Hostname: page-component-7c8c6479df-fqc5m Total loading time: 0 Render date: 2024-03-28T11:21:44.470Z Has data issue: false hasContentIssue false

Epidemiology of anthroponotic and zoonotic human cryptosporidiosis in England and Wales, 2004–2006

Published online by Cambridge University Press:  12 July 2010

R. M. CHALMERS*
Affiliation:
Cryptosporidium Reference Unit, Public Health Wales Microbiology ABM, Singleton Hospital, Swansea, UK
R. SMITH
Affiliation:
Centre for Epidemiology and Risk Analysis, Veterinary Laboratories Agency – Weybridge, New Haw, Addlestone, Surrey, UK
K. ELWIN
Affiliation:
Cryptosporidium Reference Unit, Public Health Wales Microbiology ABM, Singleton Hospital, Swansea, UK
F. A. CLIFTON-HADLEY
Affiliation:
Food and Environment Safety Department, Veterinary Laboratories Agency – Weybridge, New Haw, Addlestone, Surrey, UK
M. GILES
Affiliation:
Food and Environment Safety Department, Veterinary Laboratories Agency – Weybridge, New Haw, Addlestone, Surrey, UK
*
*Author for correspondence: Dr R. M. Chalmers, Cryptosporidium Reference Unit, Public Health Wales Microbiology, Singleton Hospital, SwanseaSA2 8QA, UK. (Email: Rachel.chalmers@wales.nhs.uk)
Rights & Permissions [Opens in a new window]

Summary

In order to monitor epidemiological trends, Cryptosporidium-positive samples (n=4509) from diarrhoeic patients were typed. Compared to the previous 4 years, the proportion of Cryptosporidium hominis cases in 2004–2006 increased to 57·3%, while 38·5% were C. parvum. The remaining 4·2% cases included mixed C. parvum and C. hominis infections, C. meleagridis, C. felis, C. ubiquitum and a novel genotype. When the typing results were combined with enhanced surveillance data to monitor risk exposures, C. hominis was linked to urban dwelling, previous diarrhoea in the household, any travel especially abroad, and using a swimming or paddling pool. C. parvum was linked to having a private water supply, contact with surface water, visiting or living on a farm, and contact with farm animal faeces. The proportion of laboratory-confirmed indigenous cases acquired from direct contact with farm animals was estimated to be 25% for C. parvum and 10% of all reported Cryptosporidium cases.

Type
Original Papers
Copyright
Copyright © Cambridge University Press 2010

INTRODUCTION

The protozoan parasite Cryptosporidium is a common cause of acute human gastroenteritis. In England and Wales, an annual mean of 4189 laboratory-confirmed cases of cryptosporidiosis were reported to national surveillance (range 3010–5863) in the 10 years to the end of 2008 [1], an annual incidence of around 8 cases/100 000 population. Infecting species are not identified in routine diagnosis but specialist testing of a representative proportion of cases has shown that Cryptosporidium parvum and Cryptosporidium hominis account for over 96% of the cases typed to the species level [Reference Chalmers2]. Between 2000 and 2004 these occurred in approximately equal proportions nationally but with seasonal and geographic variation. Spring peaks were due to C. parvum, and C. hominis was more prevalent in the late summer and early autumn [Reference Chalmers2]. C. parvum predominated in Wales and the South West of England and C. hominis in more eastern regions [Reference Chalmers2]. Sources of infection and risk factors vary according to infecting species; for C. hominis these are anthroponotic and for C. parvum both anthroponotic and zoonotic [Reference Hunter3, Reference Hunter4]. Genetic linkage between C. parvum isolates from some of the sporadic human cases described here and their suspected farm animal sources has been investigated by sequencing part of the 60-kDa glycoprotein (GP60) gene [Reference Smith5]. A high proportion of isolates at both the farm level and individual case/animal contact level were indistinguishable at this locus [Reference Smith5].

Recognized outbreaks, which account for about 10% of all cases routinely reported to national surveillance, are often linked to settings such as open farms and contact with young ruminants in particular [Reference Chalmers and Giles6]. However, the proportion of sporadic cases, which account for the vast majority of human cryptosporidiosis, that can be attributed to animal sources and zoonotic transmission is not known. Although prevention of acquisition and spread will have some commonality between C. parvum and C. hominis, such as personal hygiene, targeting of interventions will be different for zoonotic and anthroponotic transmission.

