Hostname: page-component-77c89778f8-gvh9x Total loading time: 0 Render date: 2024-07-22T06:20:15.340Z Has data issue: false hasContentIssue false

Legionellosis on the rise: A scoping review of sporadic, community-acquired incidence in the United States

Published online by Cambridge University Press:  28 July 2023

Michelle A. Moffa
Affiliation:
Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA
Clare Rock
Affiliation:
Division of Infectious Diseases, Department of Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA Department of Hospital Epidemiology and Infection Control, Johns Hopkins Hospital, Baltimore, MD, USA
Panagis Galiatsatos
Affiliation:
Medicine for the Greater Good, Johns Hopkins University School of Medicine, Baltimore, MD, USA Division of Pulmonary and Critical Care Medicine, Johns Hopkins University School of Medicine, Baltimore, MD, USA
Shantini D. Gamage
Affiliation:
U.S. Department of Veterans Affairs, National Infectious Diseases Service, Veterans Health Administration, Washington, DC, USA Division of Infectious Diseases, Department of Internal Medicine, University of Cincinnati College of Medicine, Cincinnati, OH, USA
Kellogg J. Schwab
Affiliation:
Department of Environmental Health and Engineering, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
Natalie G. Exum*
Affiliation:
Department of Environmental Health and Engineering, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
*
Corresponding author: Natalie G. Exum; Email: nexum1@jhu.edu
Rights & Permissions [Opens in a new window]

Abstract

Over the past two decades, the incidence of legionellosis has been steadily increasing in the United States though there is noclear explanation for the main factors driving the increase. While legionellosis is the leading cause of waterborne outbreaks in the US, most cases are sporadic and acquired in community settings where the environmental source is never identified. This scoping review aimed to summarise the drivers of infections in the USA and determine the magnitude of impact each potential driver may have. A total of 1,738 titles were screened, and 18 articles were identified that met the inclusion criteria. Strong evidence was found for precipitation as a major driver, and both temperature and relative humidity were found to be moderate drivers of incidence. Increased testing and improved diagnostic methods were classified as moderate drivers, and the ageing U.S. population was a minor driver of increasing incidence. Racial and socioeconomic inequities and water and housing infrastructure were found to be potential factors explaining the increasing incidence though they were largely understudied in the context of non-outbreak cases. Understanding the complex relationships between environmental, infrastructure, and population factors driving legionellosis incidence is important to optimise mitigation strategies and public policy.

Type
Original Paper
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (http://creativecommons.org/licenses/by/4.0), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2023. Published by Cambridge University Press

Introduction

Legionnaires’ disease (LD), a type of pneumonia that is often severe and has a 7–10% mortality rate, is the leading cause of drinking water disease outbreaks in the United States (USA) [Reference Soda, Barskey, Shah, Schrag, Whitney, Arduino, Reddy, Kunz, Hunter, Raphael and Cooley13]. Legionellosis, caused by infection with Legionella bacteria, includes both LD and the flu-like illness, Pontiac fever. LD is especially severe in older people, smokers, and those with compromised immune systems [Reference Soda, Barskey, Shah, Schrag, Whitney, Arduino, Reddy, Kunz, Hunter, Raphael and Cooley1]. Infections are caused by the inhalation of waterborne Legionella from aerosolised sources in the built environment such as showers, sinks, and cooling towers. Healthcare-associated legionellosis infections and travel-associated infections each account for about 20% of cases in the USA, while the remaining estimated 60–65% of cases are considered community-acquired [Reference Barskey, Derado and Edens2, 3]. National analyses suggest that up to 95% of community-acquired legionellosis cases in the USA are sporadic, meaning that they are not associated with a known outbreak, cluster, or identifiable environmental exposure [3, Reference Hicks4].

Reporting to the Centers for Disease Control and Prevention (CDC) National Notifiable Diseases Surveillance Systems (NNDSS) since 2002 shows that age-standardised legionellosis incidence has been increasing at an average annual rate of 9.3% [Reference Barskey, Derado and Edens2]. The annual increase in reported legionellosis cases is highest in the Middle Atlantic and Midwest regions [Reference Hicks4]. An investigation of the true LD burden in the USA indicates that the current observed incidence rate of 1.2 cases per 100,000 population is substantially lower than the estimated rate of 11.6 cases per 100,000 population [Reference Hicks4, Reference Cassell, Gacek, Rabatsky-Ehr, Petit, Cartter and Weinberger5].

There is a lack of epidemiological data on the underlying drivers of increasing legionellosis incidence in the USA [Reference Barskey, Derado and Edens2]. Numerous clinical and environmental factors have been hypothesised for the year-on-year increase. There is a need to differentiate between whether there is a true increase in cases or whether the increase in reported incidence is an artefact of more successful detection. Potential clinical factors include improved clinical case capture due to more accessible diagnostic tools, increased physician awareness of atypical pathogens of community-acquired pneumonia (CAP), and improved reporting practices to the CDC [Reference Murdoch6, Reference Cunha, Burillo and Bouza7]. The increase in incidence may also be due to demographic factors such as an ageing and increasingly immunocompromised population [Reference Htwe and Khardori8, Reference Cunha9]. Environmental drivers causing Legionella to be more prevalent in water systems and increasingly exposing the population may include drinking water infrastructure factors [Reference LeChevallier10, Reference Hamilton, Hamilton, Johnson, Jjemba, Bukhari, LeChevallier, Haas and Gurian11] and climate factors such as changing temperature and precipitation [Reference Pampaka, Gómez-Barroso, López-Perea, Carmona and Portero12].

In this scoping review, we aimed to synthesise evidence from across the literature regarding the drivers of sporadic, community-acquired legionellosis cases in the USA since 2006. The weight of evidence from the scoping review is then discussed to determine the impact of these factors as drivers of increasing incidence.

Methods

We conducted a scoping review to identify studies and reviews that discuss the following themes: (i) epidemiological trends and potential drivers of increasing sporadic, community-acquired legionellosis incidence and (ii) trends in clinician awareness, diagnostic testing, and reporting of legionellosis. We focused the search on studies within the USA due to the complexity of factors influencing legionellosis transmission and the major differences in population demographics, water treatment and management, and public policy between the USA and other countries. We include international studies in the discussion to help interpret and contextualise our findings.

