Laboratory-based evaluation of legionellosis epidemiology in Ontario, Canada, 1978 to 2006
© Ng et al; licensee BioMed Central Ltd. 2009
Received: 30 June 2008
Accepted: 21 May 2009
Published: 21 May 2009
Legionellosis is a common cause of severe community acquired pneumonia and respiratory disease outbreaks. The Ontario Public Health Laboratory (OPHL) has conducted most testing for Legionella species in the Canadian province of Ontario since 1978, and represents a multi-decade repository of population-based data on legionellosis epidemiology. We sought to provide a laboratory-based review of the epidemiology of legionellosis in Ontario over the past 3 decades, with a focus on changing rates of disease and species associated with legionellosis during that time period.
We analyzed cases that were submitted and tested positive for legionellosis from 1978 to 2006 using Poisson regression models incorporating temporal, spatial, and demographic covariates. Predictors of infection with culture-confirmed L. pneumophila serogroup 1 (LP1) were evaluated with logistic regression models.
1,401 cases of legionellosis tested positive from 1978 to 2006. As in other studies, we found a late summer to early autumn seasonality in disease occurrence with disease risk increasing with age and in males. In contrast to other studies, we found a decreasing trend in cases in the recent decade (IRR 0.93, 95% CI 0.91 to 0.95, P-value = 0.001); only 66% of culture-confirmed isolates were found to be LP1.
Despite similarities with disease epidemiology in other regions, legionellosis appears to have declined in the past decade in Ontario, in contrast to trends observed in the United States and parts of Europe. Furthermore, a different range of Legionella species is responsible for illness, suggesting a distinctive legionellosis epidemiology in this North American region.
Legionella species are Gram-negative bacteria that are ubiquitous in both natural aquatic and moist soil and mud environments [1, 2] and in artificial aquatic habitats . Human infection with Legionella  has two distinct forms – Legionnaires' disease, a more severe form of infection which includes pneumonia, and Pontiac Fever, a milder febrile flu-like illness without pneumonia .
Legionellosis occurs both sporadically and in outbreaks; the latter may be community-based, hospital-based, or occur in the long-term care setting [6, 7]. While major outbreaks make media headlines and prompt evaluation of point-sources of exposure , most legionellosis is likely sporadic. Although there are no clinical features unique to severe legionellosis , case fatality rates are extremely high (10–40%) and may approach 50% in nosocomial outbreaks in individuals with already compromised health status [10, 11]. Sporadic cases are reported throughout the year with summer or autumn seasonality, presumably due to enhanced proliferation of the bacteria in warmer aquatic environments [11, 12].
Both sporadic legionellosis, and large legionellosis outbreaks are known to occur in the Canadian province of Ontario [8, 13], with over 1400 cases recorded between 1978 and 2006. However, the epidemiology of legionellosis in this region has not been reviewed previously in the biomedical literature. The centralized nature of legionellosis testing in Ontario over a thirty year period, combined with the retention of culture-based testing by the Ontario Public Health Laboratory (OPHL) in conjunction with serological and antigen-based assays for disease, make laboratory records a useful database for the evaluation of legionellosis trends in this jurisdiction since testing was first introduced. Our objectives were to provide a laboratory-based review of the epidemiology of legionellosis in Ontario; to explore the diversity of Legionella species that cause disease in this province; and to evaluate spatial and temporal patterns in legionellosis case occurrence over the past three decades.
Legionellosis is a notifiable disease in Ontario, with all testing for Legionella in the context of outbreak investigation, and most testing of clinical specimens, performed at the OPHL in Toronto. Testing records for legionellosis are available from May 1978 to the present; as such, the OPHL database contains information on most identified cases of legionellosis in Ontario over the past 30 years. The database includes information on the dates of onset of illness, case reporting, and laboratory testing; patient age and sex; and the hospital or healthcare facility from which test specimens were submitted, as well as the city or public health unit in which the submitting facility is located. Data are also available on testing type performed and test results for all patients with at least one positive test. As this is a laboratory-based database, information on city or public health unit of residence for each case, clinical characteristics and outcome, and such background data such as smoking status and or the presence of co-morbid illness, was not available.