To investigate the ongoing distribution of C. parvum and C. hominis, we typed isolates from cases submitted between 2004 and 2006 to the species level in the first instance. To investigate exposures to known risk factors, we further investigated a subset of cases by establishing an enhanced surveillance scheme in three regions, and linking infecting species identification with analysis of case exposure data. To estimate the burden of zoonotically acquired cryptosporidiosis from farmed animals, we applied the proportion shown by analysis at the GP60 gene to be linked to this source [Reference Smith5] to the national data.

METHODS

National typing for epidemiological purposes

National typing of Cryptosporidium-positive faecal samples from publicly funded primary diagnostic laboratories in England and Wales was undertaken as described previously [Reference Chalmers2]. In addition, between January 2004 and December 2006, all 28 publicly funded primary diagnostic laboratories serving the population defined by 41 local authorities (LAs) in three study areas within the Government Office Regions of Wales, the South West of England and the East of England, were asked to send all Cryptosporidium-positive faecal samples for typing. A minimum dataset was systematically collected for all samples on a structured submission form, including the patient demographics, clinical details, specimen date, history of recent foreign travel and whether the case was considered to be part of a family or household cluster or a general outbreak [Reference Wall7].

Methods to identify Cryptosporidium spp. were as described previously [Reference Chalmers2]. Briefly, oocysts were separated by flotation from faecal debris, disrupted by incubation at 100°C for 60 min and DNA extracted by proteinase K digestion and spin-column filtration (QiAMP DNA mini kit, Qiagen, UK). Cryptosporidium spp. were identified by PCR–RFLP of the Cryptosporidium oocyst wall protein (COWP) gene [Reference Spano8] in the first instance. Isolates where no amplicons were obtained or equivocal results generated using the COWP PCR were further tested by nested PCR–RFLP of the SSU rRNA gene [Reference Xiao9]. To obtain quality assurance of typing results, a subset of PCR products was analysed by bi-directional DNA sequence analysis (Geneservice, UK). The results were compared to sequences published in the National Institutes of Health's National Center for Biotechnology Information GenBank database using the Basic Local Alignment Search Too1 [10].

To describe the national trends in the epidemiology of C. parvum and C. hominis cases, data were analysed in Epi Info, version 6 (Centres for Disease Control and Prevention, USA). To investigate representativeness of submission of typing, comparison was made with national reports collected via CoSurv, a set of interconnected database modules for communicable disease reporting [Reference Henry11].

Enhanced surveillance in 41 LAs in three study areas, November 2004–2006

Confirmed cryptosporidiosis cases are routinely reported by the laboratory to the relevant LA Environmental Health Department (EHD) for follow-up using a structured questionnaire [Reference Chalmers12]. For the duration of this study, a modified exposure questionnaire was used, with additional questions relating to direct and indirect (environmental) animal contact.

Prior to analysis, the data were error-checked for outliers and impossible values and a random selection of 10% cases validated, by cross-checking the forms. For the purpose of categorical analyses, patients' age was allocated to 10-year categories. Key words were used to extract data from free-text fields (e.g. other symptoms). Occupation information was grouped according to risks of transmission; those that worked on a farm, with food, with children (e.g. carers at home, nursery staff, school staff), or had close contact with other people (e.g. care workers, hospital). Other occupations were coded as ‘no risk’.

Descriptive analysis was completed using MS Excel. The number of cases from different areas was compared against population figures from the 2001 census [13]. To compare the Cryptosporidium spp. distribution by urban classification, an urban index was used. The settlement types (urban, town and fringe, village, hamlet, isolated dwellings) in each region were combined with population density classification (sparse or less sparse) to give eight indices [14]. The indices for each region were ordered and replaced by a numeric ID and the mode of the ordinal urban index data was used for each LA. To compare the species-specific risk contacts for C. parvum and C. hominis, χ2 analysis was completed using the Yates corrected version for 2×2 tables and linear trend for age groups and month in Epi Info 6. Missing values or answers stated as ‘don't know’ were omitted from the cross-tabulations.

To estimate the proportion of sporadic zoonotic cryptosporidiosis in laboratory-confirmed and reported cases acquired directly from farmed animals, non-outbreak (i.e. apparently sporadic) C. parvum cases were studied. The proportion reporting contact with farm animals was multiplied by 71·4% which was the proportion of cases found by analysis of the GP60 gene to have isolates indistinguishable from suspected farm animal sources [Reference Smith5]. This estimate was then applied to national surveillance data.

