Late-Onset Schizophrenia in Women: What Is Known

Creator: Ryan S. Sultan, MD
Published: 2026-11-12

By Ryan S. Sultan, MD
Psychiatrist · Columbia University Irving Medical Center
Published November 12, 2026 · 6-minute read

Schizophrenia can start after 40. An international consensus published in 2000 defined late-onset schizophrenia as illness onset after 40 years of age, and very-late-onset schizophrenia-like psychosis as onset after 60.¹ In a pooled analysis of 133,693 incident cases, the relative risk was greater in men up to age 39 and greater in women across ages 50 to 70.²

The same analysis found no evidence for a second incidence peak in middle-aged women.² New psychosis after 40 warrants a prompt medical evaluation.

How common onset after 40 is

The figures in circulation are secondary. A Cochrane review states as background that onset after 40 is reported in 23% of patients hospitalised with schizophrenia.³ A 2014 review gives "at least 20%" of patients, and a 2013 paper gives approximately 20% of middle-aged and older adults with schizophrenia.^{4, 5} These statements have different denominators, and a search on November 12, 2026 found no primary epidemiological source for a single figure.

The sex ratio shifts toward women with age

In one catchment-area sample of nonaffective functional psychoses, the male-to-female ratio was 1.56:1 at ages 16 to 25, reached unity around age 30, and fell to 0.38:1 at ages 66 to 75. That study described a slight second peak in rates at ages 46 to 55 and a third, more emphatic peak after 65.⁶

The pooled analysis of 43 samples found no evidence for a second incidence peak in middle-aged women, and it reported that incidence declined up to age 60 more strongly in men than in women. Overall, men had a 1.15-fold greater risk (95% CI, 1.00 to 1.31).² For onset after 65, a meta-analysis found some evidence of higher schizophrenia rates in women (odds ratio 1.6; 95% CI, 1.0 to 2.5).⁷

How late-onset illness differs

A study of 854 outpatients with schizophrenia over age 40 compared 110 with late onset and 744 with early onset. The late-onset group included more women and married individuals, had less severe positive symptoms, and took lower antipsychotic doses. Negative and depressive symptoms were similar in the two groups.⁸

Estrogen and schizophrenia: what the trials tested

One randomized estradiol trial enrolled 183 women aged 18 to 45 with schizophrenia or schizoaffective disorder. Both doses of transdermal estradiol produced greater decreases in positive, general and total symptom scores than placebo, and the largest effect size was 0.44, on positive symptoms at the higher dose.⁹ The trial included no women older than 45, so it can't speak to illness that begins after menopause.

Raloxifene has been tested as an add-on to antipsychotics, with mixed results:

In 56 women aged 40 to 70, raloxifene produced a greater reduction in total symptom score than placebo, with no significant effect on positive symptoms.¹⁰
In 200 severely ill, decompensated postmenopausal women, results favored placebo by 4.5 points.¹¹
A meta-analysis of nine studies and 561 patients, men included, found raloxifene superior to placebo on total symptom severity.¹²
A later trial of 102 patients, 28% of them female, found no main effect of raloxifene, and its authors reported female-specific benefits on negative symptoms and working memory in secondary analyses.¹³

In a Finnish cohort of 3,488 women aged 40 to 62 with schizophrenia-spectrum disorders, use of menopausal hormone therapy was associated with a 16% lower relapse risk compared with non-use (adjusted hazard ratio 0.84; 95% CI, 0.78 to 0.90). The association was present at ages 40 to 49 and 50 to 55 and absent at ages 56 to 62.¹⁴ This is register data from women with established illness. Hormone therapy is a decision for the patient and the prescribing doctor.

Antipsychotic response after menopause

In national register data, women were consistently hospitalized for psychosis more often than their male peers starting at age 45 to 50, and antipsychotic effectiveness declined in women after 45.¹⁵ Time since menopause was negatively associated with antipsychotic response and explained almost 42% of the variance in the model.¹⁶

Both studies concern women with established schizophrenia, and both report associations. A 2026 analysis of individual participant data from acute trials found no evidence of sex differences in drug effects on overall symptoms, though it didn't examine menopause.¹⁷ A change in symptoms around menopause is a reason to see the prescriber, who decides whether the dose should change.

