On this page
Study design
The retrospective analysis began with 91 people with Parkinson’s disease who were cognitively normal; 45 developed impairment during follow-up. Researchers trained a TabPFN model on routine clinical and laboratory data and tested it in the independent PPMI cohort.
What the study found
The model’s area under the curve was 0.822 in internal validation and 0.742 in external validation. Age and baseline cognitive score were influential predictors; triglyceride measures also contributed. Separate analyses found differing triglyceride trajectories between participants who developed impairment and those who remained stable.
Limits and interpretation
Prediction performance does not show that triglycerides cause cognitive decline. The internal sample was small, and retrospective cohort results can reflect selection and measurement differences. The paper calls for prospective multicenter validation before clinical implementation.
This sourced news summary and translation were prepared with AI assistance. Source links, the news date and research limitations are disclosed; this is not a claim of independent medical expert review. News production method.
Sources
Research news is not medical advice or a recommendation to use a substance. Read findings in the context of the study design, participants and limitations. Editorial standards.
← All news