Seven-Year Case-Control Study in California of Risk Factors for Infant Botulism

Abstract

To ascertain possible risk factors for infant botulism, the intestinal infectious form of human botulism, in the years immediately following its initial recognition in California in1976. Parents of 159 California laboratory-confirmed cases of infant botulism from 1976 to 1983 and 318 healthy controls were interviewed using a comprehensive (>300 factors) questionnaire. "Neighborhood controls" (n=184) were matched on date of birth, sex, race/ethnicity, and neighborhood of residence. "County controls" (n=134) were matched only on date of birth, sex, and county of residence. Age-stratified bivariate and multivariate conditional logistic regression analyses were performed using SAS. All cases required hospitalization. Bivariate analyses identified several risk factors that in multivariate analyses were not significant. In multivariate analyses, risk factors differed with stratification by age. For the ≤2 month-old neighborhood controls comparison, birth order >1, cesarean delivery, ≤1 bowel movements (BMs) per day, and windy residence area were associated with illness hospitalization, and for the county controls comparison, only pacifier use was associated. For the <2 month-old neighborhood controls comparison, <1 bowel movements (BMs) per day, cesarean delivery, birth order >1, and windy residence area were associated with illness hospitalization, and for the county controls comparison, pets in the home was an additional risk factor. With the exception of the ≤2-month-old county controls group, slower intestinal transit time (≤1 BM/d) was associated with illness. Otherwise, our case-control investigation identified few physiologic, environmental, and maternal factors associated with infant botulism hospitalization in California.