What this study adds
A nascent body of research links industrial food animal production—a source of air pollution—with adverse respiratory outcomes in nearby communities. This study contributes to the field by analyzing residential proximity to poultry operations and community-acquired pneumonia in Pennsylvania. To our knowledge, no prior studies have assessed this relationship in the United States, despite a high prevalence of industrial poultry production and the substantial burden of pneumonia on mortality and hospitalization rates. Using a more thorough measurement approach than past epidemiologic studies, we found that living closer to more and larger poultry operations may increase individual risk of community-acquired pneumonia.
In the United States, pneumonia is a leading cause of death and illness-related hospitalization.1 , 2 Young children and older adults face the highest burden of community-acquired pneumonia (CAP),3 , 4 but it is also a common and costly infection among working-age adults.5 Risk factors for CAP include lifestyle factors (e.g., smoking, poor dental hygiene), inhalation and oxygen therapy, and comorbid conditions (e.g., chronic respiratory disease, HIV infection).6 , 7
Ambient air pollution also increases risk of lower respiratory infections, including CAP.8–10 Exposure to air pollutants such as particulate matter induces oxidative stress in pulmonary macrophages and epithelial cells, reducing the lung’s defenses against bacterial pathogens, thereby increasing susceptibility to respiratory infections.10 , 11 Exposure to air pollution may also disrupt the composition of the oropharyngeal microbiota, which is believed to play an essential role in respiratory health. Through the use of available nutrients and niches, a balanced microbiota provides resistance to the acquisition and establishment of pathogens and prevents the overgrowth of pathobionts (potential pathogens endemic to the respiratory tract) such as Streptococcus pneumoniae, the most common cause of CAP.12 , 13 Prior research found an absence of protective bacteria among pneumonia patients, leading to an imbalanced oropharyngeal microbiota and creating openings for overgrowth of pathogenic bacteria.14
As a source of air pollution, industrial food animal production (IFAP) can compromise respiratory health.15 These large, homogeneous, densely packed livestock operations emit particulate matter, endotoxins, and other pollutants, which spread downwind through ventilation fans and emissions from decomposing manure.15–18 Adverse effects on lung function and increased respiratory symptoms have been reported among individuals living near IFAP, particularly among susceptible groups.19–21 A group conducted a series of studies in livestock-intensive areas of the Netherlands to examine associations between living near animal farms and pneumonia.22–25 These studies reported a largely consistent finding of increased pneumonia risk among individuals living near goat farms and a less consistent relationship for poultry farm exposure. However, three of the four studies used data gathered during a Coxiella burnetti outbreak, an infection generally associated with goats, but not poultry, that leads to Q-fever and can manifest in pneumonia.23–25 In addition to this limitation, only one study accounted for accumulated risk from multiple farms in proximity to individuals’ home addresses,25 and none incorporated animal density in their exposure metrics.
Limited existing evidence in settings with low Q-fever warrants greater understanding of the relationship between living near poultry IFAP and lower respiratory infections, particularly studies that comprehensively account for poultry operation proximity and size. Furthermore, to our knowledge, no studies have examined relationships between poultry IFAP and CAP in the United States, despite a high prevalence of poultry IFAP. Thus, the goal of this study was to evaluate the association between individual-level residential proximity to high-density poultry operations and CAP diagnosis.
We conducted a nested case–control study among patients of Geisinger, a large integrated health system in Pennsylvania. To identify CAP cases, we used electronic health record (EHR) data for 501,843 children and adults with an outpatient, inpatient, or emergency department contact between January 2004 and July 2015. The study area included 38 Pennsylvania counties comprising Geisinger’s primary care market and bordering counties. Geisinger’s primary care patients represent the age and sex distribution of the general population in central and northeastern Pennsylvania.26 The latitude and longitude of subjects’ addresses were geocoded using ArcGIS version 10.1 (Esri, Redlands, CA).27 The Geisinger Institutional Review Board approved the study and waived informed consent.
