All Issue

2022 Vol.40, Issue 5 Preview Page

Article

A Study on the Effect of Urban Environment on Obesity Using Multinomial Logistic Regression Models

다항 로지스틱회귀모형을 활용한 도시환경이 비만에 미치는 영향에 관한 연구

Songhyeon SHIN1
,
Jahun KOO2
,
Sujae KIM3
,
Sangho CHOO4*
신 송현1
,
구 자헌2
,
김 수재3
,
추 상호4*
1M.S. Student, Department of Urban Planning, Hongik University, Seoul 04066, Korea
2Ph.D. Student, Department of Urban Planning, Hongik University, Seoul 04066, Korea
3Research Professor, Hongik University, Seoul 04066, Korea
4Professor, Department of Design and Planning, Hongik University, Seoul 04066, Korea
1홍익대학교 도시계획과 석사과정
2홍익대학교 도시계획과 박사과정
3홍익대학교 연구교수
4홍익대학교 도시공학과 교수
*Corresponding Author
31 October 2022. pp. 717-734
PDF XML Citation
Abstract

Recently, research for eco-friendly and sustainable city design has been actively attempted. In this situation, the health city theory based on the relationship between the built environment and individual health levels has been proposed to improve the resident’s health at the urban level. In this regard, we aimed to identify the relationship between the built environment and individual health levels by using community health survey data from 2018 to 2020. The individual health level was classified under three categories: low weight, normal, and obesity using BMI (Body Mass Index). Then we tried to investigate how the built environment and individual factors affect individual health levels using multinomial logistic regression. As a main result of the analysis, it was found that car drivers and motorcycle drivers were more likely to be obese than normal weight. The higher the proportion of park and public sports facilities, the lower likely to be obese. In addition, the higher volume of public transit, availability of sharing bike and sharing PM (Personal Mobility) services were significantly inversely associated with obesity. Furthermore, using these factors we predicted the individual health level to use as an evaluation index for health city. To this end, DNN (Deep Neural Network) algorithm, which is widely used in classification problems recently, and the multinomial logistic regression were performed. These two methods were compared by calculating the evaluation using the Confusion Matrix. The performance of the DNN model was found to be superior to that of the multinomial logistic regression model, particularly in the precision of obesity prediction. These findings can be used as basic data and evaluation index when designing a health city through the prediction of the individual health levels and obesity population.

Keywords
body mass index
confusion matrix
deep neural network
health city
multinomial logistic regression

최근 친환경적이며 지속가능한 도시 설계를 위한 연구들이 활발히 진행되고 있다. 이러한 상황 속에서, 제안된 건강도시 이론은 지역의 도시환경과 개인의 건강수준 간의 관계를 기반으로 도시적 차원에서 시민들의 건강을 증진시키고자 하는 이론이다. 이에 본 연구에서는 최근 3개년 간의 지역사회건강조사자료를 활용하여 도시환경이 개인의 건강수준에 미치는 영향 분석을 통해 건강도시 설계에 기여하고자 하였다. 체질량 지수를 활용하여 개인의 건강수준을 저체중, 정상, 비만으로 분류하였으며, 다항 로지스틱회귀모형을 활용하여 정상을 기준으로 저체중과 비만에 미치는 영향요인들을 분석하였다. 주요 분석 결과, 자동차 운전자, 오토바이 운전자인 경우 정상에 비해 비만일 가능성이 높게 나타났으며, 공원 면적비율, 공공체육시설 면적비율이 높은 지역에 거주할수록 비만일 가능성이 낮게 나타났다. 또한, 대중교통 이용량이 많은 지역에 거주할수록 비만일 가능성이 낮게, 공공자전거 서비스 지역에 거주하는 경우, 공유 PM 서비스 지역에 거주하는 경우 비만일 가능성이 낮게 나타났다. 나아가, 장래 건강도시 평가지표로 활용하기 위해 도출된 영향요인들을 활용하여 개인의 건강수준을 예측하였다. 이를 위해 다항 로지스틱회귀모형과 더불어 최근 분류문제에 널리 활용되고 있는 DNN(Deep Neural Network) 알고리즘을 활용하였다. 그리고 두 방법론을 비교 및 평가하기 위해 Confusion Matrix를 활용하여 평가지표를 산출하였다. 분석결과, DNN 모델의 성능이 다항 로지스틱회귀모형에 비해 우수한 것으로 나타났으며 특히, 비만 예측의 정밀도가 크게 우수한 것으로 나타났다. 이러한 결과는 영향요인을 고려한 건강도시 설계의 기초자료로 활용될 수 있을 것이며, 개인의 건강수준 예측을 통한 비만인구 예측으로 건강도시 설계 시 평가지표로 활용될 수 있을 것으로 기대된다.

