15 January 2009
Long-term Cryptosporidium typing reveals the aetiology and species-specific epidemiology of human cryptosporidiosis in England and Wales, 2000 to 2003
To improve understanding of the aetiology and epidemiology of human cryptosporidiosis, over 8,000 Cryptosporidium isolates were submitted for typing to the species level over a four year period. The majority were either Cryptosporidium parvum (45.9%) or Cryptosporidium hominis (49.2%). Dual infection occurred in 40 (0.5%) cases and six other known Cryptosporidium species or genotypes were found in 67 (0.9%) cases. These were Cryptosporidium meleagridis, Cryptosporidium felis, Cryptosporidium canis, and the Cryptosporidium cervine, horse and skunk genotypes. The remaining 3.5% were not typable. Epidemiology differed between infecting species. C. parvum cases were younger, although C. hominis was more prevalent in infants under one year and in females aged 15 to 44 years. Spring peaks in cases reported to national surveillance were due to C. parvum, while C. hominis was more prevalent during the late summer and early autumn as well as in patients reporting recent foreign travel. Temporal and geographical differences were observed and a decline in C. parvum cases persisted from 2001. Typing of isolates allowed outbreaks to be more clearly delineated, and demonstrated anthroponotic spread of C. parvum as well as C. hominis. Our findings suggest that national surveillance for Cryptosporidium should be conducted at the species level.
The aetiological agent of the diarrhoeal disease cryptosporidiosis in humans has been traditionally ascribed to the protozoan parasite Cryptosporidium parvum, on the basis of microscopical identification of oocysts or detection of oocyst wall antigens in faeces, and the assumption that all oocysts detected were monospecific . However, C. parvum variants, recognised initially phenotypically (designated human or H and cattle or C types) and latterly through genomic studies, segregate into two genotypes (1 and 2), of which genotype 1 is host-adapted for humans . This is now assigned species status and called Cryptosporidium hominis and genotype 2 remains C. parvum . Although these are the most commonly found species in human cryptosporidiosis worldwide, the distribution varies temporally and geographically . Six other Cryptosporidium species have also been found in this host (Cryptosporidium meleagridis, Cryptosporidium felis, Cryptosporidium canis, Cryptosporidium suis, Cryptosporidium muris and Cryptosporidium andersoni), as have C. hominis monkey genotype, C. parvum mouse genotype and the Cryptosporidium cervine, chipmunk genotype I, skunk, horse and rabbit genotypes [4, 5, 6]. The site-specific occurrence and pathogenicity of these unusual Cryptosporidium species/genotypes in humans appears to depend on a combination of endemnicity, exposure and parasite-related factors rather than host immune status .
Discrimination between Cryptosporidium species/genotypes is not possible by methods traditionally applied in routine diagnostic laboratories and national cryptosporidiosis surveillance is usually undertaken and reported without account of the aetiology. One exception is Scotland, where reference laboratory typing results have been incorporated in national surveillance since 2004 . In England and Wales, the parasite is routinely identified at the genus level only  and surveillance data show that in the ten years between 1998 and 2007, the number of laboratory confirmed cases reported annually ranged from 3,010 to 5,863 . More cases are reported in one to two year old children and cases are unevenly distributed over time, with peaks in the spring and autumn .
Although data have been published on the species identification and occurrence of Cryptosporidium spp. in human isolates, numbers studied are often small and / or from selected patient groups, and are rarely representative of community cases routinely seeking medical assistance . The distribution of C. parvum and C. hominis cases mainly in England between 1998 and 1999, has been shown to vary geographically and temporally . C. parvum was detected in 56.1%, C. hominis in 41.7%, and the remaining 2.2% comprised C. meleagridis, C. felis, C. andersoni, C. canis, C. suis and the Cryptosporidium cervine type, and samples containing both C. parvum and C. hominis . While these studies contributed to knowledge of the epidemiology and transmission of Cryptosporidium species, national surveillance remained at the genus-level.
In order to improve our understanding of the aetiology and epidemiology of human cryptosporidiosis, and investigate changes over time, an on-going, representative, national collection of Cryptosporidium oocysts was established for the whole of England and Wales in January 2000. Here we describe the establishment, baseline aetiology and epidemiological analysis of the national collection for the first four years (2000 to 2003), and assess the value of Cryptosporidium typing for epidemiological and surveillance purposes.