Case ascertainment and control selection
We identified incident CAP cases using Geisinger system and International Classification of Diseases (ICD, 9th and 10th Revisions) diagnostic codes. ICD-10 codes were internally converted to ICD-9 during data extraction. Cases were defined as subjects with an ICD-9 diagnostic code for pneumonia (480.x, 481, 482.x, 483.x, 484.x, 485, 486, 487.0) and a Current Procedural Terminology code for chest radiological imaging.28 We identified cases from codes from the following encounter types: outpatient (53% of cases), inpatient (23%), and emergency department (21%); an additional 3% of cases were identified via medication records. To exclude nosocomial pneumonia cases,29 we omitted subjects with hospitalization for more than 2 days in the 90 days preceding pneumonia diagnosis; residence in an institution (i.e., nursing home, assisted living facility, residential treatment center, or prison); diagnosis of ventilator-associated pneumonia or Pneumocystis jirovecii pneumonia in the year prior to pneumonia diagnosis; diagnosis with an excluding condition (i.e., recurrent pneumonia, solid organ or bone marrow transplant, immunodeficiency, AIDS, hematologic malignancies, primary lung cancer, or tuberculosis); chemotherapy or radiation therapy in the 6 months prior to pneumonia diagnosis; or outpatient dialysis in the 30 days prior to pneumonia diagnosis.
Randomly selected outpatient controls were frequency-matched (5:1) to cases by age category, sex, and encounter in the same year as CAP diagnosis. Exclusion criteria for controls included a prior pneumonia diagnostic code, diagnosis with an excluding condition (listed above), or residence in an institution. Subjects could serve as controls over multiple years, and cases were eligible to serve as controls until the year of CAP diagnosis.
Poultry operation data
Poultry operation data were identified using publicly available nutrient management plans (NMPs), which are required for livestock operations that exceed either: (1) two animal equivalent units (AEUs, 1000 pounds of live weight on an annualized basis) per acre and have greater than eight AEUs (per Pennsylvania Act 38); or (2) 1000 total AEUs (per US Clean Water Act). Considering AEUs are adjusted for the proportion of time during the year that animals are present on an operation, eight AEUs would be equivalent, for example, to 5050 five-pound broilers (chickens raised for meat) raised over the course of six 6-week cycles (252 production days). We obtained NMPs for the year 2015 for operations in the study area from County Conservation Districts. We used operation addresses to assign latitude and longitude and Google Earth to confirm the presence of a poultry house.30
Poultry types included broilers, layers (egg-laying hens), pullets (young hens produced for breeding), turkeys, and ducks. Air pollutant emissions from poultry operations likely vary by poultry type, the local climate of the poultry house, ventilation type, and manure management system.31
Poultry operation activity metric assignment
We created a poultry operation activity metric to account for the intensity and proximity of poultry operations using total number of poultry operations in the study area, distance between each subject’s residence and poultry operations, and number of poultry AEUs at each operation. We used inverse distance-squared gravity measures32 to derive the activity metric for each subject as previously reported:33
where n was the number of operations, a i was poultry AEUs at operation i, and d ij 2 was the squared-distance (m2) between operation i and subject j. The activity metric was modeled as quartiles; quartile one represented the lowest activity and served as a reference category. The metric was not time varying as IFAP operations in the region are largely static.34 We used R version 3.3.1 (R Foundation for Statistical Computing) to compute the metric.
We obtained sex, age at diagnosis/encounter, race/ethnicity, history of Medical Assistance (measure of low socioeconomic status35), and smoking status from the EHR. We used medication records to create an indicator (yes versus no) for any antibiotic order and orders for a macrolide, tetracycline, or quinolone (antibiotics commonly used to treat CAP29) in the 30 days and 6 months prior to diagnosis/encounter. We examined indications for antibiotic orders when available. Covariates created using subjects’ geocoded addresses included distance in meters between subject residence and nearest Geisinger clinic or hospital, presence of a Geisinger hospital in the subject’s county of residence (yes versus no), community type (i.e., townships, boroughs, cities), and community socioeconomic deprivation (CSD). Community type was identified using a mixed definition of place that generally captures a rural to urban gradient.27 CSD was based on six US Census indicators as previously described.36 We defined season of diagnosis as winter (December–February), spring (March–May), summer (June–August), and fall (September–November).
