Reacción a "Reactions: one in eight deaths in 2019 were linked to bacterial infections"

Bacterial infections are one of the leading causes of death worldwide, in many cases due to increasing antibiotic resistance, which means that we have fewer and fewer therapeutic options and that the number of deaths and complications increases every year. They therefore pose a threat to global health that requires in-depth research to provide us with effective tools for their control. The GBD 2019 Antimicrobial Resistance Collaborators study provides a wealth of data that allows us to have an overview of the problem, although the data collection is quite uneven both by country and by the sources used. This study describes S. aureus, E. coli, S. pneumoniae, K. pneumoniae and A. baumannii as the most frequent cause of death from infections in the world, which is nothing new as they have been included for years in the WHO priority lists - as well as the most worrying infections they cause such as lower respiratory tract infections, sepsis and intra-abdominal infections. 

However, the handling of this large amount of data means that important aspects such as risk factors and the effect of infection by antibiotic-resistant strains on patient deaths are not explored in detail, leading to general conclusions that may not be applicable in all cases. Furthermore, it seems that they do not find any effect of gender on the infections studied, which is contradictory since there is evidence of increased mortality in some processes (such as sepsis due to S. aureus), which may be an example of the biases in the inclusion of women in clinical studies and the use of undifferentiated databases. In fact, the same article already highlights these limitations in the analysis: biases in data collection, possibility of overestimation of some pathogens, high percentage of deaths considered to be due to sepsis in patients with serious non-infectious diseases that may have been the cause of death even without infection being present, etc.  

In conclusion, conclusive results would require a very carefully selected sample that is comparable across all sources used. The situations in countries with economic, cultural and social differences that influence the prevalence and evolution of infections and how health processes are addressed should not be generalised.