Fig. 3|. Incident suicidal ideations in the study population with T2DM.

Comparison of incident (first-time experience) suicidal ideation In the study population with T2DM and no history of suicidal ideation before the index event of the first prescription of semaglutide or other non-GLP1R agonist anti-diabetes medications that occurred from 1 December 2017 through to 31 May 2021 between the propensity score-matched semaglutide and non-GLP1R agonist anti-diabetes medication groups within the 6-month time window after the index event. For each group, the overall risk (number of cases) is shown, where overall risk is defined as the number of patients with outcomes during the 6-month time window divided by number of patients in the group at the beginning of the time window. HRs were calculated using Kaplan–Meier analysis to estimate the probability of outcome at daily time intervals with censoring applied.