Between January 2000 and December 2003, faecal samples in which Cryptosporidium was detected during routine diagnosis of diarrhoeal disease in publicly funded laboratories through out England and Wales were referred to the Cryptosporidium Reference Unit (CRU) in Swansea for typing to the species level. Briefly, oocysts were separated from faecal debris by salt flotation, and disrupted by boiling, and DNA was extracted by a spin column technique (QIAamp DNA Mini Kit, Qiagen Ltd.) as described previously . The Cryptosporidium oocyst wall protein (COWP) gene was investigated for all isolates using polymerase chain reaction – restriction fragment length polymorphism (PCR-RFLP)  and isolates where no amplicons were obtained were further tested by PCR-RFLP analysis of the small sub-unit (SSU) rRNA gene . If amplicons were still not obtained, the stool was examined by microscopy following modified Ziehl-Neelsen staining of fixed smears  or immunofluorescence staining (Crypto-Cel, TCS Water Sciences) according to the manufacturer’s instructions. PCR products with equivocal or unusual RFLP profiles were purified (Qiaquick, Qiagen Ltd), sequenced in both directions (GeneService Ltd) and edited, consensus sequences compared with published sequences in the GenBank database using the National Institutes of Health National Centre for Biotechnology Information basic local alignment search tool (http://www.ncbi.nlm.nih.gov/BLAST). Sequences were verified and >99.5% similarity, in the region targeted by the PCR, to a published sequence was considered homologous.
Patient demographics (locality, date of birth or age, and sex), clinical details, specimen collection 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, was collected from the diagnostic laboratory on a form submitted with each sample and outbreaks verified with the investigating authority. For specimens where the collection date was missing, a proxy date was calculated by deducting from the date of receipt at the CRU the modal time delay for this interval (five days). Cases were geographically located using the Government Office Region of the submitting laboratory. Countries visited by patients reporting recent foreign travel were grouped according to the health advice provided in Health Information for Overseas Travel .
To confirm that the submitted samples were representative, the dataset was compared by specimen date, patient age and sex distribution with primary diagnostic laboratory surveillance reports to the Health Protection Agency, using the Mantel-Haenszel version of the chi-squared test and Mann-Whitney two sample test for sex and age distribution respectively.
Differences in demographic data and patient history of first time, confirmed cases of C. parvum and C. hominis in the whole dataset were compared by univariate logistic regression analysis and age distribution investigated using the Mann-Whitney two-sample test. Patients infected with C. parvum were designated as “cases” while patients infected with C. hominis were designated as “controls”. Further analyses were undertaken separately for each infecting species. Patients co-infected with both species were excluded from these analyses. All statistical analyses were undertaken using EpiInfo (Version 6, Centers for Disease Control and Prevention, Atlanta, GA) and STATA 7 (Stata Corporation, College Station, TX).
During the four year period from 1 January 2000 to 31 December 2003, a total of 8,075 faecal specimens were received from 133 primary diagnostic laboratories throughout England and Wales, representing 44.3% of the 18,235 Cryptosporidium cases reported to national surveillance over the same time period. The monthly distribution of submitted isolates reflected the number of cases reported to national surveillance (Figure 1).
Figure 1. Monthly total numbers of cases of Cryptosporidium in humans in England and Wales, 2000 to 2003, comparing laboratory surveillance reports and C. parvum and C. hominis cases identified in the sub-set submitted for typing
The specimen collection date was available for 7,732 of the 8,075 (95.8%) specimens, and the time delay to date of receipt by the CRU ranged from 1 to 311 days (mean = 6 days, mode = 5 days, median = 5 days). The age of the patient was known for 8,003 (99.1%) specimens. The youngest patient was two months old and the oldest 98 years (mean = 16 years, mode = 1 year, median = 9 years). This was not significantly different from the cases reported to national surveillance (Mann-Whitney two-sample test =-0.031, df=1, p=0.9752).
Of the 8,075 specimens received by the CRU, 3,965 (49.1%) specimens were from males, 4,072 (50.4%) were from females and for just 38 (0.5%) the sex of the patient was not known. This was not significantly different from the ratio of male to female cases (1:1.02) reported to national surveillance (λ2=0.06, P>0.05, df=1). Foreign travel was indicated on the submission form for 1,049 (13.0%) specimens in the CRU collection compared with 3% of cases reported to national surveillance.
