Open Journal of Psychiatry, 2012, 2, 305-316 OJPsych
http://dx.doi.org/10.4236/ojpsych.2012.24043 Published Online October 2012 (http://www.SciRP.org/journal/ojpsych/)
Elderly patients with very late-onset schizophrenia-like
psychosis and early-onset schizophrenia: Cross-sectional
and retrospective clinical findings
Caroline Girard1,2, Martine Simard1,2*
1School of Psychology, Laval University, Quebec City, Canada
2Research Centre at Mental Health University Institute of Quebec, Quebec City, Canada
Received 15 June 2012; revised 18 July 2012; accepted 29 July 2012
ABSTRACT
Objectives: The aim of this study was to characterize
the symptoms at onset/past and current symptoms of
patients with Very Late-Onset Schizophrenia-Like
Psychosis (VLOSLP; first onset of psychotic symp-
toms at/or after 60 years old) with those of elderly
patients diagnosed with schizophrenia before the age
of 40 years old (Early-Onset Schizophrenia—EOS) in
order to validate the clinical nosology proposed by
the International Late-Onset Schizophrenia Group.
Methods: This is a between-patient comparison study
with retrospective and current data taken from an
historical cohort that was conducted from May/2005
to August/2008. Seventeen VLOSLP and 17 EOS
were included. Schizophrenia and schizophrenia-like
psychotic disorders were initially diagnosed by board-
certified psychiatrists with the Diagnostic and Statis-
tical Manual Criteria at use at onset of the disorders.
Patients’ symptoms were assessed with the Scale for
the Assessment of Positive Symptoms (SAPS) and the
Scale for the Assessment of Negative Symptoms
(SANS). The general scores on the SAPS/SANS were
the primary outcomes. Results: Both groups had hal-
lucinations and delusions at onset of the disease, but
the following symptoms were more present and se-
vere in EOS than in VLOSLP: hallucinations (p =
0.001); assiduity loss (p < 0.001); grandiosity (p =
0.001); reference (p < 0.001) and influence (p = 0.001)
delusions. VLOSLP had mostly persecutory delusions.
At current evaluation (follow-up of cohort), most pa-
tients in the two groups presented residual symptoms
of anhedonia and apathy, but EOS, presented more
symptoms of friendship poverty (d = 1.42, larg e effect
size) than VLOSLP. The neuroimaging studies (when
available) at follow-up demonstrated greater vascular
cerebral lesions/vulnerability in VLOSLP than in
EOS patients. Conclusion: This study showed that
both VLOSLP and EOS had positive and negative
symptoms in the past/at onset of the disease, but they
were more severe in EOS than in VLOSLP. However,
the positive symptoms of both groups at follow-up of
the cohort (current evaluation) responded relatively
well to neuroleptics.
Keywords: Psychosis; Negative Symptoms;
Schizophrenia; Very-Late-Onset
1. INTRODUCTION
A recent study found that the one-year prevalence of all
schizophrenia spectrum disorders in individuals aged 60
years and older was 0.71% [1]. Clearly, with the aging of
population, these late-onset disorders will become a
more important health issue [2,3]. Indeed the greatest
per-capita expenditures for individuals with schizophre-
nia were reported among the older beneficiaries [4]. At
the beginning of the new millennium, The International
Late-Onset Schizophrenia (ILOS) Group attempted to
disentangle inconsistencies regarding the diagnostic sys-
tems and nomenclature of late-onset schizophrenia/psy-
chosis disorders [5]. Following a Medline literature re-
view and two days of debate, this group achieved con-
sensus on diagnosis and recommended a nosology for
these disorders: The diagnosis of late-onset schizophre-
nia (LOS) for an illness onset after the age of 40 and the
diagnosis of very-late-onset schizophrenia-like psychosis
(VLOSLP) for an illness onset after the age of 60. The
International Late-Onset Schizophrenia Group proposed
that early-onset schizophrenia or EOS (onset before the
age of 40) and LOS (onset after 40 years old) would be
more similar than different in terms of symptoms [5,6]
whereas VLOSLP (onset after 60 years old) would pre-
sent with a lower prevalence of formal thought disorder
and affective blunting, and a higher prevalence of visual
hallucinations than EOS [5]. This nomenclature may
*Corresponding author.

C. Girard, M. Simard / Open Journal of Psychiatry 2 (2012) 305-316 306
have had face validity and clinical utility, but required
further investigation in several domains, especially re-
garding the differential clinical symptomatology with
EOS [5,7].
Following the publication of the nomenclature pro-
posed by the ILOS Group, numerous studies were con-
ducted to differentiate the clinical profile between LOS
and EOS patients [6,8-17]. However, because the present
paper will rather focus on VLOSLP/EOS than on LOS/
EOS, the LOS/EOS differential nomenclature will not be
further developed; the interested readers shall consult
Pearman and Batra [6] for a recent review on this par-
ticular topic.
Studies on VLOSLP alone [18-23], as well as studies
comparing the profile of patients with VLOSLP with that
of patients with LOS [14,24,25], elderly EOS [14,24,
26-29] and healthy controls [30,31] were also realized to
better differentially characterize the demographics, clini-
cal symptoms and evolution [14,20,29] of VLOSLP pa-
tients.
In terms of positive symptoms, mild paranoid delu-
sions [14,26] mostly persecutory [18,20,21,24,25], audi-
tory [18,21,25,26] and visual hallucinations [18,21,25],
as well as all the Scale for the Assessment of Positive
Symptoms (SAPS) [32] behavioral disturbances except
for psychomotor abnormalities [25] were described in
VLOSLP. These symptoms were sometimes as common
as in LOS [25]. However, presence of formal thought
disorders in VLOSLP generated contradictory results.
Some authors reported formal thought disorders in 100%
of French-speaking VLOSLP patients [25], while other
authors working mostly on English- and Deutsch-speak-
ing patients reported a low prevalence [18], and a lower
risk of formal thought disorder in VLOSLP than in EOS
patients [14]. Formal thought disorder was found to be
absent in VLOSLP patients in the remaining studies [20,
24].
The presence of negative symptoms was also very
controversial [14,20,21,25]. In a previous study realized
by our team [25], both VLOSLP and LOS patients pre-
sented with several negative symptoms as measured by
the Scale for the Assessment of Negative Symptoms
(SANS) [33]; only apathy was more frequent in LOS
than in VLOSLP. However, in most studies, negative
symptoms were absent in VLOSLP [14,20] while they
were present in EOS and to a lesser extent in LOS [14].
Negative symptoms were not at all or barely mentioned
in several studies on VLOSLP [18,21,24,26].