Treatment evidence specific to late onset is thin

A Cochrane review found one acceptable trial, an eight-week comparison of risperidone and olanzapine in 44 inpatients, and concluded: "There is no trial-based evidence upon which to base guidelines for the treatment of late-onset schizophrenia."³

The ATLAS trial recruited 101 participants with onset at 60 or later. Their mean age was 80.2 years, and most were female. Improvement on the Brief Psychiatric Rating Scale at 12 weeks was 7.7 points greater with low-dose amisulpride than with placebo. Serious extrapyramidal side effects occurred in three patients on amisulpride and none on placebo.¹⁸

A review of treatment in older adults states that antipsychotics should generally be used in lower doses, a judgment the prescriber makes for each patient.⁵ I've written separately about schizophrenia and psychosis treatment.

The boxed warning concerns dementia-related psychosis

The Risperdal (risperidone) label carries this boxed warning: "Elderly patients with dementia-related psychosis treated with antipsychotic drugs are at an increased risk of death. RISPERDAL is not approved for the treatment of patients with dementia-related psychosis." Section 5.1 of the label cites analyses of 17 placebo-controlled trials in which the risk of death in drug-treated patients was between 1.6 and 1.7 times that in placebo-treated patients.¹⁹

The warning is about dementia-related psychosis, and none of the sources gathered on November 12, 2026 measured antipsychotic mortality risk in late-onset schizophrenia without dementia.

In a Swedish cohort, 15,409 people with very-late-onset schizophrenia-like psychosis had a higher rate of dementia than 154,090 matched people without it (hazard ratio 4.22; 95% CI, 4.05 to 4.41). The association remained after a two-year washout for possible misdiagnosis (hazard ratio 2.22).²⁰ A second national cohort reported dementia in 19.5% of the very-late-onset group and 6.4% of the comparison group.²¹ These are associations in people with onset after 60, and the Swedish authors suggest that psychosis may be an early prodromal change for some individuals.²⁰

Limitations of the evidence

This evidence has limitations. First, the age definitions come from a 2000 consensus paper, and the proportion figures are secondary. Second, the estradiol trial enrolled women aged 18 to 45, and the menopause findings come from women with established illness. Nevertheless, the pooled analysis found a greater relative risk in women across ages 50 to 70.

A first episode of psychosis after 40 calls for a medical evaluation, a careful diagnosis, and ongoing symptom monitoring.

Frequently asked questions

Can schizophrenia start at 40 or 50?

Yes. A 2000 international consensus defined late-onset schizophrenia as illness onset after 40 years of age and very-late-onset schizophrenia-like psychosis as onset after 60. New psychosis after 40 warrants a prompt medical evaluation.¹

Is late-onset schizophrenia more common in women?

In a pooled analysis of 133,693 incident cases, the relative risk was greater in men up to age 39 and greater in women across ages 50 to 70. In one catchment-area sample of nonaffective functional psychoses, the male-to-female ratio fell from 1.56:1 at ages 16 to 25 to 0.38:1 at ages 66 to 75.^{2, 6}

Can menopause trigger schizophrenia?

The pooled analysis of 133,693 cases found no evidence for a second incidence peak in middle-aged women, although relative risk was greater in women across ages 50 to 70. One single-sample study described a slight second peak in rates at ages 46 to 55.^{2, 6}

Does estrogen help schizophrenia symptoms?

In a trial of 183 women aged 18 to 45, added transdermal estradiol produced greater symptom decreases than placebo, with a largest effect size of 0.44. The trial included no women older than 45, and raloxifene trials are mixed, including one in 200 postmenopausal women that favored placebo by 4.5 points. Hormone therapy is a decision for the patient and the prescribing doctor.^{9, 11}

Does late-onset schizophrenia turn into dementia?

In a Swedish cohort, very-late-onset schizophrenia-like psychosis (onset after 60) was associated with a higher rate of dementia (hazard ratio 4.22), and the association remained after a two-year washout (hazard ratio 2.22). This is an association.²⁰

In a crisis: call or text 988, or go to the nearest emergency department. A first episode of psychosis belongs in a hospital-based or coordinated specialty care program.

I start and manage clozapine in an outpatient practice in Chelsea, Manhattan. Details are on the schizophrenia care in NYC page. Clozapine should not be stopped, skipped, or restarted without the prescriber. This page is general information and does not replace an evaluation.