Journal of Exposure Science and Environmental Epidemiology (2012) 22, 533 -- 538 & 2012 Nature America, Inc.

All rights reserved 1559-0631/12

www.nature.com/jes

ORIGINAL ARTICLE

Population study on chronic and acute conjunctivitis associated with ambient environment in urban and rural areas
Chun-Chi Chiang1,2, Chiang-Chang Liao3,4, Pei-Chun Chen4,5, Yi-Yu Tsai1,2 and Yu-Chun Wang6 The objective of this study is to evaluate whether daily clinic visits for conjunctivitis are associated with the ambient environment in urban and rural areas of Taiwan. The incidences of acute and chronic conjunctivitis (International Classication of Disease 9 Clinical Modication 372.0 and 372.1) in two urban cities and two rural counties and their relative risks (RRs) are associated with air pollutants (nitrogen oxides (NOx), sulfur dioxide, ozone, and particulate matter o10 mm in aerodynamic diameter) and/or weather statuses were assessed from the insurance reimbursement claims of a representative 1 million people from 2000 to 2007. The patients resided in rural counties were approximately eight time more likely to have acute complains and 41.3 time more likely to have chronic complaints than the patients lived in the capital, Taipei. Per 10 1C increment of the daily average temperature increased the risk of acute conjunctivitis and chronic conjunctivitis with RRs of 1.06 (95% condence interval (CI): 1.03--1.09) and 1.05 (95% CI: 1.04--1.07), respectively. A 10-p.p.b. increase in NOx concentration also increased the risk of acute conjunctivitis (RR 1.03, 95% CI: 1.02--1.04) and chronic conjunctivitis (RR 1.06, 95% CI: 1.05--1.06). Residents in rural counties, females, the elderly, and children have higher risk of conjunctivitis. Ambient temperature and NOx concentration can cause greater signicant risks on the diseases. Journal of Exposure Science and Environmental Epidemiology (2012) 22, 533 -- 538; doi:10.1038/jes.2012.37; published online 20 June 2012 Keywords: conjunctivitis; air pollution; NOx; temperature; urban and rural

INTRODUCTION Conjunctivitis is a common disease presented at eye clinics, mostly caused by virus infection, allergic reaction, or atopic effect.1 Environmental factors are also associated with the incidence of conjunctivitis.2 Progressive industrialization and increase in number of motor vehicles have resulted in high levels of air pollution. Johnson2 has declared that, due to constant contact with the ocular surface of the eyes, toxins may access directly these ocular structures and cause conjunctivitis-like syndromes. Studies have linked respiratory allergic diseases to exposure to air pollution, particularly, in urban areas of developed countries.3--6 Air pollutants may exacerbate existing respiratory disorders and allergic diseases.5 Riediker et al.7 have also found that air pollution may worsen eye irritation among pollen-allergic individuals suffering from allergic inammation.7 Both allergens and pollutants can directly initiate mucosal inammation because of oxidative stress, pro-inammatory cytokine production, and cyclooxygenase, lipoxygenase, and/or protease activation.8 Moreover, studies have found that conjunctiva may exhibit symptoms similar to those of allergic conjunctivitis in the absence of allergic response.9,10 The common signs and symptoms include redness, itching, and burning sensation due to allergy, dry eyes, or toxicity; sometimes, these signs and symptoms may not appear. Leonardi and Lanier6 have reported that individuals exposed to high temperatures, low wind speed, and high pollution levels may

experience itchy and red eyes without any underlying apparent allergic mechanism and medical history of allergy.6 They referred to this as urban eye allergy to highlight the conditions wherein airborne allergens are not typically prevalent. Although the above-mentioned studies have been useful in demonstrating the association between pollution and eye symptoms, they have been conducted based on specic study subjects. Some of them were limited to the small sample size or localized area data.11 To evaluate the relationship between conjunctivitis and air pollutants, we analyzed data from Taiwans universal insurance claims. Our study aims are to investigate the chronological trends of acute and chronic conjunctivitis; to determine whether air pollution increases the risk of the diseases; and to differentiate between cases reported in urban areas and in rural counties. To our knowledge, this is the rst study to examine specically the longitudinal trend of the disease. We also examined the risk differences of acute and chronic conjunctivitis between urban and rural areas. METHODS AND MATERIALS Data source
This study used information on health insurance claims obtained from the National Health Research Institute (NHRI); daily meteorological records obtained from the Central Weather Bureau (CWB); and daily air pollution