Regarding the evolution of VLOSLP, Mazeh and col-
leagues [29] reported that the majority of VLOSLP pa-
tients did not present cognitive and functional deteriora-
tion over a mean follow-up period of 30 months, on the
contrary to EOS patients (42% with functional decline
and 16% with cognitive decline). However, the research-
ers only administered the Clinical Global Impression
(CGI) scale and a telephone interview in which they
asked 3 questions to the caregiver in order to assess the
cognitive and functional status of the patient. Köhler and
colleagues [14] also found that generally, later age of
onset was associated with better outcome, although there
was a significant gender by age of onset interaction, with
men displaying poorer outcome in the EOS/LOS group
and women tending to have a worse outcome in the
VLOSLP.
Good response to antipsychotics [24,26,34,35], even
better than in LOS [24] and elderly EOS patients [24,26],
was often registered in VLOSLP patients. VLOSLP in-
dividuals, especially inpatients [35], responded well to
atypical antipsychotics [20,35], and in particular to ami-
sulpride [34].
In terms of sociodemographics, there was generally a
female preponderance in VLOSLP [1,14,18,20,22,24-26,
29]. Regarding education, in one instance patients with
VLOSLP were described as being more educated than
elderly EOS [26] but in another study they were reported
to be less educated than LOS [25]. Regarding social
abilities, VLOSLP patients were described as having a
better premorbid social functioning than elderly EOS,
because they had been more married [26] and/or wi-
dowed/divorced [22,25] than single. However, other au-
thors found that a majority of VLOSLP patients living
in the community were alone [20]. Male patients with
VLOSLP might be more socially isolated than women
[23].
Interestingly, some studies have shown that ethnicity
might play a role on the demographics and symptoms’
profiles of VLOSLP [20-23]. Some authors reported that
Black Caribbean [20,21] as well as migrant patients with
different ethnic origins [22,23] diagnosed with VLOSLP
were younger and more likely to be male than their Brit-
ish counterparts. Black Caribbean VLOSLP patients
tended to present more with delusions of a “bizarre” or
mystical nature and with a somatic or hypochondriacal
content than British patients [21].
In summary, studies on VLOSLP patients over the last
decade have consistently reported the presence of para-
noid delusions and generally less severe positive symp-
toms than in EOS. However, the presence/absence and
severity of negative symptoms remains a controversial
issue. Regarding the evolution of VLOSLP, gender might
have an impact on the outcome, with VLOSLP women
possibly having a poorer prognosis than men; but few
studies addressed this question so further research are
required. Finally, ethnicity might play a role in demo-
graphic characteristics as well as in the clinical profile of
VLOSLP patients. As there were very few studies con-
ducted on a French-speaking population with VLOSLP,
the differential clinical and sociodemographic profile of

C. Girard, M. Simard / Open Journal of Psychiatry 2 (2012) 305-316 307
French-speaking patients with VLOSLP versus that of
French-speaking patients with EOS shall be further in-
vestigated. Indeed, do VLOSLP patients share the same
symptomatology/etiology than EOS, or do they have a
neurodegenerative and/or neurovascular substrate to their
disorder as it has been suggested by some authors [36-
38]?
To address these issues, our team performed a histori-
cal cohort study on the cognitive and clinical profile/
evolution of elderly patients with VLOSLP versus those
of patients with life-long EOS. The detailed method and
cognitive results were presented in a previous report [39].
In summary, although VLOSLP and EOS patients did
not meet the diagnostic criteria for dementia at onset of
the psychotic disorder, the results showed that the groups
presented at follow-up with cognitive deficits in several
domains, and 20% of VLOSLP had developed dementia
overtime (after onset of psychotic disorders) compared to
5.9% of EOS [39].
The objective of the present paper is thus to present
the clinical results of this historical cohort study, i.e. to
characterize and compare the current (follow-up) and
past (at onset of the psychotic disorder) positive and
negative symptoms of patients with VLOSLP with those
of patients aged 55 years and older who received a diag-
nosis of EOS. In terms of clinical profile, VLOSLP were
hypothesized to present with less severe current (follow-
up) and past positive symptoms than EOS, to present
with some negative symptoms, but of less severity than
those of EOS patients. In terms of demographics, VLO-
SLP were hypothesized to be predominantly female, to
have a better premorbid social functioning, i.e. to be
more married/widowed/divorced than EOS (more single),
and to be less educated than EOS, per the findings of a
previous study on a French-speaking population by our
team [25].
2. METHODS
Research design and settings: This was a between-
patient comparison study made at follow-up, with a ret-
rospective component (onset of the psychotic disorder),
following the historical cohort [39]. Patients were re-
cruited in psychiatric/geriatric psychiatry units of five
hospitals in the province of Quebec, Canada.
Participants: Seventeen elderly patients presenting
for the first time at/or after 60 years old with either schi-
zophrenia, schizophreniform disorder, delusional disor-
der, non-specific psychotic disorder or paraphrenia (VLO-
SLP) were compared to 17 patients aged 55 years old
and older who had received a diagnosis of schizophrenia
before the age of 40 (EOS). Supplementary inclusion
criteria for both groups involved: 1) The absence of de-
mentia at onset of the initial psychotic episode (per the
medical chart) (onset of historical cohort); and 2) psy-
chotic symptoms at onset (of historical cohort) not ex-
plained by another psychiatric and/or neurological and/or
systemic/medical disorder. The initial diagnoses had to
be made by board-certified psychiatrists using the offi-
cial Diagnostic and Statistical Manual Criteria (DSM)
employed at the time.
Ethical considerations: This research was approved
by the Ethical Research Committee of all hospitals in-
volved in this study. Written informed consent was ob-
tained from every patient and/or legal substitute by a
research assistant not involved in the evaluation. Patients
had to: 1) Consent to participate in the evaluations of the
study; 2) Allow the investigator to review the medical file
or, if not, accept to answer a socio-demographic/medical
questionnaire; 3) Allow an investigator to contact a fa-
mily member/friend (facultative, given that patients pre-
senting with paranoid symptoms might be reluctant to
involve family members of friends in the study); 4) Con-
sent to have a summary of his/her results put in his/her
personal medical file (facultative). All data were coded/
registered anonymously in the database.
Material: Socio-demographics and medical data were
first collected using a standard questionnaire on physical
status, medical/family antecedents, current medications
(dosages) and past investigations. This information was
corroborated by the patient medical records (when con-
sent obtained, see above). A semi-structured interview
designed to investigate the current and past psychiatric
symptoms was afterwards administered. This interview
was based on the Scale for the Assessment of Positive
Symptoms (SAPS) [32], and the Scale for the Assess-
ment of Negative Symptoms (SANS) [33]. Each symp-
tom was scored using a Likert scale (from 0 = absence to
4 = very important) reflecting the severity of each symp-
tom.
