Lifestyle and chronic kidney disease: A machine learning modeling study

Abstract

Background: Individual lifestyle varies in the real world, and the comparative efficacy of lifestyles to preserve renal function remains indeterminate. We aimed to systematically compare the effects of lifestyles on chronic kidney disease (CKD) incidence, and establish a lifestyle scoring system for CKD risk identification.

Methods: Using the data of the UK Biobank cohort, we included 470,778 participants who were free of CKD at the baseline. We harnessed the light gradient boosting machine algorithm to rank the importance of 37 lifestyle factors (such as dietary patterns, physical activity (PA), sleep, psychological health, smoking, and alcohol) on the risk of CKD. The lifestyle score was calculated by a combination of machine learning and the Cox proportional-hazards model. A CKD event was defined as an estimated glomerular filtration rate <60 ml/min/1.73 m², mortality and hospitalization due to chronic renal failure, and self-reported chronic renal failure, initiated renal replacement therapy.

Results: During a median of the 11-year follow-up, 13,555 participants developed the CKD event. Bread, walking time, moderate activity, and vigorous activity ranked as the top four risk factors of CKD. A healthy lifestyle mainly consisted of whole grain bread, walking, moderate physical activity, oat cereal, and muesli, which have scored 12, 12, 10, 7, and 7, respectively. An unhealthy lifestyle mainly included white bread, tea >4 cups/day, biscuit cereal, low drink temperature, and processed meat, which have scored -12, -9, -7, -4, and -3, respectively. In restricted cubic spline regression analysis, a higher lifestyle score was associated with a lower risk of CKD event (p for linear relation < 0.001). Compared to participants with the lifestyle score < 0, participants scoring 0-20, 20-40, 40-60, and >60 exhibited 25, 42, 55, and 70% lower risk of CKD event, respectively. The C-statistic of the age-adjusted lifestyle score for predicting CKD events was 0.710 (0.703-0.718).

Conclusion: A lifestyle scoring system for CKD prevention was established. Based on the system, individuals could flexibly choose healthy lifestyles and avoid unhealthy lifestyles to prevent CKD.

Keywords: chronic kidney disease; cohort study; lifestyle; machine learning; scoring system.

Figure 1

Score of lifestyle factors. (A) Healthy and unhealthy lifestyles are categorized according to the hazard ratios (HRs) in Supplementary Figure 2. The mean decrease impurity (MDI)/1,000 was adopted as the lifestyle score for every factor. Moderate PA included walking upstairs, going the gym, jogging, energetic dancing aerobics, most sports, using heavy power tools, and other physically demanding DIY and gardening. Light DIY included pruning, watering the lawn; other exercises included swimming, cycling, keeping fit, and bowling; Heavy DIY included weeding, lawn mowing, carpentry, and digging. PA, physical activity; DIY, do-it-yourself; MET, Metabolic Equivalent Task. (B) The lifestyle score was categorized as <0, 0–20, 20–40, 40–60, corresponding to grade 0, grade 1, grade 2, grade 3 and grade 4 respectively.

Figure 2

Validation of the lifestyle score in long-term outcomes. Panel (A) shows a restricted cubic spline regression analysis, which indicates a linear relationship between the total lifestyle score (equals to the scores of healthy lifestyle factors minus the scores of unhealthy lifestyle factors) and risk of CKD events. Panel (B) shows the categorization for risk of CKD event according to the total lifestyle score. Panel (C) shows the receiver operator characteristic curves (ROC) of the age-adjusted lifestyle score. Panel (D) or (E) is a restricted cubic spline regression analysis, which indicates a linear relationship between the total lifestyle score (equals to the scores of healthy lifestyle factors minus the scores of unhealthy lifestyle factors) and the risk of CVD events or all-cause mortality.