Case Definition and Testing Methods
Ontario adheres to the national case definition for legionellosis as defined by the Public Health Agency of Canada . The culture of clinical specimens is regarded as the gold standard for diagnosis of Legionnaire's disease but immunoflourescence (direct fluorescent antibody (DFA)) and serological assays (indirect fluorescent antibody (IFA)) targeting L. pneumophila antigen and L. pneumophila-specific antibodies can also be used . Immunodiagnostic methods in our laboratory involve routine testing using a large panel of reagents for both DFA and IFA; species and groups included in routine testing include L. pneumophila serogroups 1 to 8, L. bozemanii, L. jordanis, L. micdadei, L. dumoffii, L. gormanii, L. oadridgensis, L. longbeachae serogroups 1 and 2, L. feeleii serogroup 1, L. anisa, L. wadsworthii, and L. maceachernii. In outbreak situations, up to 48 Legionella species can be identified using IFA. Reagents are prepared following standard protocols of the U.S. Centers for Disease Control (CDC); test interpretation is according to standard procedures defined by the CDC. Potential limitations of immunodiagnostic methods include inability to detect some Legionella species, difficulty with validation of infection by unusual strains (due to lack of clinical material) and the potential for cross-reactivity.
A validated, in-house urine antigen immunochromatographic test (ICT) was used at the laboratory from 1984 to 2005 ; in 2005, poor performance of this assay in the context of a large institutional legionellosis outbreak  resulted in replacement of this assay with a commercially available urine ezyme-linked immunosorbent assay (Binax™).
The OPHL can receive up to three types of specimens per patient (serum, urine and respiratory tract tissues); the laboratory typically identifies and reports individuals as legionellosis cases on the basis of information derived from multiple, complementary tests (a mean of four independent tests were been performed on individuals identified as "cases" during the study period). A diagnosis of legionellosis at the OPHL was confirmed when at least one of the following criteria was met in conjunction with a compatible clinical illness: (i) isolation of a Legionella species or detection of the antigen from respiratory secretions, lung tissue, pleural fluid, or other normally sterile fluids; or (ii) a significant (four-fold increase or greater) rise in Legionella antibody titre (both IgG and IgM together) between acute and convalescent sera; or (iii) single specimen or standing Legionella antibody titre seroconversion from >=1:256 against Legionella sp.; or (iv) detection of Legionella soluble urine antigen .
Legionellosis rates were estimated by dividing provincial or regional disease counts by the appropriate Statistics Canada population estimates [14, 18, 19]. As Ontario's current public health units were defined in 1995, PHU population estimates were not available for prior years; we estimated PHU populations from 1990 to 1995 through linear extrapolation, but did not attempt regional rate calculations for years prior to 1990, due to concerns that additional extrapolation would be inaccurate.
We evaluated temporal and spatial patterns in case occurrence using Poisson regression models; seasonality was modeled by incorporating sine and cosine components into regression models (i.e., via use of the "fast Fourier transform") . Models were used to estimate average rates of disease, as well as incidence rate ratios (IRR) for disease in population subgroups. Logistic regression was used to evaluate temporal changes in the likelihood that cases had been identified using different available testing methodologies.
