- Email: [email protected]
Environmental testing for norovirus in various institutional settings using catering companies as sentinels for norovirus prevalence among the general population
Food Control 47 (2015) 98e102
Contents lists available at ScienceDirect
Food Control journal homepage: www.elsevier.com/locate/foodcont
Environmental testing for norovirus in various institutional settings using catering companies as sentinels for norovirus prevalence among the general population €gele a, Kyara Klunder a, Ingeborg L.A. Boxman a, *, Linda Verhoef b, Geke Ha a b Nathalie A.J.M. te Loeke , Harry Vennema , Claudia C.C. Jansen a, Marion Koopmans b, c, 1 a
Laboratory for Feed and Food Safety, Food and Consumer Product Safety Authority (NVWA), P.O. Box 144, 6700 AC Wageningen, The Netherlands Laboratory for Infectious Diseases and Screening, National Institute of Public Health and the Environment (RIVM), P.O. Box 1, 3720 BA Bilthoven, The Netherlands c Department of Viroscience, ErasmusMC, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 January 2014 Received in revised form 5 June 2014 Accepted 14 June 2014 Available online 28 June 2014
Noroviruses (NoV) are among the most common causes of viral gastroenteritis (GE) worldwide and can be transmitted from person-to-person, via food or contaminated surfaces. The present study aimed to examine the prevalence of NoV RNA on surfaces in food preparation and sanitary areas in different health care settings and to compare the outcomes with the prevalence in nearby located catering companies, mainly restaurants, for general public, as sentinels. For this purpose, 1087 environmental swabs were taken for NoV analyses from surfaces in 241 institutional departments and 123 catering companies in The Netherlands without a recently reported outbreak of gastro-enteritis in high NoV season only. NoV RNA was detected in 15.1% of the 73 nonhospital health care institutions, 11.1% of the 54 hospital central kitchen departments, 14.9% of the 114 decentralized hospital kitchens (in-patient units) and 4.1% of the 123 nearby located catering companies. Twenty-five of the 49 positive environmental samples were genotyped by sequence analyses. In 7% of the investigated hospitals (4/58), NoV was detected in two or more departments. NoV prevalence was significantly lower in food preparation areas than in sanitary facilities (p < 0.05), but only in hospital central kitchen departments and non-hospital health care settings, and not in de-centralized hospital kitchens in in-patient units or in catering companies for the general public. This data suggests that there is a need for education on risks of NoV transmission by food handling of healthcare workers using in ward kitchen facilities. © 2014 Published by Elsevier Ltd.
Keywords: Swabs Non-outbreak Hospital Restaurant
1. Introduction Noroviruses (NoV) are among the most common causes of viral gastroenteritis (GE) worldwide. In the Netherlands, the incidence of NoV illness in 2009 was estimated to be 3800 cases per 100.000 inhabitants, with an estimated burden over 1600 disability-
* Corresponding author. Tel.: þ31 88 2230447. E-mail addresses: [email protected] (I.L.A. Boxman), linda.verhoef@ €gele), [email protected] rivm.nl (L. Verhoef), [email protected] (G. Ha (K. Klunder), [email protected] (N.A.J.M. te Loeke), [email protected] (H. Vennema), [email protected] (C.C.C. Jansen), [email protected], [email protected] (M. Koopmans). 1 Present address: Department of Viroscience, ErasmusMC, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands. http://dx.doi.org/10.1016/j.foodcont.2014.06.026 0956-7135/© 2014 Published by Elsevier Ltd.
adjusted life years (Verhoef, Koopmans, et al., 2013). The virus is highly infectious (Teunis et al., 2008) and often seen in outbreaks with high attack rates in health care settings, cruise ships, catered events, (EFSA, 2011; FAO/WHO, 2008; Meakins, Adak, Lopman, & O'Brien, 2003; Widdowson et al., 2004). NoV outbreaks are often prolonged, lasting longest in hospitals (19 days, range 6e92 days) and in nursing homes (16 days, range 3e44 days) in comparison to non-health care settings (7 days, range 1e26 days) (Harris, Lopman, & O'Brien, 2010). Outbreaks are recurring due to high levels of shedding in stool or vomitus and sometimes chronic shedding, especially in hospitals where people are vulnerable and there is a constant turnover of new susceptible people during outbreaks. Closure of wards is one of the control measures to control NoV spread in hospitals (Greig & Lee, 2012). Spreading is thought to be through person-to-person transmission, although environmental
I.L.A. Boxman et al. / Food Control 47 (2015) 98e102
contamination has been implicated in this transmission (Lopman et al. 2012). NoV have been demonstrated on different types of surfaces (floors, tables, doorknobs, handles, medical equipment, bed rails, carpets and curtains) in health care facilities, schools and food production facilities (Boone & Gerba, 2007; Gallimore et al., 2006; Morter et al., 2011; Wu et al., 2005). Recently, results of environmental sampling in catering companies suggested that health care institutions were more likely to be NoV contaminated than catering companies for general public (Boxman et al., 2011). This initiated further studies to examine the prevalence of NoV in different food preparation areas in various health care settings as part of assessment of risk of food contamination, and including an inventory on food handler's knowledge and working practices (Verhoef, Jaramillo Gutierrez, Koopmans, Boxman, 2013). The aim of the present study was to determine the NoV environmental presence during the winter season in healthcare settings. For the purpose of direct comparison, sampling in catering companies, mainly restaurants, nearby the institutions was performed. This was also done to reflect the background epidemiology of NoV in the general population, because of yearly differences in NoV activity (Lopman et al., 2004; Verhoef et al., 2008). In contrast to food handlers in institutions who work in a semi-closed settings prone to NoV outbreaks, food handlers in catering companies are part of the general public. Catering companies were therefore considered as sentinels for the NoV activity in the general population in this paper. In order to gain a deeper understanding of the potential sources of food-borne NoV outbreaks, outcomes were compared between food preparation areas and sanitary facilities within central and decentralized kitchens in health care institutions, and in comparison with outcomes of sampling in nearby public catering companies. 2. Materials and methods 2.1. Study rationale In the periods January 2010eApril 2010 and February 2011eMarch 2011, inspections were performed by officers of the Netherlands Food and Product Safety Authority (NVWA) in health care institutions, as well as in catering companies in the vicinity of these institutions, during the winter seasonal peak of viral gastrointestinal disease. The rationale for this was to capture environmental presence of NoV during the period of the year with most illness, in environments that may be relevant for the epidemiology of these viruses in health care settings. The healthcare settings included 58 hospitals (44.3%), 54 homes for the elderly (41.2%) as well as 11 nursing homes (8.4%). In hospitals, samples were collected in 54 central kitchen departments (32.1%) and 114 decentralized kitchens (67.9%). This was done to be able to discriminate food-handler introductions at these sites, that likely differ in the level of knowledge and stringency of food handling practices: staff in central kitchen department prepares food which is subsequently distributed to kitchens in-patient units (de-centralized kitchens), where other nursing staff reheat food to adequate temperatures, and serve meals and drinks. Sampling in nearby catering companies, mainly restaurants, was done to reflect the background epidemiology of NoV.
