Association between residential greenness and allergic diseases among adolescents in South Korea: A nationwide representative study

Abstract

Background

Residential greenness is an important environmental factor potentially influencing the development of allergic diseases in adolescents; however, its impact remains understudied in South Korea. This study aimed to examine the association between residential greenness and allergic disease prevalence using nationally representative data.

Method

We analyzed data from 1,130,598 adolescents (7–12th grade) participating in the Korean Youth Risk Behavior Web‐based Survey (2007–2024). Residential greenness was estimated using the normalized difference vegetation index (NDVI) derived from satellite imagery based on GPS coordinates. Logistic regression models assessed associations between NDVI and physician‐diagnosed allergic rhinitis (AR), atopic dermatitis (AD), and asthma, with subgroup analyses conducted by sex, school grade, household income level, and sufficiency of fatigue recovery.

Results

Among 1,130,598 adolescents (51.27% boys), 525,979 (46.52%) participants reported allergic diseases: 363,167 (32.12%) with AR, 250,462 (22.15%) with AD, and 88,584 (7.84%) with asthma. Higher residential greenness was associated with lower adjusted odds ratios of AR (0.83 [95% CI, 0.78–0.89]), AD (0.75 [0.70–0.80]), and asthma (0.45 [0.41–0.50]), with the strongest inverse association observed for asthma. The protective association of greenness was stronger in middle school students compared to high school students and in the higher‐income group compared to the low‐income group.

Conclusion

Residential greenness was associated with a reduced prevalence of allergic diseases among Korean adolescents, with stronger protective associations observed among middle school students and those from higher‐income households. These findings highlight the need to improve equitable access to green spaces for all adolescents.

Key message.

In Korean adolescents, higher residential greenness was significantly associated with reduced odds of physician‐diagnosed allergic rhinitis, atopic dermatitis, and most notably asthma. The protective effect was strongest among middle school students and those from higher‐income households. These results highlight the importance of urban green spaces in mitigating allergic disease risk during adolescence and underscore the need to ensure equitable access to greenness for all youth.

1. INTRODUCTION

Allergic diseases such as allergic rhinitis (AR), atopic dermatitis (AD), and asthma are among the most prevalent chronic conditions during childhood and adolescence and represent a substantial public health concern. ¹ Their prevalence is notably higher in South Korea than in many other countries, and they are frequently linked to psychological distress and reduced quality of life in adolescents. ² , ³

As concern grows over the environmental determinants of allergic diseases, residential greenness has emerged as a potential protective factor. ⁴ However, findings remain inconsistent, with some studies reporting null associations and others suggesting beneficial effects. ⁵ , ⁶ This heterogeneity may be explained by differences in urban development across countries and variations in study populations, including inconsistent age stratification. ⁷ Consequently, country‐specific investigations focusing on adolescents in highly urbanized settings—such as South Korea—are critical, given their vulnerability to environmental exposures and the developmental sensitivity of their immune systems.

Given these considerations, this study seeks to clarify the association between residential greenness and allergic diseases among Korean adolescents, who may be particularly susceptible to environmental risks. It also examines whether these associations differ according to demographic and lifestyle factors, including sex, school grade, household income, and perceived sufficiency of fatigue recovery. Therefore, the findings are intended to inform more targeted public health interventions and urban planning policies that promote adolescent health.

2. METHODS

2.1. Data source

This study analyzed data from a nationally representative sample of 1,130,598 participants drawn from the Korean Youth Risk Behavior Web‐based Survey (KYRBS), conducted annually between 2007 and 2024. Participants with missing values in covariates, such as body mass index (BMI), were excluded from the analysis. The KYRBS is an annual survey conducted by the Korea Disease Control and Prevention Agency (KDCA) and the Ministry of Education since 2005 to assess the health behavior status of Korean adolescents. Approximately 60,000 middle and high school students aged 12–18 participate in the survey each year, with an average response rate of approximately 95%. ⁸ Students with disabilities or dyslexia who are unable to participate independently are excluded from the survey.

