新北市都會區開放空間綠地與成年居民健康的關係

The Relationship between Urban Open Green Spaces and Health of Adults in Metropolitan Area of New Taipei City

錢建文(Jien-Wen Chien)

指導教授 ： 詹長權

國立臺灣大學/公共衛生學院/環境與職業健康科學研究所/博士(2021年)

https://doi.org/10.6342/NTU202200005

摘要

背景：都市綠地與身體健康和心理健康有許多相關性或實驗性的研究，然而大多數的研究都是在人口密度低且綠地多的地區做的；且台灣很少有研究探討城市綠地對於居民的健康效應。
研究目的：探討都市開放空間綠地和新北市都會區成年居民健康之間的關係。
方法：我們所研究的族群為新北市都會區三十歲以上居民共40,375人，在2007到2009參與了政府所舉辦的健康篩檢計畫。本研究定義的都市開放綠地空間(urban open green spaces, UOGS)為內政部國土繪測圖資之土地利用歸類為公園，綠地(排除263平方公尺以下)，國小以上學校(排除私立學校)，與運動場所(排除室內)。上述空間若小於一公頃，且至少含有以下三者：綠色植物、運動設施、以及座椅，歸類為小型公共都市開放空間(small public urban open spaces, SPUOS)。「綠」(greenness)的測量方法為利用衛星遙測資料所計算出來的標準化差異植生指標(normalized difference vegetation index, NDVI)。UOGS的可近性(accessibility)定義為研究參與者住家與UOGS的最近距離。UOGS的可利用性(availability)定義為研究參與者住家附近半徑五百公尺內的NDVI平均值。土地使用回歸模型(land use regression model)則用來評估空氣汙染的個人暴露。利用羅吉斯迴歸分析(Logistic regression)計算UOGS的可近性與可利用性對於各種疾病的勝算比的影響。NDVI與血壓值(包括收縮壓，舒張壓與平均血壓)的關係由多變量線性回歸分析來探討。細懸浮微粒(PM2.5)對於Greenness與血壓的負相關是否有中介作用則由中介分析(mediation analysis)與干擾分析(moderation analysis)來探討。
結果：研究族群的NDVI範圍為-0.22到0.26；個人PM2.5平均年暴露範圍為10.96到43.59微克每立方米(µg/m3)。體重過重，糖尿病，慢性腎病(chronic kidney disease, CKD) ，與高血壓的發生率依次為:49.5%, 7.0%, 9.6%,32.3%；若將高血壓分期，包括高血壓前期，第一期到第三期高血壓的發生率依次為:37.3%, 21.4%, 8.2%, 2.7%。我們發現CKD與SPUOS顯著相關，高血壓則與NDVI顯著相關；體重過重與糖尿病則與UOGS無相關。住家距離SPUOS兩百公尺以上與小於或等於兩百公尺比較，CKD的校正後勝算比(adjust odds ratio, OR)為1.144 (95%信賴區間:1.059-1.237)。若將公立學校加上SPUOS一起看，所得結果相近。CKD與NDVI無關。次族群分析發現此效應在以下健康的族群較為顯著:包括沒有吸菸，沒有高血壓，沒有糖尿病，和沒有高三酸甘油脂血症。在高血壓方面，NDVI第四個四分位(最綠)與第一個四分位比較，以及PM2.5每增加10µg/m3，發生高血壓的勝算比分別為0.744(95%信賴區間:0.698-0.793)與1.057(95%信賴區間:1.015-1.100)。在高血壓分期方面，NDVI第一個四分位(最不綠)與第四個四分位比較，發生高血壓前期以及第一到第三期高血壓的勝算比分別為1.145(95%信賴區間: 1.063- 1.233)，1.395(95%信賴區間:1.276-1.525)，1.528(95%信賴區間:1.353-1.724)，和1.852(95%信賴區間:1.531-2.241)。由線性回歸模式來看，NDVI每增加一個四分位距(0.03)，收縮壓，舒張壓與平均血壓分別降低0.75mmHg(95%信賴區間:-0.90,-0.61)，0.51mmHg(95%信賴區間: -0.60,-0.41)，和0.59mmHg(95%信賴區間:-0.69,-0.49)。中介分析發現NDVI與較低高血壓的相關性中，PM2.5並沒有中介效果，但是有干擾效果。由分層分析來看，NDVI與高血壓前期的負相關在下列族群較顯著:年輕者，女性，無吸菸，沒有糖尿病，沒有高膽固醇血症，沒有高三酸甘油脂血症，沒有慢性腎病，和沒有代謝症候群者。
結論：新北市都會區成年居民若居住在距離SPUOS和公立學校比較近，則發生CKD的機會較少；PM2.5與高血壓正相關，而NDVI則與高血壓前期與第一期到第三期高血壓的發生呈負相關。

關鍵字

高血壓 ； 高血壓前期 ； 慢性腎病 ； 糖尿病 ； 體重過重 ； 都市開放綠地空間 ； 小型公共都市開放空間 ； 細懸浮微粒 ； 綠 ； 標準化差異植生指標 ； 中介效果 ； 干擾效果