As visual inspection of disease data suggested nonlinear changes in disease incidence over time, we evaluated both linear, quadratic, and cubic trends in incidence by incorporating year, year-squared, and year-cubed terms into regression models ; similar non-linear components were incorporated into logistic models for evaluation of trends in testing over time. We evaluated the improvement in model fit upon incorporation of a polynomial (i.e., squared) term into the model using the likelihood ratio test. As quadratic terms significantly improved model fit, but are complex to interpret, we present linear trends broken down into three decade-long time periods: 1978 to 1987; 1988 to 1997; and 1998 onwards; these latter estimates were generated using linear splining techniques . Due to the potential for cross-reactivity of sera used in immunodiagnostic methods, we defined cases as having "definitive" speciation if the case was culture-confirmed, and "tentative" speciation if speciation was based entirely on immunodiagnostic methods. We restricted our analyses of trends and patterns of legionellosis occurrence according to species to culture-confirmed cases. Statistical analyses were performed using STATA version 9.0 (STATA Corporation, College Station, TX) and thematic maps were created using ArcGIS version 9.2 (ESRI, Redlands, CA). Due to the absence of personal identifiers in the dataset, and the fact that the study involved the use of a pre-existing dataset, the study received expedited approval from the Research Ethics Board of the Hospital for Sick Children.
Epidemiological Profile of Legionellosis in Ontario
1401 cases of legionellosis were reported between May 1978 and December 2006; cases were identified in all health regions (Figure 1). Most cases were identified through testing performed on specimens submitted from the Toronto health region (46.5%, n = 648). The average annual number of legionellosis cases identified in the province between 1978 and 2006 was 48.3 cases (range 5 to 139), for an estimated crude average incidence of infection of 0.41 per 100,000 person years.
Males were more likely to be infected than females, with estimated annual infection rates of 0.55 per 100,000 as compared to 0.35 per 100,000 in females (IRR 1.57, 95% CI 1.41–1.76) (Figure 2 and Table 1). Although more cases of legionellosis were reported in the oldest women than in the oldest men, this effect was due to greater numbers of women surviving to advanced age; the annualized rate of infection in the oldest men (11.2 cases per 100,000) was approximately double that seen in the oldest women (5.7 cases per 100,000) (Figure 2). The association between male gender and legionellosis risk was strengthened by adjustment for older average age in the female population (adjusted IRR 1.98, 95% CI 1.78–2.22, P < 0.001).
Annualized rates by OHR, during the period between 1990 and 2006 (n = 1000), ranged from 0.12 cases per year per 100,000 persons in the Eastern region to 0.87 cases per 100,000 persons in the Toronto region, with the appearance of a northeast to southwest risk gradient in the southern (most populous) part of the province (Figure 1). The highest regional rate recorded during the study period was 3.39 cases per 100,000 persons in the South West region in 1998. There was, however, a significant increase in legionellosis diagnosis in Toronto relative to non-Toronto regions (IRR 2.96, 95% CI 2.6–3.4).
Temporal Trends in Case Occurrence
Temporal Trends in Testing
Data on definitive Legionella speciation were available for 351 cases with culture confirmation. Tentative species identification, based on direct and indirect immunological methods, was available for serum, tissue, and urine specimens from 533 additional cases. The distribution of species for culture confirmed isolates was significantly different for culture confirmed cases, as compared to those without culture confirmation (χ2 squared 96.9 on 8 d.f., P < 0.001) (Table 2). Because of heterogeneity in the distribution of species depending on the presence or absence of culture confirmation, we restricted our subsequent primary analyses to culture-confirmed cases.
Among culture-confirmed cases, 10 distinct species of Legionella were represented, as well as 9 distinct serogroups of Legionella pneumophila. L. pneumophila serogroup 1 (LP1) was the most commonly isolated strain, with 66% of all isolates identified as LP1. LP6 and other strains of LP comprised an additional 22% of isolates with the remainder of cases associated with non-pneumophila species (Figure 4). Among 533 cases with tentative speciation based on immunologic methods, a smaller proportion (52%) was identified as LP1, and a greater proportion (36%) were identified as non-pneumophilia species.