99
equipment, such as the grip of a refrigerator, knife or serving spoon, and the handle of a cutting or mixing machine in central kitchens, and surfaces from the microwave or the cutlery drawer in de-centralized kitchens. The second swab was used to collect a surface sample from the soap dispenser, from the pepper-and-salt set or from a food storage box in the refrigerator. In hospital in-patient units the handle of the dish washer was given as an alternative surface. The third swab was used to collect a surface sample from the flushing chain or knob and the toilet seat (both upper and undersurfaces) in the men's (employees') bathroom only, which was chosen for reasons of systematic sampling. Sampling was done by inspectors who had no knowledge about possible outbreaks related to the sites that they visited to avoid oversampling of known contamination events. Environmental swab samples were either kept at 4 C or kept frozen at 20 C during transport and stored frozen at 20 C until they were processed at the laboratory of the NVWA. Analyses were performed on coded samples. 2.3. Extraction and analyses of swab samples Environmental swabs were extracted and analysed for the presence of NoV RNA as described previously (Boxman et al., 2009; 2011) with some modifications. In brief, RNA extractions were performed using 30 ml glassmilk (MP Biomedical) per swab extraction with no difference in detection (data not shown), as rieux) was no longer available. normal non-magnetic silica (BioMe RNA samples were analyzed using two-step reverse transcription (RT) real time PCR. The N2pol assay targeting the polymerase region was used for the detection of NoV GII types as described (Boxman et al., 2011). For the detection of NoV GI types, the N1cap assay targeting the ORF1/ORF2 region was used in a two-step RT real time PCR format instead of the N1cap nested real-time RT-PCR assay (Boxman et al., 2009). 2.4. Sequencing and phylogenetic analyses NoV presumptive positive samples, as judged by the presence of a typical S -curve in the amplification plot, were re-amplified and sequenced as described previously (Boxman et al., 2011). NoV sequences detected in the present study were genotyped using the NoV typing tool (Kroneman et al., 2011) (http://www. rivm.nl/en/Topics/N/NoroNet/Databases/) and compared to Dutch NoV outbreak strains detected in humans involved in outbreaks and for which clinical samples had been sent in for diagnosis at the Institute for Public Health and the Environment in the Netherlands. Sequence comparisons were done if an overlapping region of at least 100 nucleotides (nt) in region 5373e5577 for ORF2 of GI types and in region 4299e4494 for ORF1 of GII types was available (Kroneman et al., 2011). A multiple alignment of selected sequences was made for each genotype or, if genotype II.4 was involved, each variant. The pairwise distance was computed using a 2-parameter Kimura nucleotide substitution model.
2.2. Sampling protocol
2.5. Statistical analyses
At each setting, two mixed surface swab samples were taken from surfaces in food preparation areas and one mixed surface swab sample was taken from the bathroom for NoV analyses, as described previously (Boxman et al., 2009; 2011). In brief, the first swab was used to collect a surface sample from a food preparation
Statistical analyses were performed using Fisher-P one tailed analyses to identify differences between settings and swabbing locations. If cell counts in a cross table did not include any observations, calculations were performed by artificially setting the number of the concerning cell to one instead of zero.
100
I.L.A. Boxman et al. / Food Control 47 (2015) 98e102
3. Results A total of 364 institutional departments or catering companies was included in the present study. In the period between January and April 2010, 73 non-hospital institutions and 71 nearby located catering companies were visited. In February and March 2011 54 hospital central kitchen departments, 114 decentralized kitchens in hospital in-patient units and 52 nearby located catering companies were visited. In central kitchen departments food is prepared which is subsequently distributed to kitchens in-patient units (the de-centralized kitchens), where nursing staff reheat food to adequate temperatures, and serve meals and drinks. In total, 1087 samples were available for testing. 3.1. Detection of NoV on environmental swabs in various institutional settings 3.1.1. Institutions all In total, 42 (5.8%) out of 718 environmental swabs from 34 (14.1%) out of 241 institutional settings tested positive for NoV (Table 1). At five in-patient units and one elderly home, samples of both food preparation areas and sanitary facilities tested positive. NoV was detected nearly two times more often on surfaces in sanitary facilities (20/236, 8.5%) compared to surfaces in food preparation areas (22/482, 4.6%) (p ¼ 0.04). 3.1.2. Non-hospital institutions Most of the 73 non-hospital institutions visited were elderly homes (74.0%) followed by nursing homes (15.1%). Thirteen (6.0%) out of 215 environmental swabs from 11 (15.1%) of 73 non-hospital institutions tested positive for NoV. NoV was three times more likely to be detected on surfaces in sanitary facilities (8/69, 11.6%) compared to surfaces in food preparation areas (5/146, 3.4%), (p ¼ 0.024). 3.1.3. Hospital central kitchen departments Six (3.7%) out of 162 environmental swabs from 6 (11.1%) out of 54 central kitchen departments tested positive for NoV. NoV was ten times more likely to be detected on surfaces in sanitary facilities (5/54, 9.3%) compared to surfaces in food preparation areas (1/108, 0.9%), (p ¼ 0.016).
3.1.4. Hospital decentralized kitchens (in-patient units) Twenty-three (6.7%) out of 341 environmental swabs from 17 (14.9%) out of 114 decentralized kitchens in in-patient units tested positive for NoV. There was no significant difference between the presence of NoV on surfaces in sanitary facilities (7/113, 6.1%) compared to presence of NoV on surfaces in food preparation areas (16/228, 7.0%), (p ¼ 0.48). 3.1.5. Between the different hospital departments The overall NoV prevalence was higher in hospital decentralized kitchens (p ¼ 0.12) than in hospital central kitchen departments, as measured using environmental swabs. Food preparation areas in hospital central kitchens were less frequently contaminated with NoV than food preparation areas in in-patient units (p ¼ 0.01). In 7% of the investigated hospitals (4/58), NoV was detected in two or more departments at the same day (Table 2). In six hospitals, NoV was detected in the central kitchen departments. Of these, two hospitals had NoV also detected in one of the de-centralized kitchens and one hospital had NoV detected in two decentralized kitchens. 3.2. Detection of NoV on environmental swabs in sentinel catering companies 3.2.1. Catering companies Most of the 123 catering companies selected as sentinels for NoV prevalence among the general population were restaurants (79.7%) followed by lunchrooms (11.4%) or take away counters (8.9%). Seven (1.9%) out of 369 environmental swabs from 5 (4.1%) out of 123 catering companies tested positive for NoV. At two catering companies, samples of both food preparation areas and sanitary facilities tested positive NoV. Remarkably, NoV was only detected on samples collected from catering companies in 2010, whereas none of the samples in 52 catering companies tested positive in 2011. There was no significant difference between the presence of NoV on surfaces in sanitary facilities (3/122, 2.5%) compared to presence of NoV on surfaces in food preparation areas (4/247, 1.6%), (p ¼ 0.42). Neither was there a significant difference when only samples collected in catering companies in 2010 were taken into account (p ¼ 0.40).