Statistical analyses: Descriptive and univariate analy-
ses were performed to compare some socio-demo-
graphics and present/past symptom profile of VLOSLP
and EOS. Chi-square and Fisher’s exact test were con-
ducted to compare categorical variables. Independent t-
tests were used to compare continuous variables. Analy-
ses were performed using SPSS for Windows version 17.
This was an analytical study and thus multiple compari-
sons were made. An alpha of p < 0.05 was applied to the
socio-demographic (Table 1), medical (Table 2) and
primary variables (Table 3) which were the general
scores obtained on the SAPS/SANS (see Table 3). Only
the variables found significant on these general scores
were further analysed to specify the symptoms’ typology.
A Bonferroni correction was applied to these results in
order to avoid the type I error. As a result, an alpha of p ≤
0.001 was determined. In addition, Effect sizes (d) [40],
were calculated on all descriptive data of Tables 3-5
with the following interpretation per Cohen [40]: d =

C. Girard, M. Simard / Open Journal of Psychiatry 2 (2012) 305-316 308
Table 1. Socio-demographic data.
Variables Total sample (N = 34) VLOSLP (n = 17) EOS (n = 17) T-test Chi-square
Gender (Female/Male) 24/10 13/4 11/6 - 0.452
Age at evaluation
(years) 70.9 (8.8) Range: 50 - 84 76.3 (4.2) 65.5 (9.1) <0.001*
Age at first symptoms
(years) 50.9 (22.9) Range: 17 - 84 71.2 (5.6) 27.9 (7.4) <0.001* -
Age at diagnosis (years) 52.8 (22.8)
Range: 17 - 81 73.1 (5.3) 29.9 (7.3) <0.001* -
Remission index
1(years) 2.8 (3.5) 5.1 (5.9) 0.176 -
Disease duration (years) 19.9 (17.1) 5.3 (5.2) 36.4 (8.1) <0.001* -
Diagnosis
SZ:
Szophreniform dis.:
Delusional dis.:
NS psychotic dis.:
Paraphrenia:
17
2
6
7
2
2
6
7
2
17
- <0.001*
Education (years) 9.88 (4.15)
Range: 4 - 21 7.47 (2.09) 12.29 (4.33) <0.001* -
Civil status
Single:
Married:
Divorced:
Widowed:
14
7
6
7
3
5
2
7
11
2
4
0
- 0.004*
Number of children 1.76 (2.22)
Range: 0 - 7 2.94 (2.54) 0.59 (0.87) 0.002* -
1Number of years between last episode and evaluation; Legend: VLOSLP: Very Late-Onset Schizophrenia-Like Psychosis; EOS: Early-Onset Schizophrenia; SZ:
Schizophrenia; Szophreniform dis.: Schizophreniform disorder; Delusional dis.: Delusional disorder; NS psychotic dis.: Non-specific psychotic disorder. *p < 0.05.
Table 2. Medical data from medical records.
Variables Total sample
(N = 34) VLOSLP (n = 17) EOS (n = 16) Chi-square Fisher exact test
Positive family history
Schizophrenia 10/30 2/16 8/14 0.010 0.019
Psychiatric disorders 18/29 7/16 11/13 0.024 0.052
Dementia 7/31 5/16 2/15 0.233 0.394
Current medication
Typical neuroleptics 12/34 2/17 10/16 0.004 0.010*
Atypical neuroleptics 30/34 16/17 14/16 0.287 0.601
Benzodiazepines 24/33 9/16 15/16 0.039 0.057
Antidepressants 14/34 8/17 6/16 0.486 0.728
Mood stabilizers 3/34 1/17 2/16 0.545 1.000
Anti-parkinsonian agents 9/34 1/17 8/16 0.007 0.017
Nootropics 0/34 0/17 0/16 - -
Medical condition
High blood pressure 18/33 12/16 6/16 0.022 0.037
Diabetes 10/33 4/16 6/16 0.520 0.708
Dyslipidemia/hypercholesterolemia 15/33 7/16 8/16 0.849 1.000
Cardiac condition 13/34 7/17 6/16 0.724 1.000
Smoking 6/34 3/17 3/16 1.000 1.000
Obesity 9/29 5/16 4/13 0.978 1.000
Hypothyroidia 8/33 3/16 5/16 0.475 0.688

C. Girard, M. Simard / Open Journal of Psychiatry 2 (2012) 305-316 309
Continued
CT-Scan results
Abnormalities at CT-Scan 13/22 11/16 2/6 0.132 0.178
Positive personality trait/disorder
- Paranoid
- Schizotypic
- Passive-dependant
- Obsessive
- Mixte
13/34
2
2
3
3
3
7/17
2
0
2
2
1
6/16
0
2
1
1
2
0.410
Current anxiety symptoms 5/34 4/17 1/16 0.146 0.335
Past anxiety symptoms 27/32 13/17 14/15 0.190 0.338
Current depressive symptoms 2/34 2/17 0/16 0.145 0.485
Past depressive symptoms 22/32 11/17 11/15 0.599 0.712
Legend: VLOSLP: Very Late-Onset Schizophrenia-Like Psychosis; EOS: Early-Onset Schizophrenia; Note: Access to medical files for 17 VLOSLP and 16 EOS. *p < 0.05.
Table 3. General psychiatric profile.
VLOSLP ( n = 17)
Mean (SD) EOS (n = 17) Mean (SD) Effect size (d) T-test (p)
Number of hallucinations (/7)0.75 (1.18) 0.82 (0.95) 0.07 0.845
Number of delusions (/20) 1.06 (1.39) 2.94 (3.47) 0.77 0.059
Number of behavioral
disturbances (/4) 0.19 (0.54) 0.59 (0.71) 0.64 0.078
SAPS
Number of formal thought
disorders (/8) 0.75 (1.57) 0.47 (0.94) 0.22 0.537
Number of affective poverty
symptoms (/7) 1.56 (1.83) 2.76 (2.56) 0.55 0.133
Number of alogia symptoms
(/5) 0.88 (1.59) 1.06 (1.35) 0.12 0.721
Number of apathy symptoms
(/3) 1.19 (1.17) 1.00 (1.00) 0.18 0.623
Current condition
SANS
Number of anhedonia
symptoms (/4) 2.19 (1.52) 2.76 (1.39) 0.39 0.263
Number of hallucinations (/7)2.06 (1.52) 3.94 (1.44) 1.27 0.001*
Number of delusions (/20) 2.65 (1.62) 7.06 (3.13) 1.86 <0.001*
Number of behavioral
disturbances (/4) 1.47 (0.94) 2.47 (0.87) 1.11 0.003*
SAPS
Number of formal thought
disorders (/8) 1.53 (1.97) 2.43 (2.28) 0.42 0.248
Number of affective poverty
symptoms (/7) 1.24 (1.89) 2.00 (1.63) 0.43 0.224
Number of alogia symptoms
(/5) 0.76 (1.15) 1.21 (1.25) 0.38 0.306
Number of apathy symptoms
(/3) 1.24 (1.20) 2.18 (0.64) 1.02 0.009*
At onset/Past
condition
SANS
Number of anhedonia
symptoms (/4) 2.94 (0.66) 3.65 (0.61) 1.12 0.003*
Legend: SAPS: Scale for the Assessment of Positive Symptoms; SANS: Scale for the Assessment of Negative Symptoms; VLOSLP: Very Late-Onset Schizo-
phrenia-Like Psychosis; EOS: Early-Onset Schizophrenia. *p < 0.05.