Peer-reviewed literature in PubMed was searched in November 2022 using the following keywords: (“Legionnaires’ Disease” OR “legionellosis” OR “Legionella” OR “Legionella pneumophila” OR “Pontiac Fever”) AND (“incidence” OR “case count” OR “burden” OR “prevalence” OR “diagnos*” OR “reporting” OR “surveillance” OR “trend”). We followed the PRISMA Extension for Scoping Reviews while reporting our findings as applicable [Reference Tricco, Lillie, Zarin, O’Brien, Colquhoun, Levac, Moher, Peters, Horsley, Weeks, Hempel, Akl, Chang, McGowan, Stewart, Hartling, Aldcroft, Wilson, Garritty, Lewin, Godfrey, Macdonald, Langlois, Soares-Weiser, Moriarty, Clifford, Tunçalp and Straus13] and acknowledge that the evidence presented in this scoping review is not considered exhaustive given that systematic methods were not used to select studies.

We included only studies written in English that discussed one or both of the themes above. To focus on understanding national and regional incidence trends of sporadic, community-acquired disease, we excluded individual case reports, outbreak reports, and water monitoring reports without the inclusion of epidemiological data. We also excluded studies that exclusively discussed the following: nosocomial or travel-associated legionellosis; novel clinical diagnostic methods; microbiological studies; and legionellosis treatment, prevention, or control strategies. We restricted the search to papers published in 2006 or later, which is the first full calendar year after the CDC changed the legionellosis case definition in 2005 [Reference Barskey, Derado and Edens2].

We screened 1,738 titles resulting from the PubMed search that may include relevant information. From abstracts and full text, we then identified 18 articles that met our inclusion criteria and had information or analysis relevant to our research question. The rating systems used for the magnitude of impact and quality of evidence to evaluate the articles as potential drivers are described in Table 1.

Table 1. Rating system to evaluate the potential drivers identified in articles of scoping review

Results

Table 2 summarises the potential drivers of increasing legionellosis incidence identified in the scoping review and the ratings assigned. All 18 of the included papers present evidence relevant to the epidemiological trends and potential drivers of sporadic, community-acquired legionellosis incidence. Seven of the included papers additionally address trends in clinical awareness, diagnostic testing, and reporting of legionellosis. The included papers are described in greater detail in Supplementary Table S1.

Table 2. Potential drivers of increasing legionellosis incidence in the United States, 2006–2022: magnitude of impact and quality of evidence

Clinical case capture

The following potential drivers of improved case capture were identified from the literature: i) more consistent reporting to the CDC; ii) an increase in clinician awareness; iii) more frequent diagnostic testing; and iv) increased access to urinary antigen tests [Reference Barskey, Derado and Edens2, Reference Hicks4, Reference Alarcon Falconi, Cruz and Naumova14, Reference Simmering, Polgreen, Hornick, Sewell and Polgreen16, Reference Allgaier, Lagu, Haessler, Imrey, Deshpande, Guo and Rothberg17, Reference Schoonmaker-Bopp19, Reference Farnham, Alleyne, Cimini and Balter21].

Increased reporting

The CDC has revised the legionellosis case definition several times since reporting began in 1976 – most recently in 2005 and 2019 – but it is unlikely that these small changes have impacted reporting data [Reference Barskey, Derado and Edens2]. Alarcon Falconi et al. [Reference Alarcon Falconi, Cruz and Naumova14] found a history of underreporting diagnosed cases to the CDC that decreased over time, suggesting improvement in reporting to the CDC. More recently, in 2011–2013, the CDC instituted an Active Bacterial Core Surveillance Programme at ten U.S. sites and found that legionellosis incidence was similar between passive NNDSS reports and the cases projected from active surveillance [Reference Dooling, Toews, Hicks, Garrison, Bachaus, Zansky, Carpenter, Schaffner, Parker, Petit, Thomas, Thomas, Mansmann, Morin, White and Langley15]. Underreporting to the CDC has likely decreased over time, and this potential driver may have a null effect on increasing incidence.

Increased clinician awareness

Reported cases in the USA remained relatively stable between 1990 and 2002 at an average of 1,268 cases annually and then increased sharply to 2,223 cases in 2003 [Reference Neil and Berkelman30]. This increase coincided with the severe acute respiratory syndrome (SARS) pandemic. Several articles suggested that clinician efforts to identify the aetiology of CAP cases may have increased during the SARS pandemic, thereby increasing the number of Legionella tests performed and cases of legionellosis identified, though there is minimal evidence to support this theory [Reference Barskey, Derado and Edens2, Reference Alarcon Falconi, Cruz and Naumova14, Reference Simmering, Polgreen, Hornick, Sewell and Polgreen16]. There remains insufficient evidence to understand how clinician awareness affected legionellosis surveillance in the USA during the 2003 SARS pandemic and what, if any, impact clinician awareness has played in the continuing incidence increase since then.

Increased testing and use of urinary antigen test diagnostics

The urinary antigen test (UAT) is a rapid, simple, and widely used diagnostic method compared with culturing Legionella. UATs were used to diagnose 69–70% of cases in 1998 [Reference Barskey, Derado and Edens2, Reference Alarcon Falconi, Cruz and Naumova14, Reference Dooling, Toews, Hicks, Garrison, Bachaus, Zansky, Carpenter, Schaffner, Parker, Petit, Thomas, Thomas, Mansmann, Morin, White and Langley15], and they increased to over 90% by 2011 [Reference Dooling, Toews, Hicks, Garrison, Bachaus, Zansky, Carpenter, Schaffner, Parker, Petit, Thomas, Thomas, Mansmann, Morin, White and Langley15]. We identified two studies that included or discussed electronic health record (EHR) data from the USA showing how the frequency of legionellosis testing has changed over time. Allgaier et al. [Reference Allgaier, Lagu, Haessler, Imrey, Deshpande, Guo and Rothberg17] analysed a national EHR database of adults hospitalised with a primary diagnosis of pneumonia across 177 hospitals. From 2010 to 2015, 26% of these patients were tested for Legionella. This percentage increased over time in the Midwest and Southern regions, and in the Northeast region, the percentage increased from approximately 30% in 2010 to 40% in 2015. Though an increasing percentage of hospitalised pneumonia patients were tested for Legionella, the underlying average annual test positivity rate over this same timeframe remained relatively constant at around 1.5%, indicating that increasing testing for Legionella is a driver of increasing incidence [Reference Allgaier, Lagu, Haessler, Imrey, Deshpande, Guo and Rothberg17]. On a smaller scale, a study of UAT usage in a Wisconsin-based healthcare system found a relatively consistent number of tests that were ordered annually between 2013 and 2017 [Reference Toberna, William, Kram, Heslin and Baumgardner20]. A cross-sectional study by Gamage et al. [Reference Gamage, Ambrose, Kralovic, Simbartl and Roselle18] analysed LD data in the national U.S. Veterans Health Administration from 2014 to 2016 and found that increased use of UAT regionally often results in lower test positivity. These studies present medium-quality evidence to support the idea that increased testing and improved diagnostics are a driver of rising legionellosis, with a moderate magnitude of impact on increasing incidence.