The analysis goal was to evaluate associations between the poultry operation activity metric and CAP diagnosis. We used mixed effects logistic regression with a random intercept for subject to account for controls later serving as cases (and thus multiple observations per subject). To account for the spatial dependence of subjects living in the same communities, we used robust standard errors clustered by community. We adjusted models a priori for sex, age (centered and centered squared), race/ethnicity (non-Hispanic white versus black/Hispanic/other), Medical Assistance (yes versus no), and smoking status (never versus former, current, and unknown). Representing a perturbation to the microbiota,12 we originally hypothesized that prior antibiotic orders may increase risk of CAP but ultimately did not include antibiotics in models because their use in various time intervals before diagnosis of CAP was strongly associated with the CAP diagnosis used as our outcome. We assessed distance to a Geisinger facility and the presence of a Geisinger hospital in the county of residence as potential confounders because long distances could motivate individuals to seek care for pulmonary symptoms at local, non-Geisinger facilities or hospitals. We evaluated CSD as a confounder since it was previously linked to higher CAP incidence.37 To determine confounding, we assessed whether poultry metric/CAP associations changed by ≥10% when potential confounders were added to regression models. We report results as adjusted odds ratios with 95% confidence intervals and P values. Results were considered significant at P < 0.05 (two-tailed).
We stratified models by community type, as we hypothesized that the association between the poultry operation activity metric and CAP might vary by community type. Prior studies have demonstrated individual-level and place-level differences by community type.36 , 38 We also evaluated effect modification by Medical Assistance, hypothesizing that poorer health status among low-income populations may increase their vulnerability to IFAP-related CAP; age, given higher CAP incidence in young children and children’s vulnerability to air pollutants;4 , 39 and season of diagnosis, because CAP and exposures from poultry operations vary seasonally.3 To evaluate effect modification, we included cross-product terms of variables with the poultry operation activity metric. We used Stata version 14.0 (StataCorp, LP, College Station, TX) for analyses.
We identified 11,910 CAP cases between 2004 and 2015, matched to 59,550 controls. CAP diagnoses increased over the study period as the hospital system expanded and EHR use increased (Table 1). Approximately one-third of cases were children under age 12 years and nearly 30% of cases were older than 62 years. The majority of cases were non-Hispanic white, reflecting the region’s demographics. Cases were about three times more likely than controls to have received an antibiotic order in the 30 days prior to diagnosis and about seven times more likely to have received a macrolide, tetracycline, or quinolone order. Prior antibiotic orders were primarily linked to diagnostic codes for upper respiratory tract infections (e.g., sinusitis); cases were more likely than controls to have received an antibiotic for bronchitis, chronic airway obstruction, fever, or cough. A slightly larger proportion of cases than controls lived in cities and boroughs.
We collected NMPs for 304 high-density poultry operations, located in 16 counties (Figure 1), the majority of which were broiler (53%) or layer (24%) operations. Operations had a median (IQR) of 154 (101, 250) AEUs (e.g., an operation raising 74,000 broilers every 6 weeks for six cycles). A comparison with US Agricultural Census data40 indicated NMP data accounted for about 95% of broilers, layers, and pullets produced in the study area. Median distances between study subjects’ residences and the nearest poultry operation were 61, 29, 10, and 4 km for quartiles 1–4 of the poultry operation activity metric, respectively.
In an unadjusted model, subjects with a poultry operation activity metric in the fourth quartile (versus first) had significantly increased odds of CAP diagnosis; inferences did not change after adjustment for covariates (adjusted odds ratio [confidence interval]) Q2: 0.98 [0.74, 1.31]; Q3: 1.17 [0.93, 1.46]; Q4: 1.66 [1.27, 2.18]). We found no evidence of confounding by distance to a Geisinger facility, presence of a Geisinger hospital in the county of residence, or CSD (results not shown).
In models stratified by community type (townships and boroughs), associations between the poultry operation activity metric and CAP were confounded by the presence of a Geisinger hospital in the county of residence, which was strongly associated with CAP diagnosis (townships: 3.86 [2.98, 4.98]; boroughs: 4.48 [3.15, 6.39]). After controlling for this variable, we found no evidence of confounding by community type, with the fourth quartile of the poultry operation activity metric positively associated with CAP in both townships and boroughs (Table 2). Cities could not be evaluated in a stratified model due to small cell counts in cross-tabulations of the poultry operation activity metric and presence of a Geisinger hospital in the county of residence.
We observed little evidence of effect modification by Medical Assistance, age, or diagnosis season on associations of the poultry operation activity metric with CAP, so interaction terms for these variables were not included in final models.
