Schizophrenia and Paranoid Disorders of Late Life

Schizophrenia and paranoid disorders of late life. Topics covered:

  • Introduction
  • Clinical features
    • Schizophrenic symptoms
    • Delusions
    • Hallucinations
    • Affective symptoms
    • Cognitive deficits
  • Epidemiology
  • Aetiological factors
    • Genetic factors
    • Sex Sensory deficits
    • Premorbid personality
    • Demonstrable brain abnormality
  • Management
    • Drug treatment 
  • References


Although we are accustomed to thinking of schizophrenia as an illness with onset in late adolescence or early adulthood and which is presumed to have a neurodevelopmental basis, psychoses that satisfy diagnostic criteria for schizophrenia can arise de novo at any point in life. Indeed, if one examines the incidence of broadly defined schizophrenia and other non-affective, non-organic psychoses across the lifecycle, then both early adult life (particularly for males) and extreme old age (particularly for females) are points when incidence peaks.

The diagnostic position of the late-onset cases is controversial for a number of reasons, and neither ICD-10 or DSM-IV have included codings for late age at onset. This is probably because we are so used to thinking of schizophrenia as a young-onset illness and more than a little prejudiced against making the diagnosis in late life. There are also important differences between early- and late-onset cases, which have led old age psychiatrists to view the late-onset cases as different from schizophrenia. ICD-9 included the diagnosis ‘late paraphrenia', a term first suggested in 1952 by Roth and Morrisey (1) to describe patients who these authors believed had schizophrenia with an onset delayed until after the age of 55 or 60 years. DSM-IIIR contained a category of late-onset schizophrenia for those cases with an onset after the age of 44 years.

Late-onset schizophrenia in the United States and late paraphrenia in Europe were never really meant to be the same thing, and such lack of agreement and international consistency contributed to their disappearance from later classification systems. The loss of age-at-onset coding or an upper age limit for the onset of schizophrenia has meant that these patients are classified within paranoid schizophrenia or delusional disorder, depending upon individual symptomatology. The available evidence supports the recognition of three age-at-onset-related categories for patients with schizophrenia and schizophrenia-like psychoses.

  1. Early-onset (before 40 years) schizophrenia is the most typical form.
  2. Late-onset (40–60 years) schizophrenia represent cases of ‘true' schizophrenia with onset delayed into late middle age.
  3. Very-late-onset (over 60 years) schizophrenia-like psychoses, although sharing many of the symptoms of schizophrenia, have a different set of associated risk factors and response to treatment than the other groups.

Clinical features

Schizophrenic symptoms

Formal thought disorder is only seen in about 5 per cent of patients with late-onset schizophrenia (2) and could not be elicited from any of 101 patients with very-late-onset schizophrenia-like psychosis. (3) Schneiderian first-rank symptoms are seen, but are less prevalent in late-onset and very-late-onset patients than in early-onset schizophrenia. Thought insertion, block, and withdrawal seem to be particularly uncommon and negative symptoms are unusual. (4)


Persecutory delusions dominate the presentation, although delusions of reference, control, grandiose ability, and those of a hypochondriacal nature are also often present. Partition delusions are found in about two-thirds of very-late-onset schizophrenia-like patients and are the belief that people, animals, material, or radiation can pass through a structure that would normally constitute a barrier to such passage. This barrier is generally the door, ceiling, walls, or floor of a patient's home and the source of intrusion is usually a neighbouring residence. (5)


Simple auditory hallucinations are very common and may take the form of a hum from persecuting machinery in an upstairs flat or the indistinct hubbub of voices from next-door. About 50 per cent of patients hear distinct voices, which make obscene or derogatory remarks about them and may comment on their actions. Visual hallucinations are common too, particularly in patients with visual difficulties, and may be fantastic and bizarre. Often these experiences resemble those of the Charles Bonnet syndrome (recurrent complex visual hallucinations, often but not always associated with reduced acuity, in the absence of neuropsychiatric disorder and accompanied by full insight), but the patients have a psychosis and are generally insightless. (6) Olfactory, gustatory, and tactile hallucinations may also be present.

Affective symptoms

Depressive features are commonly seen and may indicate a better prognosis. (7)

Cognitive deficits

Paranoid and schizophrenia-like psychoses can complicate Alzheimer's disease or dementia associated with Lewy bodies, but in such cases disorders of memory and concentration can generally clearly be seen to have preceded the emergence of persecutory symptoms. Many patients with psychoses that begin in old age have subtle and apparently only slowly progressive cognitive deficits; these do not constitute dementia and are not the same as schizophrenic negative symptoms, but seem to be more significant than the cognitive symptoms that accompany early-onset schizophrenia. Even the most cognitively intact of these patients appear to be impaired on ‘executive' cognitive functions such as attention set-shifting ability and planning. (8)