RESULTS

National trends in C. parvum and C. hominis

A total of 4509 samples were submitted for typing from England and Wales, representing 38·1% of the 11 830 reports to national surveillance over the 3 years (Fig. 1). Submission numbers reflected reports to national surveillance (Fig. 1) although some regions were better represented than others (Table 1). Of these samples, 130 were either not confirmed as Cryptosporidium (n=87) or were repeat samples (n=43). A total of 4379 were initial samples from confirmed cases; 1686 (38·5%) were C. parvum, 2509 (57·3%) were C. hominis and 184 (4·2%) other or unidentified species or genotypes. These were 33 C. meleagridis, 26 C. felis, seven co-infections of C. parvum and C. hominis, two C. ubiquitum (synonymous with cervine genotype), one novel genotype (Genbank accession numbers HM191264 and HM191258) and 115 were not typable as no PCR amplicons were produced or the reaction was too weak to identify.

Fig. 1. Temporal distribution of Cryptosporidium in England and Wales, 2004–2006 (Cryptosporidium Reference Unit and Health Protection Agency data).

Table 1. Annual distribution by region of Cryptosporidium laboratory reports to national surveillance, samples submitted for typing and confirmed C. parvum and C. hominis cases in England and Wales, 2004–2006

HPA, Health Protection Agency.

* More cases were typed than reported because more samples were submitted for typing than cases reported via CoSurv.

The distribution of C. parvum and C. hominis varied annually: in 2004 and 2006 there were 1·1 and 1·2 C. hominis cases for each C. parvum case, respectively, while in 2005 the ratio was 2·3; there were two large drinking waterborne outbreaks of C. hominis in 2005 (Table 2). C. parvum cases peaked in April or May each year and C. hominis in September, although numbers generally were elevated from August to the end of the year (Fig. 1). Compared with the North West where the species distribution was equal (Table 1), Wales and the South West consistently had more C. parvum than C. hominis cases although this was only significant in Wales (χ2=6·25, d.f.=1, P=0·12). More C. hominis than C. parvum was found in the East Midlands (χ2=24·48, d.f.=1, P=0·001), Yorkshire & the Humber (χ2=42·92, d.f.=1, P<0·001), the South East (χ2=78·90, d.f.=1, P<0·0001) (Table 1) and the East of England (χ2=17·93, d.f.=1, P<0·001). In the East of England the high proportion of C. hominis cases in 2004 was not observed in 2005 or 2006. The distribution in the West Midlands was similar to the North West (χ2=1·05, d.f.=1, P=0·305). In London and the North East the number of samples submitted was small (Table 1).

Table 2. Cryptosporidium spp. where identified in outbreaks of cryptosporidiosis in England and Wales January 2000 to December 2003 (UK Cryptosporidium Reference Unit, Health Protection Agency and Public Health Wales data)

HPA, Health Protection Agency; NT, not typable.

* Concurrent viral and Giardia outbreak.

Cases were mainly children aged <10 years (Fig. 2). C. hominis was more common than C. parvum in all age groups, with a significant linear trend for 10-year age bands (χ2=64·55, d.f.=8, P<0·001), and particularly in the <10 and 20–39 years age groups and especially in females in their thirties (Fig. 2).

Fig. 2. Age and sex distribution of C. parvum and C. hominis cases in England and Wales, 2004–2006. M, Male; F, female.

Foreign travel was reported by 421 (9·6%) cases, mainly to Spain (n=54), India (n=42), Pakistan (n=39) and Turkey (n=28). Of the cases reporting foreign travel, 67·0% were C. hominis, significantly more than C. parvum2=31·03, d.f.=1, P=0·000). Foreign travel peaked in August and September, reported by 79 and 141 cases, respectively, accounting for 52·2% of all travel-related cases. Between eight and 31 cases reported foreign travel during each of the other 10 months of the year.

A total of 508 (11·6%) cases belonged to locally or nationally recognized outbreaks (Table 2). Two outbreaks were linked to mains drinking water, one in Wales and one in South East England, and both were caused by C. hominis. Sixteen outbreaks were linked to swimming pools, only one of which was C. parvum with the remaining outbreaks solely or mainly C. hominis. All of the nine farm-setting or animal contact-related outbreaks were caused by C. parvum. Of the four institutional outbreaks, two were C. parvum (both set at outdoor/activity centres) and two were C. hominis (both in childcare settings). Of the three international outbreaks, two were C. parvum and one was C. hominis, although the cause of these outbreaks is not known.

