REVIEW OF: Parabiaghi A et al, Pharmacopsychiatry 2025;58(4):159–169
STUDY TYPE: Population-based pharmacoepidemiologic cohort study

Antipsychotic adherence is a major challenge; nearly 40% of patients who begin them stop within three months. This large, real-world Italian study tracked over 42,000 adults who started an oral antipsychotic between 2001 and 2021. Researchers used pharmacy records to compare “persistence” (time to discontinuation) across and between first- and second-generation antipsychotics (FGAs/SGAs).

Patients were followed for 3 months primarily, with secondary analyses that extended up to 20 years. Persistence was adjusted for age, sex, and polypharmacy.

Results

• SGAs overall had better persistence than FGAs. The hazard ratio (HR) for discontinuation was 0.76 (95% CI: 0.73–0.79), meaning patients prescribed SGAs had a 76% likelihood of discontinuing compared to those prescribed FGAs.
• Of antipsychotics available in the US, clozapine had the lowest absolute number of prescriptions (amisulpride was the lowest overall).
• However, clozapine had the longest persistence (median 175 days vs 92–133 for other meds). It significantly outperformed all other agents, including olanzapine (HR: 0.85; 95% CI: 0.79–0.91) and risperidone (HR: 0.80; 95% CI: 0.74–0.87).
• Olanzapine and aripiprazole had longer persistence than risperidone and quetiapine.
• Quetiapine had the poorest persistence but was the second most prescribed.
• By one year, 80% of patients had discontinued their initial agent. Only 1% were re-prescribed the same drug.

CARLAT TAKE
This real-world study reported prescription data only, not diagnoses or clinical outcomes, and didn’t verify whether patients took their meds. Those are significant limitations. The researchers elucidated an interesting (but unsurprising) paradox: The most persistent medication, clozapine, was one of the least prescribed. This study is further evidence, if we needed it, that we should be using clozapine more often than we do.