Understanding and Managing Psychosis in Late Life

March 1, 2003

Although late-onset psychosis is not as common as the early-onset variety, it can still pose difficulties in diagnosing and treating patients. How are patients with late-onset psychosis different from those who have early-onset, and what sorts of issues should clinicians be aware of?

With the aging of the baby boomers, the number of people in the United States older than 65 is projected to double from about 35 million today to nearly 70 million by 2030. There will be a disproportionately greater increase in the number of elderly Americans who suffer from a mental illness--from approximately 6 million today to about 15 million by 2030 (Jeste et al., 1999a). Younger adults who have a serious mental illness, such as schizophrenia, currently have a significantly shorter life span than those without mental illness. The average life span of these patients is expected to increase, thanks to improved pharmacologic and other treatments, as well as general improvement in health and nutrition. In addition, as people in the general population live longer, the numbers of individuals who will develop psychotic disorders in later life will also grow.

One of the most disenfranchised groups in health care is older people with psychotic disorders. The health-related well-being of older patients with psychosis living in the community is comparable to (or even slightly worse than) that of outpatients with AIDS (Patterson et al., 1996).

Acute psychotic symptoms in older people may reflect delirium or metabolic causes. Chronic and persistent psychotic symptoms belong to one of two groups: primary psychotic disorders (such as schizophrenia, delusional disorder, psychotic mood disorder) or psychosis secondary to dementia or other general medical conditions. Two important chronic psychotic disorders in older people that will be discussed in this article are schizophrenia in late life and psychosis of Alzheimer's disease (AD). Table 1 highlights the main differences between these two conditions (Jeste and Finkel, 2000). (Due to copyright concerns, this table cannot be reproduced online. Please see Table 1 on p19 of the print edition--Ed.)


Schizophrenia affects about 1% of the population and is arguably the most expensive mental illness in adults. Furthermore, the cost of health care tends to be highest for the oldest of these patients (Cuffel et al., 1996). Older adults with schizophrenia often live alone, in assisted-care facilities, in homeless shelters or on the street. The proportions of those living with family or in long-term institutional settings are relatively small. These patients are usually unable to work or interact with society in a productive manner. Nonetheless, research in late-life schizophrenia shows that the century-old Kraepelinian concept of "dementia praecox"--i.e., a disorder with onset restricted to adolescence or early adulthood with progressive downhill course--is not quite accurate. Schizophrenia can manifest for the first time in middle age or later, and the course of schizophrenia in old age is typically different from that of dementia (Jeste et al., 1997).

Of the older people with schizophrenia, nearly 25% have late-onset schizophrenia (with onset of illness usually in middle age), while the remaining 75% have had schizophrenia since adolescence or early adulthood. Middle-age-onset patients are similar to early-onset patients in terms of positive symptoms (paranoid delusions and auditory hallucinations), family history of the illness, pattern of cognitive impairment (deficits in learning and abstraction but not in delayed recall), nonspecific brain-imaging abnormalities such as mild ventricular enlargement and white matter hyperintensities, chronicity of illness, and higher mortality from suicide and other causes (Jeste et al., 1997). However, late-onset schizophrenia is characterized by less severe negative symptoms such as social withdrawal and blunting of emotions, less severe deficits in learning, need for and tolerance to lower doses of antipsychotic medications, and a much higher proportion of women than men. One possible explanation for this gender difference is related to a loss of estrogen with menopause in older women. The theory is that estrogen has a protective effect against schizophrenia, and when that protection diminishes, women who are vulnerable to develop schizophrenia begin to manifest symptoms of the illness (Seeman, 1997).

The course of early-onset schizophrenia in older age is relatively stable and non-deteriorating. Most elderly people with schizophrenia are symptomatic and disabled but not demented and are living in the community. Negative symptoms persist, whereas positive symptoms tend to show a modest improvement with aging (Cohen et al., 2000).

Can schizophrenia ever remit? Most people are now familiar with the story of John Forbes Nash, Ph.D., the Nobel Prize-winning mathematician whose remission of schizophrenia around the age of 60 was portrayed in the Academy Award-winning film, A Beautiful Mind. Yet the concept of remission of schizophrenia continues to be controversial. One would not typically expect a person with schizophrenia in remission to display completely normal life functioning. The physical effects of aging, biopsychosocial stressors of old age, effects of having suffered from decades of severe psychiatric illness, side effects of treatments, social stigma and the typical history of suboptimal premorbid functioning interact to preclude such outcomes. It is worth stressing that schizophrenia is a neurodevelopmental disorder--i.e., people with schizophrenia usually have had somewhat impaired cognitive functioning since early childhood.

Nonetheless, patients in persistent clinical remission should be free of positive and negative symptoms, have reasonably age-appropriate everyday functioning, and be without psychiatric hospitalization for the previous five years or longer. Such remission may occur in up to 20% of aging patients with schizophrenia. Better premorbid functioning, female gender, later age of symptom onset, shorter duration of illness, paranoid subtype, early initiation of treatment and psychosocial support have been reported to be associated with a greater likelihood of remission.

Treatment of schizophrenia in most cases involves a combination of antipsychotic medications and psychosocial therapy.

