Zero or More: Methodological Challenges of Counting and Estimating Deaths Related to Antibiotic-resistant Infections

Estimates of the number of deaths from antimicrobial-resistant (AMR) infections are important data for clinicians and public health officials advocating for resources to prevent and treat these infections. The aims of this article are to describe the various approaches to calculating deaths from AMR infections, to compare the tally of deaths by each approach, and to explain how to interpret the results. Currently, none of the 3 methods employed by vital statistics systems to count deaths from specific causes (underlying cause of deaths, multiple causes of death, and avoidable deaths) count deaths from AMR infections. These deaths can be estimated by 4 approaches: case-fatality rate, infection-related mortality, and excess mortality using controls with antibiotic-susceptible infections or controls without antibiotic-resistant infections. When encountering discrepant estimates of AMR-related deaths, it is important to consider which method was used and whether it was the right method to answer the question being asked.