Table 1 Prevalence of NoV in institutional settings and nearby located catering companies. Year
NoV detected in any sample on company level
NoV detected in samples taken from surfaces
No./total
Kitchen
Bathroom %
No./total
All samples
No./total
%
%
No./total
%
2010
Elderly home Nursing home Other non-hospital institutions Subtotal Restaurant Catering/lunchroom Take-away Subtotal
10/54 1/11 0/8 11/73 4/57 0/6 1/8 5/71
18.5 9.1 0 15.1 7.0 0 13 7.0
5/107 0/23 0/16 5/146 3/112 0/12 1/19 4/143
4.7 0 0 3.4 2.7 0 5.3 2.8
7/52 1/9 0/8 8/69 2/57 0/6 1/7 3/70
13.0 11.1 0 11.6 3.5 0 14 4.3
12/159 1/32 0/24 13/215 5/169 0/18 2/26 7/213
7.5 3.1 0 6.0 3.0 0 7.7 3.3
2011
Hosp. central kitchen depart. Hosp. de-centr.kitchens dept. Subtotal Restaurant Catering/lunchroom Take-away Subtotal
6/54 17/114 23/168 0/41 0/8 0/3 0/52
11.1 14.9 13.7 0 0 0 0
1/108 16/228 17/336 0/82 0/16 0/6 0/104
0.9 7.0 5.1 0 0 0 0
5/54 7/113 12/167 0/41 0/8 0/3 0/52
9.3 6.1 7.2 0 0 0 0
6/162 23/341 29/503 0/123 0/24 0/9 0/156
3.7 6.7 5.8 0 0 0 0
All institutional settings All catering companies
34/241 5/123
14.1 4.1
22/482 4/247
4.6 1.6
20/236 3/122
8.5 2.5
42/718 7/369
5.8 1.9
I.L.A. Boxman et al. / Food Control 47 (2015) 98e102 Table 2 Hospitals with NoV detected in environmental samples at two or three departments. ID
Department
Location
Genogroup
Genotype
ZH04
Internal Medicine
Bathroom and kitchen Kitchen Bathoom Kitchen
GII
II.4 c.n.b.aa
GII GII GII
Non-typed GIIb Non-typed GII II.4 New Orleans
GII GII
II.4 c.n.b.a. Non-typed GII
GI and GII GI GII
I.2, non-typed GII Non-typed GI Non-types GII
ZH13
ZH19
ZH22
Neurology Central Kitchen dept In-patient (non specified) Central Kitchen dept Gynaecology Central Kitchen dept Oncology/Surgery Haematology
Bathroom Bathroom and kitchen Bathroom Kitchen Kitchen
a
c.n.b.a. GII.4 sequence of which the variant could not be assigned. Non-typed GII: sequence analyses was not successful as the signal of the presumptive positive samples was too weak. b
In 2010, the overall NoV prevalence appeared higher in nonhospital health care institutions than in nearby located catering companies, but the difference was not statistically significant (p ¼ 0.13). In 2011, NoV prevalence was, however, significantly higher in hospital central kitchen departments than in nearby located catering companies for the general population (p ¼ 0.02). 3.3. Typing of NoV RNA detected on environmental surfaces In total, 49 samples tested positive for NoV RNA. GI NoV strains were detected in three samples and GII NoV strains were detected in 46 samples. Sequence analyses were not always successful, as for some presumptive positive samples the signal was too weak. Twenty-five samples (51%) yielded a sequence suitable for genotyping and sequence analyses. GII.4 was the most frequently detected genotype (n ¼ 19). Nine of these, were sub-typed as GII.4.New Orleans 2009 variant. The remaining could not be assigned to a particular GGII.4 variant. GIIb (GII.P21) was detected three times and GIIg, GI.2 and GI.4 were detected one time each. For two hospitals, sequences were available for more than one location. This showed that the sequence types were different within single hospitals (ZH13 and ZH22), suggesting independent contamination events. 3.4. Comparative sequence analysis for environmental and reported outbreak samples Comparative sequence analysis was performed to identify possible clustering of the 25 sequences obtained from environmental samples and sequences obtained from reported outbreaks of gastroenteritis, i.e., clinical samples collected in the period from January 2008 through June 2013 (167 GII.b (GII.P21), 58 II.g, 1034 GGII.4 New Orleans 2009, and 70 GGII.4 sequences for which the variant could not be assigned). Unfortunately the genomic regions for environmental and diagnostic genogroup I detection did not overlap. Sequences of genogroup II strains as detected by environmental sampling were interspersed with those circulating in Dutch patients. Based on a minimum overlap of 100 nt, five clusters (GII.4 New Orleans 2009 (n ¼ 2), GII.b (or GII.P21) (n ¼ 2) and GII.g (n ¼ 1)), were identified consisting of 99.5% similar sequences and having at least one sequence detected in the present study and one or more Dutch outbreak clinical samples. These clusters were geographically dispersed and could not be linked based on available epidemiological information.
101
4. Discussion This study reports the prevalence of NoV in health care institutional departments in contrast to nearby located catering companies as sentinels for NoV prevalence among the general population. The findings suggest that rates of exposure in hospital based central kitchen staff is higher than in the sentinel catering company kitchen staff, pointing at potential cross transmissions of viruses within the hospital between in-patient units and central kitchen departments. NoV prevalence was significantly lower on swabs taken from food preparation areas than those from sanitary facilities (p < 0.05) in hospital central kitchen departments and non-hospital health care settings only, and not in de-centralized hospital kitchens in in-patient units or in catering companies for the general public. The finding is suggestive of a higher compliance with good hand hygiene by staff of the former two settings. The differences in the estimated NoV prevalence could be biased as all users of the employee's bathroom will touch the flush chain or knob, whereas such an equivalent, an always-touched site in the kitchen is not available. For this reason sampling surfaces was standardized to the extend possible, and twice as many samples were taken from the food handling areas than from bathroom areas. The results showed that, despite the possible bias in favour of the bathroom areas, part of the food handling areas were detected positive, which is an unwanted situation. The higher compliance with good hand hygiene in staff in central kitchen departments of hospitals or non-hospital health care settings may be due to more appropriate education in food safety. This is important as here food is prepared for vulnerable persons. In contrast, staff working in decentralized kitchens in in-patient units most often will be medically trained nursing staff. In a study of Buccheri and coworkers, more than 80% of the respondent nurses involved in food service functions did not attend any educational course on food hygiene (Buccheri et al., 2007). Specifically concerning potential transmission of viral food-borne illness by food handlers several gaps in education and training were identified, demonstrating that knowledge on NoV was low, although awareness of NoV was significantly higher among food handlers in institutional settings than in catering companies (Verhoef, Jaramillo Gutierrez, et al. 2013). Besides knowledge, health care workers (Allegranzi & Pittet, 2009) as well as food handlers in catering companies (Green et al., 2007; Pragle, Harding, & Mack, 2007) encounter difficulties in complying with hand hygiene indications. The close contact between nursing staff and patients might contribute to NoV spread over the wards (Heijne et al. 2012). Transmission of NoV via hand contact was suggested by the types of contaminated sites in hospitals identified (Morter et al. 2011). Such hot spots of contamination were patient monitoring equipment, computers and notes trolleys at nurses stations', dispensers and hand and grab rails, even post cleaning. Recently, outcomes of questionnaires and environmental testing have been combined to identify independent determinants of surface NoV contamination (Verhoef, Jaramillo Gutierrez, et al. 2013). The results showed that the likelihood for such a contamination was increased when being situated in an institutional setting or when a positive answer was given to the question whether a person would continue food handling despite sick with vomiting complaints. In the present study, seven percent of the hospitals had NoV detected in two or more departments at the same day, most likely due to independent contamination events. That multiple strains can circulate or co-circulate during the NoV season during hospital outbreaks was demonstrated by analyses of fecal samples from patients or medical staff (Morter et al., 2011; Xerry, Gallimore, Cubitt, & Gray, 2010). These studies were,
102
I.L.A. Boxman et al. / Food Control 47 (2015) 98e102
however, restricted to wards, and did not include the central kitchen departments. As expected from the many hospital outbreaks reported (reviewed by Harris et al., 2010; Greig & Lee, 2012), the present study demonstrated that NoV was more present in hospital departments than in the catering companies serving as their sentinels. Remarkably the prevalence of NoV in 2010 in catering companies was much higher (7%) than in 2011 (none), with no other explanation than that the prevalence in the catering companies reflecting yearly differences in NoV activity (Lopman et al., 2004; Verhoef et al., 2008). According to the laboratory reports of NoV in the United Kingdom and the Netherlands, (http://www. hpa.org.uk/webc/HPAwebFile/HPAweb_C/1287143931777), http:// www.rivm.nl/Onderwerpen/V/Virologische_weekstaten/Rapportages/ Open_rapportages_(virologische_weekstaten/) NoV activity was much higher in 2010 as compared to the same period in 2011, demonstrating the usefulness of such sentinel controls. 5. Conclusions The present study specifically points to differences in contamination patterns in settings without a recently reported outbreak of gastroenteritis, i.e. either NoV contamination specifically detected bathrooms as was seen in hospital central kitchen departments or NoV contamination that is more dispersed, including food preparation areas, as observed in the in-patient units and catering companies. More attention should therefore be paid to the hurdles for not being in compliance to the badly needed hand washing practices. This will not only affect NoV spread, but also will affect other food-borne disease transmitted by food handlers such as Shiga toxigenic Escherichia coli or Salmonella or hospital acquired illnesses transmitted by health care personnel such as methicillin resistant Staphylococcus aureus (MRSA), Clostridium difficile or vancomycin-resistant Enterococci (VRE). Acknowledgements We would like to thank the Dutch contributors to Noronet, which allowed us to compare observed sequences with Dutch outbreak sequences, and Annelies Kroneman for her help in using the Noronet tools for analysis of sequence comparison. References Allegranzi, B., & Pittet, D. (2009). Role of hand hygiene in healthcare-associated infection prevention. Journal of Hospital Infection, 73(4), 305e315. http:// dx.doi.org/10.1016/j.jhin.2009.04.019. Epub 2009 Aug 31. Boone, S. A., & Gerba, C. P. (2007). Significance of fomites in the spread of respiratory and enteric viral disease. Applied and Environmental Microbiology, 73(6), 1687e1696. €gele, G., Tilburg, J. J., Vennema, H., et al. Boxman, I. L., Dijkman, R., te Loeke, N. A., Ha (2009). Environmental swabs as a tool in norovirus outbreak investigation, including outbreaks on cruise ships. Journal of Food Protection, 72(1), 111e119. €gele, G., te Loeke, N. A., & Koopmans, M. Boxman, I. L., Verhoef, L., Dijkman, R., Ha (2011). Year-round prevalence of norovirus in the environment of catering companies without a recently reported outbreak of gastroenteritis. Applied and Environmental Microbiology, 77(9), 2968e2974. Buccheri, C., Casuccio, A., Giammanco, S., Giammanco, M., La Guardia, M., & Mammina, C. (2007). Food safety in hospital: knowledge, attitudes and practices of nursing staff of two hospitals in Sicily, Italy. BMC Health Services Research, 7, 45.
