INTRODUCTION: Until recently, vancomycin-resistant enterococcus (VRE) infection or colonisation was a rare occurrence in Singapore. The first major VRE outbreak involving a 1500-bed tertiary care institution in March 2005 presented major challenges in infection control and came at high costs. This study evaluates the predictors of VRE carriage based on patients' clinical and demographic profiles. MATERIALS AND METHODS: Study patients were selected from the hospital inpatient census population during the VRE outbreak (aged 16 years or more). Clinical information from 84 cases and 377 controls were analysed. RESULTS: Significant predictors of VRE carriage included: age>65 years Odds ratio (OR), 1.98; 95% CI (confidence interval), 1.14 to 3.43); female gender (OR, 2.15; 95% CI, 1.27 to 3.65); history of diabetes mellitus (OR, 1.94; 95% CI, 1.14 to 3.30), and staying in a crowded communal ward (OR, 2.75; 95% CI, 1.60 to 4.74). Each additional day of recent hospital stay also posed increased risk (OR, 1.03; 95% CI, 1.01 to 1.04). CONCLUSION: Elderly diabetic females with prolonged hospitalisation in crowded communal wards formed the profile that significantly predicted VRE carriage in this major hospital-wide outbreak of VRE in Singapore. It is imperative that active VRE surveillance and appropriate infection control measures be maintained in these wards to prevent future VRE outbreaks.
INTRODUCTION: Until recently, vancomycin-resistant enterococcus (VRE) infection or colonisation was a rare occurrence in Singapore. The first major VRE outbreak involving a 1500-bed tertiary care institution in March 2005 presented major challenges in infection control and came at high costs. This study evaluates the predictors of VRE carriage based on patients' clinical and demographic profiles. MATERIALS AND METHODS: Study patients were selected from the hospital inpatient census population during the VRE outbreak (aged 16 years or more). Clinical information from 84 cases and 377 controls were analysed. RESULTS: Significant predictors of VRE carriage included: age>65 years Odds ratio (OR), 1.98; 95% CI (confidence interval), 1.14 to 3.43); female gender (OR, 2.15; 95% CI, 1.27 to 3.65); history of diabetes mellitus (OR, 1.94; 95% CI, 1.14 to 3.30), and staying in a crowded communal ward (OR, 2.75; 95% CI, 1.60 to 4.74). Each additional day of recent hospital stay also posed increased risk (OR, 1.03; 95% CI, 1.01 to 1.04). CONCLUSION: Elderly diabetic females with prolonged hospitalisation in crowded communal wards formed the profile that significantly predicted VRE carriage in this major hospital-wide outbreak of VRE in Singapore. It is imperative that active VRE surveillance and appropriate infection control measures be maintained in these wards to prevent future VRE outbreaks.