Racial and socioeconomic inequities

Epidemiological studies identified in the scoping review frequently found racial and socioeconomic disparities in legionellosis incidence in the USA [Reference Barskey, Derado and Edens2, Reference Hicks4, Reference Dooling, Toews, Hicks, Garrison, Bachaus, Zansky, Carpenter, Schaffner, Parker, Petit, Thomas, Thomas, Mansmann, Morin, White and Langley15, Reference Allgaier, Lagu, Haessler, Imrey, Deshpande, Guo and Rothberg17, Reference Toberna, William, Kram, Heslin and Baumgardner20Reference Gleason, Ross and Greeley23]. However, race and ethnicity data are often not reported with legionellosis case data across many states in the USA [Reference Barskey, Derado and Edens2, Reference Simmering, Polgreen, Hornick, Sewell and Polgreen16], making it challenging to understand the role that racial inequities contribute to increasing incidence.

Racial inequity

Many analyses of legionellosis in the USA show incidence is highest among Black or African American individuals relative to other racial groups [Reference Barskey, Derado and Edens2, Reference Hicks4, Reference Dooling, Toews, Hicks, Garrison, Bachaus, Zansky, Carpenter, Schaffner, Parker, Petit, Thomas, Thomas, Mansmann, Morin, White and Langley15, Reference Allgaier, Lagu, Haessler, Imrey, Deshpande, Guo and Rothberg17, Reference Farnham, Alleyne, Cimini and Balter21, Reference Hunter, Salandy, Smith, Edens and Hubbard22]. In an analysis of cases reported to the CDC, Barskey et al. [Reference Barskey, Derado and Edens2] found that when comparing a 1992–2002 baseline to 2018 data, the age-standardised incidence has increased at twice the rate among Black people (from 0.47/100,000 population in 1992–2002 to 5.21/100,000 population in 2018) relative to White people (from 0.37/100,000 population to 1.99/100,000 population). In New York City, legionellosis incidence from 2002 to 2011 was higher among the non-Hispanic Black population, at 2.15/100,000 population compared with 1.56/100,000 population for non-Hispanic Whites, though the difference was not statistically significant [Reference Farnham, Alleyne, Cimini and Balter21]. Many national epidemiological studies do not stratify data specifically for Hispanic/Latino people due to missing ethnicity data for over 30% of cases reported to the CDC [Reference Barskey, Derado and Edens2, Reference Hicks4]. Despite these important racial disparities, we concluded that major data gaps and a lack of analysis linking any temporal or causal component of structural racism to legionellosis trends prevent us from determining the magnitude of impact that racial inequities have on increasing incidence.

Socioeconomic inequity

Socioeconomic demographics have shown significant relationships within multiple state-based studies in the USA [Reference Toberna, William, Kram, Heslin and Baumgardner20, Reference Farnham, Alleyne, Cimini and Balter21, Reference Gleason, Ross and Greeley23]. Income was the most studied social determinant of health identified in a narrative review of disparities in LD incidence [Reference Hunter, Salandy, Smith, Edens and Hubbard22]. In New York City, the yearly age-adjusted incidence of community-acquired legionellosis between 2002 and 2011 was 2.5 times higher in the census tract with the highest poverty level (3/100,000 population) compared with the census tract with the lowest poverty level (1.2/100,000 population) [Reference Farnham, Alleyne, Cimini and Balter21]. In New Jersey, poverty level was the strongest risk factor for legionellosis in multivariate models using various census tract-level sociodemographic variables. High-risk census tracts were also more likely to have lower education levels [Reference Gleason, Ross and Greeley23]. Given that none of these studies were national in scale or discussed how poverty dynamics may be changing over time, there was insufficient evidence about the magnitude of impact that socioeconomic inequity may have on increasing incidence.

Population susceptibility

Numerous epidemiological studies have established that legionellosis risk increases with age, smoking, and comorbidities such as chronic obstructive pulmonary disease (COPD), diabetes, kidney disease, and immunosuppression [Reference Cooley, Pondo, Francois Watkins, Shah, Schrag, Vugia, Alden, Cartter, Thomas, Ryan, Laine, Pradhan and Dunn25].

Ageing population

An ageing population is frequently cited as a contributing factor to increasing legionellosis incidence in the USA [Reference Barskey, Derado and Edens2, Reference Hicks4, Reference Cassell, Gacek, Rabatsky-Ehr, Petit, Cartter and Weinberger5, Reference Alarcon Falconi, Cruz and Naumova14, Reference Han24]. Overall, the age distribution of legionellosis cases reported to the CDC has remained relatively constant over time [Reference Alarcon Falconi, Cruz and Naumova14]. Three longitudinal epidemiological studies spanning from the 1990s through 2018 found that age-standardised incidence in the USA has increased at a slightly lower rate than crude, unadjusted incidence [Reference Barskey, Derado and Edens2, Reference Hicks4, Reference Han24], indicating that an ageing population may have a minor impact on increasing incidence, backed up by high-quality evidence.

Increase in underlying medical conditions

None of the studies identified for the scoping review analysed how medical conditions or lifestyle factors (e.g. smoking) associated with LD risk have changed over time. Therefore, there was insufficient evidence to evaluate their impact on legionellosis incidence trends.