When we evaluated predictors of the isolation of LP1 as opposed to other Legionella strains, in 351 culture-confirmed cases, Toronto and the contiguous Central East OHR isolates were most likely to have LP1 speciation (OR for Toronto 1.82, 95% C.I 1.14–2.90, P = 0.01; Central East OHR 4.06, 95% C.I 1.14–14.40, P = 0.03). An increased likelihood of LP1 speciation was seen between 1988 and 1997, compared to other decades (OR 1.73, 95% C.I 1.09 to 2.73, P = 0.02). No differences were seen with respect to gender or age and likelihood of LP1 species.
The centralized nature of testing for legionellosis in the Canadian province of Ontario allowed us to evaluate trends in this disease over a period of approximately 30 years. To our knowledge, this represents the longest time series of legionellosis cases in the biomedical literature, dating to shortly after the initial identification of L. pneumophila as a human pathogen . Our findings exhibit similarities with previously published analyses of legionellosis epidemiology, but also exhibit important differences. We affirmed the strong associations between male gender and advanced age and legionellosis risk that have been described previously [11, 12, 25, 26], and found strong evidence for summer-autumn seasonality of legionellosis in this region of Canada. Such seasonality has been observed in studies from other developed countries, and may suggest that seasonal environmental influences on environmental reservoirs (such as surface waters) influence human disease risk [12, 25, 27, 28].
However, in this study we identify two features of legionellosis epidemiology in Ontario that appear to be at odds with observations in other jurisdictions: first, there is an apparent recent decrease in disease incidence; the second difference relates to the distribution of Legionella species, and (within the L. pneumophila species) bacterial serogroups associated with disease occurrence. Our identification of a statistically significant overall decrease in legionellosis incidence in the province since 1998 (notwithstanding an extremely large outbreak that occurred in the province in 2005 ), is at variance with reported increases in legionellosis reported recently in other North American jurisdictions, and in Australia and Europe [12, 25, 26, 29].
An important question raised by investigators in these jurisdictions has been whether legionellosis incidence is truly increasing, or whether changes in test practices (particularly the use of urine antigen testing) have increased identification of this historically under-diagnosed infection. In a similar manner, differences in estimated legionellosis burden and trends in Ontario, relative to these other jurisdictions, could reflect true epidemiological differences, differences in regional testing practices (and disease reporting) , or differences in importation of travel-related legionellosis cases [31–33]. Other explanations for a true change in the burden of legionellosis in Ontario during the study period are also possible; for example, Canadian standards for construction and maintenance of hospitals and other public buildings were updated between 1999 and 2001 ; such changes may also have contributed to the amelioration of legionellosis risk observed during the last third of our observation period.
Indirect evidence suggests that differences in testing practice (particularly differences in the uptake of urinary antigen testing) are not primarily responsible for the difference in observed disease trends in Ontario, relative to other jurisdictions. As described above, there was a marked increase over time in the likelihood that Ontario legionellosis cases were identified by urine antigen testing, with or without other complementary testing methods. By contrast, the contribution of serological testing to diagnosis decreased markedly during the period under observation, and culture testing to diagnosis has changed in a non-linear fashion over time: we postulate that increases in the likelihood of culture-based diagnosis in our laboratory from the late 1980s to late 1990s may have reflected increasing technical skill over time, whereas declining probability of culture confirmation since 1998 may be attributable to changes in the use of urinary antigen testing mentioned above. In any case, the trends we describe here were robust when we restricted our analyses to cases with or without culture confirmation.
If the decline in legionellosis in Ontario reported here represents a true epidemiological shift, rather than an artifact of clinical practice patterns and laboratory testing practices, it may be worth noting that almost all major population centers in Ontario (Ottawa being the exception) lie < 30 miles from the Great Lakes. As Legionella species are abundant in surface waters, and the Great Lakes ecosystem is currently undergoing rapid ecological change , the possibility that downward trends in legionellosis relate to changes in local hydrology warrants further investigation. Although the proximate source of legionellosis is usually water distribution systems or cooling-tower related aerosols, the seasonality of legionellosis, described here and elsewhere [12, 36], is also strongly suggestive of environmental influences on legionellosis risk. We have recently demonstrated that acute changes in local watershed hydrology in Toronto are linked to changes in legionellosis risk . These associations warrant further investigation.