C. Girard, M. Simard / Open Journal of Psychiatry 2 (2012) 305-316 310
Table 4. Positive symptoms.
Variables
VLOSLP (n = 17)
Ratio1 SAPS EOS (n = 17) Ratio SAPSES (d) T-test (p)
Mean (SD) Mean (SD)
Hallucinations Current
Past
5/16
15/17 8/17 16/16
Auditory Current
Past
5/16
14/17
1.12 (1.73)
3.12 (1.54)
7/17
15/16
0.82 (1.07)
3.24 (1.30)
0.21
0.09
-
0.811
Commenting voices Current
Past
1/15
5/17
0.35 (1.00)
1.18 (1.88)
2/16
11/15
0.29 (0.69)
2.59 (1.77)
0.07
0.77 - 0.031
Conversing voices Current
Past
1/15
4/17
0.35 (1.00)
0.94 (1.75)
0/17
6/15
-
1.47 (1.84)
-
0.29
-
0.396
Ordering voices Current
Past
2/16
4/17
0.35 (1.00)
0.76 (1.56)
1/17
9/16
0.12 (0.49)
2.12 (1.97)
0.31
0.77 - 0.034
Somatic Current
Past
2/15
1/17
0.53 (1.18)
0.24 (0.97)
2/17
9/16
0.41 (1.18)
1.71 (1.76)
0.10
1.08 -
0.006
Olfactive/Gustative Current
Past
0/16
2/17
0.06 (0.24)
0.41 (1.18)
0/16
2/16
0.06 (0.24)
0.41 (1.00)
0.00
0.00
-
1.000
Visual Current
Past
1/15
5/17
0.35 (1.00)
0.94 (1.64)
2/16
11/16
0.29 (0.69)
2.24 (1.72)
0.07
0.77 - 0.031
Delusions Current
Past
8/16
16/17 11/16 17/17
Persecutory Current
Past
6/16
15/17
1.12 (1.54)
3.29 (1.45)
5/16
16/16
0.71 (1.16)
3.82 (0.73)
0.30
0.49
-
0.191
Jealousy Current
Past
0/16
5/17
0.06 (0.24)
0.88 (1.58)
0/17
1/16
-
0.29 (0.99)
-
0.46
-
0.203
Sexual/Erotomaniac Current
Past
2/16
5/17
0.41 (1.06)
1.12 (1.80)
2/17
6/16
0.47 (1.33)
1.24 (1.68)
0.05
0.07
-
0.845
Culpability Current
Past
0/16
0/17
0.06 (0.24)
-
2/16
5/17
0.41 (1.00)
1.00 (1.73)
0.57
-
-
0.030
Grandiosity Current
Past
1/16
1/16
0.18 (0.53)
0.24 (0.97)
5/17
10/16
0.82 (1.38)
2.18 (1.85)
0.67
1.38 - 0.001*
Catastrophic Current
Past
0/16
0/17
0.06 (0.24)
-
2/17
7/17
0.41 (1.18)
1.59 (1.97)
0.49
-
-
0.004
Mystic/religious Current
Past
0/16
3/17
0.06 (0.24)
0.24 (0.56)
4/16
8/15
0.65 (1.22)
1.82 (1.81)
0.81
1.33 -
0.003
Somatic Current
Past
2/16
2/16
0.47 (1.18)
0.29 (0.77)
2/16
4/15
0.47 (1.18)
1.24 (1.75)
0.00
0.75 - 0.055
Reference Current
Past
3/16
6/16
0.65 (1.32)
1.35 (1.84)
6/14
16/17
1.24 (1.48)
3.47 (1.01)
0.42
1.49 - <0.001*
Influence Current
Past
1/16
1/17
0.24 (0.75)
0.18 (0.73)
2/16
9/16
0.41 (1.06)
2.00 (1.87)
0.19
1.40 - 0.001*
Thought reading Current
Past
1/16
0/17
0.12 (0.33)
-
2/16
6/16
0.35 (0.86)
1.35 (1.80)
0.39
-
-
0.007
Thought diffusion Current
Past
1/16
1/16
0.18 (0.53)
0.29 (0.99)
7/17
8/15
1.12 (1.50)
1.88 (1.87)
0.93
1.11 - 0.005
Thought imposition Current
Past
0/16
0/17
0.06 (0.24)
-
2/17
5/16
0.29 (0.85)
1.06 (1.60)
0.42
-
-
0.015
Behavioral
disturbances Current Past
2/16
14/17 8/17 17/17
Bizarre presentation Current
Past
0/17
1/17
-
0.24 (0.97)
4/17
5/16
0.65 (1.22)
0.82 (1.24)
-
0.53
-
0.133
Inappropriate social and/or
sexual conducts
Current
Past
2/16
12/17
0.35 (0.86)
2.29 (1.65)
2/17
12/15
0.18 (0.53)
2.47 (1.63)
0.25
0.11
-
0.755
Agressive and/or agitated
behaviour
Current
Past
1/16
9/17
0.18 (0.53)
1.88 (1.87)
0/17
16/17
-
3.06 (1.09)
-
0.79 -
0.034
Repetitive/Stereotyped
behaviour
Current
Past
0/16
3/17
0.06 (0.24)
0.47 (1.13)
4/16
9/15
0.47 (0.80)
1.59 (1.37)
0.79
0.89 - 0.014
Formal thought disorders Current
Past
4/16
8/17 4/17 9/14
Loosening of associations Current
Past
2/16
7/17
0.41 (1.00)
1.41 (1.81)
4/17
8/14
0.53 (1.01)
1.94 (1.82)
0.12
0.29
-
-
Tangentiality Current
Past
3/16
7/17
0.53 (1.07)
1.29 (1.65)
1/17
8/14
0.12 (0.49)
1.82 (1.70)
0.53
0.32
-
-

C. Girard, M. Simard / Open Journal of Psychiatry 2 (2012) 305-316 311
Continued
Incoherence Current
Past
1/16
1/17
0.18 (0.53)
0.18 (0.73)
0/17
2/14
-
0.53 (1.07)
-
0.39
-
-
Illogical speech Current
Past
1/16
2/17
0.12 (0.33)
0.29 (0.85)
0/17
3/14
-
0.88 (1.54)
-
0.49
-
-
Circumstantial speech Current
Past
2/16
5/17
0.47 (1.18)
0.88 (1.45)
3/17
7/14
0.47 (1.07)
1.65 (1.73)
0.00
0.48
-
-
Logorrhea Current
Past
1/16
2/17
0.24 (0.75)
0.35 (1.00)
0/17
4/14
-
0.82 (1.24)
-
0.42
-
-
1Ratio refers to the proportion of patients presenting this symptom. Legends: VLOSLP: Very Late-Onset Schizophrenia-Like Psychosis; EOS: Early-Onset Schi-
zophrenia; ES = Effect Size; SD = Standard Deviation. *p < 0.001.