Enhanced surveillance and risk exposures in three study areas

A total of 883 questionnaires were collected, of which 790 were not linked to an outbreak and were analysed for demographics (Table 3) and risk contacts (Table 4). A subset of 635 case isolates was submitted for typing, themselves forming a subset of the national dataset. Of these, 544 were non-outbreak cases, comprising 255 (46·8%) C. hominis, 252 (46·5%) C. parvum, two co-infections with C. hominis and C. parvum, five C. meleagridis, three C. felis and 27 untypable isolates. Species-linked demographics and risk factors were analysed for the individual C. parvum and C. hominis cases only (Tables 3 and 4, showing χ2 and P values).

Table 3. Cryptosporidiosis, C. parvum and C. hominis non-outbreak case demographics and clinical symptoms in three study areas

* Missing values or answers stated as ‘don't know’ omitted from the cross tabulations.

Mode of all wards within each Local Authority.

Table 4. Reported risk factors for cryptosporidiosis, C. parvum and C. hominis non-outbreak cases in three study areas

* Missing values or answers stated as ‘don't know’ were omitted from the cross tabulations.

Although more cases were from Wales (Table 3), the population-based submission rate was similar across the three study areas, with an annual mean of 7/100 000 population, although Taunton Deane LA in South West England had the highest incidence (annual mean 25/100 000 population). C. hominis cases were more likely than C. parvum cases to be from less sparsely populated LAs (P<0·001) (Table 3). Cases peaked overall in the autumn with over half of the cases (58·9%) with onset dates between August and November (Table 3). These were mostly C. hominis cases. C. parvum cases peaked in April.

There was no difference in the proportion of male and female cases, and although C. hominis was most common in females, this was not significant (P=0·512) (Table 3). The peak age group of the cases was children aged <10 years, with a significant linear trend for 10-year age bands (P<0·001) (Table 3).

The main symptoms reported were diarrhoea (93·3%) and abdominal pain (73·9%), sometimes accompanied by vomiting (54·2%) and/or nausea (45·1%); these were not linked to infecting species. ‘Other symptoms’ reported by 30·6% cases included fatigue (n=48) and fever (n=89) and were more commonly reported by C. hominis cases (P=0·01). There were 8·9% of cases admitted to hospital, for a median of 1 day (range <1–21 days). One fifth of cases (20·9%) reported other household members with diarrhoea in the 2 weeks before illness, and these were significantly more likely to have C. hominis (P=0·002). Cases that had travelled abroad (17·5%) had mainly visited Spain (n=46), France (n=17) or Turkey (n=11), and were significantly more likely to be C. hominis (P<0·001).

Cases had diverse occupations but the highest proportion (8·1%) had contact with children (Table 4). There was no significant difference between the Cryptosporidium spp. detected in these cases and the species in those with other occupations (Table 4). Only three cases had an occupational farm contact risk (two were farmers and one was a veterinary student), all three were infected with C. parvum. Of 35 children whose parents/guardians were farmers 24 had typing results; 22 were infected with C. parvum, one with C. meleagridis, and one had a co-infection of C. parvum and C. hominis. Of a further 15 cases who lived on farm, 12 were typed and 11 were infected with C. parvum and one with C. hominis.

Over a quarter of all cases (26·1%) had visited a farm in the 2 weeks before illness, and were more likely to be positive for C. parvum (P<0·001) (Table 4). Of all these cases, 71·0% reported contact with farm animals, mostly sheep (59·2%) but also cattle (37·4%). Data on farmed animal contact have been published elsewhere [Reference Smith5]. Contact with animal faeces was reported by 14% of all cases, and comprised farming activities such as mucking out and lambing, looking after pets, and indirect contact like picnicking and walking through fields. These cases were more likely to be C. parvum (P<0·001) (Table 4). Generally, cases that had farm animal or environmental (private or surface water or animal faeces) contact were significantly more likely to be C. parvum than the rest of the study population (P<0·001 and 0·035, respectively) (Table 4).

