https://www.nature.com/articles/s41386-018-0278-3
The current analysis included patients with a diagnosis of first-episode schizophrenia or schizoaffective disorder who met the following criteria: (1) referral to the First-Episode Psychosis Program between 2003 and 2013; (2) treatment with an oral second-generation antipsychotic according to a standardized treatment algorithm; (3) positive symptom remission; (4) subsequent relapse (i.e., second episode) in association with non-adherence; and (5) reintroduction of antipsychotic treatment with the same agent used to achieve response in the first episode.
there were significant episode-by-time interactions for all outcomes of antipsychotic treatment response over 1 year in favor of the first episode compared to the second episode (50% response rate: 48.7 vs. 10.4% at week 7; 88.2 vs. 27.8% at week 27, respectively).
TL;DR Patients who relapsed after going off meds after a first episode had lower rates of medication response to the medication they were previously on. Pretty significantly, with 88% responding in the first episode, but only 27% in the second. Adds to anecdotal evidence here that it’s risky to go off meds…