Table 5. Negative symptoms.
Variables VLOSLP (n = 17)
Ratio1 SANS EOS (n = 17)
Ratio SANS ES (d) T-test (p)
Mean (SD) Mean (SD)
Affective poverty Current
Past
8/16
7/17 11/17 12/16
Restricted face
expression
Current
Past
6/16
6/17
0.88 (1.17)
1.06 (1.52)
10/17
7/15
1.47 (1.33)
1.47 (1.63)
0.47
0.26
-
-
Gesture poverty Current
Past
4/16
2/17
0.59 (1.06)
0.29 (0.85)
7/17
1/14
0.88 (1.11)
0.35 (0.79)
0.27
0.07
-
-
Visual contact poverty Current
Past
1/16
1/17
0.18 (0.53)
0.18 (0.73)
4/17
3/15
0.59 (1.18)
0.65 (1.22)
0.48
0.48
-
-
Reduced emotional
reactivity
Current
Past
7/16
7/17
1.12 (1.36)
1.47 (1.84)
11/17
12/15
1.88 (1.58)
2.88 (1.69)
0.52
0.79 - -
Inappropriate affect Current
Past
1/16
2/17
0.18 (0.53)
0.35 (1.06)
3/17
7/15
0.41 (0.94)
1.76 (1.95)
0.31
0.94
-
-
Voice monotony Current
Past
3/16
1/17
0.41 (0.80)
0.18 (0.73)
6/17
0/14
0.82 (1.19)
0.18 (0.39)
0.41
0.00
-
-
Alogia Current
Past
5/16
6/17 8/17 8/14
Speech poverty Current
Past
5/16
5/17
0.88 (1.32)
0.94 (1.52)
5/17
6/14
0.65 (1.06)
1.18 (1.33)
0.19
0.17
-
-
Speech content
poverty
Current
Past
4/16
5/17
0.82 (1.43)
1.00 (1.62)
4/17
4/14
0.65 (1.22)
0.88 (1.27)
0.13
0.08
-
-
Speech freezing Current
Past
2/16
3/17
0.41 (1.00)
0.47 (1.07)
4/17
4/13
0.47 (0.87)
0.94 (1.25)
0.06
0.41
-
-
Apathy Current
Past
10/16
11/17 10/17 17/17
Presentation and
personal hygiene
neglected
Current
Past
4/16
4/16
0.76 (1.35)
0.82 (1.47)
1/17
6/17
0.18 (0.73)
0.88 (1.36)
0.56
0.04
-
0.904
Assiduity loss Current
Past
5/16
9/17
1.12 (1.76)
1.65 (1.69)
8/17
17/17
1.35 (1.58)
3.59 (0.51)
0.14
1.71 -
<0.001*
Physical anergia Current
Past
10/16
8/16
1.65 (1.41)
1.53 (1.63)
8/17
14/16
1.35 (1.58)
2.94 (1.34)
0.20
0.95 - 0.010
Anhedonia Current
Past
13/16
17/17 15/17 17/17
Restricted interests
and hobbies
Current
Past
10/16
14/15
1.94 (1.68)
2.82 (1.19)
10/17
17/17
1.59 (1.54)
3.53 (0.51)
0.22
0.84 - 0.031
Incapacity with
intimate relations
Current
Past
7/15
14/17
1.47 (1.66)
2.76 (1.48)
13/17
17/17
2.47 (1.55)
-
0.62
-
-
0.005
Friendship poverty Current
Past
12/16
15/16
2.53 (1.74)
3.29 (1.26)
13/17
16/17
3.94 (0.24)
3.71 (0.99)
1.42
0.37
-
0.297
1Ratio refers to the proportion of patients presenting this symptom. Legends: VLOSLP: Very Late-Onset Schizophrenia-Like Psychosis; EOS: Early-Onset Schi-
zophrenia; ES = Effect Size; SD = Standard Deviation. *p < 0.001.

C. Girard, M. Simard / Open Journal of Psychiatry 2 (2012) 305-316 312
0.2 (small effect size), d = 0.5 (medium effect size); d =
0.8 (large effect size).
3. RESULTS
Of the 60 patients (VLOSLP: n = 33; EOS: n = 27) re-
ferred for recruitment between May 2005 and August
2008, 19 refused to participate, 5 were considered non
eligible following detailed investigations [cerebral tu-
mour (n = 1); bipolar disorder (n = 1); psychopathic per-
sonality with paraphilia (n = 1); impossibility to deter-
mine the onset of first symptoms (n = 1); palliative stage
(n = 1)], and 2 were not reachable, resulting in a final
sample of 17 VLOSLP and 17 EOS.
Table 1 presents the socio-demographic data. VLO-
SLP had significantly more children than EOS. A match-
ing procedure for age and education was not possible,
resulting in significant differences between the groups,
with VLOSLP being significantly older and less edu-
cated than elderly EOS. A visual inspection of Table 1
also reveals that both groups included more women than
men, that EOS had longer disease duration than VLO-
SLP (which is linked to the diagnoses), and that more
VLOSLP were married/widowed than EOS.
Table 2 presents the medical data. Patients with EOS
took significantly more typical antipsychotics than pa-
tients with VLOSLP. A visual inspection of the results
demonstrates that 84.6% of EOS had psychiatric disor-
ders in their family compared to 43.8% in the VLOSLP.
