To test this, they enrolled patients with invasive breast cancer and took vitamin D measurements at three points: before treatment, during treatment, and after treatment, with an average follow-up of 38 months. To analyze the data, they used group-based trajectory modeling, a statistical method that identifies distinct subgroups of patients who share similar patterns of change over time, allowing them to move beyond static measurements and characterize how each patient’s vitamin D status evolved throughout treatment.