Climate and hydrometeorological factors

Climate change is expected to have a promotive effect on waterborne infectious diseases such as legionellosis. The consistent seasonality observed with legionellosis incidence in the USA points to environmental factors as important potential drivers of disease. We identified five studies that examined longitudinal legionellosis case data with hydrometeorological factors to understand the impact of weather and climate on increasing incidence [Reference Simmering, Polgreen, Hornick, Sewell and Polgreen16, Reference Han24, Reference Gleason, Kratz, Greeley and Fagliano26Reference Passer, Danila, Laine, Como-Sabetti, Tang and Searle28].

Precipitation

With robust multivariate regression modelling, there was a consistently significant promotive effect of increased precipitation on the risk of disease [Reference Simmering, Polgreen, Hornick, Sewell and Polgreen16, Reference Gleason, Kratz, Greeley and Fagliano26, Reference Passer, Danila, Laine, Como-Sabetti, Tang and Searle28]. Gleason et al. [Reference Gleason, Kratz, Greeley and Fagliano26] found a strong positive association between precipitation and increased legionellosis rates in New Jersey for the monthly, but not daily, precipitation variable, indicating that an extended latency period may be required to have a promotive effect on Legionella proliferation in the environment. Similarly, in Passer et al. [Reference Passer, Danila, Laine, Como-Sabetti, Tang and Searle28], when lag effects were introduced to create a 14-day lagged average precipitation variable, a strong effect was found for the increase in the risk of sporadic infection. In Simmering et al. [Reference Simmering, Polgreen, Hornick, Sewell and Polgreen16], a case–control study design found that there was nearly 20 mm more rain (80.4 mm) for cases than for controls (61.7 mm). These studies give confidence to the classification that precipitation is a major driver of sporadic, community-acquired legionellosis infections, backed up by high-quality evidence.

Temperature

Temperature was found to have a promotive effect on legionellosis incidence rates in 33 Eastern U.S. states and the District of Columbia [Reference Han27]. A state-wide time-series study in New Jersey found that temperature was less predictive of legionellosis risk than wet, humid weather though a positive association between temperature and disease risk was found in a univariate model [Reference Gleason, Kratz, Greeley and Fagliano26]. Han [Reference Han24] found that mean temperature had a promotive effect on legionellosis incidence when the analysis was restricted to data in the peak season of May to October. Passer et al. [Reference Passer, Danila, Laine, Como-Sabetti, Tang and Searle28] found an increased risk of sporadic infections when lag effects were introduced to create a 14-day lagged average temperature variable. Overall, there was high-quality evidence to classify temperature as having a moderate impact as a driver of increasing legionellosis incidence.

Relative humidity

Relative humidity (RH) was found to be positively associated with an increased risk of legionellosis in numerous studies. In Simmering et al. [Reference Simmering, Polgreen, Hornick, Sewell and Polgreen16], regardless of precipitation, warm (60–80 °F) and humid weather was a major risk factor for disease, especially in the very humid months (RH >80%). Gleason et al. [Reference Gleason, Kratz, Greeley and Fagliano26] used a time-series design to show that average RH was positively associated with disease risk. Passer et al. [Reference Passer, Danila, Laine, Como-Sabetti, Tang and Searle28] found a positive association for RH in a fully adjusted model though the effect was weaker than the effect of precipitation. The quality of evidence presented was considered high to classify RH as moderate for its magnitude of impact.

Built environment

Few articles analysed the relationship between the built environment and legionellosis incidence in the context of sporadic, community-acquired infections.

Water infrastructure

Two state-level spatial analyses of legionellosis incidence included water source variables. In Connecticut, the incidence was higher in regions with higher rates of private well water usage, with presumably no disinfectant residuals to limit microbial growth in pipes [Reference Cassell, Gacek, Warren, Raymond, Cartter and Weinberger29]. In New Jersey, there was no statistically significant association between LD incidence and whether the public drinking water source was groundwater versus surface water [Reference Gleason, Ross and Greeley23]. Given the lack of epidemiological nationwide analyses or analyses of specific water infrastructure characteristics, there was insufficient evidence to classify its magnitude of impact on incidence.

Older housing stock and urbanisation

Several studies found that legionellosis incidence is higher in urban environments, but there was limited discussion about mechanisms driving incidence in urban contexts [Reference Allgaier, Lagu, Haessler, Imrey, Deshpande, Guo and Rothberg17, Reference Passer, Danila, Laine, Como-Sabetti, Tang and Searle28, Reference Cassell, Gacek, Warren, Raymond, Cartter and Weinberger29]. In a spatial analysis of census tracts in New Jersey, Gleason et al. [Reference Gleason, Ross and Greeley23] found that having a greater percentage of housing units built pre-1970 was strongly associated with legionellosis incidence; the effect was even stronger with pre-1950 housing. Higher percentages of vacant housing and renter-occupied housing were also associated with higher incidence, though the latter relationship did not hold when controlling for poverty and pre-1950 housing [Reference Gleason, Ross and Greeley23]. Populations living in higher-poverty neighbourhoods in New York City were also found to have an increased risk of acquiring legionellosis, indicating that neighbourhood-level factors need to be examined in greater depth [Reference Farnham, Alleyne, Cimini and Balter21]. The strength of association found in Gleason et al. [Reference Gleason, Ross and Greeley23] regarding older housing stock classified this potential driver as having a major impact on increasing incidence though the quality of evidence was low given the singular geographically limited study on the topic.

Discussion

This scoping review assessed the available evidence for legionellosis risk factors and potential differing impacts on driving the increase in incidence observed across the USA since 2006. Although there is a lack of strong evidence to explain the trend of increasing incidence, the included literature suggests that it represents a true increase in disease and cannot be solely explained by improved clinical case capture. Numerous clinical, environmental, infrastructure, and demographic factors are likely playing interconnected roles in driving legionellosis incidence.

Clinical case capture

While improved reporting and diagnostic testing in the early 2000s likely contributed to the rise in reported legionellosis cases in that period, this does not explain the more recent rise in incidence. Between 2002 and 2018, the largest increase in age-standardised incidence occurred in 2016–2018, which cannot be explained by increased clinical testing and reporting trends [Reference Barskey, Derado and Edens2]. In the VA healthcare system with extensive testing for LD in patients that may be more susceptible to infection, increased use of UAT regionally often results in lower test positivity [Reference Gamage, Ambrose, Kralovic, Simbartl and Roselle18]. The increase in testing volume may be due to changes in clinician awareness, health-seeking behaviour, or other factors. Though official clinical practice guidelines about CAP diagnosis by ATS/IDSA recommend Legionella testing only in adult patients with severe CAP or when there are epidemiological indicators such as recent travel or a Legionella outbreak, implementation of the guidelines may miss over 40% of legionellosis cases [Reference Decker, Harris, Muder, Hong, Singh, Sonel and Clancy31, Reference Hollenbeck, Dupont and Mermel32].

