Open Access Open Badges Research

Spatial epidemiology of dry eye disease: findings from South Korea

Sun-Bi Um1, Na Hyun Kim1, Hyung Keun Lee2, Jong Suk Song3 and Hyeon Chang Kim4*

Author Affiliations

1 Department of Public Health, Yonsei University Graduate School, Seoul, South Korea

2 Institute of Vision Research, Department of Ophthalmology, Yonsei University College of Medicine, Seoul, South Korea

3 Department of Ophthalmology, Korea University College of Medicine, Seoul, South Korea

4 Department of Preventive Medicine, Yonsei University College of Medicine, 50-1Yonsei-ro, Seodaemun-gu, Seoul 120-752, Republic of Korea

For all author emails, please log on.

International Journal of Health Geographics 2014, 13:31  doi:10.1186/1476-072X-13-31

Published: 15 August 2014



DED rate maps from diverse regions may allow us to understand world-wide spreading pattern of the disease. Only few studies compared the prevalence of DED between geographical regions in non-spatial context. Therefore, we examined the spatial epidemiological pattern of DED prevalence in South Korea using a nationally representative sample.


We analyzed 16,431 Korean adults aged 30 years or older of the 5th Korea National Health and Nutrition Examination Survey. DED was defined as previously diagnosed by an ophthalmologist as well as symptoms experienced. Multiple logistic regression analysis was used to assess the spatial pattern in the prevalence of DED, and effects of environmental factors.


Among seven metropolitan cities and nine provinces, three metropolitan cities located in the southeast of Korea revealed the highest prevalence of DED. After adjusting for sex, age and survey year, people living in urban areas had higher risk of having DED. Adjusted odds ratio for having previously diagnosed DED was 1.677 (95% CI 1.299-2.166) for metropolitan cities and 1.580 (95% CI 1.215-2.055) for other cities compared to rural areas. Corresponding odds ratio for presenting DED symptoms was 1.388 (95% CI 1.090-1.766) for metropolitan cities and 1.271 (95% CI 0.999-1.617) for other cities. Lower humidity and longer sunshine duration were significantly associated with DED. Among air pollutants, SO2 was associated with DED, while NO2, O3, CO, and PM10 were not.


Our findings suggest that prevalence of DED can be affected by the degree of urbanization and environmental factors such as humidity and sunshine duration.

Air pollutants; Dry eye disease; Meteorological factors; Prevalence; Spatial epidemiology