키워드
체질량지수
오차행렬
심층신경망
건강도시
다항로지스틱회귀모형
References
1
Ayon S. I., Islam M. M., Hossain M. R. (2020), Coronary Artery Heart Disease Prediction: A Comparative Study of Computational Intelligence Techniques, IETE Journal of Research, Volume: 10.1080/03772063.2020.1713916 10.1080/03772063.2020.1713916
2
Barbara B. B., Ikuho Y., Ken R. S., Cathleen D. Z., Lori K. J., Jessie X. F. (2009), Mixed Land Use and Walkability: Variations in Land Use Measures and Relationships with BMI, Overweight, and Obesity, Health & Place, 15(4), 1130-1141. 10.1016/j.healthplace.2009.06.00819632875PMC2778756
3
Bayraci S., Susuz O. (2019), A Deep Neural Network (DNN) Based Classification Model in Application to Loan Default Prediction, Asociatia Generala a Economistilor din Romania, Theoretical and Applied Economics, 4(621), 75-84.
4
Beak J. U. (2000), Six Sigma and Statistics, The Korean Society for Quality Management, Journal of the Korean Society for Quality Management, 28(3), 114-123.
5
Choi M. H., Jung H. Y. (2020), A Study on the Influencing Factor of Intention to Use Personal Mobility Sharing Services, J. Korean Soc. Transp., 38(1), Korean Society of Transportation, 1-13. 10.7470/jkst.2020.38.1.001
6
Chun H. J. (2016), A Study on the Effect of Urban Environment Using GIS and Spatial Econometric Models on Residents, Residential Environment Institute Of Korea, Residential Environment: Journal of The Residential Environment Institute of Korea, 14(2), 109-118.
7
Community Health Survey, 2018, Korea Centers for Disease Control and Prevention.
8
Community Health Survey, 2019, Korea Centers for Disease Control and Prevention.
9
Community Health Survey, 2020, Korea Centers for Disease Control and Prevention.
10
Daniel F., Steven C., Stephen A. M. (2013), Does Transportation Mode Modify Associations between Distance to Food Store, Fruit and Vegetable Consumption, and BMI in Low-income between Neighborhoods?, The American Journal of Clinical Nutrition, 97(1), 167-172. 10.3945/ajcn.112.03639223193006PMC3712124
11
Feng S., Bingyu L., Yik W. L., Paul S. F. Y. (2019), Associations between Commuting and Well-being in the Context of a Compact City with a Well-developed Public Transport System, Journal of Transport & Health, 13, 103-114. 10.1016/j.jth.2019.03.016
12
Han J., J. Pei M. Kamber (2011), Data Mining: Concepts and Techniques, Elesvier.
13
Jang J. M., Son W. B. (2017), Analysis of Association between Traffic and Local Variables Affecting Body Mass Index, Gyeonggi Research Institute, GRI REVIEW, 19(2), 79-100.
14
Jang M. H., Seok G. H., Park H. S. (2012), The Relationship between Existence of Spouses and Obesity for Korean Adults, Korea Institute of Electronic Communication Science, The Journal of the Korea Institute of Electronic Communication Science, 7(3), 679-685.
15
KDCA (2021), Korea Community health at a Glance 2020: Korea Community Health Survey (KCHS).
16
Kim D. H., Lee K. H., Jung K. T. (2012), The Effects of Smoking on Obesity and Its Implications for Health Insurance, Korea Insurance Research Institute, Journal of Insurance and Finance, 67(0), 111-132.
17
Kim D. Y., Lee K. S. (2015), Gender Difference in the Relationship between Marital Status and Obesity: Propensity Score Matching Approach, The Korean Association of Health Economics and Policy, The Korean Journal of Health Economics and Policy, 21(2), 27-47.
18
Kim E. J., Kang M. G. (2011), Effects of Built Environmental Factors on Obesity and Self-reported Health Status in Seoul Metropolitan Area Using Spatial Regression Model, Korea Research Institute for Human Settlements, The Korea Spatial Planning Review, 68, 85-98. 10.15793/kspr.2011.68..005
19