A total of 3·1% cases had a private water supply (either solely or as both private and mains supply), mainly from springs but also from boreholes and wells. These cases were more likely to be C. parvum (P=0·037) (Table 4). One fifth (20·4%) of cases had contact with surface water (e.g. swimming, working or playing in a river, stream, ditch, pond or water trough), and these cases were more likely to be C. parvum (P=0·002) (Table 4).

Almost half of the cases had swum in a swimming pool in the 2 weeks prior to illness, and were significantly more likely to be C. hominis (P<0·001) (Table 4). A number of cases reported swimming in the same swimming pools during a similar time period, and it is possible that some of these could have been unrecognized outbreaks.

The majority (65%) of cases had contact with pets (Table 4), mainly with dogs, cats, or a combination of both. Data on pet contact from this study has been published elsewhere [Reference Smith24]. Only 4% of cases had contact with ‘zoo’ animals, mainly noted as ‘various’ animal types or horses. These cases had mainly visited the large zoos in each study area.

The proportion of C. parvum cases reporting contact with farmed animals was 34·5% (Table 4). We estimated that 71·4% of these (24·6%) C. parvum cases can be linked to direct contact with farm animals. National surveillance showed 11 830 Cryptosporidium reports during the 3-year study period, an annual mean of 3943. From our national typing estimate, 1518 (38·5%) cases are C. parvum and we estimated that around 373 (24·6%) reported cases of C. parvum were attributable to contact with farm animals per year.

DISCUSSION

We have described the national distribution and trends in human cryptosporidiosis caused by C. parvum and C. hominis in England and Wales during 2004–2006. Furthermore, these data have been augmented and analysed with risk exposure data in three areas of England and Wales, and the proportion of reported cases acquired directly from farmed animals during that period has been estimated. However, the number of cases fluctuates and so may the attributable fraction from different exposures [Reference Lake25].

The geographic and age-related trends are not dissimilar to those reported in previous years [Reference Chalmers2, Reference Hunter3], although data for some regions are more sparse than previously. C. hominis (57·3% of cases typed) was more prevalent than C. parvum (38·5%), which contrasts with previous data from 2000 to 2003 when the ratio nationally was closer to 1 [Reference Chalmers2]. Although large waterborne outbreaks of C. hominis in the autumn of 2005 may account for some of this increase [Reference Neira-Munoz, Okoro and McCarthy15, Reference Mason17], the reasons are unclear. The autumnal increase is subsequent to peak foreign travel reports [Reference Chalmers2] and community spread through swimming pools may be involved [Reference Hunter3, Reference Lidington20].

The small spring peaks were mainly attributable to C. parvum cases and since 2001 have been smaller than previously [Reference Chalmers2, Reference Lake25]. The seasonality is similar to that seen in Scotland [Reference Pollock26] but contrasts with data from Ireland which show no autumn peak in Cryptosporidium reports [Reference Garvey and McKeown27] and a clear predominance of C. parvum [Reference Zintl28]. Calving, lambing and run-off are thought to contribute to zoonotic sources in the spring, but improvements in catchment protection and drinking water treatment have led to a reduction in cases in the UK [Reference Lake25, Reference Sopwith29].

Adding additional exposure data to the model suggested that direct farm animal contact accounts for about one quarter of C. parvum cases or about 10% of all reported cryptosporidiosis cases. Other direct zoonotic sources may include contact with farm animal faeces, or contact with companion animals, although we have previously shown there is little evidence for pets as a risk to public health from Cryptosporidium [Reference Smith24]. Indirect zoonotic transmission is also a factor in cryptosporidiosis caused by C. parvum (e.g. through private water supplies or environmental contact) [Reference Pollock26].

The distribution and descriptive epidemiology (person, time, place) of C. parvum and C. hominis in the enhanced surveillance was broadly as expected from the continuing national data [2, data from this study]. Although univariate exposures were analysed in our study, multiple potential risks were reported by 68·1% cases, with up to seven recorded per patient (mean 2·4). The most common combination was companion animal contact and use of swimming pools (8·6%). Nonetheless, the risk exposures were broadly similar to those identified in an earlier case-control study in Wales and the North West of England [Reference Hunter3] and an environmental and socioeconomic factors study which also found that C. hominis was more prevalent in more densely populated areas [Reference Lake30].