1 Department of Ophthalmology, China Medical University Hospital, Taichung, Taiwan; 2School of Medicine, China Medical University, Taichung, Taiwan; 3Department of Anesthesiology, Taipei Medical University Hospital, Taipei, Taiwan; 4China Medical Hospital Management Ofce for Health Data, Taichung, Taiwan; 5Institute of Clinical Medical Science, China Medical University, Taichung, Taiwan; 6Department of Bioenvironmental Engineering, Chung Yuan Christian University College of Engineering, Chung Li, Taiwan. Correspondence: Professor Yu-Chun Wang, Department of Bioenvironmental Engineering, Chung Yuan Christian University College of Engineering, 200 Chung-Pei Road, Chung Li 320, Taiwan. Tel.: 886 3 265 4916. Fax: 886 3 265 4949. E-mail: ycwang@cycu.edu.tw Received 2 December 2011; revised 26 February 2012; accepted 2 April 2012; published online 20 June 2012

Conjunctivitis and air pollution Chiang et al

534
monitoring records obtained from the Taiwan Environmental Protection Administration (TEPA).12--14 Over 96% of 23 million population in Taiwan has been covered in the National Health Insurance (NHI) program since 1996.15 The Taiwan NHRI has established an electronic reimbursement claims data set with a representative sample of 1 million people, randomly selected from all insured citizens. Information on two urban cities (Taipei and Kaohsiung) and two rural counties (Yunlin and Yilan) for years 2000--2007 were retrieved. Scrambled patient identication numbers are available for linking the NHI electronic database and secure patient privacy. This study is exempted from ethical review. Information consisted of gender, birthdates, and health care received by the insured, date on outpatient visit, inpatient admission and discharge, and medical institutions providing health care services. The diagnoses of diseases were coded according to the ninth revision of the International Classication of Diseases Clinical Modication (ICD-9-CM). For this study, we retrieved data on the newly diagnosed acute conjunctivitis (ICD9 CM 372.0) and chronic conjunctivitis (ICD9 CM 372.1). Medical records for readmission with the same diagnosis within 7 days were exclusive in this study. CWB provided daily the 24-h weather data (average temperature, relative humidity, rainfall, and wind speed) from 25 real-time weather monitoring stations in Taiwan.12 This study used daily weather measurements from Taipei, Kaohsiung, Chiayi (most close to Yunlin), and Yilan weather stations from 2000 to 200716 (Figure 1). The Taiwan Air Quality Monitoring Network established by TEPA in 1993 included 74 stationary monitoring stations distributed throughout the island.13,17 Concentrations of particulate matters o10 mm in aerodynamic diameter (PM10), nitrogen oxides (NOx), sulfur dioxide (SO2), and ozone (O3) were determined and recorded hourly at each station. This study analyzed the daily average data for PM10, NOx, SO2, and O3 monitored at 28 general ambient stations, 13 in Taipei, 10 in Kaohsiung, 3 in Yunlin, and 2 in Yilan (Figure 1). All the subjects were assumed to be exposed to the same levels of air pollutants measured by monitoring stations.

Statistical analysis
Our data analyses rst compared demographic characteristics, weather and air pollution status, and average annual clinic visits per 1,000 persons for acute and chronic conjunctivitis for the four study areas. The annual trends plotted from the daily visits for these complains referred to years 2000--2007. We used the generalized linear model with Poisson regression18 to measure associations between daily visits for acute and chronic conjunctivitis and covariates and ambient weather and pollution. The pooled and area-age-specic analyses were expressed as:

Figure 1. Locations for real-time monitoring stations of ambient air pollution in the studied areas.

logYi a b1 X 1 b2 X 2 . . . bp X p

where Yi denotes the daily clinic visits for acute or chronic conjunctivitis; X1--X4 are the daily concentrations of PM10, SO2, NOx, and O3; and X5--X8 are the daily measurements of average temperature, relative humidity, and cumulative rainfall and wind speed. Relative risks (RRs) associated with area, sex, and age (o15, 15--39, 40--64, and 464 years) were also assessed controlling for holidays (including weekends), and calendar months and years in the models. This study evaluated lag effects of air pollutants on lag 0, 1, and 2 days. We further considered the moving averages (lag 0--2 days) of air pollutants as optional covariates. We selected optimized model to evaluate the association between acute/chronic conjunctivitis and ambient environment based on model selection criteria --- deviance divided by the degree of freedom, where if the value is more close to 1 indicates a better model. The RRs of clinic visits for acute and chronic conjunctivitis, and their 95% condence intervals (CIs), were estimated based on the exponential transformation of the bi estimate. All statistical analyses were performed by SAS version 9.1 (GENMOD procedure for Poisson regression analyses) (SAS Institute, Cary, NC, USA). The level of statistical signicance (alpha level) was 0.05.