EFSA [European Food Safety Authority]. (2011). Scientific Opinion on an update on the present knowledge on the occurrence and control of foodborne viruses. EFSA Journal, 9, 2190. http://dx.doi.org/10.2903/j.efsa.2011.2190. (96 pp.) Available from http://www.efsa.europa.eu/en/efsajournal/doc/2190.pdf (Accessed 17.02.12). FAO/WHO [Food and Agriculture Organization of the United Nations/World Health Organization]. (2008). Viruses in Food: Scientific advice to support risk management activities: Meeting report. In Microbiological risk assessment series, No. 13. Available from http://www.who.int/foodsafety/publications/micro/mra13/ en/index.html (Accessed 17.02.12). Gallimore, C. I., Taylor, C., Gennery, A. R., Cant, A. J., Galloway, A., IturrizaGomara, M., et al. (2006). Environmental monitoring for gastroenteric viruses in a pediatric primary immunodeficiency unit. Journal of Clinical Microbiology, 44(2), 395e399. Green, L. R., Radke, V., Mason, R., Bushnell, L., Reimann, D. W., Mack, J. C., et al. (2007). Factors related to food worker hand hygiene practices. Journal of Food Protection, 70(3), 661e666. Greig, J. D., & Lee, M. B. (2012). A review of nosocomial norovirus outbreaks: infection control interventions found effective. Epidemiology and Infection, 140(7), 1151e1160. http://dx.doi.org/10.1017/S0950268811002731. Epub 2012 Jan 4. Harris, J. P., Lopman, B. A., & O'Brien, S. J. (2010). Infection control measures for norovirus: a systematic review of outbreaks in semi-enclosed settings. Journal of Hospital Infection, 74(1), 1e9. http://dx.doi.org/10.1016/j.jhin.2009.07.025. Epub 2009 Oct 12. Heijne, J. C., Rondy, M., Verhoef, L., Wallinga, J., Kretzschmar, M., Low, N., et al. (2012). Quantifying transmission of norovirus during an outbreak. Epidemiology, 23(2), 277e284. http://dx.doi.org/10.1097/EDE.0b013e3182456ee6. Kroneman, A., Vennema, H., Deforche, K., van der Avoort, H., Penaranda, S., Oberste, M. S., et al. (2011). An automated genotyping tool for enteroviruses and noroviruses. Journal of Clinical Virology, 51(2), 121e125. ~ aduy, P., Park, G. W., Hall, A. J., Parashar, U. D., & Vinje , J. (2012). Lopman, B., Gastan Environmental transmission of norovirus gastroenteritis. Current Opinion in Virology, 2(1), 96e102. http://dx.doi.org/10.1016/j.coviro.2011.11.005. Epub 2011 Dec 11. Lopman, B., Vennema, H., Kohli, E., Pothier, P., Sanchez, A., Negredo, A., et al. (2004). Increase in viral gastroenteritis outbreaks in Europe and epidemic spread of new norovirus variant. Lancet, 363(9410), 682e688. Meakins, S. M., Adak, G. K., Lopman, B. A., & O'Brien, S. J. (2003). General outbreaks of infectious intestinal disease (IID) in hospitals, England and Wales, 1992-2000. Journal of Hospital Infection, 53(1), 1e5. Morter, S., Bennet, G., Fish, J., Richards, J., Allen, D. J., Nawaz, S., et al. (2011). Norovirus in the hospital setting: virus introduction and spread within the hospital environment. Journal of Hospital Infection, 77, 106e112. Pragle, A. S., Harding, A. K., & Mack, J. C. (2007). Food workers' perspectives on handwashing behaviors and barriers in the restaurant environment. Journal of Environmental Health, 69(10), 27e32. Teunis, P. F., Moe, C. L., Liu, P., Miller, S. E., Lindesmith, L., Baric, R. S., et al. (2008). Norwalk virus: how infectious is it? Journal of Medical Virology, 80(8), 1468e1476. Verhoef, L., Depoortere, E., Boxman, I., Duizer, E., van Duynhoven, Y., Harris, J., et al. (2008). Emergence of new norovirus variants on spring cruise ships and prediction of winter epidemics. Emerging Infectious Diseases, 14(2), 238e243. Verhoef, L., Jaramillo Gutierrez, G., Koopmans, M., & Boxman, I. L. A. (2013). Reported behavior, knowledge and awareness toward the potential for norovirus transmission by food handlers in Dutch catering companies and institutional settings in relation to the prevalence of norovirus. Food Control, 34, 420e427. Verhoef, L., Koopmans, M., Van Pelt, W., Duizer, E., Haagsma, J., Werber, D., et al. (2013). The estimated disease burden of norovirus in the Netherlands. Epidemiology and Infection, 141, 496e506. Widdowson, M. A., Cramer, E. H., Hadley, L., Bresee, J. S., Beard, R. S., Bulens, S. N., et al. (2004). Outbreaks of acute gastroenteritis on cruise ships and on land: identification of a predominant circulating strain of noroviruseUnited States, 2002. Journal of Infectious Diseases, 190(1), 27e36. Epub 2004 Jun 8. Wu, H. M., Fornek, M., Schwab, K. J., Chapin, A. R., Gibson, K., Schwab, E., et al. (2005). A norovirus outbreak at a long-term-care facility: the role of environmental surface contamination. Infection and Control Hospital Epidemiology, 26(10), 802e810. Xerry, J., Gallimore, C. I., Cubitt, D., & Gray, J. J. (2010). Tracking environmental norovirus contamination in a pediatric primary immunodeficiency unit. Journal of Clinical Microbiology, 48(7), 2552e2556. http://dx.doi.org/10.1128/ JCM.00066e10.