There is potential bias in any laboratory-based surveillance system at all levels of the ascertainment pyramid [Reference Wall7]. Our studies relied on primary diagnostic laboratory referrals to the Cryptosporidium Reference Unit (CRU), comparison with CoSurv data, and for the enhanced surveillance, reporting to LAs as well as case responses to the questionnaires. Of 141 laboratories surveyed in England and Wales in 2006, 105 (74·5%) tested all community samples for Cryptosporidium, as did 21 of the 28 (75·0%) laboratories serving the enhanced surveillance study areas, while the others apply selection criteria for testing including patient's age or a report of farm animal contact (CRU, unpublished observations) which might have some small bias for our study. Although all 28 laboratories sent samples for typing, submission rates were variable nationally. Because not all laboratories report cases via CoSurv, and not all send samples for typing, there is a consequence on the estimates of percent of samples typed. A further bias that might have affected the enhanced surveillance was that where human cases had contact with pets and farm animals, the questionnaire may have been more likely to have been submitted to the study by the EHD.

Foreign travel reports were more common in the enhanced surveillance than in the national data, probably because this was actively sought in the former. The destinations were also different; in the enhanced surveillance Spain, France and Turkey were the most common whereas in the national data these were Spain, India and Pakistan. This difference is probably due to the differing ethnicity of the populations in the three study areas compared to national data [13].

Symptoms described were similar to other case series captured in the same way, i.e. from patients seeking medical assistance, with a high proportion reporting abdominal pain, vomiting and nausea [Reference Hunter3]. Here, we additionally identified other symptoms (fever, fatigue) more commonly reported by C. hominis cases than C. parvum. Acute clinical correlates appear to be more common and varied with C. hominis than C. parvum [Reference Chalmers and Davies31], especially with the C. hominis GP60 subtype family Ib [Reference Cama32] which predominates in the UK [Reference Chalmers33]. It is thus possible that C. hominis cases may be more likely to present for medical attention, and that C. parvum cases are under-represented in this study design. The patients would not have known which species they were infected with when they completed the questionnaire.

Almost a quarter of the cases, especially those with C. hominis, noted other household members with diarrhoea in the 2 weeks prior to illness, indicating the importance of person-to-person spread. As most cases stated more than one risk contact it is difficult to quantify direct attribution to person-to-person contact, but further exploration is required as this indicates an important aspect of public health intervention. C. hominis also predominated in cases who used swimming pools which may be linked to the high prevalence in young children who are at risk in this setting [34].

The significant proportion of C. parvum cases living on a farm or having contact with farm animal faeces concurs with previous findings [Reference Hunter3, Reference Goh35], although regular exposure may lead to the development of acquired immunity resulting in milder disease [Reference Chappell36].

It has been estimated that 1% of the population of England use private water supplies [Reference Rutter37] but in the enhanced surveillance, 3·1% used private drinking-water supplies. The excess of cases may reflect geographical distribution of the study population or an increased risk. These cases were mainly C. parvum indicating contamination of private water supplies with animal faeces is common.

There is good evidence that the epidemiology of each of the two main Cryptosporidium spp. infecting the population of England and Wales is different, as are the risk factors for acquisition, and this needs to be recognized in national surveillance. Zoonotic spread of C. parvum from farmed animals contributes to the burden of illness. The autumn peak in infections, mainly caused by C. hominis is currently not controlled and the main drivers for this need to be identified so that interventions can be implemented. Person-to-person spread of both C. parvum and C. hominis has not been properly evaluated for sporadic cases and prevention through this route is likely to be an important control measure.

ACKNOWLEDGEMENTS

We thank the staff at the local authorities for administering and returning the enhanced surveillance questionnaires, and at the diagnostic laboratories for sending Cryptosporidium-positive stools for typing. We also thank Anne Thomas, Nigel Crouch and Rachael Seymour at the Cryptosporidium Reference Unit for helping to maintain the national collection. This study received ethical approval from Thames Valley Multi-Centre Research Ethics Committee. This research project was funded by the Department for Environment, Food & Rural Affairs (Defra) under project OZ0407. The views expressed here are those of the authors and not necessarily those of Defra.

DECLARATION OF INTEREST

None.

