. 2021 Jun 29;137(4):660–671. doi: 10.1177/00333549211026781

Table.

Articles on health disparities in Legionnaires’ disease, by study period (N = 19)

Study design	Geographic area	Study population/sample size	Outcome measures	Primary results	Primary SDH examined
Burton et al,⁸ 2003-2004
Active, population-based surveillance for invasive bacterial disease (ABCs)	Connecticut (entire state), Minnesota (entire state), California (1 county), Colorado (5 counties), Georgia (20 counties), Maryland (6 counties), New York (15 counties), Oregon (3 counties), and Tennessee (11 counties)	4870 cases of invasive community-acquired bacterial pneumonia among adults aged ≥18	Annual incidence rate of invasive bacterial pneumonia in each census tract	The RR comparing incidence between Black and White adults was 2.40 (95% CI, 2.24-2.57). The overall incidence rate among Hispanic adults was 2.40 (95% CI, 2.24-2.57). Incidence among Black residents of census tracts with ≥20% of people living in poverty (most impoverished) was 4.4 times the incidence among White residents of census tracts with <5% of people living in poverty (least impoverished).	Economic stability (population employed in working-class occupations, median annual household income, percentage of low-income housing, percentage of high-income housing)
Bush et al,⁷ 2007-2017
Routine surveillance	New York, New York	Cases of Legionnaires’ disease/sociodemographic analysis on 1011 cases from 2015-2017	Annual incidence rate of Legionnaires’ disease	Legionnaires’ disease incidence (average annual age-adjusted rate per 100 000 residents) was highest among Black residents (6.0) compared with Latino (2.9), White (2.7), and Asian (1.8) residents. Legionnaires’ disease incidence was highest in neighborhoods with the highest poverty levels (5.7) compared with neighborhoods with the lowest poverty levels (3.0). 11% of Legionnaires’ disease patients reported a health care exposure in the 10 days before they developed symptoms.	Economic stability (percentage of residents living below the FPL); health care access and quality (comorbidities [ie, diabetes, smoking status] and health care–associated exposure) Neighborhood and built environment (neighborhood borough)
Chapman et al,¹⁸ 2006-2011
Routine surveillance	Northeast England	1351 cases of IPD stratified by age (<16, 16-29, 30-44, 45-64, ≥65)	IPD incidence rate	IPD incidence increased linearly with increasing deprivation from 7.0 to 13.6 per 100 000 population. Incidence of IPD was positively correlated with increasing deprivation quintile for income deprivation, employment deprivation, health deprivation and disability, education, skills and training deprivation, crime, and living environment domains.	Economic stability (income deprivation, employment deprivation) Health care access and quality (health deprivation and disability) Education access and quality (education skills and training deprivation) Neighborhood and built environment (barriers to housing and services, living environment deprivation, crime; urban/rural classification)
Corrado et al,¹⁹ 2010-2014
Retrospective analysis	New York, New York	New York City residents aged ≥18; 4 614 108 hospitalizations of adult residents, 283 927 pneumonia-associated	Pneumonia-associated hospitalizations	Neighborhoods with >30% of residents living below the FPL had significantly higher mean annual age-adjusted rates of pneumonia-associated hospitalizations than neighborhoods with <10% of residents living below the FPL (RR = 2.5; 95% CI, 2.5-2.6).	Economic stability (poverty level) Health care access and quality (health insurance type, facility type)
Farnham et al,⁶ 2002-2011