Using EHR data from a large patient population in a region with many IFAP facilities, we evaluated the association between residential proximity to high-density poultry operations and CAP diagnoses. We found 66% increased odds of CAP diagnosis for individuals living in the highest quartile of the poultry operation activity metric (versus the lowest). Hypothesizing this association might vary across community types (townships, boroughs, cities), which differ on a range of place-level characteristics (e.g., population density, traffic, level of commercial development) and individual-level characteristics (e.g., socioeconomic status, race/ethnicity) in this region, we stratified our analysis but found a consistent relation between the highest quartile of the poultry operation activity metric and CAP diagnosis. The presence of significant elevated associations in only the fourth quartile of the poultry operation activity metric is consistent with a threshold relation. Our findings suggest that living in proximity to a greater number of and larger poultry operations may increase the risk of acquiring CAP, regardless of community type.
Prior research from the Netherlands related to poultry IFAP and CAP showed inconsistent findings. Three studies coincided with a goat-related C. burnetti infection outbreak.23–25 In a large EHR-based study, Smit et al25 demonstrated that individuals living within 1.15 km of a poultry farm had 11% increased risk per farm of CAP diagnosis, confirming their earlier study results.24 In contrast, a case–control study of 408 adult CAP patients with laboratory testing found the presence of sheep and number of goats within 1 km of patient home addresses, but not poultry, was associated with CAP caused by C. burnetii.23 The study used a more stringent definition of CAP, likely excluding cases with milder symptoms that Smit et al may have included in their analyses.23 After the C. burnetti outbreak concluded, a cross-sectional study among 2426 adults reported no association between living near poultry operations and pneumonia (defined as either self-reported physician-diagnosed pneumonia or EHR-recorded pneumonia).22 However, in a sensitivity analysis, the study reported an association between poultry farms within 1000 m of an individual’s residence and EHR-based pneumonia.
Our findings align most closely with those of Smit et al’s,25 perhaps due to similarities in study design. They used a data-driven kernel analysis that assessed accumulated risk from multiple farms in proximity to individuals’ home addresses, providing a similar advantage to our gravity measure, which took into account the number, animal density, and distance of all poultry operations in the study area. Additionally, neither study required laboratory or radiographic confirmation of pneumonia. This reduced specificity of case definitions, potentially capturing subjects without CAP or with less severe cases of CAP. Such cases are relevant for assessment if poultry IFAP and CAP associations are explained by increased susceptibility to lower respiratory infections resulting from microbiota dysbiosis due to air pollution, as evidence from Smit et al’s25 research suggests. They found that CAP patients living <1 km to one or more poultry farms had an altered oropharyngeal microbiota, with increased abundance of S. pneumoniae, as compared to those living ≥1 km to farms, which they hypothesized reflected an imbalanced oropharyngeal microbiota resulting from farm-related air pollutants. They noted that although poultry farms can be a source of Chlamydia psittaci (the cause of psittacosis, a disease resembling pneumonia), very few cases of CAP in their study area were caused by an avian pathogen. Similarly, our findings are not likely explained by avian-transmitted infections, as we found only one instance of psittacosis in our patient population.
Antibiotics also alter the oropharyngeal microbiota composition, creating a disequilibrium that increases vulnerability to colonization by pathogens and susceptibility to pneumonia.12 , 13 This could explain why CAP cases in our study were more likely than controls to have received an antibiotic order in the 30 days and 6 months prior to diagnosis. Yet compared to controls, CAP cases were even more likely to have received an order for antibiotics commonly used to treat pneumonia. Additionally, antibiotic orders were largely linked to diagnostic codes for pulmonary symptoms and conditions. This indicates antibiotic orders prior to CAP diagnosis may have reflected a clinical response to a prodromal phase during which subjects experienced symptoms of airway inflammation and other respiratory infections. These prodromal respiratory symptoms could stem from IFAP-related air pollutants, suggesting that residential proximity to poultry operations may be a risk factor for a variety of respiratory health outcomes, as shown for swine and dairy/veal IFAP in this region for asthma.34
Strengths of this study include a large sample size and the use of EHR data to classify CAP diagnoses. Additionally, the poultry operation activity metric incorporated information on subjects’ residential distance from poultry facilities, as well as the number of facilities and animal density, and comparison with US Agricultural Census data indicated that NMP data accounted for nearly all poultry production in the study area. Our study also had limitations. Although we adjusted for history of Medical Assistance, this measure does not account for all dimensions of socioeconomic status35; thus, our results are subject to residual confounding by unmeasured socioeconomic status. If poorer subjects were more likely to live near IFAP and have a CAP diagnosis, as indicated by higher levels of Medical Assistance among CAP cases, then risk associated with IFAP may have been overestimated. However, subjects living in the highest quartile of the poultry operation activity metric had the least Medical Assistance use, suggesting they were not the most socioeconomically disadvantaged. We also did not incorporate information on swine or bovine IFAP that may contribute to air pollution, which could be done in subsequent studies. Additionally, we lacked environmental samples from poultry operations and around residences and subjects’ microbiota data, preventing an evaluation of the hypothesis that air pollution from poultry operations increases vulnerability to CAP by altering the oropharyngeal microbiota. Our study provides initial evidence that residential proximity to poultry operations may increase risk for CAP, but additional research is needed to confirm these findings and elucidate the underlying environmental and biological pathways.