RESULTS Interpretation of medical data For this study, 75,488 patients (139,124 claims) with acute conjunctivitis and 158,878 patients (554,722 claims) with chronic conjunctivitis were identied from the claims data in 2000--2007 (data not shown). Among them, there were 40,171 patients diagnosed with both diseases. Most patients with conjunctivitis received medical service at clinics in their communities (92.8% of acute cases and 85.5% chronic cases). Over 93% of the second eye clinic visits were made 7 days apart, and only 2.03% of visits occurred within 3 days. Readmission rate was 4.47% for acute conjunctivitis, and 1.24% for chronic conjunctivitis within 7 days. Disease trends Figure 2 shows the overall chronological trends of acute and chronic conjunctivitis in the four study areas. The average daily visits for acute conjunctivitis ranged from 0.002 to 2.60 per 1,000 (mean 0.28 per 1,000) with a peak in the summer, and from 0.002 to 6.50 per 1,000 (mean 0.99 per 1,000) for chronic conjunctivitis with two peaks annually. An outbreak of acute conjunctivitis appeared from June to August 2002, resulting in a daily average of 0.32 clinic visits per 1,000. Table 1 shows the population characteristics, weather status, mean air pollution levels, and mean clinic visits for conjunctivitis among the study areas. Compared with the rural counties of Yunlin and Yilan, the urban areas of Taipei and Kaohsiung had residents with higher income and had less elderly population. The concentrations of PM10, SO2, and O3 were the highest in
& 2012 Nature America, Inc.

Journal of Exposure Science and Environmental Epidemiology (2012), 533 -- 538

Conjunctivitis and air pollution Chiang et al

535
Average (SD) daily measures of temperature, air pollutants levels, and selected characteristics of urban Taipei and Kaohsiung and rural Yunlin and Yilan in 2000--2007.34 Taipei Population density (persons/km ) Low income ratio (%)a Average income (USD/person/year) Elderly (%) in 2005b Average temperature (1C) Relative humidity (%) PM10 (mg/m3) SO2 (p.p.b.) NOx (p.p.b.) O3 (p.p.b.) Mean annual clinic visits (per 1,000) Acute conjunctivitis Chronic conjunctivitis
2

Table 1.

Kaohsiung 5,173 0.95 8,737 9.0 25.3 (3.84) 75.4 (7.22) 76.7 (36.6) 7.75 (3.01) 28.7 (12.4) 29.2 (12.2) 161 463

Yunlin 562 1.49 6,977 13.7 23.5 (4.75) 78.3 (6.06) 68.4 (34.2) 3.31 (1.52) 18.1 (7.38) 29.1 (10.0) 226 349

Yilan 215 1.43 7,202 12.0 22.9 (4.83) 79.4 (7.39) 39.7 (19.4) 2.04 (1.30) 16.1 (4.95) 24.8 (9.35) 227 560

5,672 1.02 10,340 9.2 23.4 (5.27) 75.9 (9.09) 48.3 (22.9) 3.86 (1.91) 31.9 (13.7) 26.2 (9.12) 49.2 273

Note: Taipei and Kaohsiung are urban cities and Yunlin and Yilan are rural counties. a The governments standard for low income: annual income less than approximate 3,850 USD per person. Low income ratio (%) (area-specic low income residents/area-specic total residents) 100%. b Elderly: resident aged 464 years.

