Factors For Abundance Of Floor Fleas In Households In The Western Of Yunnan Province

[Objectives] To know the distribution and proportion of floor fleas in households in the Western of Yunnan Province and to establish a epidemiological model to predict the abundance of floor fleas via analyzing the factors affecting the abundance of floor fleas. A new method for control and prevention of floor fleas abundance in households was explored to provide a theory basis for control flea-borne diseases. [Methods] 1.Study samples and objects: Eight hundred households(20 households each village) were randomly selected in 40 villages of 10 counties from flea-borne diseases in natural foci in the Western of Yunnan Province as study unit. Floor fleas and their potential factors were collected via observation and face to face interview. 2.Floor fleas Collection: Floor fleas were collected using water plate method. Ten water plates each household were placed for 3 continuous nights. The floor fleas each household were collected and saved in the bottle with 75% ethanol.3.Collecting potential factors affecting floor fleas abundance: Face to face interview was done to collect potential factors by self-made questionnaires(each household), including social economy, human intervention, ecological environment, geographical landscape and so on. 4.Data management and analysis: Data set were established in Epi Data 3.02 software and the file of Stata format was exported. Data were analyzed using R 3.0.1 software and statistical packages used including epicalc, pscl, lmtest and so on. Hurdle negative binomial regression model was applied to explore the factors affecting floor fleas abundance. The number of floor fleas in the households was regarded as dependent variables and all potential factors in the questionnaires were regarded as independent variables. Univariate analysis for each factor(variable) was conducted and the variables with P value less than 0.2 were chosen to enter the prototype multiple regression model. The final models were refined using backward elimination to reduce independent variable(α=0.05 as statistical significance level). Odds ratios(OR) and abundance ratios(AR) were calculated via the exponentiation for each coefficient of variable on logistic regression component and the negative binomial component. OR shows the probability of the infection of floor fleas in household. AR shows the probability on the number of floor fleas in those households with floor fleas. And 95% confidence interval(CI) of OR and AR were estimated. [Results] 1. Two thousand four hundred and six floor fleas were collected in 800 households and were divided into 5 species, 5 genera, 3 families, which included Pulex irritans(1750), Ctenocephalides felis felis(546), Xenopsylla cheopis(84), Monopsyllus anisus(21), and Letopsylla segnis(5). The number of Pulex irritants and Ctenocephalides felis felis accounted for 72.7% and 22.7%, respectively. The average of floor fleas was 60 in 40 villages. Of 800 households, floor fleas were captured in 346 households and the general infection rate of floor fleas was 43.3%.2.The negative binomial regression model for the number of floor fleas was established. The following factors were confirmed: The probability of capturing floor fleas and the number of captured floor fleas in households with Ethnic groups(Bai, Wa, Naxi, et al.) were 3.2 and 2.5 times higher than those with Han ethnic group. The probability of capturing floor fleas and the number of capturing floor fleas in households with cats were 1.6 and 2.6 times than those households without cats. The probability of capturing floor fleas in households with dogs was 2.0 times than those without dog. The probability of captured floor fleas in households with goats was 1.9 times higher than those without goat. The probability of captured floor fleas in households with cement floors decreased by 62% compared with those without cement floor. The number of captured floor fleas in households with the wooden floors increased 2.2 times compared with having not wooden floors. The number of captured floor fleas in households that surrounded by other houses was 2.7 times higher than those which don’t. The number of captured floor fleas in households that surrounded by roads was 1.4 times than those which having not road. The number of captured floor fleas in households with crops growing surroundings was 2.0 times compared with those houses without crops growing. The number of captured floor fleas in households with rice growing surroundings was about 11 times higher than those without rice growing. [Conclusions] 1. Two thousand four hundred and six floor fleas were collected in 800 households in Western of Yunnan Province and were divided into 5 species, 5 genera, 3 families. Pulex irritans was the dominant species, which acounted for 72.7%. The largest number of floor fleas was collected in Deqin and Lanping county and the flea index was 0.248 and 0.243, respectively. The transmitting risks of flea-borne diseases still existed. Control measures should be conducted to control the number of floor fleas and rodents to reducing the risk of flea-borne diseases. And surveillance on infectious diseases should be strengthened to find the occurrence of diseases earlier.2.The households with Ethnic group(Bai, Wa, Naxi, et al.)with cats, dogs, sheep, and paths surroundings are the factors that increase the probability of capturing floor fleas. The households with Ethnic group(Bai, Wa, Naxi, et al.), cats, dogs, the wooden floors, houses, crops and rice surroundings as factors that lead to increase the number of floor fleas. The households with above characteristics should be concerned and measures for control fleas should be done. In addition, some measures(such as choosing cement floor) contributing to reduce the number of floor fleas in the houses should be implemented to decrease the occurrence of fleas-borne diseases.