The sample selection process has three stages. Initially, the population is stratified by region and school level to reduce sampling error. Secondly, the sample is allocated to 400 middle and 400 high schools nationwide using a proportional allocation method. Last, stratified cluster sampling is used, with schools as the primary units and classes as the secondary units. Sampling weights are applied to ensure the data reflect the entire population of Korean adolescents. The study was conducted in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines. ⁹ KYRBS data were anonymous, and the study protocol was approved by the Institutional Review Board of the Korean Disease Control and Prevention Agency (2014‐06EXP‐02‐P‐A) and by the local law of the Population Health Promotion Act 19 (117058) from the Korean government. All the participants (or their parents or legal guardians in the case of children under 16) provided written informed consent. This study was conducted in accordance with the principles of the Declaration of Helsinki.

2.2. Definition of allergic diseases

This study aimed to examine the association between the normalized difference vegetation index (NDVI) and the prevalence of AR, AD, and asthma among Korean adolescents from 2007 to 2024. Each allergic condition was identified based on responses to physician‐diagnosis questions. Specifically, AR was defined as a positive response to “Have you ever been diagnosed with allergic rhinitis by a doctor?”; AD was defined as a positive response to the question, “Have you ever been diagnosed with atopic dermatitis by a doctor?”; and asthma was defined as a positive response to “Have you ever been diagnosed with asthma by a doctor?” Based on these responses, the presence or absence of each allergic disease was determined. In addition, information was collected on potential risk factors associated with the onset of allergic diseases, including age, sex, and socioeconomic status. ¹⁰

2.3. Exposure to greenness

To evaluate regional levels of green space, we utilized the satellite‐derived NDVI. NDVI is calculated by dividing the difference between near‐infrared and red visible light reflectance by their sum, yielding values ranging from −1.0 to 1.0. ¹¹ Higher values indicate denser vegetation, while negative values represent surfaces without vegetation, such as bodies of water or those with spectral characteristics opposite to vegetation. NDVI data was obtained from the National Aeronautics and Space Administration's Terra satellite, specifically using the Moderate Resolution Imaging Spectroradiometer product MOD13A3 (version 6.1). ¹² This global dataset is provided monthly at a spatial resolution of 1 km and is projected in a sinusoidal grid format. Monthly MOD13A3 data were aggregated according to administrative regions in South Korea to calculate annual average NDVI values from January 2007 to December 2024. All geospatial processing and analyses related to NDVI were conducted using QGIS (version 3.34; QGIS Development Team).

2.4. Covariates

The covariates included in this study were selected based on prior research indicating their potential associations with allergic diseases. ¹³ These covariates were sex (boys and girls), school grade (middle school [grades 7–9] and high school [grades 10–12]), region of residence (Seoul, Busan, Daegu, Incheon, Gwangju, Daejeon, Ulsan, Sejong, Gyeonggi, Gangwon, Chungbuk, Chungnam, Jeonbuk, Jeonnam, Gyeongbuk, Gyengnam, and Jeju), ¹⁴ BMI group (underweight, normal, overweight, and obesity), household income level (low, middle, and high), physical activity (no days, 1–4 days/week, and over 5 days/week), type of residence (with family, with relatives, with friends/alone/in a dormitory, and in a facility), subjective sufficiency of fatigue recovery (sufficient, neutral, and insufficient), fast food consumption (no days, 1–4 days/week, and over 5 days/week), current smoking status (yes and no) and agricultural area (agricultural and non‐agricultural). For adolescents, BMI classification was based on the 2017 Korean Children and Youth Growth Charts, as standard adult BMI criteria are not applicable. ¹⁵ The BMI categories were defined as underweight (<5 percentile), normal (5–84 percentile), overweight (85–94 percentile), and obesity (≥ 95 percentile). Current smoking was defined as having smoked at least once during the past month.

