Background: Cryptosporidium, an apicomplexan protozoa, is a leading contributor to diarrheal morbidity and mortality in children under five years old worldwide. As there is no vaccine and no approved drug for Cryptosporidium spp. in young children, a focus on prevention of infection is critical. We undertook a pilot case-control study to define the extent of person-to-person transmission of cryptosporidiosis within families in an urban and rural community in Bangladesh. Methods: We enrolled 48 case families with a Cryptosporidium-infected child aged 6-18 months. Controls were age-sex matched Cryptosporidium-negative children (n=12). Once children were identified, we enrolled all household members. We then followed these individuals for 8 weeks, with weekly surveillance stools and testing with qPCR for Cryptosporidium spp. Findings: In the 48 case families, the rate of secondary infections with Cryptosporidium was 18.6% (22/118) compared to 0 new infections (0/35) in the 12 control families. In the 22 urban Mirpur households, the secondary attack rate was 30% (18/60) in cases compared to 0% (0/14) in controls (chi-square p = 0.018). In contrast, in the 21 rural Mirzapur households, the secondary attack rate was 6.9% (4/58) in case households compared to 0% (0/21) in controls (chi-square p = 0.22). Genotyping by gp60 demonstrated infection with the same subspecies in four of six families. Serologic response to Cryptosporidium infection was associated with younger age, longer duration of infection, and C hominis gp60_IbA9G3R2 infection. Interpretation: The high rate of secondary infection in Mirpur suggests that person-to-person transmission is likely a major source of Cryptosporidium infection for young children living in this region. GP 60 genotyping demonstrated direction of infection in 2 households, and concurrent infection in five household. Further work is needed to understand differences in transmissibility and differences in immunity to different genotypes.