A second important difference between the epidemiological data reported here and those reported from other jurisdictions was the identification of L. pneumophila 1 (LP1) in < 70% of cases with definitive speciation via culture methods (though exclusive use of culture-based methods may also favor identification of L. pneumophila species ); LP1 was identified in an even smaller proportion of cases with tentative speciation based on immunodiagnostic methods without culture confirmation. By contrast, the proportion of legionellosis cases attributable to LP1 has been reported to be > 90% in studies of legionellosis performed elsewhere in North America and Europe [39–41]. This may suggest that the parallel use of type-specific immunodiagnostic methods (such as serum IFA and tissue DFA, which are used in parallel with urine antigen testing and culture test methods at our laboratory whenever appropriate specimens are available) are important in the elucidation of the true diversity of pathogenic Legionella species, and that exclusive reliance on urine antigen testing methods may artificially inflate the proportion of cases due to LP1. Alternatively, this finding may again suggest that the epidemiology of legionellosis in this North American region is unique, and may be heavily influenced by the proximity of major population centers to the Great Lakes.
As might be expected given the geographic and demographic size and diversity of Ontario, we found evidence for important within-province heterogeneity in disease epidemiology as well. Disease incidence in the Eastern OHR (where Ottawa represents the major population aggregation) was approximately one tenth that seen in the Toronto OHR. Although some clinical laboratory urine antigen testing for legionellosis is performed in tertiary care centers in Ottawa and Kingston, the magnitude of this difference may, again, suggest that proximity to the Great Lakes is an important modulator of disease epidemiology.
Like any observational study, this study has important limitations. Principle among these is our use of a laboratory database for epidemiological purposes; this has the effect of inflating case estimates for regions with high concentrations of tertiary care centers (such as Toronto), as geographic coding reflects the locale of the healthcare institution that submitted patient specimens for testing, rather than patients' home addresses, and also means that desirable covariates (such as immune compromise, smoking status, comorbid illnesses, and case outcomes) are unavailable. Nonetheless we do not believe that such reporting effects would explain the magnitude in observed difference in disease burden between Toronto and the Ottawa region (itself a major site of tertiary care beds in the province). Furthermore, we believe that the wide geographic base, prolonged interval of data collection, detailed information on testing practices, and performance of all testing by a single laboratory distinguish this study from those of shorter duration conducted in Europe and the U.S. [11, 26], and from previous Canadian sentinel surveillance efforts , and also provide new insights into the distinctive epidemiology of this disease in Canada's most populous province.
In conclusion, we evaluated trends in legionellosis case occurrence and use of diagnostic modalities from the vantage of a major public health service laboratory over a period of three decades. While observed legionellosis epidemiology in this jurisdiction bore some similarities to that described in other jurisdictions, the apparent decline in legionellosis in the past decade in Ontario, and the wide variety of Legionella species identified in association with illness, suggest that the epidemiology of legionellosis in this North American region may be distinctive. Molecular epidemiologic studies in progress are likely to provide further insights into the phylogenetic characteristics of pathogenic Legionella species in Ontario.
This work was supported in part by a grant from the U.S. National Institutes of Allergy and Infectious Diseases (NIAID) (R21-AI065826) and an Ontario Early Researcher Award, both to Dr. Fisman. The contents of this paper are solely the responsibility of the authors and do not necessarily represent the official views of the NIAID, the Ontario Agency for Health Protection and Promotion, or the Government of Ontario.
