Table 2.
Global assessment through adapted Bradford Hill Criteria
| Criteria | Description | Source/method |
|---|---|---|
| Strength of the association | Although ICs are not measures of risk, the strength of the disproportionality both in primary (vs. all other drugs) and false negative analysis suggests a negative signal | Disproportionality analysis |
| Analogy | The irrelevance was also demonstrated for other anti-diabetic/anti-obesity drugs (empagliflozin and Orlistat), which were used as a negative control in this study | Disproportionality analysis and labels |
| Biological plausibility/empirical evidence | GLP-1RAs are also proposed to have pro-cognitive effects. Particularly in terms of dual therapeutic mechanisms potentially improving both central nervous system deficits and metabolic burden [36]. There is no evidence to support that GLP-1RA will cause SSIBs | Disproportionality analysis and literature |
| Consistency | Results of disproportionality approaches were consistent in false negative | Disproportionality analysis |
| Coherence | A randomized, controlled trial reports three adolescent cases associated with suicidal ideation/behavior using liraglutide. However, the participant who committed suicide, who was in the liraglutide group, had a history of attention deficit–hyperactivity disorder and there was one suicide attempt in the liraglutide and placebo groups, respectively [37]. An exploratory pooled analysis reported 34/3291 suicidal ideation with liraglutide. But no between-treatment imbalances in suicidal ideation/behavior or depression were noted through prospective questionnaire assessments [38]. A case report describes two instances of depression associated with semaglutide [39]. | literature search |
| Specificity | The results of primary and false-negative analysis showed no association between GLP-1RAs and SSIBs. The co-medication analysis indicated that the occurrence of SSIBs was more likely to be related to the patient's own mental state. | Disproportionality and co-medication analysis |
| Temporal relationship | Available data suggested that there were cases of SSIBs after discontinuation of GLP-1RA, but further studies could not be performed due to missing data | Time-to-onset analysis |
| Reversibility | This criterion is of limited value here as there is not enough data on rechallenge and de-challenge in the FAERS database | Descriptive |