The Association of Waist Circumference with the Prevalence and Survival of Digestive Tract Cancer in US Adults: A Population Study Based on Machine Learning Methods

Abstract

Aims: This paper aims to investigate the relationship of waist circumference (WC) with digestive tract cancer morbidity and mortality.

Methods: Based on the data from a nationally representative US population survey, we summarized the prevalence of digestive tract cancer and all-cause mortality of cancer patients across WC quartiles. Adjusted logistic regression and restricted spline curve were used to analyze WC and the prevalence of digestive tract cancer. Moreover, Cox regression and the Kaplan-Meier curve were applied to investigate the association of WC with all-cause mortality. We also attempted to make a model to predict cancer happening.

Results: This paper included a total of 34,041 participants, with digestive tract cancer observed in 265 (0.7%) individuals. WC was positively associated with digestive tract cancer morbidity after full adjustment of covariates (OR: 1.72 and 95% CI: 1.41-2.10). Also, individuals in the highest WC group had a higher risk of digestive tract cancer (Q4, OR: 2.71 and 95% CI: 1.48-5.00). Moreover, no significant association was observed in upper digestive cancer, and WC was associated with a longer survival time once diagnosed (hazard ratio (HR): 0.50 and 95% CI: 0.28-0.92). Finally, the model we made proved to be effective.

Conclusion: High WC is a risk factor for digestive tract cancer with or without adjusting for body mass index, especially those located in the lower digestive tract. However, once digestive tract cancer has been diagnosed, patients with higher WC showed better survival outcomes. Moreover, machine learning methods can be used to predict digestive tract cancer risk in the future.

Figure 1

Flow chart of selection of eligible participants. NHANES: National Health and Nutrition Examination Survey.

Figure 2

(a) Restricted cubic spline plots of the association between waist circumference and digestive tract cancer. The association was adjusted for age (<80 and ≥80), gender, smoking, drinking, HbA1c, and BMI. The median of the waist circumference was set as the reference for this figure. (b) Restricted cubic spline plots of the association between waist circumference and lower digestive tract cancer. (c) Restricted cubic spline plots of the association between waist circumference and upper digestive tract cancer. BMI: body mass index; CI: Confidence interval.

Figure 3

Sensitivity analysis on the association between waist circumference and digestive tract cancer risk based on logistic regression analysis. The association was adjusted for BMI, age, gender, smoking, drinking, and HbA1c. CI: confidence interval; BMI: body mass index.

Figure 4

A Kaplan-Meier curve of the association between waist circumference and all-cause mortality of digestive tract cancer in the following 60 months (5 years). The waist circumference was divided by median into two groups (≤97 cm and >97 cm), and the survival comparison among groups was adjusted by the Bonferroni-Holm method. WC: waist circumference.

Figure 5

The receiver operating characteristic curve of the identification ability of the proposed nomogram. The nomogram showed relatively strong identification ability with an area under the curve of 0.71 (95% confidence interval, 0.46-0.94), specificity of 0.708, sensitivity of 0.750, and accuracy of 0.972.