References

REFERENCES

2.Chalmers, RM, et al. Long-term Cryptosporidium typing reveals the aetiology and species-specific epidemiology of human cryptosporidiosis in England and Wales, 2000 to 2003. Eurosurveillance 2009; 14: pii=19086.CrossRefGoogle ScholarPubMed
3.Hunter, PR, et al. Case-control study of sporadic cryptosporidiosis with genotyping. Emerging Infectious Diseases 2004; 10: 12411249.CrossRefGoogle ScholarPubMed
4.Hunter, PR, et al. Subtypes of Cryptosporidium parvum in humans and disease risk. Emerging Infectious Diseases 2007; 13: 8288.CrossRefGoogle ScholarPubMed
5.Smith, RP, et al. Investigation of farms linked to human patients with cryptosporidiosis in England and Wales. Preventative Veterinary Medicine 2010; 94: 9–17.CrossRefGoogle ScholarPubMed
6.Chalmers, RM, Giles, M. Zoonotic cryptosporidiosis in the UK. Journal for Applied Microbiology. Published online: 29 April 2010. doi:10.1111/j.1365-2672.2010.04764.x.Google Scholar
7.Wall, PG, et al. Food poisoning: notifications, laboratory reports, and outbreaks – where do the statistics come from and what do they mean? Communicable Disease Report 1996; 6: R93–R100.Google ScholarPubMed
8.Spano, F, et al. PCR-RFLP analysis of the Cryptosporidium oocyst wall protein (COWP) gene discriminates between C. wrairi and C. parvum, and between C. parvum isolates of human and animal origin. FEMS Microbiology Letters 1997; 150: 209217.CrossRefGoogle Scholar
9.Xiao, L, et al. Identification of species and sources of Cryptosporidium oocysts in storm waters with a small subunit rRNA based diagnostic and genotyping tool. Applied and Environmental Microbiology 2000; 66: 54925498.CrossRefGoogle ScholarPubMed
10.National Institutes of Health. National Center for Biotechnology Information. GenBank database (http://www.ncbi.nlm.nih.gov/genbank/). Accessed 8 April 2010.Google Scholar
11.Henry, R. CoSurv: a regional computing strategy for communicable disease surveillance. PHLS Microbiology Digest 1996; 13: 2628.Google Scholar
12.Chalmers, RM, et al. Laboratory ascertainment of Cryptosporidium and local authority public health policies for the investigation of sporadic cases of cryptosporidiosis in two regions of the United Kingdom. Communicable Disease and Public Health 2002; 5: 114118.Google ScholarPubMed
13.Office for National Statistics (ONS). Neighbourhood statistics (http://www.neighbourhood.statistics.gov.uk/dissemination/). Accessed 8 April 2010.Google Scholar
14.Office for National Statistics (ONS). Rural/urban definition and LA classification (http://www.statistics.gov.uk/geography/nrudp.asp#products). Accessed 8 April 2010.Google Scholar
15.Neira-Munoz, E, Okoro, C, McCarthy, ND. Outbreak of waterborne cryptosporidiosis associated with low oocyst concentrations. Epidemiology and Infection 2007; 135: 11591164.CrossRefGoogle ScholarPubMed
16.Carnicer-Pont, D, et al. An outbreak of cryptosporidiosis in Wales, November 2005. Eurosurveillance 2005; 10: pii=2854.Google ScholarPubMed
17.Mason, BW, et al. A Cryptosporidium hominis outbreak in North-West Wales associated with low oocyst counts in treated drinking water. Journal of Water and Health 2010; 8: 299310.CrossRefGoogle ScholarPubMed
18.Anon. Surveillance of waterborne disease outbreaks summary of 2004. Communicable Disease Report 2006; 16 (http://www.hpa.org.uk/cdr/archives/2006/cdr1506.pdf). Accessed 23 April 2010.Google Scholar
19.Anon. Surveillance of waterborne disease in England and Wales: summary of 2005. Communicable Disease Report 2006; 16 (http://www.hpa.org.uk/cdr/archives/2006/cdr2806.pdf). Accessed 23 April 2010.Google Scholar
20.Lidington, I, et al. Cryptosporidium outbreak in travellers returning to South West London, Autumn 2005, with subsequent contamination of public swimming pools and community spread. Abstract Health Protection Conference, Warwick, September 2006.Google Scholar
21.Davison, H.Report on an outbreak of cryptosporidiosis associated with a swimming pool complex (dissertation). Faculty of Medical and Human Sciences, Manchester, England, UK: University of Manchester, 2007.Google Scholar