In conclusion, this study demonstrated that residing closer to more and larger poultry operations was associated with CAP, a cause of significant morbidity and mortality. As the first study to evaluate associations of poultry IFAP and CAP in the US, our findings highlight the need for additional research on the contribution of IFAP to air pollution and associated health risks for nearby communities. Understanding such localized risks could better inform clinical decision-making, including monitoring of patients susceptible to lower respiratory tract infections living in proximity to IFAP.
Conflict of interest statement
The authors declare that they have no conflicts of interest with regard to the content of this report.
We thank Mona Mohammed, Shai Gerstle, and Matthew Geiger at Bucknell University for their assistance with data collection, as well as Joseph DeWalle of Geisinger for his assistance developing variables and maps.
To protect patient privacy, individual-level medical record data are not available.
This research was funded by a Fisher Center Discovery Program grant from the Sherrilyn and Ken Fisher Center for Environmental Infectious Diseases in the Johns Hopkins School of Medicine.
1. National Center for Health StatisticsHealth, United States, 2016: With chartbook on long-term trends in health 2017. Hyattsville, MD
2. Pfuntner A, Wier LM, Stocks C. Most frequent conditions in U.S. hospitals, 2011. HCUP Statistcal Brief #162 2013. Rockville, MDAgency for Healthcare Research and Quality
3. Jain S, Self WH, Wunderink RG, Fakhran S, Balk R, Bramley AM, Reed C, Grijalva CG, Anderson EJ, Courtney DM, Chappell JD, Qi C, Hart EM, Carroll F, Trabue C, Donnelly HK, Williams DJ, Zhu Y, Arnold SR, Ampofo K, Waterer GW, Levine M, Lindstrom S, Winchell JM, Katz JM, Erdman D, Schneider E, Hicks LA, McCullers JA, Pavia AT, Edwards KM, Finelli L, Team CES. Community-acquired pneumonia requiring hospitalization among U.S. adults. N Engl J Med 2015; 373415–27
4. Jain S, Williams DJ, Arnold SR, Ampofo K, Bramley AM, Reed C, Stockmann C, Anderson EJ, Grijalva CG, Self WH, Zhu Y, Patel A, Hymas W, Chappell JD, Kaufman RA, Kan JH, Dansie D, Lenny N, Hillyard DR, Haynes LM, Levine M, Lindstrom S, Winchell JM, Katz JM, Erdman D, Schneider E, Hicks LA, Wunderink RG, Edwards KM, Pavia AT, McCullers JA, Finelli L; Team CESCommunity-acquired pneumonia requiring hospitalization among U.S. children. N Engl J Med 2015; 372835–45
5. Broulette J, Yu H, Pyenson B, Iwasaki K, Sato R. The incidence rate and economic burden of community-acquired pneumonia in a working-age population. Am Health Drug Benefits 2013; 6494–503
6. Almirall J, Bolibar I, Serra-Prat M, Roig J, Hospital I, Carandell E, Agusti M, Ayuso P, Estela A, Torres A; Community-Acquired Pneumonia in Catalan Countries (PACAP) Study GroupNew evidence of risk factors for community-acquired pneumonia: a population-based study. Eur Respir J 2008; 311274–84
7. Torres A, Peetermans WE, Viegi G, Blasi F. Risk factors for community-acquired pneumonia in adults in Europe: a literature review. Thorax 2013; 681057–65
8. Mehta S, Shin H, Burnett R, North T, Cohen AJ. Ambient particulate air pollution and acute lower respiratory infections: a systematic review and implications for estimating the global burden of disease. Air Qual Atmos Health 2013; 669–83
