The significance of the number of submitted samples and patient-related factors for faecal bacterial diagnostics

The sensitivity of bacteriological testing of faecal samples from patients with diarrhoea has not been properly determined. The present study analysed the association between the results of stool sample examinations and the number of samples examined per patient and other patient-related factors. Data concerning faecal specimens referred for culture for enteric bacterial pathogens (Campylobacter, Salmonella, Shigella and Yersinia) to the central microbiological laboratory in Denmark between 1995 and 2003 were analysed. In total, 620 000 samples were sorted into 277 000 sample-series, i.e., samples submitted from the same individual on the same day. Data were analysed by multivariate logistic regression, with the outcome being a positive sample-series, i.e., one or more positive samples per series. Overall, 11.9% of the sample-series were positive. For adults (aged > or =18 years), the OR for a positive diagnosis was 1.20 (95% CI 1.18-1.21) for each additional sample. Positive diagnoses were also more likely during summer, if the patient was male, or if the patient was neither very young nor very old. The added diagnostic effect of additional samples was more pronounced for the group of patients with persistent (>2 weeks) diarrhoea. Overall, the probability of finding common pathogenic bacteria in faecal samples was found to vary according to the number of samples, the season and the patient's age and gender. Analysis of more than one sample improves the sensitivity of faecal culture by at least 20% for each additional sample.