Microbiological and genotyping results
Cryptosporidium was confirmed by microscopy or PCR in 7,829 (97.0%) specimens. Of the remaining 246 (3.0%) specimens, seven were identified by microscopy as Cyclospora cayetanensis, 44 were insufficient in volume for confirmation and the remaining 195 contained yeast cells, mushroom spores, pollen grains or unidentified staining artefacts.
Of the 7,829 confirmed isolates, 7,560 (96.6%) were typable by PCR-RFLP. The positivity rate for COWP PCR-RFLP was 88% on initial test, rising to 92% when a repeat test was included. The overall positivity rate rose to 96.6% following testing of COWP negative samples by SSU rDNA PCR-RFLP. The remaining 3.4% of specimens were confirmed by microscopy, but were not amplified or showed equivocal results (e.g. bands too faint to assign to species/genotypes or multiple bands present) by the PCR methods described here. A total of 141 repeat specimens were received from 70 patients. None of these sequential samples demonstrated a change in the Cryptosporidium species from that detected in the initial specimen.
Of the 7,758 first specimens from each patient, 3,817 (49.2%) were C. hominis, 3,564 (45.9%) were C. parvum, 40 (0.5%) were dual infections with C. parvum and C. hominis, other Cryptosporidium species/genotypes were identified in 67 (0.9%) and 270 (3.5%) were not typable. The unusual species/genotypes were C. meleagridis (n=56), C. felis (n=4), Cryptosporidium cervine genotype (n=4), C. canis (n=1), Cryptosporidium horse genotype (n=1) and Cryptosporidium skunk genotype (n=1). The finding of the horse and skunk genotypes has been described by Robinson et al.,  and the epidemiology of cases other than C. parvum and C. hominis is being prepared for publication elsewhere.
The patient demographics for C. parvum and C. hominis are compared in Table 1. The mean age of C. parvum cases (15 years, range 0 to 92 years, median 8 years, mode 1 year) was lower than that of C. hominis cases (17 years, range 0 to 97 years, median 9 years, mode 1 year) (Mann-Whitney two-sample test=9.69, df=1, p=0.002). Both species were linked to young age (0 to 9 years). There was an excess of C. parvum in 10 to 19 year olds, whereas C. hominis was common in adults, particularly those between 30 and 39 years of age. More detailed examination of the age-related data (Figure 2) showed that C. hominis was also more prevalent than C. parvum in infants under one year of age. Although C. parvum and C. hominis cases overall did not differ with regard to sex, this was affected by age with more C. parvum in young boys and more C. hominis, especially in females, in the 30 and 39 years age group (Figure 2). There was no difference in the distribution of these Cryptosporidium species in immunocompetent and immunocompromised patients.
Table 1. Comparison of demographics and history of 7,381 cases with Cryptosporidium parvum and Cryptosporidium hominis in England and Wales over four years from 2000 to 2003: analysis using case-control methodology
Figure 2. Age and sex distribution of Cryptosporidium parvum and Cryptosporidium hominis cases in England and Wales over four years from 2000 to 2003 (n= 7,381)
More patients with C. parvum belonged to recognised outbreaks but fewer belonged to family or household clusters where C. hominis was more common.
C. parvum cases were less likely to have reported travel outside the United Kingdom (UK) prior to illness than C. hominis cases. The locations visited were Europe (207 C. parvum; 378 C. hominis), Indian subcontinent (18 C. parvum; 42 C. hominis), North Africa and the Middle East (18 C. parvum; 42 C. hominis), sub-Saharan and southern Africa (13 C. parvum and 27 C. hominis), the Caribbean (6 C. parvum and 18 C. hominis), South East Asia and Far East (4 C. parvum and 2 C. hominis), North America, Australia and New Zealand (5 C. parvum and 7 C. hominis), Central America (3 C. parvum and 8 C. hominis), South America ( 2 C. parvum and 6 C. hominis), mixed locations or country not stated (29 C. parvum and 36 C. hominis).
Regional differences were observed when compared with the West Midlands which had similar numbers of C. parvum and C. hominis cases. Government Office Regions on the eastern side of the country (i.e. London, South East, Yorkshire and the Humber, North East, East of England and the East Midlands) were more likely to have increased numbers of C. hominis while Wales, on the western side, had more C. parvum cases. The proportion of C. parvum and C. hominis cases in the North West and the South West were similar.
