An Automated Malnutrition Screening Tool Using Routinely Collected Data for Older Adults in Long-Term Care:...

Area of Science:

• Gerontology
• Nutritional Science
• Health Informatics

Background:

• Malnutrition is prevalent in long-term care settings, impacting resident health and increasing healthcare costs.
• Accurate and efficient screening tools are crucial for timely intervention.
• Routinely collected data offers a valuable resource for developing such tools.

Purpose of the Study:

• To develop and internally validate an automated malnutrition screening tool (AutoMal) using routinely collected data.
• To assess the performance of AutoMal in identifying malnutrition in long-term care residents.

Main Methods:

• A diagnostic prediction model was developed and internally validated using data from 539 residents across 10 Australian long-term care facilities.
• Logistic regression and bootstrapping techniques were employed to select candidate variables and develop the final model.
• The Subjective Global Assessment served as the reference standard for malnutrition diagnosis.

Main Results:

• The AutoMal model incorporated body mass index and 6-month weight change, identified in 100% of bootstrapped samples.
• AutoMal demonstrated excellent discrimination, with an area under the curve (AUC) of 0.8378.
• The tool achieved a sensitivity of 78% and specificity of 77% at the optimal threshold.

Conclusions:

• The AutoMal tool effectively identifies malnutrition in long-term care settings.
• Its reliance on common electronic health record variables enables automated and efficient malnutrition screening.
• AutoMal has the potential to improve early detection and management of malnutrition in this vulnerable population.