Nomogram for predicting nutritional risk of cognitive impairment

Area of Science:

• Geriatric Medicine
• Nutritional Science
• Clinical Prediction Modeling

Background:

• Cognitive impairment patients are highly susceptible to malnutrition, exacerbating cognitive decline.
• Subjective nutrition screening is unreliable in patients with cognitive impairment.
• Objective indicators are essential for accurate nutritional risk assessment in hospitalized patients with cognitive impairment.

Purpose of the Study:

• To develop a predictive nomogram for nutritional risk in patients with cognitive impairment.
• To identify objective indicators for assessing nutritional status in this population.

Main Methods:

• Least Absolute Shrinkage and Selection Operator (LASSO) regression for variable selection.
• Multivariable logistic regression to build the final prediction model.
• Internal validation using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis.

Main Results:

• Six predictive factors were identified: body mass index, age, triglyceride, cognitive-improving drug use, controlling nutritional status, and geriatric nutritional risk index.
• The nomogram achieved an area under the ROC curve of 0.91 in the training cohort and 0.88 in the validation cohort.
• Decision curve analysis indicated a favorable net benefit within a threshold range of 0.00-0.80.

Conclusions:

• A risk nomogram incorporating six objective factors can effectively predict nutritional risk in cognitive impairment patients.
• This tool aids in the objective evaluation of nutritional status for hospitalized individuals with cognitive impairment.