- Costa J, Tiago I, da Costa MS, Verissimo A: Presence and persistence of Legionella spp. in groundwater. Appl Environ Microbio. 2005, 71 (2): 663-671. 10.1128/AEM.71.2.663-671.2005.View ArticleGoogle Scholar
- Wallis L, Robinson P: Soil as a source of Legionella pneumophila serogroup 1 (Lp1). Australian and New Zealand Journal of Public Health. 2005, 29 (6): 518-520. 10.1111/j.1467-842X.2005.tb00242.x.View ArticlePubMedGoogle Scholar
- Declerck P, Behets J, van Hoef V, Ollevier F: Detection of Legionella spp. and some of their amoeba hosts in floating biofilms from anthropogenic and natural aquatic environments. Water Research. 2007, 41 (14): 3159-3167. 10.1016/j.watres.2007.04.011.View ArticlePubMedGoogle Scholar
- Burnsed LJ, Hicks LA, Smithee LM, Fields BS, Bradley KK, Pascoe N, Richards SM, Mallonee S, Litrell L, Benson R, et al: A Large, Travel-Associated Outbreak of Legionellosis among Hotel Guests: Utility of the Urine Antigen Assay in confirming Pontiac Fever. Clin Infect Dis. 2007, 44: 222-228. 10.1086/510387.View ArticlePubMedGoogle Scholar
- Fields BS, Benson RF, Besser RE: Legionella and Legionnaire's Disease: 25 Years of Investigation. Clin Microbiol Rev. 2002, 15 (3): 506-526. 10.1128/CMR.15.3.506-526.2002.View ArticlePubMedPubMed CentralGoogle Scholar
- Sala M, Arias C, Oliva J, Pedrol A, Roura P, Dominguez A: Community outbreak of Legionnaires disease in Vic-Gurb, Spain in October and November 2005. European Surveillance. 2007, 12 (3): 223-Google Scholar
- Joseph C: Legionnaires' disease in Europe 2000–2002. Epidemiol Infect. 2004, 132 (3): 417-424. 10.1017/S0950268804002018.View ArticlePubMedPubMed CentralGoogle Scholar
- Abbas Z, Nolan L, Landry L, Galanis E, Egan C: Investigation of an outbreak of Legionnaires' disease in a hospital under contruction: Ontario, September-October 2002. Canada Communicable Disease Report. 2003, 29 (17): 145-152.PubMedGoogle Scholar
- Kirby BD, Snyder KM, Meyer RD, Findgold SM: Legionnaires' disease: report of sixty-five nosocomially acquired cases of review of the literature. Medicine (Baltimore). 1980, 59 (3): 188-205.View ArticleGoogle Scholar
- Benin A, Benson R, Besser R: Trends in legionnaires disease, 1980–1998: declining mortality and new patterns of diagnosis. Clin Infect Dis. 2002, 35: 1039-1046. 10.1086/342903.View ArticlePubMedGoogle Scholar
- Marston BJ, Lipman HB, Breiman RF: Surveillance for Legionnaires' disease. Arch Intern Med. 1994, 154: 2417-2422. 10.1001/archinte.154.21.2417.View ArticlePubMedGoogle Scholar
- Fisman DN, Lim S, Wellenius GA, Johnson C, Britz P, Gaskins M, Maher J, Mittleman MA, Spain V, Hass CN, et al: Its Not the Heat, It's the Humidity: Wet Weather Increases Legionellosis Risk in the Greater Philadelphia Metropolitan Area. J Infect Dis. 2005, 195: 2066-2073. 10.1086/498248.View ArticleGoogle Scholar
- Marrie TJ, De Carolis E, Yu VL, Stout JE: Legionnaire's disease – Results of a multicentre Canadian study. Can J Infect Dis. 2003, 14 (3): 154-158.View ArticlePubMedPubMed CentralGoogle Scholar
- StatsCanada: Population and dwelling counts, for Canada, Provinces and Territories, and census divisions, 2006 and 2001 censuses. 2007, [http://www12.statcan.ca/english/census06/data/popdwell/Table.cfm?T=702&PR=35&SR=1&S=3&O=D]Google Scholar
