Host competence—the ability to acquire, harbor, and transmit infections—drives pathogen spread and persistence in multi-host communities. Evaluating species-specific competence is critical for predicting transmission, particularly for generalist fungal pathogens like Batrachochytrium dendrobatidis (Bd). Despite its central role in disease dynamics, we lack an epidemiologically grounded competence metric that rigorously accounts for how infection intensity affects a host's competence. This knowledge gap limits our ability to compare mechanisms across species and assess their roles in pathogen persistence. To address these challenges, we developed a novel, load-dependent competence metric using host-pathogen Integral Projection Models (IPMs) that integrates variation in susceptibility, within-host pathogen growth, and pathogen shedding dynamics.

We applied this metric to lab-based challenge experiments with three common North American amphibians (Notophthalmus viridescens, Rana clamitans, and Rana catesbeianus) that persist endemically with Bd. Using dose-response assays and repeated pathogen shedding measurements across species, we asked: i) is there a consistent, non-linear relationship between infection intensity and pathogen shedding across species? and ii) which load-based traits best predict host competence? We quantified four of five components of host competence—susceptibility, pathogen growth, pathogen survival, and load-dependent shedding—and used these to parameterize species-specific IPMs, integrating competence into a single relative metric across species.

We found that Bd shedding increased non-linearly with infection intensity, contradicting the standard assumption that Bd shedding is linearly related to infection intensity. N. viridescens and R. catesbeianus were the most competent hosts but through distinct pathways: high susceptibility in N. viridescens and elevated shedding rates in R. catesbeianus. In contrast, density dependent reductions in pathogen growth and shedding limited R. clamitans competence. Thus, species-level competence is not determined by a single trait, but emerges from interactions among multiple load-based processes.

Our results demonstrate that variation in competence emerges from distinct, species-specific processes across multiple dimensions of competence. By linking individual infection dynamics to population-level transmission potential, our integrative framework provides a more mechanistic approach to predicting host contributions to community-level pathogen persistence

We conducted a controlled dose-response experiment to assess Bd susceptibility and shedding across species (details in Appendix S1). After confirming Bd-free status via qPCR following heat treatment, adult R. clamitans, N. viridescens, and R. catesbeianus were exposed to one of four Bd zoospore doses ($1\times10^4$ to $1\times10^7$) or a control. All exposures occurred at $21^\circ C$ in individual containers under standardized housing and feeding conditions.

We exposed individuals to a Tennessee Bd isolate (FMB003, GPL-1), with zoospore concentrations enumerated using a hemocytometer. Individuals were exposed for 24 hours, then monitored for infection and zoospore shedding repeatedly. Skin swabs to measure Bd infection load were taken every 6 days, followed by 24-hour water exposure in separate containers to quantify shedding. Experiment duration varied among species (56-79 days post-exposure). Swab and shedding samples were processed for Bd detection via qPCR using established protocols [@Boyle2004; @Blooi2013]. DNA extractions and Bd quantification methods are detailed in Appendix S1.