Fatty Liver Index Association with COPD Prevalence and Severity in Two Cohorts

Yiben Huang published a dual-cohort study on figshare analyzing the association between the fatty liver index (FLI) and chronic obstructive pulmonary disease (COPD). The study uses data from the NHANES 2017–2020 survey (n=15,560) and a clinical cohort from Wenzhou Third Affiliated Hospital, China (n=296). Machine learning algorithms were used for feature selection, and logistic regression models evaluated associations.

Use Cases

• Training predictive models for COPD prevalence based on metabolic biomarkers like FLI.
• Analyzing non-linear relationships between biomarkers and disease severity using restricted cubic splines.
• Developing clinical nomograms to visualize the contribution of key predictive factors for COPD.
• Evaluating model accuracy for disease prediction using receiver operating characteristic curves and confusion matrices.

Strengths

• Utilizes two independent cohorts totaling over 15,800 participants.
• Machine learning-optimized models demonstrated predictive accuracy, with an AUC of 0.819 reported for the NHANES model.
• Identified a specific non-linear threshold relationship with an inflection point at FLI = 1.045.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count for the underlying data tables is unknown, which may limit suitability assessment.
• The dataset is stored as a DOCX file (8.9 MB), which may not be a standard format for structured data analysis.

Provenance

Source

National Health and Nutrition Examination Survey (NHANES) and Wenzhou Third Affiliated Hospital, China.

Collection Method

Data collected from a national survey and a hospital clinical cohort.

Time Range

NHANES data covers 2017–2020.

Freshness

Last updated 2026-05-05 14:32:21; freshness should be verified.

Geography

United States (NHANES) and China (Wenzhou).