While included analyses of EHR data from the USA do not give a clear picture of the impact of increased testing incidence, two international analyses present convincing evidence that incidence is increasing at a faster rate than testing [Reference Leung, Lam, Cheung, Chan and Chuang33, Reference Romay-Lema, Corredoira-Sánchez, Ventura-Valcárcel, Iñiguez-Vázquez, García Pais, García-Garrote and Rabuñal Rey34]. An epidemiological study of LD in Hong Kong found that between 2010 and 2015, UAT frequency increased by 127% and LD case frequency by 230% [Reference Leung, Lam, Cheung, Chan and Chuang33]. Similarly, an epidemiological study of community-acquired Legionella infections in a Spanish hospital found that the number of UAT orders increased 3.4 times and case incidence increased 5.9 times when comparing 2001–2005 with 2011–2015 data [Reference Romay-Lema, Corredoira-Sánchez, Ventura-Valcárcel, Iñiguez-Vázquez, García Pais, García-Garrote and Rabuñal Rey34]. Further, a qualitative study about Swiss physicians’ approach to CAP diagnosis concluded that awareness about LD is generally high and that changing awareness does not appear to explain increasing incidence trends in Switzerland [Reference Fischer, Deml and Mäusezahl35]. Because these international studies may not be generalisable to U.S. health care, there is a need for future quality analyses of EHR data and clinician knowledge, attitudes, and practices towards Legionella testing to better understand clinical trends in the USA. Reported legionellosis incidence may increase in future decades as diagnostic methods such as PCR – which can detect species and serogroups other than L. pneumophila serogroup 1 – become more widespread. This shift was demonstrated in a New Zealand analysis that found a marked increase in incidence corresponding with the uptake of molecular diagnostic methods [Reference Graham, Harte, Zhang, Fyfe and Baker36]; future studies in the USA will need to account for the impact of these changes.

Racial and socioeconomic inequalities and population susceptibility

There is insufficient evidence in the reviewed literature to characterise the impact of racial and socioeconomic inequities on increasing legionellosis incidence in the USA. The high proportion of legionellosis cases reported to the CDC with unknown race – nearly one-fifth of cases from 2000 to 2009 – limits our ability to fully understand the role race plays in risk [Reference Hicks4]. It is evident that factors driving increased risk are disproportionately affecting Black people. The racial disparity has been suggested to be largely driven by economic differences [Reference Barskey, Derado and Edens2, Reference Farnham, Alleyne, Cimini and Balter21]. Disparities in housing stock or community-level infrastructure such as proximity to industrial buildings and cooling towers, ageing water infrastructure and premise plumbing, and the percentage of vacant buildings may drive racial and socioeconomic inequities in environmental exposures [Reference Dooling, Toews, Hicks, Garrison, Bachaus, Zansky, Carpenter, Schaffner, Parker, Petit, Thomas, Thomas, Mansmann, Morin, White and Langley15, Reference Toberna, William, Kram, Heslin and Baumgardner20, Reference Farnham, Alleyne, Cimini and Balter21]. Racial inequities in access to preventive healthcare and higher rates of underlying medical conditions may also contribute to legionellosis as a health disparity issue [Reference Barskey, Derado and Edens2, Reference Hunter, Salandy, Smith, Edens and Hubbard22]. Although there is a lack of evidence about how these dynamics are changing over time with respect to legionellosis, health equity research has found that both racial and socioeconomic health inequities in the USA have persisted [Reference Mahajan, Caraballo, Lu, Valero-Elizondo, Massey, Annapureddy, Roy, Riley, Murugiah, Onuma, Nunez-Smith, Forman, Nasir, Herrin and Krumholz37, Reference Zimmerman and Anderson38]. Our review demonstrates an imperative for further examination of the interaction of legionellosis incidence with health disparities and population susceptibility to better implement equitable interventions.

Climate and hydrometeorological factors

We found that climate and hydrometeorological factors, notably precipitation, underlying the seasonality of LD are major drivers of increasing incidence. Temperature and RH were found to have more moderate impacts with high quality of evidence. The relationships between these climate factors and legionellosis were strengthened when analyses were restricted to the Midwest and Northeast regions of the USA [Reference Han27]. This finding raises concerns that climate change will continue to increase legionellosis incidence in the USA under current trends of more frequent days with temperatures above 90 °F and increased precipitation [Reference Karl, Melillo and Peterson39]. Numerous international studies have similar findings that increased precipitation [Reference Beauté, Sandin, Uldum, Rota, Brandsema, Giesecke and Sparén40Reference Braeye, Echahidi, Meghraoui, Laisnez and Hens43], elevated temperature [Reference Beauté, Sandin, Uldum, Rota, Brandsema, Giesecke and Sparén40Reference Karagiannis, Brandsema and Van der Sande42, Reference Fischer, Saucy, Vienneau, Hattendorf, Fanderl, de Hoogh and Mäusezahl44, Reference Graham, Kim, Baker, Fyfe and Hales45], and higher relative humidity [Reference Halsby, Joseph, Lee and Wilkinson41Reference Braeye, Echahidi, Meghraoui, Laisnez and Hens43, Reference Ricketts, Charlett, Gelb, Lane, Lee and Joseph46] are associated with increased community-acquired legionellosis cases and clusters. Future research is needed to understand the exposure pathways between changing hydrometeorological factors and LD infections.

