Risk factors for community-acquired pneumonia were studied by collecting data from all pneumonia patients (n = 201) in a defined child population between 3 months and 15 years of age during a surveillance period of 12 months, and from randomly selected healthy controls under 15 years of age (n = 250) from the same area. A structured questionnaire was used, with 409 (176 patients and 233 controls) families responding in total. The chi-square test, and the univariate and multivariate logistic regression analyses were used. The confounding effects of gender, age and place of acquisition were standardized, and the possible interactions between these variables and each individual risk factor were calculated. In the risk factor analysis, the responders were classified into 2 age groups: under 5 years and 5-14 years of age. In children under 5 years of age the significant risk factors were a history of recurrent respiratory infections during the past year [odds ratio (OR) 5.5], a history of wheezing episodes (OR 5.3), and a history of otitis media and tympanocentesis before the age of 2 years (OR 3.6). In the older children, the significant risk factors were a history of recurrent respiratory infections during the previous year (OR 3.0), and a history of wheezing periods at any age (OR 2.1). To sum up, a susceptibility to respiratory infections was found to be significantly associated with community-acquired pneumonia, and no interactions with age, gender or place of acquisition were significant. This trend was reflected by a history of wheezing and that of acute otitis media.