Contents lists available at ScienceDirect
Food Control journal homepage: www.elsevier.com/locate/foodcont
Environmental testing for norovirus in various institutional settings using catering companies as sentinels for norovirus prevalence among the general population €gele a, Kyara Klunder a, Ingeborg L.A. Boxman a, *, Linda Verhoef b, Geke Ha a b Nathalie A.J.M. te Loeke , Harry Vennema , Claudia C.C. Jansen a, Marion Koopmans b, c, 1 a
Laboratory for Feed and Food Safety, Food and Consumer Product Safety Authority (NVWA), P.O. Box 144, 6700 AC Wageningen, The Netherlands Laboratory for Infectious Diseases and Screening, National Institute of Public Health and the Environment (RIVM), P.O. Box 1, 3720 BA Bilthoven, The Netherlands c Department of Viroscience, ErasmusMC, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands b
a r t i c l e i n f o
a b s t r a c t
Article history: Received 13 January 2014 Received in revised form 5 June 2014 Accepted 14 June 2014 Available online 28 June 2014
Noroviruses (NoV) are among the most common causes of viral gastroenteritis (GE) worldwide and can be transmitted from person-to-person, via food or contaminated surfaces. The present study aimed to examine the prevalence of NoV RNA on surfaces in food preparation and sanitary areas in different health care settings and to compare the outcomes with the prevalence in nearby located catering companies, mainly restaurants, for general public, as sentinels. For this purpose, 1087 environmental swabs were taken for NoV analyses from surfaces in 241 institutional departments and 123 catering companies in The Netherlands without a recently reported outbreak of gastro-enteritis in high NoV season only. NoV RNA was detected in 15.1% of the 73 nonhospital health care institutions, 11.1% of the 54 hospital central kitchen departments, 14.9% of the 114 decentralized hospital kitchens (in-patient units) and 4.1% of the 123 nearby located catering companies. Twenty-five of the 49 positive environmental samples were genotyped by sequence analyses. In 7% of the investigated hospitals (4/58), NoV was detected in two or more departments. NoV prevalence was significantly lower in food preparation areas than in sanitary facilities (p < 0.05), but only in hospital central kitchen departments and non-hospital health care settings, and not in de-centralized hospital kitchens in in-patient units or in catering companies for the general public. This data suggests that there is a need for education on risks of NoV transmission by food handling of healthcare workers using in ward kitchen facilities. © 2014 Published by Elsevier Ltd.
Keywords: Swabs Non-outbreak Hospital Restaurant
1. Introduction Noroviruses (NoV) are among the most common causes of viral gastroenteritis (GE) worldwide. In the Netherlands, the incidence of NoV illness in 2009 was estimated to be 3800 cases per 100.000 inhabitants, with an estimated burden over 1600 disability-
* Corresponding author. Tel.: þ31 88 2230447. E-mail addresses: [email protected] (I.L.A. Boxman), linda.verhoef@ €gele), [email protected] rivm.nl (L. Verhoef), [email protected] (G. Ha (K. Klunder), [email protected] (N.A.J.M. te Loeke), [email protected] (H. Vennema), [email protected] (C.C.C. Jansen), [email protected], [email protected] (M. Koopmans). 1 Present address: Department of Viroscience, ErasmusMC, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands. http://dx.doi.org/10.1016/j.foodcont.2014.06.026 0956-7135/© 2014 Published by Elsevier Ltd.
adjusted life years (Verhoef, Koopmans, et al., 2013). The virus is highly infectious (Teunis et al., 2008) and often seen in outbreaks with high attack rates in health care settings, cruise ships, catered events, (EFSA, 2011; FAO/WHO, 2008; Meakins, Adak, Lopman, & O'Brien, 2003; Widdowson et al., 2004). NoV outbreaks are often prolonged, lasting longest in hospitals (19 days, range 6e92 days) and in nursing homes (16 days, range 3e44 days) in comparison to non-health care settings (7 days, range 1e26 days) (Harris, Lopman, & O'Brien, 2010). Outbreaks are recurring due to high levels of shedding in stool or vomitus and sometimes chronic shedding, especially in hospitals where people are vulnerable and there is a constant turnover of new susceptible people during outbreaks. Closure of wards is one of the control measures to control NoV spread in hospitals (Greig & Lee, 2012). Spreading is thought to be through person-to-person transmission, although environmental
I.L.A. Boxman et al. / Food Control 47 (2015) 98e102
contamination has been implicated in this transmission (Lopman et al. 2012). NoV have been demonstrated on different types of surfaces (floors, tables, doorknobs, handles, medical equipment, bed rails, carpets and curtains) in health care facilities, schools and food production facilities (Boone & Gerba, 2007; Gallimore et al., 2006; Morter et al., 2011; Wu et al., 2005). Recently, results of environmental sampling in catering companies suggested that health care institutions were more likely to be NoV contaminated than catering companies for general public (Boxman et al., 2011). This initiated further studies to examine the prevalence of NoV in different food preparation areas in various health care settings as part of assessment of risk of food contamination, and including an inventory on food handler's knowledge and working practices (Verhoef, Jaramillo Gutierrez, Koopmans, Boxman, 2013). The aim of the present study was to determine the NoV environmental presence during the winter season in healthcare settings. For the purpose of direct comparison, sampling in catering companies, mainly restaurants, nearby the institutions was performed. This was also done to reflect the background epidemiology of NoV in the general population, because of yearly differences in NoV activity (Lopman et al., 2004; Verhoef et al., 2008). In contrast to food handlers in institutions who work in a semi-closed settings prone to NoV outbreaks, food handlers in catering companies are part of the general public. Catering companies were therefore considered as sentinels for the NoV activity in the general population in this paper. In order to gain a deeper understanding of the potential sources of food-borne NoV outbreaks, outcomes were compared between food preparation areas and sanitary facilities within central and decentralized kitchens in health care institutions, and in comparison with outcomes of sampling in nearby public catering companies. 2. Materials and methods 2.1. Study rationale In the periods January 2010eApril 2010 and February 2011eMarch 2011, inspections were performed by officers of the Netherlands Food and Product Safety Authority (NVWA) in health care institutions, as well as in catering companies in the vicinity of these institutions, during the winter seasonal peak of viral gastrointestinal disease. The rationale for this was to capture environmental presence of NoV during the period of the year with most illness, in environments that may be relevant for the epidemiology of these viruses in health care settings. The healthcare settings included 58 hospitals (44.3%), 54 homes for the elderly (41.2%) as well as 11 nursing homes (8.4%). In hospitals, samples were collected in 54 central kitchen departments (32.1%) and 114 decentralized kitchens (67.9%). This was done to be able to discriminate food-handler introductions at these sites, that likely differ in the level of knowledge and stringency of food handling practices: staff in central kitchen department prepares food which is subsequently distributed to kitchens in-patient units (de-centralized kitchens), where other nursing staff reheat food to adequate temperatures, and serve meals and drinks. Sampling in nearby catering companies, mainly restaurants, was done to reflect the background epidemiology of NoV.