480 400 Daily clinic visits 320

Acute conjunctivitis Chronic conjunctivitis Smoothing line of acute conjunctivitis Smoothing line of chronic conjunctivitis

Table 2. Multivariable Poisson regression analysis measured RRs for clinic visits of acute and chronic conjunctivitis in association with area, age, and combined ambient status in studied areas, 2000--2007.
Acute RR 95% CI -3.01--3.15 5.51--5.81 8.06--8.51 2.13--2.19 --1.11--1.14 2.58--2.67 1.15--1.18 1.03--1.09 1.00--1.01 0.80--0.86 1.02--1.04 0.99--1.01 RR 1.0 1.54* 1.32* 2.44* 2.79* 1.0 1.41* 4.04* 1.51* 1.05* 0.99 0.91* 1.06* 1.02* Chronic 95% CI -1.53--1.56 1.30--1.34 2.40--2.48 2.77--2.81 --1.40--1.42 4.01--4.08 1.49--1.51 1.04--1.07 0.99--0.99 0.89--0.93 1.05--1.06 1.02--1.02

240 160 80 0 2000

Area Taipei Kaohsiung Yunlin Yilan Age, years o15 15--39 40--64 464 Sex (F to M) Average temperature (10 1C) PM10 (10 mg/m3) SO2 (10 p.p.b.) NOx (10 p.p.b.) O3 (10 p.p.b.)

1.0 3.08* 5.66* 8.28* 2.16* 1.0 1.13* 2.63* 1.16* 1.06* 1.00 0.83* 1.03* 1.00

2001

2002

2003

2004 Year

2005

2006

2007

2008

Figure 2. Daily clinic visits for acute and chronic conjunctivitis in studied areas (i.e., Taipei, Kaohsiung, Yunlin, and Yilan) from 2000 to 2007.

Kaohsiung, and the concentration of NOx was the highest in Taipei. The average annual clinic visit rate for acute conjunctivitis was higher in rural areas. Yilan had the highest rate of average clinic visits for chronic conjunctivitis, followed by Kaohsiung. Taipei had the lowest clinic visit rates. Associated risk factors Compared with Taipei, the risk of acute infection was almost eight times greater in rural counties (Table 2); Kaohsiung also had an RR of 3.08. The risk of chronic conjunctivitis was the highest in Yilan, followed by Kaohsiung, and Yunlin. Children and the older population were at higher risks for both types of conjunctivitis. Females were also more prone to elevated risks. On lag 0 day, the average temperature and concentrations of NOx were most signicantly associated with the clinic visits for these two diseases. A 10-p.p.b. increase of O3 also independently increased the risk of chronic conjunctivitis. There was no signicant association between the complains and levels of PM on lag 0 day. Table 3 shows the Poisson model estimated RRs of conjunctivitis associated with age, sex, and ambient environmental status
& 2012 Nature America, Inc.

Abbreviations: CI, condence interval; F, female; M, male; NOx, nitrogen oxides; O3, ozone; PM10, particulate matter o10 mm in aerodynamic diameter; RRs, relative risks; SO2, sulfur dioxide. Note: This analysis also controlled for relative humidity, wind speed, cumulative rainfall, and public holiday (including weekend), calendar months and years. *Po0.0001.

on lag 0 day among four areas. Young and older residents were at higher risk compared with those aged 15--39 years. The risk associated with temperature was the highest in Kaohsiung and the lowest in Taipei. The risk of conjunctivitis was signicantly associated with the increment of NOx and O3 in urban areas. Residents in Yilan were also positively associated with NOx levels. Figure 3 shows age-specic RRs of acute and chronic conjunctivitis associated with 10 p.p.b. NOx increment on lag 0 day by area. The association was signicant for chronic

