Predictors for presence and abundance of small mammals in households of villages endemic for commensal rodent plague in Yunnan Province, China.

BACKGROUND: Ninety-one rodent plague epidemics have occurred in Lianghe county, Yunnan Province, China, between 1990 and 2006. This study aimed to identify predictors for the presence and abundance of small mammals in households of villages endemic for rodent plague in Lianghe county. RESULTS: Rattus flavipectus and Suncus murinus were the two species captured in 110 households. Keeping cats decreased the number of captures of R. flavipectus by one to two thirds and the chance of reported small mammal sightings in houses by 60 to 80%. Food availability was associated with fewer captures. Keeping food in sacks decreased the small mammal captures, especially of S. murinus 4- to 8-fold. Vegetables grown around house and maize grown in the village reduced the captures of S. murinus and R. flavipectus by 73 and 45%, respectively. An outside toilet and garbage piles near the house each reduced R. flavipectus captures by 39 and 37%, respectively, while raising dogs and the presence of communal latrines in the village increased R. flavipectus captures by 76 and 110% but were without detectable effect on small mammal sightings. Location adjacent to other houses increased captures 2-fold but reduced the chance of sightings to about half. In addition, raising ducks increased the chance of sighting small mammals 2.7-fold. Even after adjusting for these variables, households of the Dai had higher captures than those of the Han and other ethnic groups. CONCLUSION: Both species captures were reduced by availability of species-specific foods in the environment, whereas other predictors for capture of the two species differed. Other than the beneficial effect of cats, there were also discrepancies between the effects on small mammal captures and those on sightings. These differences should be considered during the implementation and interpretation of small mammal surveys.