並列摘要

Background: There are many associated or experimental studies exploring the beneficial effects of urban open green spaces on residential physical or mental health. However, most studies were conducted in the low populated area with high greenery. Besides, there are few such studies in Taiwan.
Purpose: The aim of the study is to explore the relationship between urban open green spaces and health of adults living in the metropolitan area of New Taipei City.
Methods: A total of 40,375 participants older than 30-year-old attended the health screening program during 2007 to 2009 in the Metropolitan area in the New Taipei City. We defined urban open green spaces (UOGS) if land use belongs to parks, green, plaza, public schools and sport venues. Small public urban open space (SPUOS) was defined as UOGS with area less than 1 hector and with at least three of the followings: vegetations, sport facilities and benches. Greenness was defined as the mean Normalized Difference Vegetation Index (NDVI). Accessibility of UOGS was defined as proximity of participants’ home to UOGS. Availability of UOGS was defined as mean value of NDVI within 500m buffer of home. Personal exposures to air pollutants were accessed by land use regression model. Logistic regression model was used to calculate odds ratios of diseases for proximity to UOGS or greenness. Multivariate linear regression model was used to evaluate the association between NDVI and blood pressures, including systolic (SBP), diastolic (DBP) and mean (MBP) pressures. The mediation and moderation analysis were used to assess the mediation effect of PM2.5 on the association between greenness and HTN.
Results: The ranges of NDVI and annual average PM2.5 exposures were -0.22 to 0.26 and 10.96-43.59 μg/m3 , respectively. The prevalence of overweight, diabetes, CKD and HTN were 49.5%, 7.0%, 9.6% and 32.3%, respectively. The prevalence of prehypertension and stage 1-3 hypertensions were 37.3%, 21.4%, 8.2% and 2.7%, respectively. We found that CKD was significantly associated with proximity to SPUOS. No relationship was found between proximity to UOGS or NDVI and overweight or diabetes. The adjusted odds ratio of CKD comparing distance to SPUOS > 200m and  200m was 1.144 (95% CI, 1.059-1.237). The effect of open space on CKD was similar if we incorporated public schools to SPUOS. Greenness was not associated with CKD. Subgroups analyses revealed the effect of SPUOS on CKD was more prominent in health residents, including in those who never smoke, no hypertension, no diabetes and no hypertriglyceridemia. The ORs of HTN were 0.744 (95% CI: 0.698-0.793) for NDVI (quartile 4 vs. quartile 1) and 1.057 (1.015-1.100) for 10 μg/m3 increase in PM2.5, respectively. The ORs of prehypertension and stage 1-3 HTN for NDVI (quartile 1 vs. quartile 4) were 1.145 (1.063-1.233), 1.395 (1.276 -1.525), 1.528 (1.353 - 1.724) and 1.852 (1.531-2.241), respectively. With each IQR increase of NDVI (0.03), we found SBP, DBP and MBP were decreased by 0.75mmHg (-0.90 - 0.61), 0.51mmHg (-0.60 - 0.41) and 0.59mmHg (-0.69 - 0.49), respectively, in linear regression models. Mediation analysis revealed greenness effect was not mediated by PM2.5. On the contrary, PM2.5 can moderate the effect of greenness on SBP. Stratified analysis found the negative relationship between greenness and prehypertension was more prominent for people who were younger, female, never smoking, without diabetes, without hypercholesterolemia, without hypertriglyceridemia, without CKD, and without metabolic syndrome.
Conclusion: Proximity to SPUOS or schools was associated with lower risk of CKD for adults living in the Metropolitan area in New Taipei City. PM2.5 was positively associated with HTN, and greenness was negatively associated with prehypertension and stage 1 to 3 HTN for urban adult residents.

並列關鍵字

hypertension ； prehypertension ； chronic kidney disease ； diabetes ； overweight ； urban open green space ； small public urban open space ； PM2.5 ； greenness ； NDVI ； mediation effect ； moderation effect

參考文獻

Abelt K, McLafferty S. Green streets: urban green and birth outcomes. International journal of environmental research and public health 2017;14:771.

Al-Kindi SG, Brook RD, Biswal S, Rajagopalan S. Environmental determinants of cardiovascular disease: lessons learned from air pollution. Nature Reviews Cardiology 2020;17:656-72.

Astell-Burt T, Feng X, Kolt GS. Green space is associated with walking and moderate-to-vigorous physical activity (MVPA) in middle-to-older-aged adults: findings from 203 883 Australians in the 45 and Up Study. Br J Sports Med 2014;48:404-6.

Bagheri Z, Nadoushan MA, Abari MF. Evaluation the effect of green space on air pollution dispersion using satellite images and landscape metrics: a case study of Isfahan City. Fresenius Environ Bull 2017;8135.

Banay RF, Bezold CP, James P, Hart JE, Laden F. Residential greenness: current perspectives on its impact on maternal health and pregnancy outcomes. Int J Womens Health 2017;9:133-44.