9. MacIntyre EA, Gehring U, Molter A, Fuertes E, Klumper C, Kramer U, Quass U, Hoffmann B, Gascon M, Brunekreef B, Koppelman GH, Beelen R, Hoek G, Birk M, de Jongste JC, Smit HA, Cyrys J, Gruzieva O, Korek M, Bergstrom A, Agius RM, de Vocht F, Simpson A, Porta D, Forastiere F, Badaloni C, Cesaroni G, Esplugues A, Fernandez-Somoano A, Lerxundi A, Sunyer J, Cirach M, Nieuwenhuijsen MJ, Pershagen G, Heinrich J. Air pollution and respiratory infections during early childhood: an analysis of 10 European birth cohorts within the ESCAPE Project. Environ Health Perspect 2014; 122107–13
10. Neupane B, Jerrett M, Burnett RT, Marrie T, Arain A, Loeb M. Long-term exposure to ambient air pollution and risk of hospitalization with community-acquired pneumonia in older adults. Am J Respir Crit Care Med 2010; 18147–53
11. Olivieri D, Scoditti E. Impact of environmental factors on lung defences. Eur Respir Rev 2005; 14(95)51–56
12. de Steenhuijsen Piters WA, Sanders EA, Bogaert D. The role of the local microbial ecosystem in respiratory health and disease. Philos Trans R Soc Lond B Biol Sci 2015; 370(1675)pii: 20140294
13. Dickson RP, Erb-Downward JR, Huffnagle GB. Towards an ecology of the lung: new conceptual models of pulmonary microbiology and pneumonia pathogenesis. Lancet Respir Med 2014; 2238–46
14. de Steenhuijsen Piters WA, Huijskens EG, Wyllie AL, Biesbroek G, van den Bergh MR, Veenhoven RH, Wang X, Trzcinski K, Bonten MJ, Rossen JW, Sanders EA, Bogaert D. Dysbiosis of upper respiratory tract microbiota in elderly pneumonia patients. ISME J 2016; 1097–108
15. Casey JA, Kim BF, Larsen J, Price LB, Nachman KE. Industrial food animal production and community health. Curr Environ Health Rep 2015; 2259–71
16. Schulze A, van Strien R, Ehrenstein V, Schierl R, Kuchenhoff H, Radon K. Ambient endotoxin level in an area with intensive livestock production. Ann Agric Environ Med 2006; 1387–91
17. Seedorf J, Hartung J, Schroder M, Linkert KH, Phillips VR, Holden MR, Sneath RW, Short JL, White RP, Pedersen S, Takai H, Johnsen JO, Metz JHM, Koerkamp PWGG, Uenk GH, Wathes CM. Concentrations and emissions of airborne endotoxins and microorganisms in livestock buildings in Northern Europe. J Agric Eng Res 1998; 7097–109
18. Viegas S, Faisca VM, Dias H, Clerigo A, Carolino E, Viegas C. Occupational exposure to poultry dust and effects on the respiratory system in workers. J Toxicol Environ Health A 2013; 76(4–5)230–9
19. Radon K, Schulze A, Ehrenstein V, van Strien RT, Praml G, Nowak D. Environmental exposure to confined animal feeding operations and respiratory health of neighboring residents. Epidemiology 2007; 18300–8
20. Borlee F, Yzermans CJ, Aalders B, Rooijackers J, Krop E, Maassen CBM, Schellevis F, Brunekreef B, Heederik D, Smit LAM. Air pollution from livestock farms is associated with airway obstruction in neighboring residents. Am J Respir Crit Care Med 2017; 1961152–61
21. Borlee F, Yzermans CJ, van Dijk CE, Heederik D, Smit LA. Increased respiratory symptoms in COPD patients living in the vicinity of livestock farms. Eur Respir J 2015; 461605–14
22. Freidl GS, Spruijt IT, Borlee F, Smit LA, van Gageldonk-Lafeber AB, Heederik DJ, Yzermans J, van Dijk CE, Maassen CB, van der Hoek W. Livestock-associated risk factors for pneumonia in an area of intensive animal farming in the Netherlands. PLoS One 2017; 12e0174796
23. Huijskens EG, Smit LA, Rossen JW, Heederik D, Koopmans M. Evaluation of patients with community-acquired pneumonia caused by zoonotic pathogens in an area with a high density of animal farms. Zoonoses Public Health 2016; 63160–6