Built environment

Our review did not identify any studies that examined deficiencies in drinking water systems and their impacts on sporadic, non-outbreak legionellosis incidence. A narrative review of social determinants of health and legionellosis similarly concluded that this relationship has been understudied for sporadic, community-acquired cases [Reference Hunter, Salandy, Smith, Edens and Hubbard22]. At a county level, Rhoads et al. [Reference Rhoads, Garner, Ji, Zhu, Parks, Schwake, Pruden and Edwards47] analysed an outbreak of legionellosis cases in the context of a source water switch that depleted disinfectant residual in the piped network in Flint, Michigan. Low levels of residual chlorine in municipal water systems, changes in corrosion control, and increases in source water turbidity have similarly been linked to legionellosis clusters [Reference Rhoads, Keane, Spencer, Pruden and Edwards48, Reference Cohn, Gleason, Rudowski, Tsai, Genese and Fagliano49]. These studies were excluded from our review because they were conducted in the context of outbreaks, but they highlight the impact that drinking water supplies can have on LD incidence. In a national analysis, Holsinger et al. [Reference Holsinger, Tucker, Regli, Studer, Roberts, Collier, Hannapel, Edens, Yoder and Rotert50] reported legionellosis outbreaks linked to buildings served by public water systems and concluded most outbreaks linked back to healthcare and hotel buildings and buildings served by large (>10,000 people) water systems. The authors did not analyse sporadic cases as they noted that source attribution is lower for such cases [Reference Holsinger, Tucker, Regli, Studer, Roberts, Collier, Hannapel, Edens, Yoder and Rotert50]. We identified a single study attributing legionellosis incidence to older housing units in New Jersey, which may be due to ageing community water infrastructure or housing units’ complex water systems [Reference Gleason, Ross and Greeley23]. Future epidemiological research should examine correlations between sporadic, community-acquired incidence and potentially relevant water system characteristics such as source water type and quality, disinfection type and level, utility age and maintenance, and water main breaks [Reference Simmering, Polgreen, Hornick, Sewell and Polgreen16, Reference Rhoads, Garner, Ji, Zhu, Parks, Schwake, Pruden and Edwards47, Reference Gleason and Cohn51]. There is also a need for future investigation to understand the risk factors that cause Legionella from public water systems to colonise and proliferate in the plumbing of single-family residential homes or other buildings [Reference Gleason and Cohn51] and how these built environment risk factors may be changing over time.

Limitations

The scoping review is limited by the small number of peer-reviewed articles examining the epidemiology of sporadic, community-acquired legionellosis. Outbreak cases represent a smaller proportion of cases in the USA [3, Reference Hicks4] but constitute a large amount of legionellosis literature, thereby leading to conclusions about the drivers of increasing incidence that may not be relevant to sporadic, community-acquired legionellosis. The review also excluded papers that solely addressed travel-associated or nosocomial cases though many of the included papers do not differentiate between these categories. Although we searched for both Legionnaires’ disease and Pontiac fever, we did not identify any papers about Pontiac fever that fit our inclusion criteria.

Summary

This review synthesised and assessed the available evidence about factors that may be driving the increasing incidence of sporadic, community-acquired legionellosis in the USA since 2006. Given the large number of studies screened and the comparatively small number of articles identified that met the inclusion criteria, more work is needed to understand the mechanisms that underlie the more than fivefold increase in reported legionellosis cases in the past two decades, amounting to approximately 10,000 cases in 2018 [Reference Barskey, Derado and Edens2]. This review found clear evidence for climate change as one of these major drivers of increasing incidence. Future research should investigate exposure pathways from the built environment and how racial and socioeconomic inequities are impacting disease incidence. A better understanding of the major factors driving the increase in sporadic, community-associated legionellosis incidence in the USA would inform targets for improved preventive measures and assist in optimising public policy.

Supplementary material

The supplementary material for this article can be found at https://doi.org/10.1017/S0950268823001206.

Data availability statement

Readers may replicate the findings in this scoping review by conducting the same search of peer-reviewed literature in PubMed using November 2022 as the end date and the same keywords described in the Methods section.

Author contribution

Methodology: C.R., S.G., K.J.S., M.A.M., P.G., N.G.E.; Writing – review & editing: C.R., S.G., K.J.S., P.G., N.G.E.; Conceptualization: S.G., N.G.E.; Validation: S.G., N.G.E.; Funding acquisition: K.J.S., N.G.E.; Data curation: M.A.M., N.G.E.; Formal analysis: M.A.M.; Investigation: M.A.M., N.G.E.; Writing – original draft: M.A.M., N.G.E.; Project administration: N.G.E.; Supervision: N.G.E.

Financial support

This work was supported by the Osprey Foundation of Maryland.