Journal of Exposure Science and Environmental Epidemiology (2012), 533 -- 538

Conjunctivitis and air pollution Chiang et al

536
Table 3. Multivariable Poisson regression analysis measured RRs of clinic visits for acute and chronic conjunctivitis combined in association with age, sex, and ambient environment of lag 0 day by area from 2000 to 2007.
Taipei RR Age, years o15 15--39 40--64 464 Sex (F to M) Average temperature (10 1C) PM10 (10 mg/m3 increase) SO2 (10 p.p.b. increase) NOx (10 p.p.b. increase) O3 (10 p.p.b. increase) 3.18* 1.00 1.34* 3.26* 1.52* 1.03+ 0.99 0.97 1.06* 1.01+ 95% CI 3.16--3.21 -1.33--1.35 3.23--3.30 1.51--1.53 1.01--1.04 0.99--1.00 0.94--1.00 1.06--1.07 1.00--1.02 RR 1.82* 1.00 1.31* 4.02* 1.45* 1.24* 1.00 0.95 1.06* 1.01* Kaohsiung 95% CI 1.80--1.85 -1.30--1.33 3.98--4.07 1.41--1.45 1.20--1.26 1.00--1.00 0.93--0.98 1.05--1.07 1.01--1.02 RR 1.94* 1.00 1.36* 3.28* 1.23* 1.13* 1.00 0.95 1.01 1.00 Yunlin 95% CI 1.88--2.01 -1.32--1.40 3.18--3.38 1.22--1.27 1.07--1.19 1.00--1.01 0.84--1.07 0.98--1.04 0.99--1.02 RR 2.29* 1.00 1.38* 3.49* 1.23* 1.10+ 0.99 1.03 1.04+ 1.01 Yilan 95% CI 2.22--2.36 -1.34--1.42 3.39--3.61 1.20--1.25 1.05--1.16 0.99--1.00 0.92--1.16 1.00--1.07 0.98--1.02

Abbreviations: CI, condence interval; F, female; M, male; NOx, nitrogen oxides; O3, ozone; PM10, particulate matter o10 mm in aerodynamic diameter; RRs, relative risks; SO2, sulfur dioxide. Note: This analysis also controlled for relative humidity, wind speed, cumulative rainfall, and public holiday (including weekend), calendar months and years. *Po0.0001. **Po0.01.

conjunctivitis in all age groups in urban areas only (Taipei and Kaohsiung). In addition to NOx, this study also found that chronic conjunctivitis was approximate signicantly associated with O3 on lag 1 and 2 days, RR was approximate 1.01/10 p.p.b. increment in pooled four area analysis (data not shown). However, the RR of O3 would be 1.04 (95% CI: 1.02--1.05) per 10 p.p.b. increment in Kaohsiung on lag 1 day. On the other hand, the effects of PM10 were greater on acute conjunctivitis; RR was approximate 1.01/ 10 mg/m3 increment in the pooled four area analysis, and with RR of 1.03 (95% CI: 1.02--1.04) per 10 p.p.b. increment in Kaohsiung on lag 1 day.
Relative risk

1.4 1.3 1.2 1.1 1.0 0.9 0.8

Acute conjunctivitis

Taipei Kaohsiung Yunlin Yilan

DISCUSSION The relationship between air pollution and respiratory symptoms has been well documented.19--21 As the eyes are always exposed to the living environment, it is logical to relate the symptoms of conjunctivitis to the effect of air pollutants.22 To the best of our knowledge, this is the rst study to use a nationwide insured cohort to evaluate the chronological trends of the ocular symptoms and its relationship with ambient environmental factors. Compared with acute conjunctivitis, chronic conjunctivitis has two peaks annually as shown in Figure 2. The number of patients with these two types of complaint is higher in warm months. Two peaks for chronic complaints reect the seasonality in etiology and the variation of environment. The associated environmental factors are likely different between urban cities and rural counties. Results of Poisson regression analysis showed a much greater risk of conjunctivitis in the rural counties than in the cities. However, the air pollution-associated conjunctivitis risk was higher for the two cities. Therefore, other etiological factors should be considered for the rural counties, such as personal hygiene. It is possible that conjunctivitis among residents in Yunlin and Yilan were dominant biologically and etiologically. As the rural population are more prevalent with the elderly and farmers, the incidence of conjunctivitis might be resulted from chronic blepharitis and contamination during farming or gardening such as pollens associated with agricultural crops and other plants. These results imply the accurateness of the hygiene hypothesis, which proposed the reverse association between biological infection and individual allergic sensitization. It is to say that higher risks of conjunctivitis in rural areas, but higher sensitization

<15 1.3 1.2 1.1 1.0 0.9 0.8 <15

15-39

40-64

>64

Chronic conjunctivitis

15-39 Age

40-64

>64

Figure 3. Relative risks of acute and chronic conjunctivitis associated with 10 p.p.b. NOx increment by area and age, 2000--2007. Taipei and Kaohsiung are urban cities and Yunlin and Yilan are rural counties.

to air pollutants in urban cities. Hygiene hypothesis may also explain children and females are at higher risk of conjunctivitis.23 Children are more likely to expose to dirt and women are more likely to manage chores at home and farm. Among the air pollutants, NOx has the strongest association with conjunctivitis symptoms, both for acute and chronic,
& 2012 Nature America, Inc.

