Predicting Obesity in Adults Using Machine Learning Techniques: An Analysis of Indonesian Basic Health Research 2018

Abstract

Obesity is strongly associated with multiple risk factors. It is significantly contributing to an increased risk of chronic disease morbidity and mortality worldwide. There are various challenges to better understand the association between risk factors and the occurrence of obesity. The traditional regression approach limits analysis to a small number of predictors and imposes assumptions of independence and linearity. Machine Learning (ML) methods are an alternative that provide information with a unique approach to the application stage of data analysis on obesity. This study aims to assess the ability of ML methods, namely Logistic Regression, Classification and Regression Trees (CART), and Naïve Bayes to identify the presence of obesity using publicly available health data, using a novel approach with sophisticated ML methods to predict obesity as an attempt to go beyond traditional prediction models, and to compare the performance of three different methods. Meanwhile, the main objective of this study is to establish a set of risk factors for obesity in adults among the available study variables. Furthermore, we address data imbalance using Synthetic Minority Oversampling Technique (SMOTE) to predict obesity status based on risk factors available in the dataset. This study indicates that the Logistic Regression method shows the highest performance. Nevertheless, kappa coefficients show only moderate concordance between predicted and measured obesity. Location, marital status, age groups, education, sweet drinks, fatty/oily foods, grilled foods, preserved foods, seasoning powders, soft/carbonated drinks, alcoholic drinks, mental emotional disorders, diagnosed hypertension, physical activity, smoking, and fruit and vegetables consumptions are significant in predicting obesity status in adults. Identifying these risk factors could inform health authorities in designing or modifying existing policies for better controlling chronic diseases especially in relation to risk factors associated with obesity. Moreover, applying ML methods on publicly available health data, such as Indonesian Basic Health Research (RISKESDAS) is a promising strategy to fill the gap for a more robust understanding of the associations of multiple risk factors in predicting health outcomes.

Obesity Risk-Factor Variation Based on Island Clusters: A Secondary Analysis of Indonesian Basic Health Research 2018

“…A recent systematic review reported various machine learning techniques that were performed to predict adult obesity from nationwide and large cross-sectional data, finding that logistic regression analysis had the highest accuracy in predicting obesity [ 5 , 6 ]. This finding is in line with our previous study [ 7 ], which found that logistic regression had the highest performance in predicting and measuring obesity. Predicting obesity risk factors by considering determinant variables can be advantageous to design and modify local existing nutrition programs and policies better for controlling the obesity problem.…”

“…In order to calculate the adjusted odds ratios (ORs), multivariate logistic regression analyses, which includes other variables associated with obesity, were performed. The selection of multivariate logistic regression to develop a predictive model was based on our prior study that showed a high-performance, including accuracy, specificity, precision, Kappa, and [ 7 ]. Multivariate logistic regression was performed using SPSS version 27 (IBM Corp, Armonk, NY, USA).…”

Obesity Risk Factor Variation Based on Islands Cluster: A Secondary Analysis of Indonesian Basic Health Research 2018

“…A recent systematic review reported various machine learning techniques that had been performed to predict adult obesity from nation-wide and large cross-sectional data and found that logistic regression analysis had the highest accuracy in predicting obesity [5,6]. In line with our previous study [7], which found that logistic regression has the highest performance in predicting and measuring obesity. Predicting obesity risk factors by considering determinant variations can be advantageous to better design and modify local existing nutrition programs and policies for controlling the obesity problem.…”

Using Machine Learning to Predict Obesity Based on Genome-Wide and Epigenome-Wide Gene–Gene and Gene–Diet Interactions

“…This aligns with the concept of personalized nutrition. The same level of dietary Comparing our performance of predicting overweight and obesity with previous research, our ROC value ∼0.70 is not greater than that of previous research (Mukhopadhyay et al, 2015;Montanez et al, 2017;Ferdowsy et al, 2021;Thamrin et al, 2021). We wish to emphasize that we did not include any anthropometric and clinical phenotypes as predictive features and simply used genome-wide genotype and DNA methylation data in combination with dietary and a few other lifestyle factors without any pre-selection based on prior knowledge.…”