2.5. Statistical analysis

All statistical analyses were performed using data from the KYRBS, incorporating sampling weights, stratification, and clustering to account for the complex survey design. ¹⁶ Logistic regression models were employed to evaluate the associations between the NDVI and physician‐diagnosed allergic diseases, including AR, AD, and asthma. ¹⁷ NDVI was treated as a continuous variable, and both unadjusted and adjusted odds ratios (ORs) with 95% confidence intervals (CIs) were calculated. The adjusted models included covariates such as sex, age (used as a proxy for school grade), BMI, household income, physical activity, type of residence, subjective sufficiency of fatigue recovery, fast food consumption, agricultural area, and current smoking status. To test the robustness of our findings, sensitivity analyses were also performed by including additional covariates. Additionally, subgroup analyses were conducted stratified by sex, school grade, household income, and subjective sufficiency of fatigue recovery to explore potential effect modification. ¹⁸ In each stratified analysis, models were adjusted for the remaining covariates except the stratification criterion. In addition, interaction analyses were conducted to assess whether the association between NDVI and allergic diseases differed across subgroups. A two‐sided test was applied, and a p‐value <.05 was considered statistically significant. All studies were performed using SAS software (version 9.4; SAS Institute, Cary, NC, USA).

3. RESULTS

This study included a final sample of 1,130,598 adolescents. Of the 1,174,993 survey participants over the entire study period, 44,395 were excluded from the analysis due to missing data on BMI, household income, or physical activity (Figure 1). The weighted characteristics of the study population were as follows: 579,635 boys (51.27%) and 550,963 girls (48.73%); 580,795 students in grades 7–9 (51.37%) and 549,803 students in grades 10–12 (48.63%; Table 1). All stratified prevalence estimates by condition are presented in Table S1. The number of participants with each allergic disease among the total study population is as follows: 363,167 (32.12%) with AR, 250,462 (22.15%) with AD, and 88,584 (7.84%) with asthma. Among participants with allergic conditions, sex‐specific distributions revealed that the prevalence of AR and asthma was higher in boys, whereas the prevalence of AD was higher in girls (AR, boys: 184,452 [16.31%]; AD, girls: 140,032 [12.39%]; asthma, boys: 50,962 [4.51%]). Regarding school grade, AR was more prevalent among high school students, while AD and asthma were more common among middle school students (AR, 10–12th: 187,708 [16.60%]; AD, 7–9th: 126,911 [11.23%]; asthma, 7–9th: 46,002 [4.07%]).

FIGURE 1.

Study population flowchart.

TABLE 1.

Crude characteristics of Korean adolescents based on data obtained from the KYRBS, 2007–2024 (n = 1,130,598).

Variables	Total
Overall, n (%)	1,130,598
Age, years, mean (SD)	15.00 (1.75)
Sex, n (%)
Boy	579,635 (51.27)
Girl	550,963 (48.73)
School grade, n (%)
7–9th (middle school)	580,795 (51.37)
10–12th (high school)	549,803 (48.63)
Region of residence, n (%)
Seoul	154,487 (13.66)
Busan	73,474 (6.50)
Daegu	64,720 (5.72)
Incheon	67,961 (6.01)
Gwangju	50,194 (4.44)
Daejeon	47,839 (4.23)
Ulsan	42,325 (3.74)
Sejong	9244 (0.82)
Gyeonggi	215,778 (19.09)
Gangwon	43,414 (3.84)
Chungbuk	45,742 (4.05)
Chungnam	52,058 (4.60)
Jeonbuk	50,542 (4.47)
Jeonnam	48,683 (4.31)
Gyeongbuk	61,290 (5.42)
Gyeongnam	73,782 (6.53)
Jeju	29,065 (2.57)
BMI group, n (%) a
Underweight	58,496 (5.17)
Normal	919,839 (81.36)
Overweight	105,132 (9.30)
Obesity	47,131 (4.17)
Household income, n (%)
Low	204,219 (18.06)
Middle	537,950 (47.58)
High	388,429 (34.36)
Physical activity, n (%)
No days	319,504 (28.26)
1–4 days/week	643,584 (56.92)
Over 5 days/week	167,510 (14.82)
Type of residence, n (%)
With family	1,077,943 (95.34)
With relatives	10,147 (0.90)
With friends/alone/in a dormitory	37,952 (3.36)
In a facility	4552 (0.40)
Subjective sufficiency of fatigue recovery, n (%)
Sufficient	295,658 (26.15)
Neutral	372,874 (32.98)
Insufficient	462,066 (40.87)
Fast food consumption, n (%)
No days	311,394 (27.54)
1–4 days/week	765,994 (67.75)
Over 5 days/week	53,210 (4.71)
Current smoking status, n (%)
Yes	94,542 (8.36)
No	1,036,056 (91.64)
Agricultural area, n (%)
Agricultural	212,573 (18.80)
Non‐agricultural	918,025 (81.20)
Allergic Rhinitis, n (%)
Yes	363,167 (32.12)
No	767,431 (67.88)
Atopic Dermatitis, n (%)
Yes	250,462 (22.15)
No	880,136 (77.85)
Asthma, n (%)
Yes	88,584 (7.84)
No	1,042,014 (92.16)

Abbreviations: BMI, body mass index; N/A, not applicable, KYRBS, Korea Youth Risk Behavior Web‐based Survey.