The annual proportion of cases of C. hominis (49.2% in 2000, 57.5% in 2001, 46.0% in 2002 and 45.1% in 2003) and C. parvum (47.1% in 2000, 35.7% in 2001, 49.3% in 2002 and 49.9% in 2003) was approximately equal each year, with the exception of 2001 when there was a much lower proportion of C. parvum cases, particularly in the spring. Because of this change over time, and the epidemiological differences highlighted here between C. parvum and C. hominis, the following data are presented annually and separately for each infecting Cryptosporidium species.
All ages were affected by the spring decline in C. parvum cases in 2001 (Figure 3), and the spring peak was only partially restored in 2002 and 2003 (Figure 1). During each of the four years most isolates were received during September, this peak being mainly composed of C. hominis and to a lesser extent C. parvum (Figure 1).
Figure 3. Age distribution of Cryptosporidium parvum and Cryptosporidium hominis cases in England and Wales, by month, in 2000, 2001, 2002 and 2003 (n=7,381)
The spring peak in C. parvum was almost exclusively composed of indigenous cases, whereas the late summer / autumn C. hominis peak included patients who had reported foreign travel (Figure 4). In 2003 there was a substantial peak in C. parvum, probably linked to an outbreak originating among holiday makers in Majorca (Table 2). The younger ages particularly were affected by the unusual autumnal peak in C. parvum in 2003 (Figure 4).
Table 2. Cryptosporidium species identified in outbreaks of cryptosporidiosis in England and Wales, from January 2000 to December 2003
Figure 4. Distribution of Cryptosporidium hominis and Cryptosporidium parvum in cases reporting travel and not reporting travel outside the United Kingdom over four years from 2000 to 2003
Specimens were received from 508 cases linked to 29 locally or nationally recognised outbreaks of cryptosporidiosis during the four year period (Table 2). Outbreaks were caused by C. hominis (13 outbreaks), C. parvum (10 outbreaks) and both species were detected in six outbreaks. Public drinking water supplies were associated with two outbreaks caused by C. hominis, one caused by C. parvum and one outbreak where both species were detected. Three outbreaks were linked to private water supplies and all three were caused by C. parvum.
Although more swimming pool-associated outbreaks in England and Wales were caused by C. hominis (n=6) than C. parvum (n=2), the largest indigenous outbreak linked to a swimming pool was caused by C. parvum. Both species were detected in three outbreaks linked to swimming pools. All swimming pool-associated cryptosporidiosis outbreaks except one were at indoor pools, the most common type in the UK. Furthermore, two international outbreaks were investigated, both linked to hotel pools in Majorca, one was caused by C. hominis in 2000 and the other by C. parvum in 2003, which may have influenced the subsequent increase in C. parvum in the autumn that year (Figure 1).
Water features were associated with two outbreaks, one linked to a fountain in a public park caused by C. hominis and the other associated with an interactive water feature at an adventure park featuring a petting zoo caused by both C. parvum and C. hominis. Three outbreaks linked to open or residential farms were caused by C. parvum. One outbreak linked to environmental contact was caused by C. parvum and C. hominis. Three outbreaks at day care nurseries were caused by C. hominis.
In this paper, long term, Cryptosporidium species-specific epidemiological analysis is described for the first time at a national level, demonstrating that aetiological identification of a large proportion of cryptosporidiosis cases is possible, and furthermore, enhances the surveillance data provided by routine genus-level reporting. The epidemiology of C. parvum and C. hominis differs, and there is evidence for distinct sources and transmission routes. C. parvum infections occur all year round but mainly in the spring, although the spring peaks have declined since 2001. Outbreaks caused by C. parvum were linked to farm visits, environmental contact, drinking and recreational water. C. hominis cases occur mainly in the summer and autumn, in infants under one year of age and in adult females between 30 and 39 years, and in people who travelled abroad. C. hominis outbreaks were linked to day care nurseries, drinking and recreational waters. Some of these risks were identified in a case-control study of sporadic cases undertaken in 2001 which found contact with farmed animals as the significant risk factor for C. parvum and travel abroad, contact with another person with diarrhoea and changing young children’s nappies to be significant risk factors for C. hominis . Thus the epidemiology supports human transmission of C. hominis, and both zoonotic and anthroponotic transmission of C. parvum. Although household clusters of cases were more commonly caused by C. hominis, C. parvum was also involved. Human transmission of C. parvum was also demonstrated in outbreaks linked to indoor swimming pools, indicating a human source of contamination and infection with this species.