Both groups presented with vascular risk factors, and
69% of the VLOSLP for whom neuroimaging data were
available at follow-up presented abnormalities, mostly
vascular. Only 6/17 EOS had CT-scans and abnormal
findings were reported in 33% of these patients (see Gi-
rard et al. [39], for further details and dosages of medica-
tions). Few active symptoms of depression and/or anxi-
ety were present in a small number of patients at the cur-
rent evaluation (follow-up) whereas most patients suf-
fered from significant anxious and depressive symptoms
in the past.
Tables 3-5 illustrate the patients’ symptomatic pro-
files. Regarding positive symptoms at onset of disease/in
the past, EOS presented significantly more hallucinations,
delusions, and behavioural disturbances than VLOSLP
per the SAPS assessment. Reference and influence delu-
sions, and grandiosity were especially more common in
EOS than in VLOSLP at the time (Table 4). Auditory
hallucinations, persecutory delusions and social/sexual
inadequate conducts were present in ≥75% of patients at
onset of disorders, regardless of the group, but EOS had
more commenting and ordering voices, mystic/religious
and thought diffusion delusions than VLOSLP (d ≥ 0.8;
large effect size). In addition, EOS had more aggressive/
agitated and repetitive/stereotyped behaviours than VLO-
SLP (d ≥ 0.8; large effect size).
Data regarding the current profile of positive symp-
toms showed that patients were mostly in remission with
no symptom being actually endorsed by a majority of
patients (≥75%) in both groups. However, EOS had more
residual symptoms of mystic/religious and thought diffu-
sion delusions as well as more repetitive/stereotyped be-
haviours than VLOSLP (d ≥ 0.8).
Regarding the negative symptoms as measured by the
SANS, EOS patients presented significantly more apathy
and anhedonia than VLOSLP patients but only at the
onset of the psychotic disorder (in the past). The detailed
analysis of the negative symptoms (Table 5) revealed
that only assiduity loss was significantly more present at
onset of the disorder (past) in EOS than in VLOSLP.
Symptoms of apathy and anhedonia were relatively com-
mon in the current (follow-up) condition of most patients
regardless of the group.
4. DISCUSSION
This study compared the positive and negative symptoms
at onset of the psychotic disorder/past condition and at
current (follow-up) evaluation of patients with VLOSLP
versus those with EOS. Different past and current symp-
tomatologies between the two groups were described,
with an overall more severe psychotic disorder in EOS
than in VLOSLP.
We had hypothesized that VLOSLP patients would
present with less severe past (at onset of the disorder)
positive symptoms than EOS. The results confirmed this
hypothesis.
EOS presented at onset/in the past more types of hal-
lucinations, delusions, and behavior disturbances than
VLOSLP. Regarding the delusion profile, EOS patients
presented more grandiosity, influence and reference de-
lusions than VLOSLP. These symptoms are common in
schizophrenia. Presence of auditory hallucinations and
persecutory delusions as principal symptoms of the two
groups is compatible with the observations of prior re-
ports on VLOSLP [24,25,35]. Nonetheless, the history of
more florid delusions in EOS compared with VLOSLP
cannot be ascertained without doubt given that the period
considered for the coding of past symptoms referred to a
longer interval of time in EOS (more than 36.4 years)
than in VLOSLP (mean of 5.3 years). However, the facts
that VLOSLP had a better premorbid social functioning
(more marriages and children) and took less typical an-
tipsychotics than EOS seemingly support the more se-
vere profile of EOS. Typical neuroleptics are mostly pre-
scribed in treatment-resistant cases. Taken altogether,
these data partly corroborate the nosology and the hy-
pothesis of milder schizophrenia symptoms in VLOSLP
than in EOS as proposed by the International Late-onset

C. Girard, M. Simard / Open Journal of Psychiatry 2 (2012) 305-316 313
Schizophrenia Group [5], and more recent reports [14,
26].
The results of this article are inconsistent with previ-
ous observations of more formal thought disorders in
EOS than in VLOSLP [5]. This could be explained by a
sample bias as the present community volunteer EOS
were perhaps less affected by the pathology than patients
in other studies, and conversely, the VLOSLP or a pro-
portion of VLOSLP were probably more cognitively
impaired at follow-up. Indeed, 80% - 93.3% of the
VLOSLP patients (depending on the psychiatrist or neu-
ropsychologist who made the diagnosis see [39] for the
method of consensus dementia/cognitive impairment
diagnosis at follow-up) had cognitive deficits at follow-
up and 20% evolved towards dementia over a mean pe-
riod of 2.4 (3.1 SD) years after the onset of psychosis
[39]. In addition, the ILOS Group proposed that visual
hallucinations would be more prevalent in VLOSLP than
in EOS [5]. The present study invalidated this hypothe-
sis.
Regarding the past negative symptoms, we had hy-
pothezied that VLOSLP patients would present with
some negative symptoms, but of less severity than the
negative symptoms of EOS patients. In accordance with
this hypothesis, the results showed that EOS had pre-
sented more apathy and anhedonia symptoms in the past
than VLOSLP, and especially more assiduity loss. This
difference in the clinical profile between VLOSLP and
EOS could reflect a more severe negative pathology in
the elderly EOS than previously suggested by some au-
thors, especially in women [28]. This could also be the
result of a bias regarding different disease duration in the
two groups.
Data regarding current symptoms suggests some re-
mission and/or symptom improvement in both groups.
Good response to atypical neuroleptics in VLOSLP is
consistent with previous data [20,26,34,35]. Indeed few
positive symptoms were actually endorsed in the current
condition (follow-up) by the patients, with delusions
being the most present in 50% of the VLOSLP and 69%
of the EOS. Persecutory delusions were the most promi-
nent residual symptom in VLOSLP while elderly EOS
presented various types of delusions. This observation is
in accordance with a more florid versus more systema-
tized/restricted delusions described, respectively, in EOS
and VLOSLP [5,14,26]. The current low prevalence of
positive symptoms in the groups might have produced a
floor effect reducing the possibility to observe other be-
tween-group differences.
The current negative profile highlights residual symp-
toms of affective poverty, apathy and anhedonia in both
groups which are consistent with previous reports in
VLOSLP [14,25]. These residual symptoms might also
be explained by a lesser efficacy of neuroleptics to con-
trol negative symptoms [41]. It is also possible that some
symptoms such as reduced facial expression and energy
might be extrapyramidal side effects of the medication
rather than intrinsic negative symptoms [42]. However,
this is less likely given that extrapyramidal side effects
are more common in typical than in atypical neuroleptics,
and that most VLOSLP (16/17) received atypical neuro-
leptics at an inferior percentage of maximum dose than
EOS (22.8% versus 50.2% for, respectively, VLOSLP
and EOS [39]).