^a According to 2017 Korean Children and Youth Listing Charts, BMI is divided into four groups: underweight (<5 percentile), normal (5–84 percentile), overweight (85–94 percentile), and obesity (≥95 percentile).

Residential greenness levels varied across the administrative regions of South Korea. The average NDVI levels between 2007 and 2024 ranged from 0.36 (Seoul) to 0.64 (Gyeongnam), with a median of 0.55 (IQR, 0.52–0.59; Table S2). In the fully adjusted models, an increase in NDVI levels was associated with a lower likelihood of allergic diseases prevalence. Specifically, the adjusted OR for AR was 0.83 (95% CI, 0.78–0.89), for AD 0.75 (0.70–0.80), and for asthma 0.45 (0.41–0.50). Among the three conditions, asthma showed the strongest inverse association with residential greenness (Table 2).

TABLE 2.

Stratified odds ratios for allergic rhinitis, atopic dermatitis, and asthma in relation to NDVI, according to sex, school grade, household income, and subjective sufficiency of fatigue recovery.

Variables	Allergic rhinitis			Atopic dermatitis			Asthma
	OR (95% CI)	Adjusted OR (95% CI)	p‐Value	OR (95% CI)	Adjusted OR (95% CI)	p‐Value	OR (95% CI)	Adjusted OR (95% CI)	p‐Value
NDVI
Overall a	0.73 (0.68–0.77)	0.83 (0.78–0.89)		0.76 (0.71–0.82)	0.75 (0.70–0.80)		0.55 (0.50–0.60)	0.45 (0.41–0.50)
Sex b
Boy	0.70 (0.64–0.77)	0.79 (0.72–0.87)	.178	0.73 (0.67–0.81)	0.73 (0.66–0.80)	.241	0.52 (0.46–0.59)	0.43 (0.38–0.49)	.171
Girl	0.76 (0.69–0.83)	0.87 (0.79–0.95)		0.80 (0.73–0.87)	0.77 (0.70–0.84)		0.59 (0.51–0.67)	0.48 (0.42–0.56)
School grade c
7–9th (middle school)	0.56 (0.59–0.72)	0.76 (0.69–0.84)	.003	0.68 (0.62–0.75)	0.68 (0.62–0.74)	<.001	0.46 (0.40–0.53)	0.38 (0.33–0.44)	.091
10–12th (high school)	0.80 (0.73–0.88)	0.89 (0.81–0.97)		0.84 (0.76–0.93)	0.82 (0.75–0.91)		0.66 (0.57–0.76)	0.55 (0.48–0.63)
Household income d
Low	0.85 (0.75–0.98)	0.95 (0.82–1.09)	.001	0.90 (0.78–1.04)	0.94 (0.81–1.09)	.009	0.78 (0.63–0.95)	0.64 (0.52–0.80)	.014
Middle	0.76 (0.70–0.83)	0.87 (0.79–0.95)		0.77 (0.69–0.84)	0.77 (0.70–0.85)		0.52 (0.45–0.60)	0.42 (0.36–0.49)
High	0.66 (0.60–0.72)	0.76 (0.69–0.84)		0.70 (0.63–0.77)	0.65 (0.59–0.72)		0.53 (0.46–0.62)	0.43 (0.37–0.51)
Subjective sufficiency of fatigue recovery e
Sufficient	0.59 (0.53–0.66)	0.73 (0.65–0.82)	<.001	0.70 (0.62–0.79)	0.73 (0.65–0.83)	.059	0.53 (0.44–0.64)	0.45 (0.37–0.54)	.248
Neutral	0.70 (0.64–0.77)	0.83 (0.75–0.92)		0.71 (0.64–0.80)	0.73 (0.65–0.82)		0.50 (0.42–0.58)	0.41 (0.35–0.49)
Insufficient	0.80 (0.73–0.87)	0.89 (0.81–0.98)		0.80 (0.72–0.88)	0.77 (0.70–0.85)		0.60 (0.52–0.59)	0.49 (0.42–0.57)

Note: The numbers in bold indicate significant differences (p < .05).