Kim J. Y., Kim S. J., Lee G. J., Choo S. H. (2021), Estimating a Mode Choice Model Shared E-scooter Service: Focused on Access Travel and Neighborhood Travel, The Korea Institute of Intelligent Transport Systems, The Journal of The Korea Institute of Intelligent Transport Systems, 20(1), 22-39. 10.12815/kits.2021.20.1.22
20
Kim K. O., Jeon Y. H., Kim Y. S. (2014), The Effects of Mental Health on Obesity among Korean Adolescents, Journal of Digital Convergence, 12(10), 467-476. 10.14400/JDC.2014.12.10.467
21
Kim M. J., Kim H. K. (2020), Development of Driver’s Behavior Model Based on Introduction of Vehicle Mileage Traveled Tax: Focusing on Busan Metropolitan City, J. Korean Soc. Transp., 38(6), Korean Society of Transportation, 507-519. 10.7470/jkst.2020.38.6.507
22
Kim S. M. (2000), body fat measurement, The Korean Academy of Clinical Geriatrics, Korean Journal of clinical geriatrics, 1(2), 23-27.
23
Kim S. M. (2004), Stress and Obesity, The Korean Society of Stress Medicine, The Korean Journal of Stress Research, 12(3), 17-20.
24
Korea Health Industry Development Institute Report (2019), Health Risk Signs Such as Adult Men and Obesity That Exceed 40hours a Week.
25
Korean Society for the Study of Obesity (2020), Quick Reference Obesity treatment Guideline.
26
Lee J. Y., Kang S. N., Kim S. A., Son D. M., Lee B. G., Ham O. K. (2019), Factors Influencing Obesity among Rural and Urban Adolescent: Analysis of 2013 Korean Youth Health Behavior Survey, Korea Society of public Health Nursing, Journal of Korean Public Health Nursing, 33(1), 73-84.
27
Lee S. W., Min S. H., Park J. Y., Yoon S. D. (2005), The Practice on Logit & Probit Model.
28
Min I. S., Choi P. S. (2012), Advanced Panel Data Analysis.
29
Moon O. R., Kang J. H., Lee S. Y., Jeong B. G., Lee S. J. et al. (2001), Increasing Prevalence of Obesity Related Disease for Koreans Associated with Overweight and Obesity, The Korean Society for Preventive Medicine, Journal of Preventive Medicine and Public Health, 34(4), 309-315.
30
Murphy M., Koohsari M., Badland H., Giles-Corti B. (2017), Supermarket Access, Transport Mode and BMI: The Potential for Urban Design and Planning Policy across Socio-economic Areas, Public Health Nutrition, 20(18), 3304-3315. 10.1017/S136898001700233628879832
31
Oh S. W. (2009), Effects of Alcohol on Obesity and Metabolic Syndrome, Korean Society for the Study of Obesity, The Korean Journal of obesity, 18(1), 1-7.
32
Ohsaki M., Wang P., Matsuda K., Katagiri S., Watanabe H., Palescu A. (2017), Confusion-Matrix-Based Kernel Logistic Regression for Imbalanced Data Classification, IEEE Transactions on Knowledge and Data Engineering, IEEE, 29(9), 1806-1819. 10.1109/TKDE.2017.2682249
33
Shin K. S., Sung H. G., Rho J. H. (2009), Analysis of Causal Relationship between Commuting Behavior of office worker and Body Mass Index using Path Analysis, Korea Planning Association, Journal of Korea Planning Association, 44(1), 197-209.
34
Sung H. G. (2018), Sharing Economy and Sharing Mobility, Public Official Benefit Association, Urban affairs, 53(599), 36-39.
35
World Health Organization (2000), The Asian-Pacific Perspective: Redefining Obesity and Its Treatment, Geneva, Switzerland, WHO Western Pacific Region.
Information
  • Publisher :Korean Society of Transportation
  • Publisher(Ko) :대한교통학회
  • Journal Title :Journal of Korean Society of Transportation
  • Journal Title(Ko) :대한교통학회지
  • Volume : 40
  • No :5
  • Pages :717-734
  • Received Date : 2022-05-26
  • Revised Date : 2022-06-29
  • Accepted Date : 2022-09-21
  • DOI :https://doi.org/10.7470/jkst.2022.40.5.717
Journal Informaiton Journal of Korean Society of Transportation Journal of Korean Society of Transportation
Archives
: 2015 Vol.34~
  • scopus
  • NRF
  • KOFST
  • crossref cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close