Multi-locus genotyping of a subset of C. parvum isolates from this collection, analysed with enhanced patient data, has demonstrated a predominance of some alleles linked to anthroponotic transmission, and others linked to zoonotic transmission .
The temporal distribution, with C. parvum predominating in the spring and C. hominis in the autumn, which has been reported in some other temperate climates  are shown to have changed over time in England and Wales. Although the national reduction in the spring peak in 2001, driven by C. parvum, showed strong association with control measures for the foot and mouth disease epidemic that year  it clearly continued after the control measures were lifted and data from this archive for the North West of England demonstrated links to improvements in drinking water quality . This demonstrates the value of typing isolates in identifying interventions for disease reduction. The regional differences observed, reflecting population densities, have been further explored in analysis of the socio-economic risk factors . This showed significant association between C. hominis and higher social economic status, young children and urban areas, and for C. parvum faecal application to land .
Although the cases in our dataset were representative of those reported to national surveillance, a higher proportion of our cases reported foreign travel. This is not considered to be submission bias but due to improved reporting since our submission form actively sought this information whereas it is reported passively to national surveillance. Travel data is under-reported in national surveillance and to a lesser extent to CRU, compared with enhanced data collection in a case-control study . Travel-related cryptosporidiosis was mainly caused by C. hominis but this is influenced by the most frequently visited areas and differences may reflect variations in the endemic Cryptosporidium species of the host countries (about which little is known), or differences in behaviour and exposure during travel to different destinations. It is also possible that outbreaks among holiday makers may occur independently of the indigenous population, particularly if hotel swimming pools are involved . It appears that foreign travel has a role in initiating the autumn peak, although this has not been investigated and should be studied further to investigate community spread and identify risk factors and interventions for disease reduction.
The typing methods used in this study enabled investigation of a vast number of specimens with very little loss in resolution . Potential mis-identifications in the COWP assay that are currently known include the Cryptosporidium rabbit genotype confounding for C. hominis  and the mouse genotype mistaken for C. parvum . Enhanced testing of a subset of our isolates indicates that the rabbit genotype is a rare human infection (unpublished data) and there is only one report from elsewhere of human infection with the mouse genotype . Cryptosporidium species/genotypes not amplified by the COWP primers were further investigated at the SSU rRNA locus. PCR amplification of isolates not typable in this algorithm may have been inhibited by substances in the faecal samples or represent genotypes not amenable to amplification with the primer sets used in this study.
We identified 40 (0.5%) cases with dual C. parvum and C. hominis infections. This proportion is comparable with that found previously in England . The disadvantage of any PCR-based system using common primers is the probable under-ascertainment of dual or multiple alleles within a sample. However, a subset of our isolates have been tested using separate species-specific primers and by multi-locus typing and showed little evidence of mixed infection . The likelihood of dual infections is also driven by the endemicity of the parasite and exposure, as higher proportions have been detected in high-prevalence regions of the UK . Unlike studies investigating only immunocompromised patients, we investigated both immunocompetent and immunocompromised populations and found no difference in the distribution of C. parvum and C. hominis, and other species/genotypes were not more prevalent in immunocompromised patients (unpublished data).
Cryptosporidium species-specific risk factors have been identified as a result of this work. Although zoonotic risks regarding handling animals have been well described, indirect exposures are less well documented and in January 2004, the focus of national collection was changed to a sentinel laboratory scheme for the study of zoonotic cryptosporidiosis. The work presented in this paper facilitates the development of more rapid methods for Cryptosporidium species identification is facilitated by this work, not only providing an archive of material for assay development and evaluation but also by identifying that the key targets in the UK, and probably elsewhere in northern Europe, are C. parvum and C. hominis. In conclusion, species-level analyses are critical to the investigation and explanation of changes in incidence over time.
We thank UK Cryptosporidium Reference Unit staff Stephen Hadfield, Guy Robinson and Cathy Bentley for scientific support, Sian Wood, David Gomez and Rachael Seymour for administrative and technical assistance. Gordon Nichols, Health Protection Agency, for providing national surveillance data, Health Protection Teams and Units in England and Wales for providing outbreak data, Roland Salmon CDSC Wales for helpful comments on the manuscript, and primary diagnostic laboratories in England and Wales for contributing specimens to the National collection.