Abbreviation: CI, confidence interval; NDVI, normalized difference vegetation index; OR, odds ratio.

^a The overall multivariable logistic regression model was adjusted for age, sex (boy and girl), BMI group (underweight, normal, overweight, and obesity), household income (low, medium, and high), physical activity (no days, 1–4 days/week, over 5 days/week), type of residence (with family, with relatives, with friends/alone/in a dormitory), subjective sufficiency of fatigue recovery (sufficient, neutral, insufficient), fast food consumption (no days, 1–4 days/week, over 5 days/week), current smoking status (yes and no) and agricultural area (agricultural and non‐agricultural).

^b The sex‐specific multivariable logistic regression model was adjusted for age, BMI group, household income, physical activity, type of residence, subjective sufficiency of fatigue recovery, fast food consumption, current smoking status, and agricultural area.

^c The school grade‐specific multivariable logistic regression model was adjusted for sex, BMI group, household income, physical activity, type of residence, subjective sufficiency of fatigue recovery, fast food consumption, current smoking status, and agricultural area.

^d The household income‐specific multivariable logistic regression model was adjusted for sex, age, BMI group, physical activity, type of residence, subjective sufficiency of fatigue recovery, fast food consumption, current smoking status, and agricultural area.

^e The subjective sufficiency of fatigue recovery‐specific multivariable logistic regression model was adjusted for sex, age, BMI group, household income, physical activity, type of residence, fast food consumption, current smoking status and agricultural area.

Figure 2 and Table 2 present the associations between sociodemographic variables and the prevalence of AR, AD, and asthma among Korean adolescents. Across all three allergic conditions, the protective associations between residential greenness and disease prevalence were consistently stronger in boys than in girls. For instance, in the case of AR, the adjusted OR was 0.79 (95% CI, 0.72–0.87) in boys and 0.87 (0.79–0.95) in girls. However, the interaction between NDVI and sex was not statistically significant, indicating that the observed sex‐based difference in effect may not be robust.

FIGURE 2.

Adjusted odds ratios for allergic rhinitis, atopic dermatitis, and asthma according to sex, school grade, household income, and subjective sufficiency of fatigue recovery. CI, confidence interval; OR, odds ratio. *Hollow squares represent non‐significant values.

The protective association of residential greenness was consistently stronger in middle school students (grades 7–9) compared to high school students (grades 10–12) for AR and AD. For AR, the adjusted OR was 0.76 (95% CI, 0.69–0.84) in middle school students and 0.89 (0.81–0.97) in high school students. A similar difference was found for AD (middle school: 0.68 [95% CI, 0.62–0.74]; high school: 0.82 [0.75–0.91]). For asthma, while the adjusted ORs indicated stronger protective associations in middle school students (0.38 [95% CI, 0.33–0.44]) than in high school students (0.55 [0.48–0.63]); however, the interaction analysis showed statistical significance only for AR and AD, indicating that the difference in greenness effects between school grades was not robust in the case of asthma.

Stratified analyses by household income revealed a dose–response pattern in AR and AD. For AR, the adjusted OR was 0.87 (95% CI, 0.79–0.95) in the middle‐income group and 0.76 (0.69–0.84) in the high‐income group. For AD, the corresponding ORs were 0.77 (95% CI, 0.70–0.85) and 0.65 (0.59–0.72), respectively. In the case of asthma, protective associations were evident across all income levels, with adjusted ORs of 0.64 (95% CI, 0.52–0.80), 0.42 (0.36–0.49), and 0.43 (0.37–0.51) in the low‐income, middle‐income, and high‐income groups, respectively. The interaction between NDVI and household income was statistically significant, indicating that the effect of greenness varied by income level.

