Guideline for the Prevention and Control of Norovirus Gastroenteritis Outbreaks in Healthcare Settings, 2011

Q2: What are the best methods to identify a norovirus outbreak in a healthcare setting?

Evidence Table Q2 - Specimen collection

Author, Yr (Ref)	Study Design Quality	Study Objective	Population and Setting N	Results	Comments	Ref ID_Data extracted by
Duizer, E; 2007 117	Diagnostic study   2,3	To use statistical analysis in determining 1) the minimum number of positive stool samples using RTPCR or ELISA (IDEIA) compared to a hypothetical gold standard needed to declare norovirus as the causative agent of a gastroenteritis outbreak and 2) the probability of finding this minimum number of positive samples for varying numbers of tested samples.	N/A	# Positive samples needed to assign norovirus as the causative agent   ELISA: 1 positive for 2-6 samples tested   RT-PCR: 1 positive for 2-4 samples tested 2 positive for 5-11 samples tested   Sensitivity (%) for detecting a norovirus outbreak for various numbers of tested samples ELISA: 57% for 2 tested samples 72% for 3 tested samples 88% for 5 tested samples 92% for 6 tested samples   RT-PCR: 84% for 2 tested samples >90% for 3 tested samples 92% for 5 tested samples 96% sensitivity for 6 tested samples	Parameters Defined outbreak as caused by norovirus if the prevalence is >8% Hypothetical gold standard: sensitivity 100%; specificity 100%.   RT-PCR: sensitivity 72%; specificity 99%.   IDEIA: sensitivity 41%; specificity 98%.   Minimum # positive samples needed is the number of positive samples where there is >95% probability of attaining a prevalence ≥8%   IDEIA NLV (Dakocytomation Ltd., Ely, UK).	044_IL
Gray, JJ; 2007 118	Diagnostic study   2,3	To determine test characteristics for IDEIA and RIDA-SCREEN.	Stool samples from patients with symptoms of gastroenteritis collected during the 2004-2005 and 2005-2006 norovirus seasons and evaluated in this European multicenter study.   2,254 samples from 273 outbreaks.   274 samples collected in sporadic cases.   144 samples had other enteric pathogens identified.	Test characteristics IDEIA: Sensitivity 58.93% (95% CI 56.12-61.68%) Specificity 93.91% (95% CI 92.23-95.25%) PPV 92.30% NPV 64.90%   RIDA-SCREEN: Sensitivity 43.81% (95% CI 41.01-46.65%) Specificity 96.27% (95% CI 95.00-97.38%) PPV 93.70% NPV 58.20%   Sensitivity for differing number of samples tested The sensitivity for outbreak diagnosis improved when ≥6 samples tested. IDEIA: 3 vs. 6 samples tested (z=±3.191; p=0.0014) RIDA-SCREEN: 3 vs. 6 samples tested (z=±3.828; p=0.0001)   Range of norovirus genotypes detected All samples: Genotype - IDEIA vs. RIDASCREEN No [(%) samples genotype detected (95% CI)]; p value GI-1 – 4 [80.00% (37.55-96.36%)] vs. 3 [60.00% (23.07-88.24%)]; 0.49 GI-2 – 11 [84.62% (57.77-95.67%)] vs. 2 [15.38% (4.33-42.23%)]; 0.0002 GI-3 – 12 [42.86% (26.51-60.93%)] vs. 9 [32.14% (17.93-50.66%)]; 0.4 GI-4 – 2 [100.00% (34.24-100.00%)] vs. 0 [0.00% (0.00-65.76%)]; 0.3 GI-5 – 3 [37.50% (13.68-69.43%)] vs. 0 [0.00% (0.00-32.44%)]; 0.2 GI-6 – 5 [71.43% (35.89-91.78%)] vs. 0 [0.00% (0.00-35.43%)]; 0.02 GI-7 – 0 [0.00% (0.00-79.35%)] vs. 0 [0.00% (0.00-79.35%)]; >0.5 GII-1 – 7 [87.50% (52.91-97.76%)] vs. 0 [0.00% (0.00-32.44%)]; 0.0024 GII-2 – 8 [50.00% (28.00-72.00%)] vs. 4 [25.00% (10.18-49.50%)]; 0.2 GII-3 – 30 [57.69% (44.19-70.13%)] vs. 11 [21.15% (12.24-34.03%)]; 0.0003 GII-4 – 203 [67.44% (61.96-72.49%)] vs. 186 [61.79% (56.19-67.10%)]; 0.17 GII-5 – 2 [33.33% (9.68-70.00%)] vs. 3 [16.67% (3.01-56.35%)]; >0.5 GII-6 – 2 [22.22% (6.32-54.74%)] vs. 0 [0.00% (0.00-29.91%)]; 0.4 GII-7 – 20 [68.97% (50.77-82.72%)] vs. 5 [17.24% (7.6-34.55%)]; 0.002 GII-8 – 0 [0.00% (0.00-79.35%)] vs. 0 [0.00% (0.00-79.35%)]; >0.5 GIV-1 – 0 [0.00% (0.00-48.99%)] vs. 0 [0.00% (0.00-48.99%)]; >0.5 rGII – 10 [52.63% (31.71-72.67%)] vs. 2 [10.53% (2.94-31.39%)]; 0.01   IDEIA showed reactivity to a broader range of genotypes than the RIDASCREEN norovirus assay, which showed genotype-dependent sensitivities.	IDEIA norovirus (Oxoid; Thermo Fisher Scientific, Ely, UK). RIDASCREEN norovirus (R-Biopharm, Darmstadt, Germany) RT-PCR was the reference standard.	053_IL
Richards, A; 2003 119	Diagnostic Study   1,2	To determine the test characteristics of ELISA and EM in detecting norwalk-like virus (NLV) infection when compared with PCR	Fecal samples collected from patients involved in outbreaks of gastroenteritis in the UK   531 fecal samples	Test characteristics (%) of ELISA vs. PCR Sensitivity – 55.5(51.1-60.0) Specificity – 98.3(97.1-99.9) PPV – 95.0(CI not reported) NPV – 76.9 (CI not reported)   Test characteristics (%) of EM vs. PCR Sensitivity – 23.9(19.5-28.1) Specificity – 99.2(98.3-100) PPV – 93.9(CI not reported) NPV – 70.7(CI not reported)   Identification of NLV as the cause of an outbreak (% of outbreaks) When the causative agent was defined by ≥ 2 positive samples EM – 7.2 ELISA – 18.6 PCR – 41.5 When the causative agent was defined by ≥ 1 positive samples EM – 19.6 ELISA – 47.8 PCR – 62.8   Sensitivity; Specificity of ELISA based on number of samples collected 2 samples – 52.9; 100 ≥4 – 69.2; 100 ≥6 – 71.4; 100   Other results Agreement between ELISA and PCR – 81.8% (Kappa = 0.57) Sensitivity of ELISA was significantly increased when compared with EM (P<0.01)	Power and sample size not reported.	848_RA

Next Table