Development of an algorithm to predict serum vitamin D levels using a simple questionnaire based on sunlight exposure

Area of Science:

• Nutritional Science
• Public Health

Background:

• Sun exposure is the primary factor influencing vitamin D production.
• Accurate assessment of vitamin D status is crucial for public health.
• Current methods often rely on serum 25(OHD) measurements, which can be resource-intensive.

Purpose of the Study:

• To develop and validate an algorithm for estimating individual vitamin D status.
• To utilize a simple questionnaire focusing on sunlight exposure.
• To identify individuals who may require serum 25(OHD) testing.

Main Methods:

• A questionnaire assessing sunlight exposure, seasonality, age, sex, BMI, and lifestyle factors was administered to 620 adults in Southern Italy.
• A two-step statistical procedure involving linear regression and model validation was employed.
• Vitamin D status was categorized into four classes based on serum 25(OHD) concentrations.

Main Results:

• Seasonality, daily sunlight exposure, and beach holidays were significant predictors of vitamin D status.
• The developed algorithm achieved high accuracy, correctly classifying 90.2% of subjects in the validation set.
• The algorithm demonstrated strong predictive ability for vitamin D status.

Conclusions:

• A simple questionnaire-based algorithm can effectively estimate vitamin D status.
• This approach can aid in selecting individuals for serum 25(OHD) measurement.
• The findings support the use of sunlight exposure data for vitamin D assessment.