Invited commentary: assessing mechanistic interaction between coinfecting pathogens for diarrheal disease

Area of Science:

• Epidemiology
• Infectious Disease Research
• Biostatistics

Background:

• Mechanistic interaction, defined by Rothman's sufficient-cause framework, is crucial for understanding coinfecting pathogen dynamics.
• Assessing mechanistic interaction often relies on tests for additive interaction, which can require strong monotonicity assumptions.
• Recent methods offer alternatives to monotonicity assumptions for detecting mechanistic interaction.

Purpose of the Study:

• To evaluate evidence for mechanistic interaction between coinfecting diarrheal pathogens using existing interaction estimates.
• To determine if additive interaction magnitudes reported in previous studies support mechanistic interaction without monotonicity assumptions.

Main Methods:

• Utilized interaction estimates from Bhavnani et al. (2012) concerning diarrheal disease coinfections.
• Applied the definition of mechanistic interaction: an outcome occurs with both exposures but not with either exposure alone.
• Assessed the sufficiency of reported additive interaction for concluding mechanistic interaction, considering monotonicity assumptions.

Main Results:

• The additive interaction estimates reported by Bhavnani et al. were of sufficient magnitude.
• Strong evidence for mechanistic interaction was found between rotavirus and Giardia.
• Strong evidence for mechanistic interaction was also found between rotavirus and Escherichia coli/Shigellae.

Conclusions:

• The study provides strong evidence for mechanistic interaction between specific coinfecting diarrheal pathogens.
• The findings support the presence of synergism between rotavirus and Giardia, and rotavirus and E. coli/Shigellae.
• Mechanistic interaction can be inferred from substantial additive interaction, even without monotonicity assumptions.