- Health_Canada: Case Definitions for Diseases Under National Surveillance. 2000, Ottawa, Canada: Health Protection Branch, Health CanadaGoogle Scholar
- Tang P, Toma S, Rajkumar W: Detection of urinary antigens of Legionella pneumophila serogroup 12 by broad-spectrum enzyme-linked immunosorbent assay. J Clin Micobiol. 1989, 27 (4): 783-784.Google Scholar
- Gilmour M, Bernard K, Tracz D, Olson A, Corbett C, Burdz T, Ng B, Wiebe D, Broukhanski G, Boleszczuk P, et al: Molecular typing of a Legionella pneumophila outbreak in Ontario, Canada. J Med Microbiol. 2007, 56: 336-341. 10.1099/jmm.0.46738-0.View ArticlePubMedPubMed CentralGoogle Scholar
- StatsCanada: Estimates of population (2001 Census and administrative data), by age group and sex, Canada, Provinces, Territories, health regions (June 2005 boundaries) and peer groups, annual (number). 2007, [http://cansim2.statcan.ca/cgi-win/CNSMCGI.EXE?Lang=E&ArrayId=01095315&Array_Pick=1&Detail=1&ResultTemplate=CII/CII___&RootDir=CII/&TblDetail=1&C2SUB=HEALTH]Google Scholar
- StatsCanada: Estimates of population, by age group and sex for July 1, Canada, provinces and territories, annual. 2008, [http://cansim2.statcan.ca/cgi-win/cnsmcgi.exe?Lang=E&RootDir=CII/&ResultTemplate=CII/CII_pick&Array_Pick=1&ArrayId=051-0001]Google Scholar
- Diggle PJ: Spectral Analysis. Time Series: A Biostatistical Introduction. 1990, Oxford, UK: Oxford University Press, 94-133. 1Google Scholar
- Woodward M: Modelling quantitative outcome variables. Epidemiology: study design and data analysis. 2005, Boca Raton, Florida: Chapman & Hall/CRC, 427-513. 2Google Scholar
- Samoli E, Analitis A, Touloumi G, Schwartz J, Anderson H, Sunyer J, Bisanti L, Zmirou D, Vonk J, Pekkanen J, et al: Estimating the exposure-response relationships between particulate matter and mortality within the APHEA multicity project. Environ Health Perspect. 2005, 113 (1): 88-95.View ArticlePubMedGoogle Scholar
- Hicks LA, Rose CE, Fields BS, Drees ML, Engel JP, Jenkins PR, Rouse BS, Blythe D, Khalifah AP, Feikin DR, et al: Increased rainfall is associated with increased risk for legionellosis. Epidemiol Infect. 2006, 1-7.Google Scholar
- McDade JE, Shepard CC, Fraser DW, Tsai TR, Redus MA, Dowdle WR: Legionaires' disease: isolation of a bacterium and demonstration of its role in other respiratory disease. N Engl J Med. 1977, 297 (22): 1197-1203.View ArticlePubMedGoogle Scholar
- Li JS, O'Brien ED, Guest C: A review of national legionellosis surveillance in Australia, 1991 to 2000. Commun Dis Intell. 2002, 26 (3): 461-468.Google Scholar
- Ricketts KD, Joseph CA: Legionnaires disease in Europe: 2005–2006. Euro Surveill. 2007, 12 (12): E7-8.PubMedGoogle Scholar
- Straus W, Plouffe J, File T, Lipman H, Hackman B, Salstrom S, Benson R, Breiman R: Risk factors for domestic acquisition of legionnaires disease. Ohio legionnaires Disease Group. Arch Intern Med. 1996, 156 (15): 1685-1692. 10.1001/archinte.156.15.1685.View ArticlePubMedGoogle Scholar
- Bhopal R, Fallon R: Seasonal variation of Legionnaires' disease in Scotland. J Infect. 1991, 22 (2): 153-160. 10.1016/0163-4453(91)91569-J.View ArticlePubMedGoogle Scholar
