Mr. B, age 73, repeatedly complained to his landlord that people were trying to poison him by pumping noxious gas into his apartment. He barricaded himself inside, taped up all air vents and windows, and left only when absolutely necessary. At night, he could hear people working the “apparatus” that pumped the gas, and could “smell” the vapors.
On examination, he was physically well but suffered from mild neural deafness and myopia. He was suspicious and guarded but oriented and not cognitively impaired. He expressed paranoid beliefs and experienced auditory and olfactory hallucinations. There was no evidence of affective disturbance.
At first he refused psychiatric care but eventually agreed to take risperidone, 0.5 mg at night. He tolerated the agent well, and his psychotic symptoms slowly resolved.
As Mr. B’s case illustrates, schizophrenia—once thought to be strictly an early-onset disorder—commonly manifests late in life (Box). Too often, however, very late-onset schizophrenia goes undiagnosed because older patients with the disorder tend to be socially isolated. Their symptoms of paranoia and reluctance by family members to intervene also can prevent them from receiving treatment that could control psychotic symptoms and improve their quality of life.
Most clinical samples of patients with schizophrenia cite few cases of onset after age 60, reflecting the confused and changing nosology of very late-onset schizophrenia.
DSM-III (1980) stated that the schizophrenia label could apply only if onset occurred before age 45. This stipulation was dropped in DSM-III-R (1987), but it undoubtedly led psychiatrists to believe that schizophrenia simply did not begin in late life. The International Late-Onset Schizophrenia Group1 today recognizes the disorder’s late-onset version as “verylate-onset schizophrenia-like psychosis.”
General population studies report rates of “late paraphrenia” of around 1%, but these studies probably underestimate the true prevalence. One presumes that persons with paranoia are less likely than those without to participate in such a study.
The Camberwell Register First Episode Study,2 performed in London, is one of the few to determine rates of nonaffective psychosis across all ages of onset. In this study, 12% of the 513 patients studied across 20 years had illness onset after age 60. Researchers suspect a similar incidence in the U.S. population.
Psychosis presenting at any age, but especially in later life, requires careful evaluation to exclude organic pathology. Very late-onset schizophrenia differs substantially from psychosis associated with dementia, as in Alzheimer’s disease, both in terms of neuropsychological and brain imaging findings.
Although limited, data on late-life psychosis offer clues to its diagnosis and treatment. This article will address:
- risk factors and clinical presentations associated with late-onset schizophrenia
- pharmacologic and psychosocial treatment options based on available evidence.
Clinical presentation of schizophrenia with onset after age 60 differs from that of early-onset schizophrenia (Table 1). To those familiar with early-onset cases, the most obvious differences in late-onset patients are negligible rates of primary negative symptoms and formal thought disorder.
Persecutory delusions are common in both types and often are elaborate. The so-called “partition” delusion, which leads the patient to believe that people or objects can transgress impermeable barriers and access his or her home with malign intent, is more common in late-onset than in early-onset schizophrenia.3
Hallucinations in very late-onset schizophrenia are often prominent and can occur in multiple modalities, including auditory, visual, and olfactory. Sometimes the hallucination and delusion are clearly linked; for example, a patient claims to smell the noxious gas he believes is being pumped into his home.
Does the difference in presentation between early- and very late-onset schizophrenia reflect distinct disease processes or the disorder’s impact at different stages of brain maturation and degeneration? To answer that question, researchers have compared late-onset patients with young early-onset patents and with older patients who developed schizophrenia in their youth. Similar phenomena have been found in both early-onset groups,4 suggesting that age of onset causes the differences in clinical presentation.
As with early-onset schizophrenia, family history is the most common cause of very late-onset schizophrenia. Despite their limitations, family history studies almost all show a familial risk of very late-onset schizophrenia lower than that of early-onset patients but greater than that of the general population.5 Published studies do not tell us whether age of onset is genetically determined, in part because not all patients at risk for very late-onset schizophrenia live long enough to manifest its symptoms.
Family history has been associated with affective disorder in some patients with very late-onset schizophrenia. One casecontrolled series of family interviews6 found an approximate 1.3% rate of schizophrenia in relatives —about the same rate as that of the control group. The rate of depression among relatives of patients with very late-onset schizophrenia was 16.3%, compared with only 4.4% for controls (p = 0.003). Thus, late-onset psychosis and affective disorders may have etiologic links.7