Finally, stratified analysis by subjective sufficiency of fatigue recovery revealed a statistically significant interaction for AR, along with a pattern suggestive of dose–response. The adjusted ORs for AR were 0.73 (95% CI, 0.65–0.82) among individuals reporting sufficient fatigue recovery, 0.83 (0.75–0.92) for those reporting neutral recovery, and 0.89 (0.81–0.98) for those reporting insufficient recovery. These findings suggest a dose–response pattern, where the protective association between greenness and AR became weaker as subjective fatigue recovery decreased.

4. DISCUSSION

4.1. Key findings

This study is the first to analyze the association between residential greenness and allergic diseases in Korean adolescents using nationally representative data. Among 1,130,598 adolescents (7–12th grade; 51.27% boys), 525,979 (46.52%) participants reported having a diagnosis of allergic diseases: 363,167 (32.12%) with AR, 250,462 (22.15%) with AD, and 88,584 (7.84%) with asthma. Higher levels of greenness, as measured by NDVI, were significantly associated with a lower prevalence of AR (adjusted OR, 0.83 [95% CI, 0.78–0.89]), AD (0.75 [0.70–0.80]), and asthma (0.45 [0.41–0.50]), with the strongest protective association observed for asthma. Middle school students in grades 7–9 showed a more pronounced protective effect of greenness exposure than high school students in grades 10–12. Additionally, the protective effect of greenness on AR was more pronounced among individuals who reported a higher income group and sufficient fatigue recovery.

4.2. Plausible underlying mechanisms

Higher levels of residential greenness were associated with a lower prevalence of AR, AD, and asthma. Several plausible biological mechanisms support these associations. Greenness reduces exposure to airborne pollutants (e.g., PM₁₀) by acting as a natural filter and enhancing ambient air ventilation, thereby mitigating airway epithelial damage and pro‐inflammatory responses. ¹⁹ , ²⁰ In parallel, greener environments harbor greater microbial diversity compared to urban built environments. ²¹ Exposure to environmental microbes through skin contact, inhalation, or ingestion may contribute to immune education by promoting the development of immune tolerance and suppressing Th2‐skewed inflammatory pathways implicated in allergic diseases. ²² Additionally, greenspace exposure has been linked to reductions in psychological stress, which can influence systemic immune regulation through modulation of the hypothalamic–pituitary–adrenal axis and gut microbiota composition. ²³ Consistent with these mechanisms, previous studies have reported that individuals residing in greener areas exhibit a higher relative abundance of beneficial genera (e.g., Lactobacillus, Bifidobacterium) and soil‐associated genera (e.g., Actinomycetospora, Brachybacterium), along with a lower relative abundance of genera that may include pathogenic species (e.g., Holdemania, Streptococcus). ²⁴

Interestingly, the protective effects of greenness were more pronounced among middle school students, which may largely reflect differences in outdoor activity levels. Younger adolescents are generally more engaged in outdoor activities compared to older students, resulting in greater exposure to the benefits of greenness. ²⁵ , ²⁶ In addition, differences by income group may reflect disparities in access to green spaces. Adolescents from higher‐income households are more likely to live in neighborhoods with greater availability, better quality, and closer proximity to green spaces, facilitating more frequent interaction with these environments. ²⁷ Conversely, adolescents from lower‐income neighborhoods often face greater physical and social barriers to accessing green spaces, including longer travel distances and lower quality environments, which may limit their exposure and attenuate the potential health benefits of residential greenness. ²⁷

While some subgroup differences reached statistical significance, their clinical or public health relevance may vary. For example, sex‐specific associations were statistically significant within each group; however, the interaction term was not significant, indicating that differences in effect sizes between boys and girls may not be substantial. In contrast, household income differences were both statistically significant and showed larger variations in effect size, which could warrant further exploration in the context of targeted public health strategies.

4.3. Comparison of previous studies

Previous studies have investigated the association between residential greenness and allergic diseases; however, the findings have been inconsistent. Several cohort studies and meta‐analyses reported a protective association ⁵ , ²⁸ , ²⁹ of higher greenness exposure against AR, AD, or asthma, while others found no significant association. ⁷ , ³⁰ Moreover, some studies even suggested that greenness could exacerbate allergic diseases through increased exposure to environmental allergens. ³¹ , ³² , ³³ In addition, many of these studies were limited by small sample sizes (n = 1050, ⁵ n = 522, ³⁴ n = 219 ³⁵ vs. n = 1,140,405 in the present study), regionally restricted cohorts, ³⁶ and a focus exclusively on adult populations. ²⁸ Furthermore, most prior studies focused on a single allergic disease ³⁷ , ³⁸ rather than evaluating multiple conditions simultaneously, making it difficult to comprehensively understand the broader impact of greenness on allergic health.