Psychosis of AD

Psychosis of AD is a serious problem for caregivers and often results in institutionalization of the patient (Jeste and Finkel, 2000). Although estimates of the prevalence of psychotic symptoms in dementia vary, approximately 50% of patients experience delusions or hallucinations within the first three years after a clinical diagnosis of AD is made (Paulsen et al., 2000). The usual symptoms of psychosis of AD are simple paranoid delusions of stealing or hiding things, jealousy, or infidelity on the part of the spouse. Visual hallucinations are more frequent than auditory ones (Jeste and Finkel, 2000). Psychosis is more likely to present during intermediate stages of dementia than in very early or very late stages. Psychotic symptoms are rarely the initial manifestation of AD. As the severity of dementia increases, psychotic symptoms improve--it is unclear if this is true remission or merely a result of worsening cognition and speech making it difficult for the patient to express their delusions.

The optimal treatment of psychosis in AD may include pharmacotherapy and/or psychosocial interventions. Pharmacotherapy is necessary when the psychotic symptoms are severe enough to cause distressing agitation or aggression or otherwise disrupt the patient's functioning. The doses of antipsychotics should be considerably lower in patients with dementia than in those with schizophrenia, as can be seen in Table 2 (Jeste et al., 1999b). Psychosocial approaches to treatment emphasize caregiver education and environmental modifications to increase patients' orientation and decrease confusion. Such interventions may include optimization of social contact, increase in structured activities, environmental enrichment, prevention of overstimulation and light therapy (Cohen-Mansfield, 2001).

Treatment Considerations

Antipsychotic medications usually constitute the primary pharmacotherapy for psychotic disorders (Jeste et al., 1999b). Guidelines for the dosing of antipsychotic medications are presented in Table 2. In older people, the heightened risk of sedation, postural hypotension, anticholinergic side effects, extrapyramidal symptoms and, especially, tardive dyskinesia makes conventional antipsychotic medications less suitable in most cases. The atypical antipsychotics should usually be considered the first-line treatments because of their greater safety. Nonetheless, even with these medications, sedation, hypotension, weight gain, diabetes and cardiac conduction changes are causes for concerns. For older patients with schizophrenia or psychosis of AD, risperidone (Risperdal), olanzapine (Zyprexa) and quetiapine (Seroquel) are currently the three most frequently prescribed antipsychotics. Studies of ziprasidone (Geodon) in this population have not yet been published. Aripiprazole (Abilify) has recently been approved as an antipsychotic by the U.S. Food and Drug Administration.

Psychosocial treatments are essential for older people with psychoses. As mentioned previously, some nonpharmacologic interventions have empirical support for efficacy in psychosis of AD. For older patients with schizophrenia, a number of different individual and group therapies are beginning to be studied. In our center, we have focused on the following interventions (Granholm et al., 2002):

  • Cognitive-behavioral social skills training to target delusional thinking, social skills and coping strategies;
  • Medication adherence therapy to provide education and medication management skills to promote adherence to antipsychotic medication regimens;
  • Functional adaptation skills training, a group training program that teaches everyday living skills (e.g., financial management, use of transportation);
  • Work rehabilitation, using a supported employment model.

There is still much work to be done in terms of optimizing treatment of psychotic symptoms while reducing side effects, increasing optimal health behaviors (e.g., medication adherence, smoking cessation, nutrition and exercise) and minimizing disparities in access to health care for older patients with psychoses. However, there are many reasons to be optimistic about future therapies for older patients with psychoses, as the numbers of clinicians and researchers in the field of geriatric psychiatry continue to grow.




Cohen CI, Cohen GD, Blank K et al. (2000), Schizophrenia and older adults. An overview: directions for research and policy. Am J Geriatr Psychiatry 8(1):19-28.


Cohen-Mansfield J (2001), Nonpharmacologic interventions for inappropriate behaviors in dementia: a review, summary, and critique. Am J Geriatr Psychiatry 9(4):361-381.


Cuffel BJ, Jeste DV, Halpain M et al. (1996), Treatment costs and use of community mental health services for schizophrenia by age cohorts. Am J Psychiatry 153(7):870-876.


Granholm E, McQuaid JR, McClure FS et al. (2002), A randomized controlled pilot study of cognitive behavioral social skills training for older patients with schizophrenia. Schizophr Res 53(1-2):167-169 [letter].


Jeste DV, Alexopoulos GS, Bartels SJ et al. (1999a), Consensus statement on the upcoming crisis in geriatric mental health. Research agenda for the next 2 decades. Arch Gen Psychiatry 56(9):848-853.


Jeste DV, Finkel SI (2000), Psychosis of Alzheimer's disease and related dementias. Diagnostic criteria for a distinct syndrome. Am J Geriatr Psychiatry 8(1):29-34 [see comment].


Jeste DV, Rockwell E, Harris MJ et al. (1999b), Conventional vs. newer antipsychotics in elderly patients. Am J Geriatr Psychiatry 7(1):70-76.


Jeste DV, Symonds LL, Harris MJ et al. (1997), Nondementia nonpraecox dementia praecox? Late-onset schizophrenia. Am J Geriatr Psychiatry 5(4):302-317 [see comment].


Patterson TL, Kaplan RM, Grant I et al. (1996), Quality of well-being in late-life psychosis. Psychiatry Res 63(2-3):169-181.


Paulsen JS, Salmon DP, Thal LJ et al. (2000), Incidence of and risk factors for hallucinations and delusions in patients with probable AD. Neurology 54(10):1965-1971 [see comment].


Seeman MV (1997), Psychopathology in women and men: focus on female hormones. Am J Psychiatry 154(12):1641-1647 [see comment].