99
equipment, such as the grip of a refrigerator, knife or serving spoon, and the handle of a cutting or mixing machine in central kitchens, and surfaces from the microwave or the cutlery drawer in de-centralized kitchens. The second swab was used to collect a surface sample from the soap dispenser, from the pepper-and-salt set or from a food storage box in the refrigerator. In hospital in-patient units the handle of the dish washer was given as an alternative surface. The third swab was used to collect a surface sample from the flushing chain or knob and the toilet seat (both upper and undersurfaces) in the men's (employees') bathroom only, which was chosen for reasons of systematic sampling. Sampling was done by inspectors who had no knowledge about possible outbreaks related to the sites that they visited to avoid oversampling of known contamination events. Environmental swab samples were either kept at 4 C or kept frozen at 20 C during transport and stored frozen at 20 C until they were processed at the laboratory of the NVWA. Analyses were performed on coded samples. 2.3. Extraction and analyses of swab samples Environmental swabs were extracted and analysed for the presence of NoV RNA as described previously (Boxman et al., 2009; 2011) with some modifications. In brief, RNA extractions were performed using 30 ml glassmilk (MP Biomedical) per swab extraction with no difference in detection (data not shown), as rieux) was no longer available. normal non-magnetic silica (BioMe RNA samples were analyzed using two-step reverse transcription (RT) real time PCR. The N2pol assay targeting the polymerase region was used for the detection of NoV GII types as described (Boxman et al., 2011). For the detection of NoV GI types, the N1cap assay targeting the ORF1/ORF2 region was used in a two-step RT real time PCR format instead of the N1cap nested real-time RT-PCR assay (Boxman et al., 2009). 2.4. Sequencing and phylogenetic analyses NoV presumptive positive samples, as judged by the presence of a typical S -curve in the amplification plot, were re-amplified and sequenced as described previously (Boxman et al., 2011). NoV sequences detected in the present study were genotyped using the NoV typing tool (Kroneman et al., 2011) (http://www. rivm.nl/en/Topics/N/NoroNet/Databases/) and compared to Dutch NoV outbreak strains detected in humans involved in outbreaks and for which clinical samples had been sent in for diagnosis at the Institute for Public Health and the Environment in the Netherlands. Sequence comparisons were done if an overlapping region of at least 100 nucleotides (nt) in region 5373e5577 for ORF2 of GI types and in region 4299e4494 for ORF1 of GII types was available (Kroneman et al., 2011). A multiple alignment of selected sequences was made for each genotype or, if genotype II.4 was involved, each variant. The pairwise distance was computed using a 2-parameter Kimura nucleotide substitution model.
2.2. Sampling protocol
2.5. Statistical analyses
At each setting, two mixed surface swab samples were taken from surfaces in food preparation areas and one mixed surface swab sample was taken from the bathroom for NoV analyses, as described previously (Boxman et al., 2009; 2011). In brief, the first swab was used to collect a surface sample from a food preparation
Statistical analyses were performed using Fisher-P one tailed analyses to identify differences between settings and swabbing locations. If cell counts in a cross table did not include any observations, calculations were performed by artificially setting the number of the concerning cell to one instead of zero.
100
I.L.A. Boxman et al. / Food Control 47 (2015) 98e102
3. Results A total of 364 institutional departments or catering companies was included in the present study. In the period between January and April 2010, 73 non-hospital institutions and 71 nearby located catering companies were visited. In February and March 2011 54 hospital central kitchen departments, 114 decentralized kitchens in hospital in-patient units and 52 nearby located catering companies were visited. In central kitchen departments food is prepared which is subsequently distributed to kitchens in-patient units (the de-centralized kitchens), where nursing staff reheat food to adequate temperatures, and serve meals and drinks. In total, 1087 samples were available for testing. 3.1. Detection of NoV on environmental swabs in various institutional settings 3.1.1. Institutions all In total, 42 (5.8%) out of 718 environmental swabs from 34 (14.1%) out of 241 institutional settings tested positive for NoV (Table 1). At five in-patient units and one elderly home, samples of both food preparation areas and sanitary facilities tested positive. NoV was detected nearly two times more often on surfaces in sanitary facilities (20/236, 8.5%) compared to surfaces in food preparation areas (22/482, 4.6%) (p ¼ 0.04). 3.1.2. Non-hospital institutions Most of the 73 non-hospital institutions visited were elderly homes (74.0%) followed by nursing homes (15.1%). Thirteen (6.0%) out of 215 environmental swabs from 11 (15.1%) of 73 non-hospital institutions tested positive for NoV. NoV was three times more likely to be detected on surfaces in sanitary facilities (8/69, 11.6%) compared to surfaces in food preparation areas (5/146, 3.4%), (p ¼ 0.024). 3.1.3. Hospital central kitchen departments Six (3.7%) out of 162 environmental swabs from 6 (11.1%) out of 54 central kitchen departments tested positive for NoV. NoV was ten times more likely to be detected on surfaces in sanitary facilities (5/54, 9.3%) compared to surfaces in food preparation areas (1/108, 0.9%), (p ¼ 0.016).
3.1.4. Hospital decentralized kitchens (in-patient units) Twenty-three (6.7%) out of 341 environmental swabs from 17 (14.9%) out of 114 decentralized kitchens in in-patient units tested positive for NoV. There was no significant difference between the presence of NoV on surfaces in sanitary facilities (7/113, 6.1%) compared to presence of NoV on surfaces in food preparation areas (16/228, 7.0%), (p ¼ 0.48). 3.1.5. Between the different hospital departments The overall NoV prevalence was higher in hospital decentralized kitchens (p ¼ 0.12) than in hospital central kitchen departments, as measured using environmental swabs. Food preparation areas in hospital central kitchens were less frequently contaminated with NoV than food preparation areas in in-patient units (p ¼ 0.01). In 7% of the investigated hospitals (4/58), NoV was detected in two or more departments at the same day (Table 2). In six hospitals, NoV was detected in the central kitchen departments. Of these, two hospitals had NoV also detected in one of the de-centralized kitchens and one hospital had NoV detected in two decentralized kitchens. 3.2. Detection of NoV on environmental swabs in sentinel catering companies 3.2.1. Catering companies Most of the 123 catering companies selected as sentinels for NoV prevalence among the general population were restaurants (79.7%) followed by lunchrooms (11.4%) or take away counters (8.9%). Seven (1.9%) out of 369 environmental swabs from 5 (4.1%) out of 123 catering companies tested positive for NoV. At two catering companies, samples of both food preparation areas and sanitary facilities tested positive NoV. Remarkably, NoV was only detected on samples collected from catering companies in 2010, whereas none of the samples in 52 catering companies tested positive in 2011. There was no significant difference between the presence of NoV on surfaces in sanitary facilities (3/122, 2.5%) compared to presence of NoV on surfaces in food preparation areas (4/247, 1.6%), (p ¼ 0.42). Neither was there a significant difference when only samples collected in catering companies in 2010 were taken into account (p ¼ 0.40).