Routine surveillance	New York, New York	1449 legionellosis cases	Incidence of legionellosis	Average incidence per 100 000 population per year for non-Hispanic Black people (2.15 cases) was higher than that for non-Hispanic White people (1.56 cases; P = .13) and significantly higher than that for Hispanic (1.02 cases: P = .003) or Asian/Pacific Islander (0.41 cases; P < .001) people; incidence in the highest poverty areas (3.0 average yearly cases per 100 000 population) was 2.5 times higher than that for the lowest poverty areas (1.2 average yearly cases per 100 000 population); 1278 (88.2%) patients had >1 underlying medical condition that was a known risk factor for legionellosis; current or past smoking (n = 879; 60.7%) and diabetes (n = 506; 34.9%) were the most frequently reported underlying conditions.	Economic stability (occupation, poverty level) Health care access and quality (comorbidities/medical risk factors)
Feemster et al,¹⁰ 2005-2008
Population-based surveillance network for pneumococcal disease	5-county Philadelphia metropolitan region	All hospitalized adult and pediatric patients with pneumococcal bacteremia (stratified by age: <18, 18-39, 40-69, ≥70)	Incidence of pneumococcal bacteremia	Increased disease incidence was associated with higher population density (incidence RR [IRR] = 1.10 [95% CI, 1.00-1.19] per 10 000 people per mile), higher percentage Black population (per 10% increase) (IRR = 1.07; 95% CI, 1.04-1.09), population aged ≤5 years (IRR = 3.49; 95% CI, 1.80-5.18), and population aged ≥65 (IRR = 1.19; 95% CI, 1.00-1.38).	Economic stability (average income, percentage below FPL, average household size)
Gleason et al,²⁰ 2003-2013
Geospatial analysis of legionellosis cases	New Jersey	1634 cases of Legionnaires’ disease	Legionnaires’ disease incidence, cluster incidence	Census tracts with the highest percentages of older age (OR = 0.38; 95% CI, 0.22-0.67); Hispanic population (OR = 0.10; 95% CI 0.05-0.19); poverty (OR = 7.21; 95% CI, 4.04-12.86); and housing units built pre-1950 (OR = 5.69; 95% CI, 2.82-11.50) were positively and significantly associated with legionellosis cluster areas.	Economic stability (poverty) Neighborhood and built environment (rent vs own home, age of housing stock, public water systems) Education ≤high school
Hayes et al,⁹ 2001-2014
Retrospective analysis	Nationally representative dataset	20 361 181 pneumonia-associated hospitalizations	Estimated numbers and rates of pneumonia-associated hospitalizations	Non-Hispanic American Indian/Alaska Native and non-Hispanic Black people had the highest average annual age-adjusted rates of pneumonia-associated hospitalization of all racial/ethnic groups at 439.2 (95% CI, 415.9-462.5) and 438.6 (95% CI, 432.5-444.7) per 100 000 population. The mean age at in-hospital death was lowest in the lowest income quartile (73.1 y [95% CI, 72.9-73.2]) and highest in the highest income quartile (76.4 y [95% CI, 76.2-76.6]). In-hospital death occurred in 10.9% (SE = 0.1%) of pneumonia-associated hospitalizations colisting an immunocompromising condition compared with 6.3% (SE = 0.03%) among those not colisting an immunocompromising condition (P < .001).	Economic stability (income) Health care access and quality (comorbidity-immune status)
Jeon et al,²¹ 2006-2008
Retrospective analysis	Referral hospital in upper Manhattan, New York	52 006 patients	Infections (bloodstream, urinary tract, and pneumonia) present on admission and health care–associated infections	Odds ratio for hospitalization related to community-acquired pneumonia was high for people with low median household income (OR = 1.90; 95% CI, 1.12-3.20), diabetes (OR = 1.09; 95% CI, 0.82-1.44), and Medicaid (OR = 1.90; 95% CI, 1.35-2.67).	Economic stability (median household income) Health care access and quality (comorbidity by ICD-9 code, primary payer health insurance type)
Kempker et al,²² 1999-2005
Longitudinal analysis	Nationally representative dataset	1523 septicemia deaths	Septicemia, influenza, and pneumonia death rates	Higher HRs related to influenza/pneumonia deaths observed for high poverty (HR = 3.0; 95% CI, 2.1-4.3), diabetes (HR = 1.46; 95% CI, 1.28-1.67), and reported receiving health care >10 times in the past year (HR = 1.87; 95% CI, 1.68-2.08).	Economic stability (income, poverty) Health care access and quality (comorbidity, self-reported use of health care)
Kim et al,²³ 2005
