First-generation antipsychotics (FGAs) and second-generation antipsychotics (SGAs) both treat conditions like schizophrenia and bipolar disorder, but they differ in how they work in the brain and, importantly, in the side effects they tend to cause. FGAs, developed in the 1950s, primarily block dopamine receptors. SGAs, which arrived in the 1990s, block dopamine receptors too but also act on serotonin receptors. That dual action is the core distinction, and it shapes everything from movement-related side effects to metabolic risks.
How Each Type Works in the Brain
All antipsychotics reduce the activity of dopamine, a chemical messenger involved in psychosis when it becomes overactive in certain brain pathways. First-generation drugs like haloperidol and chlorpromazine work almost entirely by blocking dopamine receptors. They’re effective at controlling hallucinations and delusions, but because dopamine also controls movement, blocking it broadly creates a high risk of neurological side effects.
Second-generation drugs like olanzapine, risperidone, quetiapine, and aripiprazole also block dopamine receptors, but they add a second layer: they block serotonin receptors as well. This serotonin-blocking effect is thought to partially offset the dopamine blockade in movement-related brain areas, which is why SGAs generally cause fewer movement problems. Research comparing these drug classes has confirmed that both the parent compounds and the breakdown products of SGAs maintain this dual activity, while first-generation drugs and their breakdown products do not.
Common First-Generation Drugs
There are 11 FDA-approved first-generation antipsychotics available in the U.S. The most frequently prescribed are chlorpromazine, perphenazine, and haloperidol. These range from low to high potency in that order. Higher-potency FGAs like haloperidol are more likely to cause movement problems, while lower-potency options like chlorpromazine tend to cause more sedation and blood pressure drops. Despite being older, these drugs remain in clinical use and are significantly cheaper than most SGAs.
Common Second-Generation Drugs
Nine SGAs are FDA-approved in the U.S., with quetiapine, risperidone, aripiprazole, and olanzapine being the most commonly prescribed. Each has a distinct side-effect profile. Olanzapine, for instance, tends to be effective but carries the highest metabolic risk. Aripiprazole is sometimes classified as a “third-generation” antipsychotic because it works differently from other SGAs. Rather than simply blocking dopamine receptors, it acts as a partial regulator, dialing dopamine activity up or down depending on what’s happening in a given brain circuit. This makes it less likely to cause both the movement side effects of FGAs and the weight gain associated with other SGAs.
Movement-Related Side Effects
The most well-known difference between the two generations involves movement problems, collectively called extrapyramidal symptoms. These include muscle stiffness, tremors, restlessness, involuntary movements, and sustained muscle contractions. About 50% of patients treated with high-potency FGAs like haloperidol develop these symptoms within the first several days. Roughly 55% of patients on haloperidol develop a Parkinson’s-like stiffness and slowness, compared with about 26% on olanzapine.
Tardive dyskinesia, a condition involving involuntary repetitive movements (often of the face, tongue, or jaw), is a particular concern with long-term use. The average rate of tardive dyskinesia among patients on FGAs falls between 24% and 30%. Earlier studies suggested that SGAs carried only one-quarter the risk, but more recent data indicates the risk with SGAs is actually more than half that of FGAs. One large trial found tardive dyskinesia rates ranging from 13% to 17% across both drug classes, with no statistically significant difference between individual medications. So while SGAs are safer on this front, they are not free of movement-related risks.
Metabolic Side Effects
Where SGAs tend to cause fewer movement problems, they introduce a different concern: metabolic changes. Weight gain, elevated blood sugar, and abnormal cholesterol levels are all more common with second-generation drugs, particularly olanzapine and quetiapine.
Olanzapine carries the steepest metabolic cost. In clinical studies, patients on olanzapine gained an average of 3.24 kg (about 7 pounds) more than those on placebo, and they were over nine times more likely to gain 7% or more of their body weight. Quetiapine also increases weight, with treated patients about four times more likely to gain weight than those on placebo. Risperidone falls in between, roughly doubling the odds of weight gain.
Beyond the scale, these drugs can shift blood lipids in unhealthy directions. Olanzapine raises triglycerides. Quetiapine raises both triglycerides and total cholesterol while lowering protective HDL cholesterol. Importantly, these lipid and blood sugar changes can occur even without weight gain, meaning metabolic monitoring matters for anyone on an SGA regardless of whether their weight changes.
Which Treats Symptoms Better
Both generations are effective at reducing the “positive” symptoms of schizophrenia, meaning hallucinations, delusions, and disorganized thinking. For decades, the hope was that SGAs would also be significantly better at treating “negative” symptoms like social withdrawal, flat emotions, and lack of motivation, as well as cognitive difficulties like problems with memory and planning. The reality is more nuanced.
Clinical guidelines do recommend SGAs over FGAs for cognitive dysfunction in schizophrenia, based on evidence showing modest improvements. SGAs are also associated with fewer relapses and fewer rehospitalizations during long-term maintenance treatment. But the landmark CATIE trial, one of the largest independent comparisons ever conducted, found something surprising: perphenazine, a first-generation drug, performed comparably in overall effectiveness to several SGAs including quetiapine, risperidone, and ziprasidone. Olanzapine had the longest time before patients stopped taking it, but its metabolic side effects were a significant tradeoff.
The CATIE trial also highlighted a broader problem: 74% of all patients discontinued their medication before 18 months, regardless of which drug they were on. Discontinuation rates ranged from 64% for olanzapine to 82% for quetiapine. This underscores that tolerability, not just efficacy, drives real-world outcomes.
Cost and Accessibility
First-generation antipsychotics are generally cheaper, and most have been available as generics for decades. A cost-effectiveness analysis found that conventional antipsychotics had lower costs and were associated with higher quality-adjusted life years compared to atypical antipsychotics, with more than a 50% likelihood of being cost-effective. That said, many SGAs are now available as generics too, which has narrowed the price gap considerably. The best choice often comes down to which side-effect profile is more manageable for a given person rather than cost alone.
How Prescribing Decisions Are Made
Current treatment guidelines generally favor SGAs as the starting point, largely because of lower rates of tardive dyskinesia and movement problems during long-term use. Japanese schizophrenia guidelines from 2022 recommend SGAs over FGAs for maintenance treatment, citing fewer relapses, fewer rehospitalizations, and less tardive dyskinesia. But guidelines also acknowledge that there is no significant difference in overall treatment discontinuation between the two classes, meaning neither generation clearly keeps patients on treatment longer.
In practice, the choice between generations often hinges on individual risk factors. Someone with diabetes or obesity might be steered away from olanzapine or quetiapine. Someone especially concerned about movement side effects might avoid high-potency FGAs. And for patients who haven’t responded to multiple medications, clozapine, a second-generation drug, remains the only antipsychotic with clear evidence of superior efficacy in treatment-resistant cases, though it requires regular blood monitoring due to a rare but serious effect on white blood cells.
The bottom line is that neither generation is universally better. FGAs and SGAs are roughly comparable in their ability to control psychotic symptoms, but they trade one set of risks for another. The “best” antipsychotic is the one a person can tolerate well enough to keep taking.