Physical Activity and Sedentary Time Top Factors for Gout Risk in People With Hyperuricemia

Physical activity (PA) and lesser sedentary time decreased the risk of gout in people with hyperuricemia (HUA), according to a new machine learning prediction model.1

“In this study, we utilized classic machine learning algorithms, integrating the interpretable SHAP (SHapley Additive exPlanations) algorithm, to explore how sedentary and PA duration, in interaction with individual characteristics and laboratory indices, affect gout risk in individuals with HUA, a high-risk indicator. These algorithms have demonstrated exceptional ability to capture complex nonlinear relationships within data and have been effectively applied in various contexts, such as modeling interactions among laboratory features and evaluating their impact on mortality prediction2,” lead investigator Yanliang Jiao, The Third Affiliated Hospital of Anhui Medical University, Hefei, China, and colleagues wrote. “This approach can help identify appropriate PA patterns to specific individual characteristics, providing better guidance on lifestyle behaviors for gout-prone individuals with HUA, and holds significant clinical relevance.”

Jiao and colleagues analyzed data from 8057 individuals with HUA from the National Health and Nutrition Examination Survey (NHANES) consortium for the period 2007–2018. They developed and compared 4 classic machine learning algorithms and selected the best-performing Random Forest (RF) model, combined it with the SHAP interpreting algorithm, and analyzed the dose–response relationship between PA duration, sedentary time, and gout risk. They also used the model to identify the most important factors influencing gout risk and developed a free online tool to help predict gout risk in people with HUA.

The RF model the investigators selected achieved a Receiver Operating Characteristic (ROC) of 0.957 in the training cohort and 0.799 in the testing cohort. In the test cohort, it demonstrated an accuracy of 0.778, a Kappa of 0.247, a sensitivity of 0.701, a specificity of 0.785, a positive predictive value of 0.224, a negative predictive value of 0.967, and an F1 score of 0.340.

Using SHAP analysis, Jiao and colleagues identified hypertension, serum uric acid, age, gender, and BMI as the top 5 factors for gout risk. They found that higher serum uric acid levels, age, BMI, creatinine, sedentary duration, lower PA, hypertension, male sex, and diabetes were associated with an elevated risk of gout. Importantly, they found that, regardless of age, sex, or comorbidities, 1 to 7 hours of PA per week was linked to a lower risk of gout, and over 6 hours a day of sedentary time increased gout risk.

“Future research could conduct a longitudinal follow-up study on the PA status of individuals with HUA to minimize selection bias and recall bias as much as possible. Such a design should thoroughly investigate the effects of specific activity types and lipid profiles, with the goal of providing more precise lifestyle recommendations. Additionally, future foundational research and Genome-Wide Association Study (GWAS) could explore the mechanisms underlying the impact of PA on metabolic syndrome and insulin resistance in relation to gout in individuals with HUA,” Jiao and colleagues concluded.1

REFERENCES

1. Jiao Y, Cheng Z, Lan Z, Kan S, Du Y. Exploring the Impact of PA and Sedentary Behavior on Gout Risk in Hyperuricemia: Insights From Machine Learning and SHAP Analysis. Int. J. Rheum. Dis. Published online April 23, 2025. doi.org/10.1111/1756-185X.70238

2. W. Qiu, H. Chen, A. B. Dincer, S. Lundberg, M. Kaeberlein, and S. I. Lee, “Interpretable Machine Learning Prediction of All-Cause Mortality,” Comm. Med. 2 (2022): 125. doi: 10.1038/s43856-022-00180-x.