Table 1 Prevalence of NoV in institutional settings and nearby located catering companies. Year
NoV detected in any sample on company level
NoV detected in samples taken from surfaces
No./total
Kitchen
Bathroom %
No./total
All samples
No./total
%
%
No./total
%
2010
Elderly home Nursing home Other non-hospital institutions Subtotal Restaurant Catering/lunchroom Take-away Subtotal
10/54 1/11 0/8 11/73 4/57 0/6 1/8 5/71
18.5 9.1 0 15.1 7.0 0 13 7.0
5/107 0/23 0/16 5/146 3/112 0/12 1/19 4/143
4.7 0 0 3.4 2.7 0 5.3 2.8
7/52 1/9 0/8 8/69 2/57 0/6 1/7 3/70
13.0 11.1 0 11.6 3.5 0 14 4.3
12/159 1/32 0/24 13/215 5/169 0/18 2/26 7/213
7.5 3.1 0 6.0 3.0 0 7.7 3.3
2011
Hosp. central kitchen depart. Hosp. de-centr.kitchens dept. Subtotal Restaurant Catering/lunchroom Take-away Subtotal
6/54 17/114 23/168 0/41 0/8 0/3 0/52
11.1 14.9 13.7 0 0 0 0
1/108 16/228 17/336 0/82 0/16 0/6 0/104
0.9 7.0 5.1 0 0 0 0
5/54 7/113 12/167 0/41 0/8 0/3 0/52
9.3 6.1 7.2 0 0 0 0
6/162 23/341 29/503 0/123 0/24 0/9 0/156
3.7 6.7 5.8 0 0 0 0
All institutional settings All catering companies
34/241 5/123
14.1 4.1
22/482 4/247
4.6 1.6
20/236 3/122
8.5 2.5
42/718 7/369
5.8 1.9
I.L.A. Boxman et al. / Food Control 47 (2015) 98e102 Table 2 Hospitals with NoV detected in environmental samples at two or three departments. ID
Department
Location
Genogroup
Genotype
ZH04
Internal Medicine
Bathroom and kitchen Kitchen Bathoom Kitchen
GII
II.4 c.n.b.aa
GII GII GII
Non-typed GIIb Non-typed GII II.4 New Orleans
GII GII
II.4 c.n.b.a. Non-typed GII
GI and GII GI GII
I.2, non-typed GII Non-typed GI Non-types GII
ZH13
ZH19
ZH22
Neurology Central Kitchen dept In-patient (non specified) Central Kitchen dept Gynaecology Central Kitchen dept Oncology/Surgery Haematology
Bathroom Bathroom and kitchen Bathroom Kitchen Kitchen
a
c.n.b.a. GII.4 sequence of which the variant could not be assigned. Non-typed GII: sequence analyses was not successful as the signal of the presumptive positive samples was too weak. b
In 2010, the overall NoV prevalence appeared higher in nonhospital health care institutions than in nearby located catering companies, but the difference was not statistically significant (p ¼ 0.13). In 2011, NoV prevalence was, however, significantly higher in hospital central kitchen departments than in nearby located catering companies for the general population (p ¼ 0.02). 3.3. Typing of NoV RNA detected on environmental surfaces In total, 49 samples tested positive for NoV RNA. GI NoV strains were detected in three samples and GII NoV strains were detected in 46 samples. Sequence analyses were not always successful, as for some presumptive positive samples the signal was too weak. Twenty-five samples (51%) yielded a sequence suitable for genotyping and sequence analyses. GII.4 was the most frequently detected genotype (n ¼ 19). Nine of these, were sub-typed as GII.4.New Orleans 2009 variant. The remaining could not be assigned to a particular GGII.4 variant. GIIb (GII.P21) was detected three times and GIIg, GI.2 and GI.4 were detected one time each. For two hospitals, sequences were available for more than one location. This showed that the sequence types were different within single hospitals (ZH13 and ZH22), suggesting independent contamination events. 3.4. Comparative sequence analysis for environmental and reported outbreak samples Comparative sequence analysis was performed to identify possible clustering of the 25 sequences obtained from environmental samples and sequences obtained from reported outbreaks of gastroenteritis, i.e., clinical samples collected in the period from January 2008 through June 2013 (167 GII.b (GII.P21), 58 II.g, 1034 GGII.4 New Orleans 2009, and 70 GGII.4 sequences for which the variant could not be assigned). Unfortunately the genomic regions for environmental and diagnostic genogroup I detection did not overlap. Sequences of genogroup II strains as detected by environmental sampling were interspersed with those circulating in Dutch patients. Based on a minimum overlap of 100 nt, five clusters (GII.4 New Orleans 2009 (n ¼ 2), GII.b (or GII.P21) (n ¼ 2) and GII.g (n ¼ 1)), were identified consisting of 99.5% similar sequences and having at least one sequence detected in the present study and one or more Dutch outbreak clinical samples. These clusters were geographically dispersed and could not be linked based on available epidemiological information.
101
4. Discussion This study reports the prevalence of NoV in health care institutional departments in contrast to nearby located catering companies as sentinels for NoV prevalence among the general population. The findings suggest that rates of exposure in hospital based central kitchen staff is higher than in the sentinel catering company kitchen staff, pointing at potential cross transmissions of viruses within the hospital between in-patient units and central kitchen departments. NoV prevalence was significantly lower on swabs taken from food preparation areas than those from sanitary facilities (p < 0.05) in hospital central kitchen departments and non-hospital health care settings only, and not in de-centralized hospital kitchens in in-patient units or in catering companies for the general public. The finding is suggestive of a higher compliance with good hand hygiene by staff of the former two settings. The differences in the estimated NoV prevalence could be biased as all users of the employee's bathroom will touch the flush chain or knob, whereas such an equivalent, an always-touched site in the kitchen is not available. For this reason sampling surfaces was standardized to the extend possible, and twice as many samples were taken from the food handling areas than from bathroom areas. The results showed that, despite the possible bias in favour of the bathroom areas, part of the food handling areas were detected positive, which is an unwanted situation. The higher compliance with good hand hygiene in staff in central kitchen departments of hospitals or non-hospital health care settings may be due to more appropriate education in food safety. This is important as here food is prepared for vulnerable persons. In contrast, staff working in decentralized kitchens in in-patient units most often will be medically trained nursing staff. In a study of Buccheri and coworkers, more than 80% of the respondent nurses involved in food service functions did not attend any educational course on food hygiene (Buccheri et al., 2007). Specifically concerning potential transmission of viral food-borne illness by food handlers several gaps in education and training were identified, demonstrating that knowledge on NoV was low, although awareness of NoV was significantly higher among food handlers in institutional settings than in catering companies (Verhoef, Jaramillo Gutierrez, et al. 2013). Besides knowledge, health care workers (Allegranzi & Pittet, 2009) as well as food handlers in catering companies (Green et al., 2007; Pragle, Harding, & Mack, 2007) encounter difficulties in complying with hand hygiene indications. The close contact between nursing staff and patients might contribute to NoV spread over the wards (Heijne et al. 2012). Transmission of NoV via hand contact was suggested by the types of contaminated sites in hospitals identified (Morter et al. 2011). Such hot spots of contamination were patient monitoring equipment, computers and notes trolleys at nurses stations', dispensers and hand and grab rails, even post cleaning. Recently, outcomes of questionnaires and environmental testing have been combined to identify independent determinants of surface NoV contamination (Verhoef, Jaramillo Gutierrez, et al. 2013). The results showed that the likelihood for such a contamination was increased when being situated in an institutional setting or when a positive answer was given to the question whether a person would continue food handling despite sick with vomiting complaints. In the present study, seven percent of the hospitals had NoV detected in two or more departments at the same day, most likely due to independent contamination events. That multiple strains can circulate or co-circulate during the NoV season during hospital outbreaks was demonstrated by analyses of fecal samples from patients or medical staff (Morter et al., 2011; Xerry, Gallimore, Cubitt, & Gray, 2010). These studies were,
102
I.L.A. Boxman et al. / Food Control 47 (2015) 98e102
however, restricted to wards, and did not include the central kitchen departments. As expected from the many hospital outbreaks reported (reviewed by Harris et al., 2010; Greig & Lee, 2012), the present study demonstrated that NoV was more present in hospital departments than in the catering companies serving as their sentinels. Remarkably the prevalence of NoV in 2010 in catering companies was much higher (7%) than in 2011 (none), with no other explanation than that the prevalence in the catering companies reflecting yearly differences in NoV activity (Lopman et al., 2004; Verhoef et al., 2008). According to the laboratory reports of NoV in the United Kingdom and the Netherlands, (http://www. hpa.org.uk/webc/HPAwebFile/HPAweb_C/1287143931777), http:// www.rivm.nl/Onderwerpen/V/Virologische_weekstaten/Rapportages/ Open_rapportages_(virologische_weekstaten/) NoV activity was much higher in 2010 as compared to the same period in 2011, demonstrating the usefulness of such sentinel controls. 5. Conclusions The present study specifically points to differences in contamination patterns in settings without a recently reported outbreak of gastroenteritis, i.e. either NoV contamination specifically detected bathrooms as was seen in hospital central kitchen departments or NoV contamination that is more dispersed, including food preparation areas, as observed in the in-patient units and catering companies. More attention should therefore be paid to the hurdles for not being in compliance to the badly needed hand washing practices. This will not only affect NoV spread, but also will affect other food-borne disease transmitted by food handlers such as Shiga toxigenic Escherichia coli or Salmonella or hospital acquired illnesses transmitted by health care personnel such as methicillin resistant Staphylococcus aureus (MRSA), Clostridium difficile or vancomycin-resistant Enterococci (VRE). Acknowledgements We would like to thank the Dutch contributors to Noronet, which allowed us to compare observed sequences with Dutch outbreak sequences, and Annelies Kroneman for her help in using the Noronet tools for analysis of sequence comparison. References Allegranzi, B., & Pittet, D. (2009). Role of hand hygiene in healthcare-associated infection prevention. Journal of Hospital Infection, 73(4), 305e315. http:// dx.doi.org/10.1016/j.jhin.2009.04.019. Epub 2009 Aug 31. Boone, S. A., & Gerba, C. P. (2007). Significance of fomites in the spread of respiratory and enteric viral disease. Applied and Environmental Microbiology, 73(6), 1687e1696. €gele, G., Tilburg, J. J., Vennema, H., et al. Boxman, I. L., Dijkman, R., te Loeke, N. A., Ha (2009). Environmental swabs as a tool in norovirus outbreak investigation, including outbreaks on cruise ships. Journal of Food Protection, 72(1), 111e119. €gele, G., te Loeke, N. A., & Koopmans, M. Boxman, I. L., Verhoef, L., Dijkman, R., Ha (2011). Year-round prevalence of norovirus in the environment of catering companies without a recently reported outbreak of gastroenteritis. Applied and Environmental Microbiology, 77(9), 2968e2974. Buccheri, C., Casuccio, A., Giammanco, S., Giammanco, M., La Guardia, M., & Mammina, C. (2007). Food safety in hospital: knowledge, attitudes and practices of nursing staff of two hospitals in Sicily, Italy. BMC Health Services Research, 7, 45.