The point prevalence of schizophrenia-like psychoses in the elderly is around 0.1 per cent to 0.4 per cent, (9) while annual first admission rates for schizophrenia in the over-65 age group are between 10 and 15 per 100 000 for males and 20 and 25 per 100 000 for females. (10)

Aetiological factors

Genetic factors

While the lifetime risk of the more typical early-onset schizophrenia for first-degree relatives of patients is around 10 per cent and does not seem to reduce with increasing age at onset up to about 50 years, (11) the first-degree relatives of patients with very-late-onset (> 59 years) schizophrenia-like psychoses do not have an elevated lifetime morbid risk compared to healthy, aged control subjects. (12) The frequency of the e4 apolipoprotein E allele in late-life patients is not raised (13) as it has been reported to be for other neurodegenerative disorders.


As we have already seen, the age-related incidence of schizophrenia and schizophrenia-like disorders shows marked sex differences. Females are at much higher risk in the late-life population, and reported female-to-male ratios for groups of these patients have ranged from 4:1 to 20:1. This imbalance cannot be explained by the greater numbers of females in the age group and implies that there is something particular about the female sex which carries a higher risk of psychosis at this age.

Sensory deficits

Moderate or severe deafness affects about 40 per cent of the late-life patients and is generally of long duration, conductive, bilateral, and profound. (14) Visual impairment, most commonly a consequence of cataract or macular degeneration, is also more common in elderly paranoid patients than those with affective disorder, and there is a higher coincidence of visual and hearing impairment in paranoid than affective patients.

Premorbid personality

A consistent theme in descriptions of patients who develop psychoses late in life is the presence of abnormal personality traits—most often schizoid or paranoid personality types.(15) Although these patients are known to experience relationship difficulties, reflected by low rates of marriage and reduced numbers of children born to those who do marry, this contrasts with premorbid educational and occupational adjustment which is generally good. Hence abnormalities in premorbid personality function have a specificity characterized by unsociability, reticence, suspiciousness, and hostility. Having said all this, the author's experience of many of these patients is that they can be warm, trusting, and often very lonely individuals whose view of the world has been grotesquely distorted by their psychotic experiences.

Demonstrable brain abnormality

Brain imaging studies with CT and magnetic resonance imaging have shown that, as is the case for patients with schizophrenia of younger onset, modest enlargement of the lateral ventricles is present. (16) If patients with focal neurological abnormalities or accompanying dementia are excluded, there are no more demonstrable focal cortical or subcortical abnormalities in late-life patients than are found in healthy elderly controls. (17,18)


As mentioned above, whilst these patients are often described as hostile—and relationships with neighbours, the general practitioner, and local police have frequently been affected by their psychotic symptoms by the time that psychiatric referral is made—they are often also extremely lonely. Without entering into collusion, it is always worth taking the time to listen to the patient's account of her persecution and it is not difficult to express sympathy for the distress she is experiencing.

Sometimes a brief admission to hospital or the establishment of regular community psychiatric nurse visits can be rendered acceptable as an attempt to ‘get to the bottom' of whatever is going on. Once a relationship of trust and support has been established, patients will often accept medication and visits from members of the psychiatric team without really ever developing any insight into their condition. Telling the patient directly that she has a mental illness is probably the quickest way to join a list of perceived persecutors, and hence use of the Mental Health Act should be reserved until all else has failed. Even though this may mean that the patient does not receive antipsychotic medication, the provision and maintenance of a relationship of trust with a member of staff who can absorb complaints about, for example the neighbours, will do more to relieve distress than an enforced prescription.

Relatives and friends can be advised to encourage the patient to reserve discussion of such complaints to the time that the community psychiatric nurse visits, if this is possible. There is no single strategy that is best for all patients. For most, interventions delivered to their own homes (community psychiatric nurse or volunteer visits, home helps, and meals on wheels) seem to be the most acceptable, and although some will respond well to the activities and company provided by a day hospital or centre, many will decline to attend. These patients are often persistent and able complainers and may have highly restricted and encapsulated delusional systems. Their complaints about neighbours or the home environment are sometimes taken at face value by social services staff. It is therefore not uncommon to discover that, by the time of the first psychiatric referral, a patient has been rehoused at least once in the preceding months.

As a general rule, even if it results in a brief reduction in complaints from the sufferer, the provision of new accommodation is followed within a few weeks by a re-emergence of symptoms. The obvious distress this causes is sufficient reason always to advise patients and social workers against such moves unless for non-delusional reasons or following successful treatment of psychosis.

Drug treatment

The drug treatment of very-late-onset cases is essentially the same as for young patients, except that the effective dose of a drug may only need to be around 10 per cent of what the formulary recommends for young adults. Hence, 1 to 2 mg of trifluoperazine or 0.5 to 2.0 mg of riperidone daily may control symptoms. The effectiveness of such low doses cannot be explained by the advanced age of the patients alone, since those with early-onset schizophrenia who have grown old (sometimes referred to as ‘graduates') will often need doses that are 50 to 100 per cent of those given to their younger counterparts.