Geospatial analysis	South Korea	402 979 hospitalizations due to bacterial pneumonia (ages 0 to >65)	Bacterial pneumonia hospitalization rate	Significant β coefficient related to deprivation index (1.993; P < .001).	Economic stability (deprivation index—car ownership, engaged in manual labor)
Marzouk et al,²⁴ 2008-2012
Descriptive, cross-sectional	Tunisia	14 Confirmed Legionnaires’ disease cases	Legionnaires’ disease incidence	4 of 14 (29%) confirmed cases reported high humidity in their homes.	Neighborhood and built environment (home humidity, living in community)
McLaughlin et al,²⁵ 2014-2017
Cross-sectional analysis of medical and prescription claims	United States	22 223 586 adults aged ≥65 who received PCV13	PCV13 uptake among adults aged ≥65	Lower uptake of PCV13 was observed for non-Hispanic Black (36.3%) and Hispanic (30.0%) adults (vs 45.6% for non-Hispanic White adults, P < .01), the poor (30.7% vs 54.2% among lowest vs highest income deciles, P < .01), adults with low educational status (33.0% vs 49.0% among adults with <high school education vs college educated, P < .01).	Economic stability (household income) Education level
Muhammad et al,²⁶ 1998-2009
Active, population-based surveillance of IPD	Nationally representative dataset	35 925 cases of IPD among adults	IPD prevalence	Diabetes was the third most prevalent comorbidity for IPD among adults aged 18-49 (23%) and ≥65 (19.5%); most common comorbidity for adults aged 50-64 (35%).	Health care access and quality (comorbidities)
Nowalk et al,²⁷ 2014
Descriptive, cross-sectional	Nationally representative dataset	2 193 296 patients hospitalized with pneumococcal disease	Pneumococcal disease hospitalizations	The likelihood of non-immunocompromising pneumococcal high-risk conditions was 12% higher among Black people than among non-Black people.	Health care access and quality (comorbidities)
Oggioni et al,²⁸ 2013
Case study	Malta	1 patient with Legionella coinfection and herpesvirus 3	Legionella pneumophila and human herpesvirus 3 coinfection	Legionella samples from the patient’s workplace toilet matched clinical samples, demonstrating an epidemiological link between clinical and environmental specimens.	Neighborhood and built environment (environmental samples from home and workplace)
Soto et al,¹¹ 1998-1999, 2007-2008
Descriptive, cross-sectional	Connecticut	5023 cases of IPD	IPD incidence rate and RR of IPD	IPD rates among people living in the highest-poverty census tracts were 2.2 times higher among people living in the poorest census tracts than among people living in the lowest-poverty tracts in 2007-2008; 2.4 times higher in 1998-1999.	Economic stability (census tract high poverty vs low poverty)
Storch et al,²⁹ 1976-1979
Case-control study	Nationally representative dataset	169 cases of Legionnaires’ disease	RR of Legionnaires’ disease	People with Legionnaires’ disease were significantly more likely to have lived near construction sites (RR = 4.4; 95% CI, 1.2-15.7) and have underlying disease (RR = 2.0; 95% CI, 0.9-4.7) than controls. Seven cases were construction workers, and no construction workers were among the controls.	Health care access and quality (comorbidities—underlying chronic diseases) Neighborhood and built environment (eg, living near construction sites, occupation, exposure to soil/wind-blown dust)
Wortham et al,³⁰ 1998-2009
Surveillance	Nationally representative dataset	47 449 cases of IPD (stratified by children aged ≤5 y; children and adults aged >5) in the ABCs/Emerging Infections Program network	IPD rates	IPD rates were 12.6 (95% CI, 12.2-13.1) among White people aged ≥5 years compared with 19.4 (95% CI, 18.0-20.8) among Black people.	Health care (hospitalization status, presence of underlying conditions); these variables were used to predict race where race variable was missing

Abbreviations: ABCs, Active Bacterial Core surveillance; HR, hazards ratio; ICD-9, International Classification of Diseases, Ninth Revision; IPD, invasive pneumococcal disease; OR, odds ratio; PCV, pneumococcal conjugate vaccine; RR, rate ratio; SDH, social determinant of health.