Leptospirosis is a zoonotic disease caused by bacteria of the genus Leptospira and most often acquired through contact with environments contaminated with leptospires shed in the urine of infected mammals. In urban environment, rodents are well-known as the main carriers of this bacteria, however there were no intensive study on the population structure of these animals, and how it associated with this disease. Hence, we use a case study from an outbreak in a residential area in Selangor, Malaysia, to investigate how community structure of small mammals, associated with the prevalence of Leptospira. One hundred cage traps were placed randomly in and around these houses in five phases with two months interval for a year. Community structures (species, sex, and age) were assigned for each individual, prior to screening for pathogenic Leptospira, using a partial lipL32 gene from the kidney samples. 185 small mammals from four species were captured, Rattus norvegicus (74.5%, N=138), R. rattus (20%, N=37), Tupaia glis (5%, N=9), and Suncus murinus (0.5%, N=1). From this number, 29 individuals were found PCR positive for pathogenic Leptospira (R. norvegicus, N=20; R. rattus, N=6; T. glis, N=2; S. murinus, N=1). The study shows that Leptospira occurrence in the small mammals were significantly correlated to age category and sampling phases, with Spearman Correlation (rs) p=0.02 and p=0.04 respectively. Adult individuals were significantly more prevalent with Leptospira infection, whereby March and June were found to associate with higher Leptospira prevalent among the small mammals, potentially coincide with low rainfall and relative humidity level. This information is important in designing a specific control method for rodents in Leptospira outbreak areas. In addition, intensive sampling and regular cleaning effort were found to significantly reduce the small mammal Leptospira reservoir, thus should be implemented in intervention strategies in the urban environment.