Sage Journals: Discover world-class research

Abstract

Aims: Factors that contribute to the development of overweight are numerous and form a complex structure with many unknown interactions and associations. We aimed to explore this structure (i.e. the mutual importance or hierarchy of sociodemographic and lifestyle-related risk factors of being overweight) using a machine-learning technique called random forest (RF). The results were compared with traditional logistic regression (LR) analysis. Methods: The cross-sectional FINRISK 2007 Study included 4757 Finns (aged 25–74 years). Information on participants’ lifestyle and sociodemographic characteristics were collected with questionnaires. Diet was assessed, using a validated food-frequency questionnaire. Height and weight were measured. Participants with a body mass index (BMI) ≥25 kg/m² were classified as overweight. R-statistical software was used to run RF analysis (‘randomForest’) to derive estimates for variable importance and out-of-bag error, which were compared to a LR model. Results: In total, 704 (32%) men and 1119 (44%) women had normal BMI, whereas 1502 (69%) men and 1432 (57%) women had BMI ≥25. Estimated error rates for the models were similar (RF vs. LR: 42% vs. 40% for men, 38% vs. 35% for women). Both models ranked age, education and physical activity as the most important risk factors for being overweight, but RF ranked macronutrients (carbohydrates and protein) as more important compared to LR. Conclusions: RF did not demonstrate higher power in variable selection compared to LR in our study. The features of RF are more likely to appear beneficial in settings with a larger number of predictors.

Keywords

Machine learning mutual importance obesity random forest risk factor

Get full access to this article

View all access options for this article.

References

Lim

Vos

Flaxman

et al . A comparative risk assessment of burden of disease and injury attributable to 67 risk factors and risk factor clusters in 21 regions, 1990–2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 2012;380(9859):2224–60.

Finucane

Stevens

Cowan

et al . National, regional, and global trends in body-mass index since 1980: systematic analysis of health examination surveys and epidemiological studies with 960 country-years and 9.1 million participants. Lancet 2011;377(9765):557–67.

Fleming

Robinson

et al . Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2014;384(9945):766–81.

Malik

Willett

FB.

Global obesity: trends, risk factors and policy implications. Nat Rev Endocrinol 2013;9(1):13–27.

Crawford

Ball

Behavioural determinants of the obesity epidemic. Asia Pac J Clin Nutr 2002;11:S718–21.

Fuglestad

Jeffery

Sherwood

NE.

Lifestyle patterns associated with diet, physical activity, body mass index and amount of recent weight loss in a sample of successful weight losers. Int J Behav Nutr Phys Act 2012;9: 79.

Breiman

Random forests. Mach Learning 2001;45(1):5–32.

Lunetta

Hayward

Segal

et al . Screening large-scale association study data: exploiting interactions using random forests. BMC Genet 2004;5:32.

Vartiainen

Laatikainen

Peltonen

et al . Thirty-five-year trends in cardiovascular risk factors in Finland. Int J Epidemiol 2010;39(2):504–18.

10.

Konttinen

Haukkala

Sarlio-Lähteenkorva

et al . Eating styles, self-control and obesity indicators. The moderating role of obesity status and dieting history on restrained eating. Appetite 2009;53(1):131–4.

11.

Männistö

Virtanen

Mikkonen

et al . Reproducibility and validity of a food frequency questionnaire in a case-control study on breast cancer. J Clin Epidemiol 1996;49(4):401–9.

12.

Kaartinen

Tapanainen

Valsta

et al . Relative validity of a FFQ in measuring carbohydrate fractions, dietary glycaemic index and load: exploring the effects of subject characteristics. Br J Nutr 2012;107(9):1367–75.

13.

Paturi

Tapanainen

Reinivuo

et al . The National FINDIET 2007 Survey. Helsinki: Publications of the National Public Health Institute B23/2008, 2008.

14.

Reinivuo

Hirvonen

Ovaskainen

et al . Dietary survey methodology of FINDIET 2007 with a risk assessment perspective. Public Health Nutr 2010;13(6A):915–9.

15.

Tolonen

Koponen

Aromaa

et al . Recommendations for the health examination surveys in Europe. Helsinki: Publications of the National Public Health Institute B21/2008, 2008.

16.

World Health Organization. Obesity: preventing and managing the global epidemic. Geneva: WHO Technical Report Series 894, 2000.

17.

R Core Team. R: a language and environment for statistical computing. Vienna: R Foundation for Statistical Computing, http://www.R-project.org/ (2013, accessed 3 August 2017).

18.

Willett

Stampfer

MJ.

Total energy intake: implications for epidemiologic analyses. Am J Epidemiol 1986;124(1):17–27.

19.

Breiman

Friedman

Olshen

et al . Classification and regression trees. Belmont, CA: Wadsworth, 1984.

20.

Wennman

Kronholm

Partonen

et al . Physical activity and sleep profiles in Finnish men and women. BMC Public Health 2014;14:82.

21.

Wirt

Collins

CE.

Diet quality – what is it and does it matter?

Public Health Nutr 2009;12(12):2473–92.

22.

Toschke

Beyerlein

von Kries

Children at high risk for overweight: a classification and regression trees analysis approach. Obes Res 2005;13:1270–4.

23.

Roda

Charreire

Feuillet

et al . Lifestyle correlates of overweight in adults: a hierarchical approach (the SPOTLIGHT project). Int J Behav Nutr Phys Act 2016;13:114.

24.

Casanova

Saldana

Chew

et al . Application of random forests methods to diabetic retinopathy classification analyses. PLoS One 2014;9(6):e98587.

25.

Churpek

Yuen

Winslow

et al . multicenter comparison of machine learning methods and conventional regression for predicting clinical deterioration on the wards. Crit Care Med 2016;44(2):368–74.

26.

Chen

Ishwaran

Random forests for genomic data analysis. Genomics 2012;99(6):323–9.

27.

Kwon

Rhee

Yoon

Application of classification algorithms for analysis of road safety risk factor dependencies. Accid Anal Prev 2015;75:1–15.

28.

Stice

Palmrose

Burger

KS.

Elevated BMI and male sex are associated with greater underreporting of caloric intake as assessed by doubly labeled water. J Nutr 2015;145:2412–18.

29.

Paalanen

Männistö

Virtanen

et al . Validity of a food frequency questionnaire varied by age and body mass index. J Clin Epidemiol 2006;59:994–1001.

30.

Pietiläinen

Korkeila

Bogl

et al . Inaccuracies in food and physical activity diaries of obese subjects: complementary evidence from doubly labeled water and co-twin assessments. Int J Obes 2010;34:437–45.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.66 MB

0.04 MB

Suitability of random forest analysis for epidemiological research: Exploring sociodemographic and lifestyle-related risk factors of overweight in a cross-sectional design

Abstract

Keywords

Get full access to this article

References

Supplementary Material