Weight Gain Cause Narratives from 2,463 Patients for Personalized Obesity Management

2,463 patients with overweight or obesity provided unstructured narratives about weight gain causes prior to a weight loss treatment. The narratives were automatically labeled into 12 thematic categories using a GPT4.1 large language model, achieving precision and recall of 0.906 and 0.897. The dataset, authored by Miksa M. Henkrich and last updated in April 2026, supports analysis of associations between reported causes, demographics, and treatment outcomes.

Use Cases

• Automated thematic labeling of patient narratives based on the 12 defined categories.
• Predicting weight loss treatment outcomes based on reported weight gain causes and demographic features.
• Identifying patient phenotypes via unsupervised clustering of narrative themes, age, sex, and BMI.
• Analyzing co-occurrence patterns and risk ratios between different weight gain causes.

Strengths

• Contains 2,463 patient narratives, providing a substantial corpus for analysis.
• Automated labeling achieved a precision of 0.906 and recall of 0.897 against a reference sample.
• Includes associated demographic and clinical outcome data for each patient.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown for the processed data, which may limit suitability assessment.
• The data is stored in a DOCX file format, which may require conversion for analysis.

Provenance

Source

figshare

Collection Method

Narratives collected from patients prior to starting a multidisciplinary medical-nutritional weight loss treatment.

Freshness

Last updated 2026-04-14 16:01:32