Although our present analysis is cross‐sectional, emerging longitudinal evidence supports potential causal pathways linking greenness exposure to allergic outcomes. For example, a cohort study conducted in Portugal found that higher greenness exposure at birth was significantly associated with reduced risks of asthma and AR by age 7, suggesting a temporal association that supports causal inference. ⁵ Another longitudinal study based on a U.S. prospective cohort showed that greater NDVI at age 7 was associated with improved pulmonary function (FEV₁ and FVC) and a differential risk of asthma depending on atopic sensitization, highlighting the role of early‐life environmental exposures in shaping respiratory health. ³⁹

Our current study contributes to this growing body of literature by using a large, nationally representative adolescent sample and applying an objective NDVI‐based greenness assessment with comprehensive confounder adjustment. To our knowledge, this is the first study to focus specifically on Korean adolescents in examining the association between residential greenness and allergic diseases. Furthermore, the methodological approach enables broader generalization to adolescent populations in other rapidly urbanizing environments. In addition, by simultaneously evaluating AR, AD, and asthma, we provided an integrated view of greenness associations across multiple allergic conditions. Our study offers complementary evidence to previous research by providing large‐scale, population‐level insights that are generalizable to adolescents. While prior studies have contributed important findings through longitudinal tracking or pooled estimates, our cross‐sectional approach adds additional value by capturing real‐world prevalence patterns across multiple allergic conditions in a nationally representative adolescent population.

4.4. Clinical and policy implications

Our findings highlight the potential role of residential greenness as a modifiable environmental factor for the prevention and management of allergic diseases among adolescents in South Korea. Greening efforts should be prioritized in regions with limited vegetation, where even modest increases may yield substantial health gains. Subgroup disparities highlight the need to consider individual and environmental contexts in public health strategies. To maximize the health benefits of greenness, it is critical to ensure that green spaces are not only available but also easily accessible within adolescents' daily activity ranges. Urban planning should ensure the placement of green spaces near schools and residential areas, while schools and community programs should actively promote structured outdoor activities. ²⁶ Equitable access to high‐quality green spaces is particularly important to avoid widening health disparities among vulnerable populations. ²⁷ Similar initiatives in other countries, such as London's Green Infrastructure Strategy and Finland's forest‐based outdoor education programs, have demonstrated the public health potential of greening efforts. Finally, while this study focused on adolescents, expanding green infrastructure is likely to yield broader health benefits across all age groups by supporting both physical and psychological well‐being. ⁴⁰