- Neil K, Berkelman R: Increasing incidence of legionellosis in the United States, 1990–2005: changing epidemiologic trends. Clin Infect Dis. 2008, 47 (5): 591-599. 10.1086/590557.View ArticlePubMedGoogle Scholar
- Ng V, Tang P, Fisman DN: Our Evolving Understanding of Legionellosis Epidemiology: Learning to Count. Clinical Infectious Diseases. 2008, 47 (600–2):
- Benin A, Benson R, Arnold K, Fiore A, Cook P, Williams L, Fields B, Besser R: An outbreak of travel-associated Legionnaires disease and Pontiac fever: the need for enhanced surveillance of travel-associated legionnellosis in the United States. J Infect Dis. 2002, 185 (2): 237-243. 10.1086/338060.View ArticlePubMedGoogle Scholar
- Ricketts KD, McNaught B, Joseph C: European Working Group for Legionella Infections Health Protection Agency C: Travel-related legionnaires' disease in Europe: 2004. Euro Surveill. 2006, 11 (4): 107-110.PubMedGoogle Scholar
- Ricketts KD, Joseph C: European Working Group for Legionella Infections Health Protection Agency C, London, UK: Travel associated legionnaires' disease in Europe: 2002. Euro Surveill. 2004, 9 (2): 6-9.PubMedGoogle Scholar
- Construction-related nosocomial infections inpatients in health care facilities. Decreasing the risk of Aspergillus, Legionella and other infections. Can Commun Dis Rep. 2001, 27 (Suppl 2): 1-46.
- Sellinger C, Stow C, Lamon E, Qian S: Recent water level declines in the Lake Michigan-Huron system. Environ Sci Technol. 2008, 42 (2): 367-373. 10.1021/es070664+.View ArticlePubMedGoogle Scholar
- Straus WL, Plouffe JF, File TM, Lipman HB, Hackman BH, Salstrom SJ, Benson RF, Breiman RF: Risk factors for domestic acquisition of legionnaires disease. Ohio legionnaires Disease Group. Arch Intern Med. 1996, 156 (15): 1685-1692. 10.1001/archinte.156.15.1685.View ArticlePubMedGoogle Scholar
- Ng V, Tang P, Jamieson F, Drews SJ, Brown S, Low DE, Johnson CC, Fisman DN: Going with the Flow: Legionellosis Risk in Toronto, Canada Is Strongly Associated with Local Watershed Hydrology. Ecohealth. 2009, Last accessed April 22, 2009., [http://www.springerlink.com/content/802827162620p204/fulltext.html]Google Scholar
- Lee TC, Vickers RM, Yu VL, Wagener MM: Growth of 28 Legionella species on selective culture media: a comparative study. J Clin Micobiol. 1993, 31: 2764-2768.Google Scholar
- Harrison TG, Doshi N, Fry NK, Joseph CA: Comparison of clinical and environmental isolates of Legionella pneumophila obtained in the UK over 19 years. Clin Microbiol Infect. 2007, 13: 78-85. 10.1111/j.1469-0691.2006.01558.x.View ArticlePubMedGoogle Scholar
- Yu VL, Plouffe JF, Pastoris MC, Stout JE, Schousboe M, Widmer A, Summersgill J, File T, Heath CM, Paterson DL, et al: Distribution of Legionella Species and Serogroups Isolated by Culture in Patients with Sporadic Community-Acquired Legionnellosis: An International Collaborative Survey. Journal of Infections Diseases. 2002, 186: 127-128. 10.1086/341087.View ArticleGoogle Scholar
- Pelaz C, Garcia L, Martin-bourqon C: Legionellae isolated from clinical and environmental samples in Spain (1983–90): monoclonal typing of Legionella pnemophila serogroup 1 isolates. Epidemiol Infect. 1992, 108 (3): 397-402. 10.1017/S0950268800049906.View ArticlePubMedPubMed CentralGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2334/9/68/prepub