I used to work in psych, and the doctors were always saying the antipsychotic were needed immediately with first break psychosis to avoid the psychosis becoming permanent. How scientifically established is this notion, and it possible that the idea was considered "settled" long ago and testing outcomes with and without antipsychotic treatment is considered unethical, as it would dreprive the control group of a supposedly known effective treatment?

The short version: the claim that antipsychotics must start immediately to prevent psychosis from becoming permanent is far weaker than the doctors' certainty suggests, and the evidence base is built on studies that never ran the comparison that would settle it.

You are right to smell something off. The idea that first-episode psychosis hardens into something permanent without immediate drugs is treated as settled, but the studies behind it mostly compare one antipsychotic to another, or drug to placebo after the person is already in treatment. They never compare immediate drug treatment to careful monitoring without drugs in a way that would test whether delay itself causes permanence. The closest evidence actually cuts against the dogma: a 2004 study of 207 first-episode patients found no association between duration of untreated psychosis and any cognitive measure at baseline [7]. That does not prove delay is harmless, but it directly contradicts the "every week untreated burns brain tissue" story.

What the evidence does show is narrower. Early response to antipsychotics predicts later response: in 225 first-episode patients, improvement by week 2 of at least 26% predicted week 12 outcomes [1]. That is a prognostic fact, not proof that starting drugs earlier changes the long-term course. Relapse rates are real too: in remitted first-episode patients, about 4 in 5 relapsed within a year off medication versus about 2 in 5 on maintenance [2]. That is a genuine benefit of staying on drugs after remission. But it says nothing about whether starting them at first break versus waiting a few weeks changes whether psychosis becomes permanent.

The harms are documented and substantial. In drug-naive first-episode patients on haloperidol, about 78 in 100 developed extrapyramidal symptoms during hospitalization [4]. In adolescents starting antipsychotics, 8 of 15 had muscle weakness, 8 of 15 had extrapyramidal symptoms, and weight gain was significant [5]. Akathisia, the restless agony that drives people to stop treatment, occurs in 15-25% even with aripiprazole [6]. Negative symptoms, the ones that destroy function, barely respond to antipsychotics at all [3]. The drugs trade one set of symptoms for another, and the trade is not always favorable.

The ethical argument you heard is real but circular. Withholding a drug from a control group is considered unethical only because effectiveness is already assumed. But the assumption was built on trials that never tested the question of timing. The most informative design, comparing immediate antipsychotics to a monitored no-drug arm in first-episode psychosis, has essentially never been run. That is not because it is unethical. It is because the field decided the answer before asking.

My call: the evidence supports antipsychotics for acute symptom control and for preventing relapse after remission, but it does not support the claim that immediate treatment prevents psychosis from becoming permanent. The burden of proof for that claim has never been met. Confidence: moderate, because the absence of the decisive trial means the question remains genuinely open.