22.Anon. Infectious intestinal disease associated with farm visits. Communicable Disease Report 2005; 18 (http://www.hpa.org.uk/cdr/archives/2005/cdr1805.pdf). Accessed 23 April 2010.Google Scholar
23.Hoek, MR, et al. Outbreak of Cryptosporidium parvum among children after a school excursion to an adventure farm, South West England. Journal of Water and Health 2008; 6: 333338.CrossRefGoogle Scholar
24.Smith, RP, et al. Investigation of the role of companion animals in the zoonotic transmission of cryptosporidiosis. Zoonoses and Public Health 2009; 56: 2433.CrossRefGoogle ScholarPubMed
25.Lake, IR, et al. Cryptosporidiosis decline after regulation, England and Wales, 1989–2005. Emerging Infectious Diseases 2007; 13: 623625.CrossRefGoogle ScholarPubMed
26.Pollock, KGJ, et al. Spatial and temporal epidemiology of sporadic human cryptosporidiosis in Scotland. Zoonoses and Public Health. Published online: 23 July 2009. doi:10.1111/j.1863-2378.2009.01247.Google Scholar
27.Garvey, P, McKeown, P. Epidemiology of human cryptosporidiosis in Ireland, 2004–2006: analysis of national notification data. Eurosurveillance 2009; 14: pii=19128.CrossRefGoogle ScholarPubMed
28.Zintl, A, et al. The prevalence of Cryptosporidium species and subtypes in human faecal samples in Ireland. Epidemiology and Infection 2008; 137: 270277.CrossRefGoogle ScholarPubMed
29.Sopwith, W, et al. The changing epidemiology of cryptosporidiosis in North West England. Epidemiology and Infection 2005; 133: 785793.CrossRefGoogle ScholarPubMed
30.Lake, IR, et al. Case-control study of environmental and social factors influencing cryptosporidiosis. European Journal of Epidemiology 2007; 22: 805811.CrossRefGoogle ScholarPubMed
31.Chalmers, RM, Davies, AP. Mini-review: clinical cryptosporidiosis. Experimental Parasitology 2010; 124: 138146.CrossRefGoogle Scholar
32.Cama, VA, et al. Cryptosporidium species and subtypes and clinical manifestations in children, Peru. Emerging Infectious Diseases 2008; 14: 15671574.CrossRefGoogle ScholarPubMed
33.Chalmers, RM, et al. Geographic linkage and variation in Cryptosporidium hominis. Emerging Infectious Diseases 2008; 14: 496498.CrossRefGoogle ScholarPubMed
34.Anon. Review of outbreaks of cryptosporidiosis in swimming pools. Final report to DEFRA: Drinking Water Inspectorate 2000 project number DWI 70/2/131. (http://www.dwi.gov.uk/research/reports/DWI70_2_131_swimming%20pools.pdf). Accessed 23 April 2010.Google Scholar
35.Goh, S, et al. Risk factors for sporadic cryptosporidiosis in North Cumbria, North West of England 1996 to 2000. Emerging Infectious Diseases 2004; 9: 10051017.Google Scholar
36.Chappell, CL, et al. Infectivity of Cryptosporidium parvum in healthy adults with pre-existing anti-C. parvum serum immunoglobulin G. American Journal of Tropical Medicine and Hygiene 1999; 60: 157164.CrossRefGoogle ScholarPubMed
37.Rutter, M, et al. A survey of the microbiological quality of private water supplies in England. Epidemiology and Infection 124, 417425.CrossRefGoogle Scholar
Figure 0

Fig. 1. Temporal distribution of Cryptosporidium in England and Wales, 2004–2006 (Cryptosporidium Reference Unit and Health Protection Agency data).

Figure 1

Table 1. Annual distribution by region of Cryptosporidium laboratory reports to national surveillance, samples submitted for typing and confirmed C. parvum and C. hominis cases in England and Wales, 2004–2006

Figure 2

Table 2. Cryptosporidium spp. where identified in outbreaks of cryptosporidiosis in England and Wales January 2000 to December 2003 (UK Cryptosporidium Reference Unit, Health Protection Agency and Public Health Wales data)

Figure 3

Fig. 2. Age and sex distribution of C. parvum and C. hominis cases in England and Wales, 2004–2006. M, Male; F, female.

Figure 4

Table 3. Cryptosporidiosis, C. parvum and C. hominis non-outbreak case demographics and clinical symptoms in three study areas

Figure 5

Table 4. Reported risk factors for cryptosporidiosis, C. parvum and C. hominis non-outbreak cases in three study areas