In terms of demographics, we had hypothesized that
VLOSLP patients would be predominantly female and
would be less educated than EOS. The results confirmed
our hypotheses. There were more women than men in
VLOSLP and EOS. This finding is partly consistent with
the reports of female preponderance in VLOSLP [1,5,
25]. The volunteer participation of patients in the present
study possibly explains the discrepancies; women are
usually more likely to participate than men [43]. VLO-
SLP had lower education than EOS, which is consistent
with previous data [14], and in contradiction with others
[26]. In the present study, the education difference might
be cultural as high school was not legally mandatory up
until the sixties in the province of Quebec.
The two patient groups had at follow-up overall com-
parable profiles for vascular risk factors (VRF), which is
surprising given that VLOSLP were older than EOS, and
because VRF typically increase at midlife [44]. Yet this
type of results was previously reported [45]. However, in
the present study, cerebral lesions, generally of vascular
nature, were found on CT-scans in 68.8% of VLOSLP at
current (follow-up) evaluation only. On the contrary, the
EOS patients were not suspected to present abnormalities
since the majority of these patients were not adminis-
tered neuroimaging exams by their treating physicians.
Only the third of the EOS with CT-scans (2/6) presented
abnormal results at current evaluation. These findings sug-
gest the development of greater cerebrovascular vul-
nerability in VLOSLP than in EOS, thus corroborating
neurodegenerative/neurovascular reports of previous stu-
dies [7,26,36,38].
Some methodological limitations must be mentioned.
Despite a multicentre recruitment, the relatively low pre-
valence of VLOSLP patients previously reported in the
French-speaking population of Quebec City (0.7%) [25]
compared to LOS, and the high rate of refusal (nearly
33%) restricted the sample size. This possibly limited
statistical power and sample representativeness. In addi-
tion, the team voluntarily adopted a conservative scoring
approach i.e. symptoms denied by a patient and not re-
ported in the medical file or by a relative were scored as
absent. This choice might have masked some symptoms.
The impossibility to interview a relative in 47% of the
patients did not permit to bypass the lack of insight

C. Girard, M. Simard / Open Journal of Psychiatry 2 (2012) 305-316 314
commonly mentioned in schizophrenia [46] and in late
paraphrenia [47]; thus some symptoms might have been
missed.
The SAPS/SANS were completed by a single rater,
which ensured the consistency of ratings, but could have
introduced a bias in scoring if the rater had preconcep-
tions. The absence of inter-rater procedure must also be
mentioned since there was variability in the details of
medical records filled by different physicians at onset of
the disorders. This might be explained by the fact that
patients were recruited in several settings. This short-
coming is characteristic of historical cohorts. Further-
more, negative symptoms could have been difficult to
distinguish from the confounding effects of depression,
medications, institutionalization, poverty, etc. [48].
Finally, inconsistency regarding the onset of disease
and group composition must be considered. Indeed, the
onset of symptoms was preferentially defined as the mo-
ment symptoms were first noticed by the patient/family
member. When this information was unavailable, onset
was defined as the time of diagnosis. However, this me-
thod is less reliable; first signs of psychosis may arise
many years before the diagnosis. This may have resulted
by the inclusion in the VLOSLP group of some patients
presenting first psychotic symptoms before the age of 40
years old, although the mean old age at diagnosis (mean =
73.1 ± 5.3 years) makes this hypothesis unlikely. Al-
ternatively, it is possible that patients who presented
cerebral abnormalities at follow-up were in fact suffering
from some prior cerebral dysfunction that was unrecog-
nized at the onset of psychotic symptoms or masked by
the presence of psychotic symptoms.
Despite these limitations, this study provided impor-
tant data on the current and past symptomatic presenta-
tion of VLOSLP in a French-speaking population living
in the community.
5. ACKNOWLEDGEMENTS
The authors wish to thank Drs. Robert Noiseux, François Primeau,
Nicole Robert, Michel Dugas, François Rousseau, Nadine Gagnon,
Evelyn Keller and Patrick Bernier for kindly referring the patients. This
work was supported by doctoral awards (C.G.) from Réseau Qué-
bécois de Recherche sur le Vieillissement, Fonds de la Recherche en
Santé du Québec, and Fondation de l’Université Laval (Bourse Hydro
Québec), and by a 2005 NARSAD Young Investigator Award (M.S).
Parts of this article were presented at the 2nd Schizophrenia Interna-
tional Research Society Conference, Firenze, Italy, April 10-14th, 2010.
REFERENCES
[1] Meesters, P.D., et al. (2012) Schizophrenia spectrum
disorders in later life: Prevalence and distribution of age
at onset and sex in a Dutch catchment area. American
Journal of Geriatric Psychiatry, 20, 18-28.
doi:10.1097/JGP.0b013e3182011b7f
[2] Mitford, E., Reay, R., McCabe, K., Paxton, R. and Turk-
ington, D. (2010) Ageism in first episode psychosis. In-
ternational Journal of Geriatric Psychiatry, 25, 1112-
1118. doi:10.1002/gps.2437
[3] Reeves, R.R. and Brister, J.C. (2008) Psychosis in Late
life: Emerging issues. Journal of Psychosocial Nursing,
46, 45-52.
[4] Bartels, S.J., Clark, R.E., Peacock, W.J., Dums, A.R. and
Pratt, S.I. (2003) Medicare and medicaid costs for schi-
zophrenia patients by age cohort compared with costs for
depression, dementia, and medically ill patients. Ameri-
can Journal of Geriatric Psychiatry, 11, 648-657.
[5] Howard, R., Rabins, P.V., Seeman, M.V., Jeste, D.V. and
the International Late-Onset Schizophrenia Group (2000)
Late-onset schizophrenia and very-late-onset schizophre-
nia-like psychosis: An international consensus. American
Journal of Psychiatry, 157, 172-178.
doi:10.1176/appi.ajp.157.2.172
[6] Pearman, A. and Batra, A. (2012) Late-onset schizoph-
renia: A review for clinicians. Clinical Gerontologist, 35,
126-147. doi:10.1080/07317115.2011.642943
[7] Lagodka, A. and Robert, P. (2009) Is late-onset schizoph-
renia related to neurodegenerative process? A review of
the literature. L’Encéphale, 35, 386-393.
doi:10.1016/j.encep.2008.06.008
[8] Howard, R., Castle, D., Wessely, S. and Murray, R.M.
(1993) A comparative study of 470 cases of early- and
late-onset schizophrenia. British Journal of Psychiatry,
163, 352-357. doi:10.1192/bjp.163.3.352
[9] Jeste, D.V., Harris, M.J., Krull, A., Kuck, J., McAdams,
L.A. and Heaton, R. (1995) Clinical and neuropsycho-
logical characteristics of patients with late-onset schizoph-
renia. American Journal of Psychiatry, 152, 722-730.
