Abstract
<jats:p>Understanding why parasite infection varies among hosts is central to explaining parasite distributions in multihost systems. Infection patterns may reflect characteristics of individual hosts, the broader ecological context, or their interaction. To distinguish these alternatives, we used <i>Mazamastrongylus dagestanica</i> (Nematoda, Trichostrongylidae), an abomasal parasite primarily associated with moose but increasingly reported from other cervids. Drawing on 121 wild cervids representing five host species sampled across six regions between 1984 and 2022, we formalized our causal assumptions with a directed acyclic graph and estimated the effects of host characteristics and ecological proxies on parasite occurrence and infection intensity using Bayesian regression models. Host species was the strongest predictor of both occurrence probability and infection intensity: moose consistently ranked highest in both measures, whereas other cervids showed lower and more uncertain estimates. Host age modified infection patterns in a species-specific manner: both outcomes increased with age in moose, whereas red deer and common fallow deer showed declining occurrence probabilities with age. Sampling region explained some additional variation in occurrence probability, while seasonal effects were weak and host sex showed little evidence of a consistent influence. Our results indicate that, within this multihost system, differences among host species predominate over seasonal and regional variation in explaining infection patterns, demonstrating that host identity is the primary determinant of how this parasite is distributed across its hosts.</jats:p>