References

Soda, EA, Barskey, AE, Shah, PP, Schrag, S, Whitney, CG, Arduino, MJ, Reddy, SC, Kunz, JM, Hunter, CM, Raphael, BH and Cooley, LA (2017) Vital signs: Health care-associated Legionnaires’ disease surveillance data from 20 states and a large metropolitan area – United States, 2015. Morbidity and Mortality Weekly Report 66, 584589.CrossRefGoogle Scholar
Barskey, AE, Derado, G and Edens, C (2022) Rising incidence of Legionnaires’ disease and associated epidemiologic patterns, United States, 1992-2018. Emerging Infectious Diseases 28, 527538.CrossRefGoogle ScholarPubMed
National Academies of Sciences Engineering and Medicine (2020) Management of Legionella in Water Systems. Washington, D.C.: The National Academies Press.Google Scholar
Hicks, LA, et al. (2011) Legionellosis — United States, 2000-2009. Morbidity and Mortality Weekly Report 60, 10831086.Google Scholar
Cassell, K, Gacek, P, Rabatsky-Ehr, T, Petit, S, Cartter, M and Weinberger, DM (2019) Estimating the true burden of Legionnaires’ disease. American Journal of Epidemiology 188, 16861694.CrossRefGoogle ScholarPubMed
Murdoch, DR (2003) Diagnosis of Legionella infection. Clinical Infectious Diseases 36, 6469.CrossRefGoogle ScholarPubMed
Cunha, BA, Burillo, A and Bouza, E (2016) Legionnaires’ disease. The Lancet 387, 376385.CrossRefGoogle ScholarPubMed
Htwe, TH and Khardori, NM (2017) Legionnaire’s disease and immunosuppressive drugs. Infectious Disease Clinics 31, 2942.Google ScholarPubMed
Cunha, BA (2001) Pneumonia in the elderly. Clinical Microbiology and Infection 7, 581588.CrossRefGoogle ScholarPubMed
LeChevallier, MW (2019) Occurrence of culturable Legionella pneumophila in drinking water distribution systems. AWWA Water Science 1, e1139.Google Scholar
Hamilton, KA, Hamilton, MT, Johnson, W, Jjemba, P, Bukhari, Z, LeChevallier, M, Haas, CN and Gurian, PL (2019) Risk-based critical concentrations of Legionella pneumophila for indoor residential water uses. Environmental Science & Technology 53, 45284541.CrossRefGoogle ScholarPubMed
Pampaka, D, Gómez-Barroso, D, López-Perea, N, Carmona, R and Portero, RC (2022) Meteorological conditions and Legionnaires’ disease sporadic cases-a systematic review. Environmental Research 214, 114080.CrossRefGoogle ScholarPubMed
Tricco, AC, Lillie, E, Zarin, W, O’Brien, KK, Colquhoun, H, Levac, D, Moher, D, Peters, MDJ, Horsley, T, Weeks, L, Hempel, S, Akl, EA, Chang, C, McGowan, J, Stewart, L, Hartling, L, Aldcroft, A, Wilson, MG, Garritty, C, Lewin, S, Godfrey, CM, Macdonald, MT, Langlois, EV, Soares-Weiser, K, Moriarty, J, Clifford, T, Tunçalp, Ö and Straus, SE (2018) PRISMA extension for scoping reviews (PRISMA-ScR): Checklist and explanation. Annals of Internal Medicine 169, 467473.CrossRefGoogle ScholarPubMed
Alarcon Falconi, TM, Cruz, MS and Naumova, EN (2018) The shift in seasonality of legionellosis in the U.S. Epidemiology & Infection 146, 18241833.CrossRefGoogle Scholar
Dooling, KL, Toews, KA, Hicks, LA, Garrison, LE, Bachaus, B, Zansky, S, Carpenter, LR, Schaffner, B, Parker, E, Petit, S, Thomas, A, Thomas, S, Mansmann, R, Morin, C, White, B and Langley, GE (2015) Active bacterial core surveillance for Legionellosis — United States, 2011-2013. Morbidity and Mortality Weekly Report 64, 11901193.CrossRefGoogle ScholarPubMed
Simmering, JE, Polgreen, LA, Hornick, DB, Sewell, DK and Polgreen, PM (2017) Weather-dependent risk for Legionnaires’ disease, United States. Emerging Infectious Diseases 23, 18431851.CrossRefGoogle ScholarPubMed
Allgaier, J, Lagu, T, Haessler, S, Imrey, PB, Deshpande, A, Guo, N and Rothberg, MB (2021) Risk factors, management, and outcomes of Legionella pneumonia in a large, nationally representative sample. Chest 159, 17821792.CrossRefGoogle Scholar
Gamage, SD, Ambrose, M, Kralovic, SM, Simbartl, LA and Roselle, GA (2018) Legionnaires disease surveillance in US department of veterans affairs medical facilities and assessment of health care facility association. JAMA Network Open 1, e180230.CrossRefGoogle ScholarPubMed
Schoonmaker-Bopp, D, et al. (2021) Improvements to the success of outbreak investigations of Legionnaires’ disease: 40 years of testing and investigation in New York state. Applied and Environmental Microbiology 87, 116.CrossRefGoogle Scholar
Toberna, CP, William, HM, Kram, JJF, Heslin, K and Baumgardner, DJ (2020) Epidemiologic survey of Legionella urine antigen testing within a large Wisconsin-based health care system. Journal of Patient-Centered Research and Reviews 7, 165175.CrossRefGoogle ScholarPubMed
Farnham, A, Alleyne, L, Cimini, D and Balter, S (2014) Legionnaires’ disease incidence and risk factors, New York, New York, USA, 2002-2011. Emerging Infectious Diseases 20, 17951802.CrossRefGoogle ScholarPubMed
Hunter, CM, Salandy, SW, Smith, JC, Edens, C and Hubbard, B (2022) Racial disparities in incidence of Legionnaires’ disease and social determinants of health: A narrative review. Public Health Reports 137, 660671.CrossRefGoogle ScholarPubMed
Gleason, JA, Ross, KM and Greeley, RD (2017) Analysis of population-level determinants of Legionellosis: Spatial and geovisual methods for enhancing classification of high-risk areas. International Journal of Health Geographics 16, 45.CrossRefGoogle ScholarPubMed
Han, XY (2021) Effects of climate changes and road exposure on the rapidly rising Legionellosis incidence rates in the United States. PLoS ONE 16, 119.Google ScholarPubMed
Cooley, LA, Pondo, T, Francois Watkins, LK, Shah, P, Schrag, S, Active Bacterial Core Surveillance Program of the Emerging Infections Program Network, Vugia, D, Alden, N, Cartter, M, Thomas, S (S), Ryan, PA, Laine, E, Pradhan, E and Dunn, J (2020) Population-based assessment of clinical risk factors for Legionnaires’ disease. Clinical Infectious Diseases 70, 24282431.CrossRefGoogle ScholarPubMed
Gleason, JA, Kratz, NR, Greeley, RD and Fagliano, JA (2016) Under the weather: Legionellosis and meteorological factors. EcoHealth 13, 293302.CrossRefGoogle ScholarPubMed
Han, XY (2019) Solar and climate effects explain the wide variation in Legionellosis incidence rates in the United States. Applied and Environmental Microbiology 85, e01776-19.CrossRefGoogle ScholarPubMed
