Skip to content

Survey study, 2023

Can diet and body measurements predict PMS?

Yes. A machine learning model using diet and body measurements could predict PMS with 80.3% accuracy. The strongest predictors were sodium intake, sugar intake, total fat intake, biotin, and fat distribution. Results need confirmation in other studies.

Early or limited evidence 223 Iranian women

What they did

Researchers studied 223 Iranian women. They measured body size and fat distribution and asked about diet using a food frequency questionnaire. They used machine learning to identify which dietary and body measurements could predict PMS.

What they found

The machine learning model correctly identified PMS 80.3% of the time and had an F1 score of 76.3%. The five strongest predictors were sodium intake, fat distribution around the upper hip, biotin intake, total fat intake, and total sugar intake.

What it means for you

Diet and body fat distribution appear strongly linked to PMS. However, this is a small observational study using a novel machine learning method. These findings need to be tested in other populations and larger studies before we know if they apply broadly.

223

women studied

80.3%

accuracy of machine learning model

76.3%

F1 score of model

About the study

Original title
Dietary intake of micronutrients are predictor of premenstrual syndrome, a machine learning method
Authors
Taheri R, ZareMehrjardi F, Heidarzadeh-Esfahani N et al.
Journal
Clinical nutrition ESPEN
Year
2023
Study design
Survey study
Participants
223
What was tested
None

Topics