Although co-infection by multiple groups of pathogens is the norm rather than the exception in nature, most research on the effects of pathogens on their hosts has been largely based on a single or few pathogen species. Nevertheless, the health impact of co-occurring infections is evident, and it is important that scientists should consider pathogen communities rather than single relevant pathogen species when assessing the impact of multiple infections. In this work we illustrate the consequences of neglecting different pathogen taxa (viruses, protozoa, helminths, arthropods) in the explanatory power of a set of Partial Least Squares Regression (PLS-R) models used for exploring the impact of co-infections on the body condition of 57 adult feral cats; 71·5% cats were co-infected by ⩾3 groups of pathogens. The best two PLS-R models provided a first component based on the combination of helminths, protozoa and viruses, explaining 29·15% of body-condition variability. Statistical models, partially considering the pathogen community, lost between 24% and 94% of their explanatory power for explaining the cost of multiple infections. We believe that in the future, researchers assessing the impact of diseases on host life-history traits should take into account a broad representation of the pathogen community, especially during early assessment of the impact of diseases on host health.