Journal of Exposure Science and Environmental Epidemiology (2012), 533 -- 538

Conjunctivitis and air pollution Chiang et al

followed by O3. To our surprise, SO2 appears to be protective in our pooled analysis for all areas. However, data analysis by area shows that this association disappears from not only the rural county but also from the urban city. This weak protective association might be a secondary association related with an undetected factor. To the best of our knowledge, rare study has reported the association between ambient SO2 exposure and conjunctivitis.10 This protective association needs a further study. Automobile exhaust is the main source of outdoor NOx.4 A o Paulo, Brazil, found a clear dose--response study conducted in Sa relationship between NO2 levels and goblet cell hyperplasia frequency, suggesting that this morphological change is an adaptive response of the conjunctival epithelium to chronic environmental injury, Novaes et al.24 and Brunekreef et al.25 analyzed the data of the International Study of Asthma and Allergies in Childhood to evaluate the self-reported truck trafc frequency on the street of residence and symptoms of asthma and allergic diseases. They found a signicant association between truck trafc frequency and rhino-conjunctivitis in most countries worldwide.25 This association remains after controlling for the exposure measurement error. An 1 mg/m3 increase in NO2 is associated with the prevalence ratio of 6.60 (95% CI: 1.33--32.77) for the reported conjunctivitis.26 In this study, Taipei has the highest NOx levels yet the lowest RRs. Urban residents in Taipei are likely to have better personal hygiene. The association between conjunctivitis and air pollution may be weaker than that and hygiene. O3 is an important factor in the summer smog generated at the ground level by photochemical reactions that involve ultraviolet radiations in the atmospheric mixture of NOx and hydrocarbons derived from vehicle emissions. Atmospheric levels of O3 and NOx have been linked to asthma and airway inammation diseases.27--29 Conjunctival tissue is similar to the airway mucosal tissue, which is sensitive to these pollutants. Conjunctivitis symptoms linked to exposure to these pollutants include itching, burning, and reddish eyes from exposure. Air pollutants can interact with aeroallergens in the atmosphere and induce inammation, resulting in increases in epithelial permeability. Some pollutants bypass the mucosal barrier, leading to allergen-induced responses.9,30,31 Air pollutants may exacerbate allergic symptoms of the respiratory tract, nose, and eyes during the pollen season. Riediker et al.7 followed up 15 pollen-allergic individuals throughout the pollen season using daily records of eye and nose symptoms and medications. Their observations demonstrated a strong association between allergic rhinoconjunctivitis symptoms and pollutant levels.7 Bourcier et al.10 analyzed the data on ophthalmological emergency examinations in Paris in 1999 and found that 22% of approximate 30,000 emergency admissions were related to cases on conjunctivitis.10 They observed a clear link between the severity of symptom and the levels of NOx, O3, and SO2 present in the environment. The level of air pollution is directly linked to short-term increases in the number of conjunctivitis cases. Similarly, Versura et al. (1999) followed a group of 100 patients with discomfort eye syndrome (DES) for a long period.32 They concluded that signs and symptoms of DES are signicantly associated with ocular surface sub-clinical cytological inammation because of atmospheric pollution. Changes in lacrimal pH, such as those produced by acidication of tears in an atmosphere with high oxidant power (NO, NO2, and SO2), may also cause irritation on the ocular surface. A reduced lipid layer of the tear lm has been implicated in pollution-related keratoconjunctivitis, resulting in reduced tear break-up time, increased vital staining, and increased cells in the conjunctival uid.22 The strength of this study is the use of a large database on NHI claims, which allowed us to observe the relationship between air pollution and conjunctivitis. Hence, the trends of conjunctivitis
& 2012 Nature America, Inc.

were objectively analyzed. Our results are also compatible with previous investigations on the association of conjunctivitis with air pollution. However, some limitations were presented in our study. First, the information on actual clinical examinations is unavailable in the claims data. Biological factors other than ambient air quality that can cause conjunctivitis also remain unclear. Second, the denitions of chronic and acute conjunctivitis according to the ICD-9-CM may not be precise enough to reect the real etiology of the diseases. In addition, this study used the second hand data to evaluate the exposure, disease association, which was under the assumption of the studied subjects are exposed to the same level of air pollutants monitored in their residence area. It is possible to underestimate the risk.33 In conclusion, this study identied the democratic factors, residence area, age, and sex, which are signicantly associated with daily clinic visits for acute and chronic conjunctivitis. Furthermore, both NOx concentration and temperature appear to have signicant effects on the ocular surface of the eyes, causing ocular irritation and conjunctivitis.