The choice of individual antipsychotic medication is determined more by consideration of undesirable side-effects than particular therapeutic advantage, with the exception of risperidone and olanzapine whose 5-hydroxytryptamine-2 receptor blocking action seems to be effective in the treatment of visual hallucinations. These newer drugs are also less likely to produce extrapyramidal side-effects in a group particularly vulnerable to such problems. (19)

The route of drug delivery is important in determining treatment success. Many of these patients are insightless and live alone, so compliance with oral medication is often doubtful. Extremely low doses of depot antipsychotic medication (e.g. fluphenazine decanoate 12.5 mg every 3 weeks), are well tolerated and effective, and can be given by a community psychiatric nurse who over a period of months to years can build up a close relationship with the patient. 


1. Roth, M. and Morrisey, J.D. (1952). Problems in the diagnosis and classification of mental disorders in old age. Journal of Mental Sciences, 98, 66–80.

2. Pearlson, G.D., Kreger, L., Rabins, P.V., et al. (1989). A chart review study of late-onset and early-onset schizophrenia. American Journal of Psychiatry, 146, 1568–74.

3. Howard, R., Almeida, O., and Levy, R. (1994). Phenomenology, demography and diagnosis in late paraphrenia. Psychological Medicine, 24, 397–410.

4. Almeida, O.P., Howard, R., Levy, R., and David, A. (1995). Psychotic states arising in late life (late paraphrenia). Psychopathology and nosology. British Journal of Psychiatry, 166, 205–14.

5. Howard, R., Castle, D., O'Brien, J., Almeida, O., and Levy, R. (1992). Permeable walls, floors, ceilings and doors. Partition delusions in late paraphrenia. International Journal of Geriatric Psychiatry, 7, 719–24.

6. Howard, R. and Levy, R. (1994). Charles Bonnet syndrome plus: complex visual hallucinations of Charles Bonnet type in late paraphrenia. International Journal of Geriatric Psychiatry, 9, 399–404.

7. Holden, N.L. (1987). Late paraphrenia or the paraphrenias? British Journal of Psychiatry, 150, 635–9.

8. Almeida, O.P., Howard, R., Levy, R., David, A., and Morris, R. (1995). Clinical and cognitive diversity of psychotic states arising in late life (late paraphrenia). Psychological Medicine, 25, 699–714.

9. Henderson, A.S. and Kay, D.W.K. (1997). The epidemiology of functional psychoses of late onset. European Archives of Psychiatry and Clinical Neuroscience, 247, 176–89.

10. Kay, D.W.K. (1972). Schizophrenia and schizophrenia-like states in the elderly. British Journal of Hospital Medicine, 8, 369–76.

11. Kendler, K.S., Tsuang, M.T., and Hays, P. (1987). Age at onset in schizophrenia. A familial perspective. Archives of General Psychiatry, 44, 881–90.

12. Howard, R., Graham, C., Sham, P., et al. (1997). A controlled family history study of late-onset non-affective psychosis (late paraphrenia). British Journal of Psychiatry, 170, 511–14.

13. Howard, R., Dennehey, J., Lovestone, S., et al. (1995). Apolipoprotein E genotype and late paraphrenia. International Journal of Geriatric Psychiatry, 10, 147–50.

14. Cooper, A.F. and Curry, A.R. (1976). The pathology of deafness in the paranoid and affective psychoses of later life. Journal of Psychosomatic Research, 20, 107–14.

15. Kay, D.W.K., Cooper, A.F., Garside, R.F., and Roth, M. (1976). The differentiation of paranoid from affective psychoses by patients' premorbid characteristics. British Journal of Psychiatry, 129, 207–15.

16. Rabins, P.V., Pearlson, G.D., Jayaram, G., Steele, C., and Tune, L. (1987). Ventricle-to-brain ratio in late-onset schizophrenia. American Journal of Psychiatry, 144, 1216–18.

17. Howard, R., Cox, T., Mullen, R., Almeida, O., and Levy, R. (1995). White matter signal hyperintensities in the brains of patients with late paraphrenia and the normal community-living elderly. Biological Psychiatry, 38, 86–91.

18. Symonds, L., Olichney, J., Jernigan, T., Corey-Bloom, J., Healy, J., and Jeste, D. (1997). Lack of clinically significant gross structural abnormalities in MRIs of older patients with schizophrenia and related psychoses. Journal of Neuropsychiatry, 9, 251–8.

19. Eastham, J.H. and Jeste, D.V. (1997). Treatment of schizophrenia and delusional disorder in the elderly. European Archives of Psychiatry and Clinical Neurosciences, 247, 209–18.