EFSA [European Food Safety Authority]. (2011). Scientific Opinion on an update on the present knowledge on the occurrence and control of foodborne viruses. EFSA Journal, 9, 2190. http://dx.doi.org/10.2903/j.efsa.2011.2190. (96 pp.) Available from http://www.efsa.europa.eu/en/efsajournal/doc/2190.pdf (Accessed 17.02.12). FAO/WHO [Food and Agriculture Organization of the United Nations/World Health Organization]. (2008). Viruses in Food: Scientific advice to support risk management activities: Meeting report. In Microbiological risk assessment series, No. 13. Available from http://www.who.int/foodsafety/publications/micro/mra13/ en/index.html (Accessed 17.02.12). Gallimore, C. I., Taylor, C., Gennery, A. R., Cant, A. J., Galloway, A., IturrizaGomara, M., et al. (2006). Environmental monitoring for gastroenteric viruses in a pediatric primary immunodeficiency unit. Journal of Clinical Microbiology, 44(2), 395e399. Green, L. R., Radke, V., Mason, R., Bushnell, L., Reimann, D. W., Mack, J. C., et al. (2007). Factors related to food worker hand hygiene practices. Journal of Food Protection, 70(3), 661e666. Greig, J. D., & Lee, M. B. (2012). A review of nosocomial norovirus outbreaks: infection control interventions found effective. Epidemiology and Infection, 140(7), 1151e1160. http://dx.doi.org/10.1017/S0950268811002731. Epub 2012 Jan 4. Harris, J. P., Lopman, B. A., & O'Brien, S. J. (2010). Infection control measures for norovirus: a systematic review of outbreaks in semi-enclosed settings. Journal of Hospital Infection, 74(1), 1e9. http://dx.doi.org/10.1016/j.jhin.2009.07.025. Epub 2009 Oct 12. Heijne, J. C., Rondy, M., Verhoef, L., Wallinga, J., Kretzschmar, M., Low, N., et al. (2012). Quantifying transmission of norovirus during an outbreak. Epidemiology, 23(2), 277e284. http://dx.doi.org/10.1097/EDE.0b013e3182456ee6. Kroneman, A., Vennema, H., Deforche, K., van der Avoort, H., Penaranda, S., Oberste, M. S., et al. (2011). An automated genotyping tool for enteroviruses and noroviruses. Journal of Clinical Virology, 51(2), 121e125. ~ aduy, P., Park, G. W., Hall, A. J., Parashar, U. D., & Vinje , J. (2012). Lopman, B., Gastan Environmental transmission of norovirus gastroenteritis. Current Opinion in Virology, 2(1), 96e102. http://dx.doi.org/10.1016/j.coviro.2011.11.005. Epub 2011 Dec 11. Lopman, B., Vennema, H., Kohli, E., Pothier, P., Sanchez, A., Negredo, A., et al. (2004). Increase in viral gastroenteritis outbreaks in Europe and epidemic spread of new norovirus variant. Lancet, 363(9410), 682e688. Meakins, S. M., Adak, G. K., Lopman, B. A., & O'Brien, S. J. (2003). General outbreaks of infectious intestinal disease (IID) in hospitals, England and Wales, 1992-2000. Journal of Hospital Infection, 53(1), 1e5. Morter, S., Bennet, G., Fish, J., Richards, J., Allen, D. J., Nawaz, S., et al. (2011). Norovirus in the hospital setting: virus introduction and spread within the hospital environment. Journal of Hospital Infection, 77, 106e112. Pragle, A. S., Harding, A. K., & Mack, J. C. (2007). Food workers' perspectives on handwashing behaviors and barriers in the restaurant environment. Journal of Environmental Health, 69(10), 27e32. Teunis, P. F., Moe, C. L., Liu, P., Miller, S. E., Lindesmith, L., Baric, R. S., et al. (2008). Norwalk virus: how infectious is it? Journal of Medical Virology, 80(8), 1468e1476. Verhoef, L., Depoortere, E., Boxman, I., Duizer, E., van Duynhoven, Y., Harris, J., et al. (2008). Emergence of new norovirus variants on spring cruise ships and prediction of winter epidemics. Emerging Infectious Diseases, 14(2), 238e243. Verhoef, L., Jaramillo Gutierrez, G., Koopmans, M., & Boxman, I. L. A. (2013). Reported behavior, knowledge and awareness toward the potential for norovirus transmission by food handlers in Dutch catering companies and institutional settings in relation to the prevalence of norovirus. Food Control, 34, 420e427. Verhoef, L., Koopmans, M., Van Pelt, W., Duizer, E., Haagsma, J., Werber, D., et al. (2013). The estimated disease burden of norovirus in the Netherlands. Epidemiology and Infection, 141, 496e506. Widdowson, M. A., Cramer, E. H., Hadley, L., Bresee, J. S., Beard, R. S., Bulens, S. N., et al. (2004). Outbreaks of acute gastroenteritis on cruise ships and on land: identification of a predominant circulating strain of noroviruseUnited States, 2002. Journal of Infectious Diseases, 190(1), 27e36. Epub 2004 Jun 8. Wu, H. M., Fornek, M., Schwab, K. J., Chapin, A. R., Gibson, K., Schwab, E., et al. (2005). A norovirus outbreak at a long-term-care facility: the role of environmental surface contamination. Infection and Control Hospital Epidemiology, 26(10), 802e810. Xerry, J., Gallimore, C. I., Cubitt, D., & Gray, J. J. (2010). Tracking environmental norovirus contamination in a pediatric primary immunodeficiency unit. Journal of Clinical Microbiology, 48(7), 2552e2556. http://dx.doi.org/10.1128/ JCM.00066e10.