPIs. 107 country-level documents show variation in both content and timing of decision outcomes across intervention types, with the majority of decisions taken at a time of evidence uncertainty, with only 5 documents including BAE. Seven out of eight key indicators of an evidence-to-decision process were identified more frequently among PIs than either NPI of facemask use or testing prior to travel. Overall evidence use was reported more frequently among PIs than either NPI of facemask use or travel testing (92%, 28% and 29%, respectively). Interpretation: There are limitations in the extent to which evidence use in decision-making is reported across the Latin America region. Institutionalising this process and grounding it in existing and emerging methodologies can facilitate the rapid response in an emergency setting. Funding: No funding was sourced for this work.