Table 1.

The characteristics of studies included in the meta-analysis.

Study	Location	Study Design Time-span	Study Population	Pollutant	Controlled Variables	Total Events	Lag (d/w)	Main Findings
Bourcier et al. (2003) [13]	Paris, France	Logistic regression 31/1/1999–31/12/1999	All	NO, NO2, O3, SO2, PM10	Temperature, pressure, humidity, wind speed, day of the week	1272	d: 0–2	A strong relation between NO, NO2, and conjunctivitis was observed. Atmospheric pressure, minimal humidity, and wind speed may increase the incidence of ocular surface complaints.
Larrieu et al. (2009) [14]	Bordeaux, France	Time series Poisson regression model 2000–2006	All	NO2, PM10, O3	Long-term trends, seasonality, days of the week, holidays, temperature, influenza epidemics	179,142	d: 0–3	There was a much higher effect of nitrogen dioxide on visits for conjunctivitis when delayed effects were considered. Conjunctivitis was also significantly associated with PM10 and ozone levels.
Chang et al. (2012) [4]	Taiwan, China	Case-crossover Meta-analysis 2007–2009	All	CO, NO2, SO2, O3, PM10, PM2.5	Temperature, rainfall, humidity	26,314,960	d: 0, 0–1 to 0–5	The effects on outpatient visits for nonspecific conjunctivitis were strongest for O3 and NO2. In winter, PM10 and SO2 had a more prominent impact on the risk of conjunctivitis.
Chiang et al. (2012) [29]	Taiwan, China (four cities)	Time series Generalized linear model 2000–2007	All	PM10, SO2, NOx, O3	Relative humidity, wind speed, rainfall, public holiday, calendar months and years.	234,366	d: 0	There were higher risks of conjunctivitis in rural areas, but higher sensitization to air pollutants in urban cities. Children, females, and the older population were at higher risks for both types of conjunctivitis.
Szyszkowicz et al. (2012) [27]	Edmonton, Canada	Case-crossover Logistic regression Time-stratification 1/4/1992–31/3/2002	All, Sex: male, female	O3	Long-term trends, seasonal effects, day-of-week and month-of-year effects	7526	d: 3–8	For conjunctivitis, associations of these conditions with ozone exposure were observed only in females.
Hong et al. (2016) [6]	Shanghai, China	Time series Generalized least squares 2008–2012	All, Sex: male, female Age: <18, 19–40, 41–60, >60 years	SO2, NO2, PM10, PM2.5, O3	Periodic trends	3,211,820	w: 1, 3	Research revealed that higher levels of ambient NO2, O3, and temperature increased the chances of outpatient visits for allergic conjunctivitis. Meanwhile, those older than 40 years were only affected by NO2 levels.
Szyszkowicz et al. (2016) [28]	Ontario, Canada (nine cities)	Case-crossover Time-stratified Apr 2004–Dec 2011	All, Sex: male, female Age: ≤17, ≥18 years	NO2, O3, SO2, PM2.5	Temperature, humidity	77,439	d: 0–8	There were positive associations between air pollution and ED visits for conjunctivitis, with different temporal trends and strength of association by age, sex, and season. Children and young adults were more vulnerable to conjunctivitis infections.
Fu et al. (2017) [5]	Hangzhou, China	Time-stratified Case-crossover Logistic regression 1/7/2014–30/6/2016	All, Sex: male, female Age: 0–1, 2–5, 6–18, 19–64, >65 years	PM10, PM2.5, SO2, NO2, O3, CO	Temperature, humidity, atmospheric pressure	9737	d: 0, 0–1	PM10, PM2.5, SO2, NO2, and CO were associated with the risk of conjunctivitis. SO2 was significantly associated with conjunctivitis patients between 2 and 5 years old and male. PM10 and NO2 were significantly associated with female conjunctivitis patients.
Jamaludin et al. (2017) [30]	Johor Bahru, Malaysian	Time series Poisson generalized linear model, negative binomial model 1/1/2012–31/12/2013	All	NO2, PM10, SO2	Rainfall, temperature, humidity	1396	w: 14,19,20	SO2 was the most abundant source that contributed to the eye diseases.
Lee et al. (2018) [11]	Daegu, Korea	Spatial analysis 1/6/2006–31/12/2014	All	PM10	SO2, NO2, O3, CO	769	d: 0	Incidence of conjunctivitis and keratitis varied from region to region.
Seo et al. (2018) [10]	Seoul, South Korea	Multi-level regression model 1/1/2011–31/12/2013	All	O3	Temperature, humidity sex, age	48,344	d: 0	The outpatient incidence of conjunctivitis was increased by O3.
Szyszkowicz et al. (2019) [9]	Edmonton, Canada	Case-crossover Time-stratified Logistic regression Apr 1992–Mar 2002	Sex: male, female	O3	Temperature, humidity	17,211	d: 0–9	Significant association was observed for air pollution at lag 5 day for males, and lag 1 day and lag 3 day for females.

Note: d, day; w, week; CO, carbon monoxide; NO₂, nitrogen dioxide; SO₂, sulfur dioxide; O₃, ozone; PM_2.5, particles smaller than 2.5 μm; PM₁₀, particles smaller than 10 μm.