INTRODUCTION: Childhood diarrheal illnesses are a major public health problem. In low-income settings data on disease burden and factors associated with diarrheal illnesses are poorly defined, precluding effective prevention programs. This study explores factors associated with recurrent diarrheal illnesses among children in Kabul, Afghanistan. METHODS: A cohort of 1-11 month old infants was followed for 18 months from 2007-2009. Data on diarrheal episodes were gathered through active and passive surveillance. Information on child health, socioeconomics, water and sanitation, and hygiene behaviors was collected. Factors associated with recurrent diarrheal illnesses were analyzed using random effects recurrent events regression models. RESULTS: 3,045 children were enrolled and 2,511 (82%) completed 18-month follow-up. There were 14,998 episodes of diarrheal disease over 4,200 child-years (3.51 episodes/child-year, 95%CI 3.40-3.62). Risk of diarrheal illness during the winter season was 63% lower than the summer season (HR = 0.37, 95%CI 0.35-0.39, P<0.001). Soap for hand washing was available in 72% of households and 11.9% had toilets with septic/canalization. Half of all mothers reported using soap for hand washing. In multivariate analysis diarrheal illness was lower among children born to mothers with post-primary education (aHR = 0.79, 95%CI 0.69-0.91, p = 0.001), from households where maternal hand washing with soap was reported (aHR = 0.83, 95%CI 0.74-0.92, p<0.001) and with improved sanitation facilities (aHR = 0.76, 95%CI 0.63-0.93, p = 0.006). Malnourished children from impoverished households had significantly increased risks for recurrent disease [(aHR = 1.15, 95%CI 1.03-1.29, p = 0.016) and (aHR = 1.20, 95%CI 1.05-1.37, p = 0.006) respectively]. CONCLUSIONS: Maternal hand washing and improved sanitation facilities were protective, and represent important prevention points among public health endeavors. The discrepancy between soap availability and utilization suggests barriers to access and knowledge, and programs simultaneously addressing these aspects would likely be beneficial. Enhanced maternal education and economic status were protective in this population and these findings support multi-sector interventions to combat illness. TRIAL REGISTRATION: www.ClinicalTrials.gov NCT00548379 https://www.clinicaltrials.gov/ct2/show/NCT00548379.