This study was funded by Department for Environment, Food and Rural Affairs and administered by the Drinking Water Inspectorate.
1. Cacciò SM, Pozio E. Advances in the epidemiology, diagnosis and treatment of cryptosporidiosis. Expert Rev Anti Infect Ther. 2006;4(3):429–43.
2. Xiao L, Fayer R, Ryan U, Upton SJ. Cryptosporidium Taxonomy: Recent Advances and Implications for Public Health. Clin Microbiol Rev. 2004;17(1):72–97.
3. Morgan-Ryan UM, Fall A, Ward LA, Hijawi N, Sulaiman I, Fayer R, et al. Cryptosporidium hominis n. sp. (Apicomplexa: Cryptosporidiidae). J Euk Microbiol. 2002;49(6):433-40.
4. Xiao L, Ryan U. Molecular Epidemiology. In: Fayer R, Xiao L, editors. Cryptosporidium and cryptosporidiosis. Boca Raton: CRC Press; 2007. p. 119-63.
5. Ajjampur SSR, Gladstone BP, Selvapandian D, Muliyil JP, Ward H, Kang G. Molecular and spatial epidemiology of cryptosporidiosis in children in a semiurban community in South India. J Clin Microbiol. 2007;45(3):915–20
6. Robinson G, Elwin K, Chalmers RM. Unusual Cryptosporidium genotypes in human cases of diarrhoea. Emerg Infect Dis. 2008;14(11):1800-2.
7. Cama VA, Bern C, Roberts J, Cabrera L, Sterling CR, Ortega Y, et al. Cryptosporidium species and subtypes and clinical manifestations in children, Peru. Emerg Infect Dis. 2008;14(10):1567-74.
8. Chalmers RM, Pollock KGJ. Cryptosporidium in Scotland 2006: reference laboratory data. HPS Week Rep. 2007;41(36):301-3.
9. Chalmers RM, Hughes S, Thomas AL, Woodhouse S, Thomas PD, Hunter P. 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. Comm Dis Publ Hlth. 2002;5(2):114-8.
10. Health Protection Agency. Cryptosporidium Laboratory reports England and Wales. All identifications, 1998-2007. 2008 [cited 2008 Sept 11]. Available from: http://www.hpa.org.uk.
11. Nichols G. Epidemiology. In: Fayer R, Xiao L, editors. Cryptosporidium and cryptosporidiosis. Boca Raton: CRC Press; 2008. p. 79-118.
12. McLauchlin J, Amar C, Pedraza-Díaz S, Nichols, G. Molecular epidemiological analysis of Cryptosporidium spp. in the United Kingdom: results of genotyping Cryptosporidium spp. in 1,705 fecal samples from humans and 105 fecal samples from livestock animals. J. Clin. Microbiol. 2000;38(11):3984-90.
13. Leoni F, Amar C, Nichols G, Pedraza-Diaz S, McLauchlin J. Genetic analysis of Cryptosporidium from 2414 humans with diarrhoea in England between 1985 and 2000. J. Med. Micro. 2006;55(6);703-707.
14. Elwin K, Chalmers RM, Roberts R, Guy EC, Casemore DP. The modification of a rapid method for the identification of gene-specific polymorphisms in Cryptosporidium parvum, and application to clinical and epidemiological investigations. Appl Environ Microbiol. 2001;67(12):5581-4.
15. Spano F, Putignani L, McLauchlin J, Casemore DP, Crisanti A. 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 Microbiol Letts. 1997;150(2):207-17.
16. Xiao L, Alderisio K, Limor J, Royer M, Lal AA. Identification of species and sources of Cryptosporidium oocysts in storm waters with a small-subunit rRNA based diagnostic and genotyping tool. Appl Environ Microbiol. 2000;66(12):5492-8.
17. Casemore DP. ACP Broadsheet 128. Laboratory methods for diagnosing cryptosporidiosis.J.Clin.Pathol.1991:44(6):445-451
18. Lea G, Leese J, editors. Health information for overseas travel. 2nd ed. London: The Stationery Office, 2001. Available from: http://www.nathnac.org/pro/yellowbook_revision.htm.
19. Anonymous. Surveillance of waterborne disease and water quality: January to June 2001, and summary of 2000. CDR Wkly. 2001;11(45). Available from: http://www.hpa.org.uk/cdr/archives/2001/cdr4501.pdf
20. Howe AD, Forster S, Morton S, Marshall R, Osborn KS, Wright P, et al. Cryptosporidium oocysts in a water supply associated with a cryptosporidiosis outbreak. Emerg Infect Dis. 2002; 8(6):619-24.