Passer, JK, Danila, RN, Laine, ES, Como-Sabetti, KJ, Tang, W and Searle, KM (2020) The association between sporadic Legionnaires’ disease and weather and environmental factors, Minnesota, 2011-2018. Epidemiology & Infection 148, e156.CrossRefGoogle ScholarPubMed
Cassell, K, Gacek, P, Warren, JL, Raymond, PA, Cartter, M and Weinberger, DM (2018) Association between sporadic legionellosis and river systems in Connecticut. Journal of Infectious Diseases 217, 179187.CrossRefGoogle ScholarPubMed
Neil, K and Berkelman, R (2008) Increasing incidence of Legionellosis in the United States, 1990-2005: Changing epidemiologic trends. Clinical Infectious Diseases 47, 591599.CrossRefGoogle ScholarPubMed
Decker, BK, Harris, PL, Muder, RR, Hong, JH, Singh, N, Sonel, AF and Clancy, CJ (2016) Improving the diagnosis of Legionella pneumonia within a healthcare system through a systematic consultation and testing program. Annals of the American Thoracic Society 13, 12891293.CrossRefGoogle ScholarPubMed
Hollenbeck, B, Dupont, I and Mermel, LA (2011) How often is a work-up for Legionella pursued in patients with pneumonia? A retrospective study. BMC Infectious Diseases 11, 14.CrossRefGoogle ScholarPubMed
Leung, Y-H, Lam, CK, Cheung, YY, Chan, CW and Chuang, SK (2020) Epidemiology of Legionnaires’ disease, Hong Kong, China, 2005-2015. Emerging Infectious Diseases 26, 16951702.CrossRefGoogle ScholarPubMed
Romay-Lema, E, Corredoira-Sánchez, J, Ventura-Valcárcel, P, Iñiguez-Vázquez, I, García Pais, MJ, García-Garrote, F and Rabuñal Rey, R (2018) Community acquired pneumonia by Legionella pneumophila: Study of 136 cases. Medicina Clinica 151, 265269.CrossRefGoogle ScholarPubMed
Fischer, FB, Deml, MJ and Mäusezahl, D (2022) Legionnaires’ disease – A qualitative study on Swiss physicians’ approaches to the diagnosis and treatment of community-acquired pneumonia. Swiss Medical Weekly 152, w30157.CrossRefGoogle Scholar
Graham, FF, Harte, D, Zhang, J, Fyfe, C and Baker, MG (2023) Increased incidence of Legionellosis after improved diagnostic methods, New Zealand, 2000-2020. Emerging Infectious Diseases 29, 11731182.CrossRefGoogle ScholarPubMed
Mahajan, S, Caraballo, C, Lu, Y, Valero-Elizondo, J, Massey, D, Annapureddy, AR, Roy, B, Riley, C, Murugiah, K, Onuma, O, Nunez-Smith, M, Forman, HP, Nasir, K, Herrin, J and Krumholz, HM (2021) Trends in differences in health status and health care access and affordability by race and ethnicity in the United States, 1999-2018. JAMA 326, 637.CrossRefGoogle ScholarPubMed
Zimmerman, FJ and Anderson, NW (2019) Trends in health equity in the United States by race/ethnicity, sex, and income, 1993-2017. JAMA Network Open 2, e196386. https://doi.org/10.1001/jamanetworkopen.2019.6386CrossRefGoogle Scholar
Karl, TR, Melillo, JM and Peterson, TC (2009) Global Climate Change Impacts in the United States. New York, NY: Cambridge University Press.Google Scholar
Beauté, J, Sandin, S, Uldum, SA, Rota, MC, Brandsema, P, Giesecke, J and Sparén, P (2016) Short-term effects of atmospheric pressure, temperature, and rainfall on notification rate of community-acquired Legionnaires’ disease in four European countries. Epidemiology and Infection 144, 34833493.CrossRefGoogle ScholarPubMed
Halsby, KD, Joseph, CA, Lee, JV and Wilkinson, P (2014) The relationship between meteorological variables and sporadic cases of Legionnaires’ disease in residents of England and Wales. Epidemiology and Infection 142, 23522359.CrossRefGoogle ScholarPubMed
Karagiannis, I, Brandsema, P and Van der Sande, M (2009) Warm, wet weather associated with increased Legionnaires’ disease incidence in the Netherlands. Epidemiology and Infection 137, 181187.CrossRefGoogle ScholarPubMed
Braeye, T, Echahidi, F, Meghraoui, A, Laisnez, V and Hens, N (2020) Short-term associations between Legionnaires’ disease incidence and meteorological variables in Belgium, 2011–2019. Epidemiology and Infection 148, e150.CrossRefGoogle ScholarPubMed
Fischer, FB, Saucy, A, Vienneau, D, Hattendorf, J, Fanderl, J, de Hoogh, K, Mäusezahl, D. (2023) Impacts of weather and air pollution on Legionnaires’ disease in Switzerland: A national case-crossover study. Environmental Research 233, 116327.CrossRefGoogle ScholarPubMed
Graham, FF, Kim, AHM, Baker, MG, Fyfe, C and Hales, S (2023) Associations between meteorological factors, air pollution and Legionnaires’ disease in New Zealand: Time series analysis. Atmospheric Environment 296, 119572.CrossRefGoogle Scholar
Ricketts, KD, Charlett, A, Gelb, D, Lane, C, Lee, JV and Joseph, CA (2009) Weather patterns and Legionnaires’ disease: A meteorological study. Epidemiology and Infection 137, 10031012.CrossRefGoogle ScholarPubMed
Rhoads, WJ, Garner, E, Ji, P, Zhu, N, Parks, J, Schwake, DO, Pruden, A and Edwards, MA (2017) Distribution system operational deficiencies coincide with reported Legionnaires’ disease clusters in Flint, Michigan. Environmental Science and Technology 51, 1198611995.CrossRefGoogle ScholarPubMed
Rhoads, WJ, Keane, T, Spencer, MS, Pruden, A and Edwards, MA (2020) Did municipal water distribution system deficiencies contribute to a Legionnaires’ disease outbreak in Quincy, IL? Environmental Science and Technology Letters 7, 896902.CrossRefGoogle Scholar
Cohn, PD, Gleason, JA, Rudowski, E, Tsai, SM, Genese, CA and Fagliano, JA (2015) Community outbreak of legionellosis and an environmental investigation into a community water system. Epidemiology and Infection 143, 13221331.CrossRefGoogle Scholar
Holsinger, H, Tucker, N, Regli, S, Studer, K, Roberts, VA, Collier, S, Hannapel, E, Edens, C, Yoder, JS and Rotert, K (2022) Characterization of reported legionellosis outbreaks associated with buildings served by public drinking water systems: United States, 2001-2017. Journal of Water and Health 20, 702711.CrossRefGoogle ScholarPubMed
Gleason, JA and Cohn, PD (2022) A review of legionnaires’ disease and public water systems – Scientific considerations, uncertainties and recommendations. International Journal of Hygiene and Environmental Health 240, 113906.CrossRefGoogle ScholarPubMed
Figure 0

Table 1. Rating system to evaluate the potential drivers identified in articles of scoping review

Figure 1

Table 2. Potential drivers of increasing legionellosis incidence in the United States, 2006–2022: magnitude of impact and quality of evidence

Supplementary material: File

Moffa et al. supplementary material
Download undefined(File)
File 109.4 KB