4.5. Strengths and limitations

This study has several limitations that should be considered while interpreting our results. First, as the KYRBS is a school‐based survey, it excludes adolescents not enrolled in school. According to data from the Korean Ministry of Education and Statistics Korea, approximately 1.8% of individuals aged 12–17 years were not enrolled in school. ⁸ While this proportion is relatively small, it may limit the generalizability of the findings to the entire adolescent population. Nevertheless, the KYRBS remains valuable in providing nationally representative data for school‐attending youth. Second, this study relies on self‐reported health outcomes, which may be subject to recall bias and misclassification. Moreover, due to the cross‐sectional nature of the study design, it is possible to identify associations between residential green space and allergic diseases, but not to infer causality. These limitations may be particularly pronounced among subpopulations with limited access to healthcare services or lower health literacy and should therefore be carefully considered in interpretation. Future studies should aim to better assess causal relationships through longitudinal or quasi‐experimental designs based on these findings. Third, due to the absence of precise residential address data, we were unable to assess individual‐level exposure to greenness within specific buffer zones. Instead, we used a semi‐ecological design, assigning the same greenness exposure to all individuals within the same administrative region. This approach may not account for intra‐regional variability in exposure; however, it remains useful for evaluating population‐level associations by linking representative health data with environmental indicators. Additionally, due to the cross‐sectional design of this study, the potential for reverse causality – for example, families with allergic children selectively choosing greener or less green environments – could not be ruled out. Furthermore, the absence of individual‐level geolocation data limited our ability to evaluate such selective relocation or residential self‐selection in greater detail. Fourth, the study did not distinguish between different types of greenness, such as natural forests versus artificial green spaces, nor did it consider vegetation composition or structure. As a result, unmeasured agricultural exposures (e.g., pesticides, farm dust, specific microbial environments) may contribute to residual confounding. Additionally, the use of administrative regions to assign greenness exposure may not capture finer‐scale spatial variability within regions. Nevertheless, to minimize potential confounding from agricultural factors, we adjusted for agricultural area status (agricultural vs. non‐agricultural) as a covariate in our analyses. Fifth, although we used one‐year average NDVI values to represent greenness exposure, this metric may not capture seasonal or short‐term fluctuations that could influence allergic symptoms. Furthermore, since the health outcomes were based on lifetime diagnosis, a temporal mismatch between current NDVI exposure and disease onset could lead to exposure misclassification. These issues underscore the need for studies with more temporally resolved exposure and outcome data. Finally, the cross‐sectional nature of the study limits causal interpretation. Although the repeated cross‐sectional design enabled population‐level inferences, the inability to assess within‐individual temporal dynamics remains a key limitation. Future longitudinal studies are needed to clarify causal relationships.

Importantly, this study represents the first investigation into the associations between residential greenness and allergic diseases among South Korean adolescents using nationally representative cross‐sectional data. The findings provide valuable evidence supporting the protective effects of residential greenness on AR, AD, and asthma. These results underscore the importance of incorporating green space development into urban planning and public health strategies aimed at reducing the burden of allergic diseases among youth in South Korea.

5. CONCLUSION

This nationally representative study is the first to identify that residential greenness is associated with a lower prevalence of allergic diseases—AR, AD, and asthma—among Korean adolescents. The protective effect of greenness, as measured by NDVI, was particularly strong for asthma and was more pronounced among middle school students and those from higher‐income households. These findings underscore the potential value of promoting accessible urban greenness as an environmental strategy to reduce allergic disease burden and health disparities in Korean adolescents.

AUTHOR CONTRIBUTIONS

Jeongmin M. Lee: Writing – original draft; writing – review and editing; conceptualization; methodology; formal analysis; visualization. Juyeong Kim: Conceptualization; methodology; formal analysis; visualization; writing – review and editing; writing – original draft. Kyeongeun Kim: Conceptualization; methodology; formal analysis; visualization; writing – review and editing; writing – original draft. Yesol Yim: Writing – review and editing. Yerin Hwang: Writing – review and editing. Selin Woo: Conceptualization; writing – original draft; writing – review and editing; visualization; methodology; formal analysis; supervision; funding acquisition; project administration. Dong Keon Yon: Supervision; formal analysis; conceptualization; methodology; visualization; writing – review and editing; writing – original draft; project administration; funding acquisition.

FUNDING INFORMATION

This research was supported by a grant from Kyung Hee University in 2024 (KHU‐20241061). The funders had no role in study design, data collection, data analysis, data interpretation, or writing of the report.

CONFLICT OF INTEREST STATEMENT

We declare no competing interests.

PEER REVIEW

The peer review history for this article is available at https://www.webofscience.com/api/gateway/wos/peer‐review/10.1111/pai.70199.

ETHICAL APPROVAL

KYRBS data were anonymous, and the study protocol was approved by the Institutional Review Board of the Korean Disease Control and Prevention Agency (2014‐06EXP‐02‐P‐A) and by the local law of the Population Health Promotion Act 19 (117058) from the Korean government.

Supporting information

Table S1.

Lee JM, Kim J, Kim K, et al. Association between residential greenness and allergic diseases among adolescents in South Korea: A nationwide representative study. Pediatr Allergy Immunol. 2025;36:e70199. doi: 10.1111/pai.70199

DATA AVAILABILITY STATEMENT

The data are available on reasonable request. Study protocol, statistical code: available from DKY (email: yonkkang@gmail.com). Data set: available from the Korea Disease Control and Prevention Agency (KDCA) and the Ministry of Education through a data use agreement.