21. Hunter PR, Wilkinson DC, Lake IR, Harrison FCD, Syed Q, Hadfield SJ, et al. Microsatellite typing of Cryptosporidium parvum in a waterborne outbreak. J Clin Microbiol. 2008;46(11):3866-7.
22. Chalmers RM, Ferguson C, Cacciò SM, Gasser RB, El-Osta YGA, Heijnen L, et al. Direct comparison of selected methods for genetic categorisation of Cryptosporidium parvum and Cryptosporidium hominis species. Int J Parasitol. 2005;35(4):397-410.
23. Anonymous. Surveillance of waterborne disease and water quality: January to June 2003, and summary of 2002. CDR Wkly. 2003;13(41). Available from: http://www.hpa.org.uk/cdr/archives/2003/cdr4103.pdf
24. Anonymous. Surveillance of waterborne disease and water quality: January to June 2002, and summary of 2001. CDR Wkly 2002;12(37). Available from:
25. Anonymous. Surveillance of waterborne disease and water quality: July to December 2003. CDR Wkly 2004;14(15). Available from:
26. Jones M, Boccia D, Kealy M, Salkin B, Ferrero A, Nichols G, et al. Cryptosporidium outbreak linked to interactive water feature, UK: importance of guidelines. Euro Surveill 2006;11(4):pii=612. Available from: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=612
27. Smerdon W. Cryptosporidiosis outbreak associated with Majorcan hotel. Euro Surveill. 2000;4(34):pii=1540. Available from: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=1540
28. Galmes A, Nicolau A, Arbona G, Smith-Palmer A, Hernández Pezzi G, Soler P. Cryptosporidiosis outbreak in British tourists who stayed at a hotel in Majorca, Spain. Euro Surveill. 2003;7(33):pii=2275. Available from: http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=2275
29. Hunter PR, Hughes LS, Woodhouse S, Syed Q, Verlander N, Chalmers RM, et al. Sporadic cryptosporidiosis case control study with genotyping. Emerg Infect Dis. 2004;10(7):1241-9.
30. Hunter PR, Hadfield SJ, Wilkinson D, Lake IR, Harrison FCD, Chalmers RM. Subtypes of Cryptosporidium parvum in humans and disease risk. Emerg Infect Dis. 2007;13(1):82-8.
31. Learmonth JJ, Ionas G, Pita AB, Cowie RS. Identification and genetic characterisation of Giardia and Cryptosporidium strains in humans and dairy cattle in the Waikato Region of New Zealand. Water Sci Technol. 2003;47(3):21-6.
32. Smerdon WJ, Nichols T, Chalmers RM, Heine H, Reacher M. Foot and Mouth disease in livestock and reduced cryptosporidiosis in humans, England and Wales. Emerg Infect Dis. 2003;9(1):22-8.
33. Sopwith W, Osborn K, Chalmers R, Regan M. The changing epidemiology of cryptosporidiosis in North West England. Epidemiol Infect. 2005;133(5): 785-93.
34. Lake IR, Harrison FCD, Chalmers RM, Bentham G, Nichols G, Hunter PR, et al. Case-control study of environmental and social factors influencing cryptosporidiosis. Europ J Epidemiol. 2007;22(11):805-11.
35. Cartwright RY. Food and waterborne infections associated with package holidays. J Appl Microbiol. 2003;94:(s1)12S–24S.
36. Xiao L, Limor J, Morgan UM, Sulaiman IM, Thompson RC, Lal AA. Sequence differences in the diagnostic target region of the oocyst wall protein gene of Cryptosporidium parasites. Appl Environ Microbiol. 2000;66(12):5499-5502.
37. Tanriverdi S, Grinberg A, Chalmers RM, Hunter PR, Petrovic Z, Akiyoshi DE, et al. Inferences on the global population structure of the protozoan pathogens Cryptosporidium parvum and Cryptosporidium hominis. Appl Environ Microbiol. 2008;74(23):7227-34.
38. Mallon M, Macloed A, Wastling J, Smith H, Reilly B, Tait A. Population structure and the role of genetic exchange in the zoonotic pathogen Cryptosporidium parvum. Journal of Mol Evol. 2003;56(4):407-17.