537

ABBREVIATIONS 95% CI, 95% condence interval; NOx, nitrogen oxides; PM10, particulate matter o10 mm in aerodynamic diameter; RR, relative risk CONFLICT OF INTEREST
The authors declare no conict of interest.

ACKNOWLEDGEMENTS
This study was supported in part by the Taiwan National Science Council (NSC 992221-E-033-052 and NSC 99-2621-M-039-001), the China Medical University Hospital (1MS1 and DMR-101-076), and the Taiwan Department of Health (DOH100-TD-B-111004, DOH100-TD-C-111-005). We thank the NHRI, Environmental Protection Administration and CWB for providing the research data. The interpretation and the conclusions contained herein do not represent the views of these agencies.

REFERENCES
1 American Academy of Ophthalmology. Cornea and External Eye Disease (2007). 2 Johnson G.J. The environment and the eye. Eye (2004): 18: 1235--1250. 3 DAmato G., Liccardi G., and DAmato M. Environmental risk factors outdoor air pollution and climatic changes) and increased trend of respiratory allergy. J Investig Allergol Clin Immunol (2000): 10: 123--128. 4 DAmato G., Liccardi G., DAmato M., and Holgate S. Environmental risk factors and allergic bronchial asthma. Clin Exp Allergy (2005): 35: 1113--1124. 5 Heinrich J., and Wichmann H.E. Trafc related pollutants in Europe and their effect on allergic disease. Curr Opin Allergy Clin Immunol (2004): 4: 341--348. 6 Leonardi A., and Lanier B. Urban eye allergy syndrome: a new clinical entity? Curr Med Res Opin (2008): 24: 2295--2302. 7 Riediker M., Monn C., Koller T., Stahel W.A., and Wuthrich B. Air pollutants enhance rhinoconjunctivitis symptoms in pollen-allergic individuals. Ann Allergy Asthma Immunol (2001): 87: 311--318. 8 Bernstein J.A., Alexis N., Barnes C., Bernstein I.L., Bernstein J.A., Nel A., Peden D., Diaz-Sanchez D., Tarlo S.M., and Williams P.B. Health effects of air pollution. J Allergy Clin Immunol (2004): 114: 1116--1123. 9 Bacsi A., Dharajiya N., Choudhury B.K., Sur S., and Boldogh I. Effect of pollenmediated oxidative stress on immediate hypersensitivity reactions and late-phase inammation in allergic conjunctivitis. J Allergy Clin Immunol (2005): 116: 836--843. 10 Bourcier T., Viboud C., Cohen J.C., Thomas F., Bury T., Cadiot L., Mestre O., Flahault A., Borderie V., and Laroche L. Effects of air pollution and climatic conditions on the frequency of ophthalmological emergency examinations. Br J Ophthalmol (2003): 87: 809--811. 11 Saxena R., Srivastava S., Trivedi D., Anand E., Joshi S., and Gupta S.K. Impact of environmental pollution on the eye. Acta Ophthalmol Scand 2003: 81: 491--494. 12 The Central Weather Bureau Online Information 2011a. Available at: http:// www.cwb.gov.tw/eng/index.htm (accessed date: 12 May 2011).