24. Smit LA, van der Sman-de Beer F, Opstal-van Winden AW, Hooiveld M, Beekhuizen J, Wouters IM, Yzermans J, Heederik D. Q fever and pneumonia in an area with a high livestock density: a large population-based study. PLoS One 2012; 7e38843
25. Smit LAM, Boender GJ, de Steenhuijsen Piters WAA, Hagenaars TJ, Huijskens EGW, Rossen JWA, Koopmans M, Nodelijk G, Sanders EAM, Yzermans J, Bogaert D, Heederik D. Increased risk of pneumonia in residents living near poultry farms: does the upper respiratory tract microbiota play a role? Pneumonia (Nathan) 2017; 93
26. Casey JA, Savitz DA, Rasmussen SG, Ogburn EL, Pollak J, Mercer DG, Schwartz BS. Unconventional natural gas development and birth outcomes in Pennyslvania, USA. Epidemiology 2016; 27163–72
27. Schwartz BS, Stewart WF, Godby S, Pollak J, Dewalle J, Larson S, Mercer DG, Glass TA. Body mass index and the built and social environments in children and adolescents using electronic health records. Am J Prev Med 2011; 41e17–28
28. Dean NC, Jones BE, Jones JP, Ferraro JP, Post HB, Aronsky D, Vines CG, Allen TL, Haug PJ. Impact of an electronic clinical decision support tool for emergency department patients with pneumonia. Ann Emerg Med 2015; 66511–20
29. American Thoracic Society, Infectious Diseases Society of AmericaGuidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med 2005; 171388–416
30. Poulsen MN, Pollak J, Sills DL, Casey JA, Rasmussen SG, Nachman KE, Cosgrove SE, Stewart D, Schwartz BS. Residential proximity to high-density poultry operations associated with campylobacteriosis and infectious diarrhea. Int J Hyg Environ Health 2017; 221323–33
31. Roumeliotis TS, Van Heyst BJ. Summary of ammonia and particulate matter emission factors for poultry operations. J Appl Poult Res 2008; 17305–314
32. Talen E, Anselin L. Assessing spatial equity: An evaluation of measures of accessbility to public playgrounds. Environ Plan A 1998; 30595–613
33. Casey JA, Curriero FC, Cosgrove SE, Nachman KE, Schwartz BS. High-density livestock operations, crop field application of manure, and risk of community-associated methicillin-resistant Staphylococcus aureus infection in Pennsylvania. JAMA Intern Med 2013; 1731980–90
34. Rasmussen SG, Casey JA, Bandeen-Roche K, Schwartz BS. Proximity to industrial food animal production and asthma exacerbations in Pennsylvania, 2005–2012. Int J Environ Res Public Health 2017; 14(4)pii: E362
35. Casey JA, Pollak J, Glymour MM, Mayeda ER, Hirsch AG, Schwartz BS. Measures of SES for electronic health record-based research. Am J Prev Med 2018; 54430–39
36. Nau C, Schwartz BS, Bandeen-Roche K, Liu A, Pollak J, Hirsch A, Bailey-Davis L, Glass TA. Community socioeconomic deprivation and obesity trajectories in children using electronic health records. Obesity (Silver Spring) 2015; 23207–12
37. Millett ER, Quint JK, Smeeth L, Daniel RM, Thomas SL. Incidence of community-acquired lower respiratory tract infections and pneumonia among older adults in the United Kingdom: a population-based study. PLoS One 2013; 8e75131
38. Casey JA, James P, Rudolph KE, Wu CD, Schwartz BS. Greenness and Birth Outcomes in a Range of Pennsylvania Communities. Int J Environ Res Public Health 2016; 13pii: E311
39. Bateson TF, Schwartz J. Children’s response to air pollutants. J Toxicol Environ Health A 2008; 71238–43
40. USDAQuick Stats 2017. National Agriculture Statistics Service;