Should we expect population thresholds for wildlife disease?

Host population thresholds for the invasion or persistence of infectious disease are core concepts of disease ecology and underlie disease control policies based on culling and vaccination. However, empirical evidence for these thresholds in wildlife populations has been sparse, although recent studies have begun to address this gap. Here, we review the theoretical bases and empirical evidence for disease thresholds in wildlife. We see that, by their nature, these thresholds are rarely abrupt and always difficult to measure, and important facets of wildlife ecology are neglected by current theories. Empirical studies seeking to identify disease thresholds in wildlife encounter recurring obstacles of small sample sizes and confounding factors. Disease control policies based solely on threshold targets are rarely warranted, but management to reduce abundance of susceptible hosts can be effective.

Introduction

Ideas about threshold levels of host abundance for invasion or persistence of infectious diseases are central to the theory and practice of disease ecology 1, 2, 3, but have their roots in human epidemiology. The notion of a threshold population for invasion (N_T) (see Glossary) is a founding principle of epidemiological theory 4, 5, 6, and the critical community size (ccs) required for disease persistence dates back to Bartlett's seminal analyses of measles data [7]. Evidence of population thresholds in wildlife disease systems has been described as ‘rare’ [8] and ‘weak’ [9], yet these concepts underpin all efforts to eradicate wildlife diseases by reducing the numbers of susceptible hosts through controversial methods such as culling, sterilization, or vaccination (e.g. 10, 11, 12). Recent empirical studies have sought to identify invasion and persistence thresholds in wildlife with mixed success 8, 9, 12, 13, 14, 15. Here, we consider these findings in the context of theoretical models of disease spread, which reveal that abrupt population thresholds are not expected for many disease-host systems. Moreover, even when thresholds are expected, demographic stochasticity makes them difficult to measure under field conditions. We discuss how conventional theories underlying population thresholds neglect many factors relevant to natural populations such as seasonal births or compensatory reproduction, raising doubts about the general applicability of standard threshold concepts in wildlife disease systems. These findings call into question the wisdom of centering control policies on threshold targets and open important avenues for future research.