Journal of Exposure Science and Environmental Epidemiology (2012), 533 -- 538

Conjunctivitis and air pollution Chiang et al

538
13 Taiwan Air Quality Monitoring Network 2011. Available at: http://taqm.epa.gov.tw/ taqm/en/PsiAreaHourly.aspx (accessed date: 12 May 2011). 14 Taiwan National Health Insurance Research Database. Available at: http:// w3.nhri.org.tw/nhird/en/Data_Subsets.htmlS3 (accessed date: 12 May 2011) 2011. 15 Cheng S.H., and Chiang T.L. The effect of universal health insurance on health care utilization in Taiwan. Results from a natural experiment. JAMA 1997: 278: 89--93. 16 Illustrations of Surface Observation Data 2011b. Available at: http://www.cwb. gov.tw/v7e/observe/real/ALLData.htm (accessed date: 9 May 2012). 17 The republic of China Year Book 2008: Chapter 12 Environmental Protection 2008. Available at: http://www.gio.gov.tw/taiwan-website/5-gp/yearbook/2008/ch12.html (accessed date: 12 May 2011). 18 McCullagh P., and Nelder J.A. Generalized Linear Models, 2 edn. Chapman & Hall/ CRC Boca Raton Florida, 1989. 19 Akinbami L.J., Lynch C.D., Parker J.D., and Woodruff T.J. The association between childhood asthma prevalence and monitored air pollutants in metropolitan areas, United States, 2001--2004. Environ Res 2010: 110: 294--301. 20 Laumbach R.J. Outdoor air pollutants and patient health. Am Fam Physician 2010: 81: 175--180. 21 Mar T.F., and Koenig J.Q. Relationship between visits to emergency departments for asthma and ozone exposure in greater Seattle, Washington. Ann Allergy Asthma Immunol 2009: 103: 474--479. 22 Norn M. Pollution keratoconjunctivitis. A review. Acta Ophthalmol Copenh 1992: 70: 269--273. 23 Zhang G., Spickett J., Lee A.H., Rumchev K., and Stick S. Household hygiene practices in relation to dampness at home and current wheezing and rhinoconjunctivitis among school age children. Pediatr Allergy Immunol 2005: 16: 587--592. 24 Novaes P., do Nascimento Saldiva P.H., Kara-Jose N., Macchione M., Matsuda M., Racca L., and Berra A. Ambient levels of air pollution induce goblet-cell hyperplasia in human conjunctival epithelium. Environ Health Perspect 2007: 115: 1753--1756. 25 Brunekreef B., Stewart A.W., Anderson H.R., Lai C.K., Strachan D.P., and Pearce N. Self-reported truck trafc on the street of residence and symptoms of asthma and allergic disease: a global relationship in ISAAC phase 3. Environ Health Perspect 2009: 117: 1791--1798. 26 Van Roosbroeck S., Li R., Hoek G., Lebret E., Brunekreef B., and Spiegelman D. Trafc-related outdoor air pollution and respiratory symptoms in children: the impact of adjustment for exposure measurement error. Epidemiology 2008: 19: 409--416. 27 Atkinson R.W., Anderson H.R., Strachan D.P., Bland J.M., Bremner S.A., and Ponce de Leon A. Short-term associations between outdoor air pollution and visits to accident and emergency departments in London for respiratory complaints. Eur Respir J 1999: 13: 257--265. 28 Bayram H., Sapsford R.J., Abdelaziz M.M., and Khair O.A. Effect of ozone and nitrogen dioxide on the release of proinammatory mediators from bronchial epithelial cells of nonatopic nonasthmatic subjects and atopic asthmatic patients in vitro. J Allergy Clin Immunol 2001: 107: 287--294. 29 Peters J.M., Avol E., Navidi W., London S.J., Gauderman W.J., Lurmann F., Linn W.S., Margolis H., Rappaport E., Gong H., and Thomas D.C. A study of twelve Southern California communities with differing levels and types of air pollution. I. Prevalence of respiratory morbidity. Am J Respir Crit Care Med 1999: 159: 760--767. 30 Bagarozzi Jr D.A., and Travis J. Ragweed pollen proteolytic enzymes: possible roles in allergies and asthma. Phytochemistry 1998: 47: 593--598. 31 Hitzfeld B., Friedrichs K.H., Ring J., and Behrendt H. Airborne particulate matter modulates the production of reactive oxygen species in human polymorphonuclear granulocytes. Toxicology 1997: 120: 185--195. 32 Versura P., Profazio V., Cellini M., Torreggiani A., and Caramazza R. Eye discomfort and air pollution. Ophthalmologica 1999: 213: 103--109. 33 Armstrong B.G. Effect of measurement error on epidemiological studies of environmental and occupational exposures. Occup Environ Med 1998: 55: 651--656. 34 Directorate-General of Budget, Accounting and Statistics, Executive Yuan, R.O.C. (Taiwan). National Statistics Vol. 2009 2012. Available at: http://www.dgbas. gov.tw/mp.asp?mp (accessed date: 25 January 2012).

Journal of Exposure Science and Environmental Epidemiology (2012), 533 -- 538

& 2012 Nature America, Inc.

Copyright of Journal of Exposure Science & Environmental Epidemiology is the property of Nature Publishing Group and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.