A clarification of transmission terms in host-microparasite models: numbers, densities and areas.

Transmission is the driving force in the dynamics of any infectious disease. A crucial element in understanding disease dynamics, therefore, is the 'transmission term' describing the rate at which susceptible hosts are 'converted' into infected hosts by their contact with infectious material. Recently, the conventional form of this term has been increasingly questioned, and new terminologies and conventions have been proposed. Here, therefore, we review the derivation of transmission terms, explain the basis of confusion, and provide clarification. The root of the problem has been a failure to include explicit consideration of the area occupied by a host population, alongside both the number of infectious hosts and their density within the population. We argue that the terms 'density-dependent transmission' and 'frequency-dependent transmission' remain valid and useful (though a 'fuller' transmission term for the former is identified), but that the terms 'mass action', 'true mass action' and 'pseudo mass action' are all unhelpful and should be dropped. Also, contrary to what has often been assumed, the distinction between homogeneous and heterogeneous mixing in a host population is orthogonal to the distinction between density- and frequency-dependent transmission modes.

Longitudinal monitoring of the dynamics of infections due to Bartonella species in UK woodland rodents.

Blood samples were repeatedly collected from 12 sympatric woodland rodents over a 12-month period and DNA extracts from each were incorporated into a bartonella-specific PCR targeting a fragment of the 16S/23S rRNA intergenic spacer region (ISR). The composition of each amplicon was analysed using restriction enzyme analysis (REA) and base sequence comparison. Bartonella DNA was detected in 70 of 109 samples. Eleven samples contained DNA derived from more than one strain. Sequence analysis of 62 samples found 12 sequence variants (ISR genotypes) that were provisionally assigned to 5 different species, 2 of which were newly recognized. Up to five different species were detected in each animal. On about two-thirds of occasions, a species detected I month was not there the next, but never was a genotype superseded by another of the same species. However, a genotype could be re-encountered months later in the same animal, even if interim samples contained other genotypes. Our results suggest that although most animals are bacteraemic most of the time, specific infections are often superseded and that a complex and dynamic epidemiology of bartonella bacteraemias exists in woodland rodents.

A longitudinal study of an endemic disease in its wildlife reservoir: cowpox and wild rodents.

Cowpox is an orthopoxvirus infection endemic in European wild rodents, but with a wide host range including human beings. In this longitudinal study we examined cowpox in two wild rodent species, bank voles Clethrionomys glareolus and wood mice Apodemus sylvaticus, to investigate the dynamics of a virus in its wild reservoir host. Trapping was carried out at 4-weekly intervals over 3 years and each animal caught was uniquely identified, blood sampled and tested for antibodies to cowpox. Antibody prevalence was higher in bank voles than in wood mice and seroconversion varied seasonally, with peaks in autumn. Infection was most common in males of both species but no clear association with age was demonstrated. This study provides a model for studying other zoonotic infections that derive from wild mammals since other approaches, such as one-off samples, will fail to detect the variation in infection and thus, risk to human health, demonstrated here.

Transmission dynamics of a zoonotic pathogen within and between wildlife host species.

The transmission dynamics of the cowpox virus infection have been quantified in two mixed populations of bank voles (Clethrionomys glareolus) and wood mice (Apodemus sylvaticus), through analyses of detailed time-series of the numbers of susceptible, infectious and newly infected individuals. The cowpox virus is a zoonosis which circulates in these rodent hosts and has been shown to have an adverse effect on reproductive output. The transmission dynamics within species is best described as frequency dependent rather than density dependent, contrary to the 'mass action' assumption of most previous studies, both theoretical and empirical. Estimation of a transmission coefficient for each species in each population also allows annual and seasonal variations in transmission dynamics to be investigated through an analysis of regression residuals. Transmission between host species is found to be negligible despite their close cohabitation. The consequences of this for the combining ability of hosts as zoonotic reservoirs, and for apparent competition between hosts, are discussed.

Cowpox: reservoir hosts and geographic range.

It is generally accepted that the reservoir hosts of cowpox virus are wild rodents, although direct evidence for this is lacking for much of the virus's geographic range. Here, through a combination of serology and PCR, we demonstrate conclusively that the main hosts in Great Britain are bank voles, wood mice and short-tailed field voles. However, we also suggest that wood mice may not be able to maintain infection alone, explaining the absence of cowpox from Ireland where voles are generally not found. Infection in wild rodents varies seasonally, and this variation probably underlies the marked seasonal incidence of infection in accidental hosts such as humans and domestic cats.