Due to the risk of Shiga-toxin producing Escherichia coli (STEC) transmission, current guidance advises excluding young children from childcare settings until microbiologically clear. Children can shed STEC for a prolonged period, and the cost-effectiveness of exclusion has not been evaluated. Our decision tree analysis, including probabilistic sensitivity analysis, estimated comparative health system costs and effects of exclusion until microbiological clearance versus return to childcare setting before this. Due to the risk of secondary cases, return before microbiological clearance resulted in the incremental loss of 0.019 QALYs, but savings of £156. Using the willingness-to-pay threshold of £20000 per QALY, the incremental net monetary benefit of exclusion until microbiological clearance was £215. Exclusion until microbiological clearance remained cost-effective if the total costs for managing the exclusion were below £576. Return before microbiological clearance may, therefore, become cost-effective in cases where the costs of managing exclusion until microbiological clearance are high and/or the risk of secondary cases is very low. Broadening the decision perspective, including the costs of exclusion to the families, may also impact the recommendation. Further research is needed to assess the risk of STEC transmission from children who have clinically recovered and the impact of STEC and exclusion on families of the affected children.

Exclusion of children with presumptive Verocytotoxin-producing Escherichia coli (VTEC) from childcare facilities until negative stool specimens are obtained is routine practice that disrupts families. We estimated the shedding and exclusion duration and transmission risk in such facilities. The study population comprised 225 children aged <6 years attending 201 childcare facilities in England with microbiologically confirmed VTEC in 2010–2011. We estimated the interval from onset to first negative specimen, and identified transmission events with secondary cases linked to facilities. The median duration of shedding was 31 days, and median period of exclusion was 39·5 days. Cases attending facilities while shedding VTEC did so for a median of 2 days before exclusion. Secondary cases occurred in 6/83 facilities (7%) attended by infectious cases. Despite evidence of VTEC shedding at facilities, transmission is relatively low. Revised control guidelines could consider supervised return for prolonged asymptomatic shedders.

On 24–25 October 2005 a cluster of five haemolytic uraemic syndrome (HUS) cases was reported in southwest France. An investigation was undertaken to identify the outbreak source and implement control measures. Cases were defined as individuals with HUS or diarrhoea with isolation of Escherichia coli O157:H7 in stools or a positive antibody response to E. coli O157 lipopolysaccharide, resident in southwest France with symptom onset after 19 September 2005. Sixty-nine identified patients had symptom onset between 5 October and 3 November 2005, including 17 cases of HUS. One brand of frozen beef burgers produced on 22 August 2005 was consumed by all patients in the week before symptom onset. E. coli O157:H7 strains from patients, patients' burgers and the manufacturing plant were genetically related. This is the largest community-wide outbreak of E. coli O157:H7 in France to date and the first associated with consumption of contaminated frozen beef burgers.