[10] Palmer, B.W., Bondi, M.W., Twamley, E.W., Thal, L.,
Golshan, S. and Jeste, D.V. (2003) Are late-onset schi-
zophrenia spectrum disorders neurodegenerative condi-
tions? Annual rates of change on two measures. Journal
of Neuropsychiatry and Clinical Neurosciences, 15, 45-
52. doi:10.1176/appi.neuropsych.15.1.45
[11] Sato, T., Bottlender, R., Schröter, A. and Moller, H.J.
(2004) Psychopathology of early-onset versus late-onset
schizophrenia revisited: An observation of 473 neuroleptic-
naïve patients before and after first-admission treatments.
Schizophrenia Research, 67, 175-183.
doi:10.1016/S0920-9964(03)00015-X
[12] Huang, C. and Zhang, Y. (2009) Clinical differences be-
tween late-onset and early-onset chronically hospitalized
elderly schizophrenic patients in Taiwan. International
Journal of Geriatric Psychiatry, 24, 1166-1172.
doi:10.1002/gps.2241
[13] Köhler, S., van Os, J., de Graaf, R., Vollebergh, W.,
Verhey, F. and Krabbendam, L. (2007) Psychosis risk as
a function of age at onset: A comparison between early-
and late-onset psychosis in a general population sample.
Social Psychiatry and Psychiatric Epidemiology, 42, 288-
294. doi:10.1007/s00127-007-0171-6
[14] Köhler, S., et al. (2009) Evidence that better outcome of

C. Girard, M. Simard / Open Journal of Psychiatry 2 (2012) 305-316 315
psychosis in women is reversed with increasing age at
onset: A population-based 5-year follow-up study. Schi-
zophrenia Research, 113, 226-232.
doi:10.1016/j.schres.2009.05.017
[15] Vahia, I.V., Palmer, B.W., Depp, C., Fellows, I., Golshan,
S., Kraemer, H.C. and Jeste, D.V. (2010) Is late-onset
schizophrenia a subtype of schizophrenia? Acta Psychi-
atrica Scandinavia, 122, 414-426.
doi:10.1111/ j.1600-0447.2010.01552.x
[16] Riecher-Rössler, A., Häfner, H., Häfner-Ranabauer, W.,
Löffler, W. and Reinhard, I. (2003) Late-onset schizo-
phrenia versus paranoid psychoses: A valid diagnostic
Distinction? American Journal of Geriatric Psychiatry,
11, 595-604.
[17] Brunelle, S., Cole, M.G. and Elie, M. (2011) Risk factors
for late-onset psychoses: A systematic review of cohort
studies. International Journal of Geriatric Psychiatry, 27,
240-252. doi:10.1002/gps.2702
[18] Hassett, A. (1999) A descriptive study of first presen-
tation psychosis in old age. Australian and New Zealand
Journal of Psychiatry, 33, 814-824.
doi:10.1046/j.1440-1614.1999.00651.x
[19] Quin, R.C., Clare, L., Ryan, P. and Jackson, M. (2009)
Not of this world: The subjective experience of late-onset
psychosis. Aging and Mental Health, 13, 779-787.
doi:10.1080/13607860903046453
[20] Reeves, S., Stewart, R. and Howard, R. (2002) Service
contact and psychopathology in very-late-onset schizoph-
renia-like psychosis: The effects of gender and ethnicity.
International Journal of Geriat ric Psychiat ry, 17, 473- 479.
doi:10.1002/gps.614
[21] Reeves, S., Hudson, S., Fletcher, H., Sauer, J., Stewart, R.
and Howard, R. (2003) Are black Caribbean patients like-
ly to receive an incorrect diagnosis of very-late-onset
schizophrenia-like psychosis than their white British coun-
terparts? American Journal of Geriatric Psychiatry, 11,
674-677.
[22] Romero-Rubiales, F., Reeves, S. and Howard, R. (2004)
Have risk factors for very-late-onset schizophrenia-like
psychosis changed in the last 40 years? International
Journal of Geriatric Psychiatry, 19, 803-807.
doi:10.1002/gps.1152
[23] Mitter, P., Reeves, S., Romero-Rubiales, F., Bell, P., Ste-
wart, R. and Howard, R. (2005) Migrant status, age, gen-
der and social isolation in very late-onset schizophrenia-
like psychosis. International Journal of Geriatric Psy-
chiatry, 20, 1046-1051. doi:10.1002/gps.1396
[24] Alici-Evcimen, Y., Ertan, T. and Eker, E. (2003) Case
series with late-onset psychosis hospitalized in geriatric
psychiatry unit in Turkey: Experience in 9 years. Inter-
national Psychogeria tri cs , 15, 69-72.
doi:10.1017/S1041610203008767
[25] Girard, C. and Simard, M. (2008) Clinical characteriza-
tion of late- and very-late-onset first psychotic episode in
psychiatric inpatients. American Journal of Geriatric Psy-
chiatry, 16, 478-487.
doi:10.1097/JGP.0b013e31816c7b3c
[26] Barak, Y., Aizenberg, D., Mirecki, I., Mazeh, D. and Achi-
ron, A. (2002) Very late-onset schizophrenia-like psy-
chosis: Clinical and imaging characteristics in compari-
son with elderly patients with schizophrenia. Journal of
Nervous Mental Disease, 190, 733-736.
doi:10.1097/00005053-200211000-00002
[27] Arbus, C., Clement, J.-P., Bougerol, T., Fremont, P., Lan-
crenon, S. and Camus, V. (2012) Health management of
older persons with chronically medicated disorders: The
results of a survey in France. International Psychogeriat-
rics, 24, 496-502. doi:10.1017/S1041610211001487
[28] Häfner, H., Hambrecht, M., Löffler, W., Munk-Jorgensen,
P. and Riecher-Rössler, A. (1998) Is schizophrenia a dis-
order of all ages? A comparison of first episodes and
early course across the life-cycle. Psychological Medicine,
28, 351-365. doi:10.1017/S0033291797006399
[29] Mazeh, D., Zemishani, C., Aizenberg, D. and Barak, Y.
(2005) Patients with very-late-onset schizophrenia-like
psychosis: A follow-up study. American Journal of Geri-
atric Psychiatry, 13, 417-419.
[30] Howard, R.J., et al. (1997) A controlled family study of
late-onset non-affective psychosis (late paraphrenia). Bri-
tish Journal of Psychiatry, 170, 511-514.
doi:10.1192/bjp.170.6.511
[31] Bentall, R.P., et al. (2009) The cognitive and affective
structure of paranoid delusions. Archives of General
Psychiatry, 66, 236-247.
doi:10.1001/archgenpsychiatry.2009.1
[32] Andreasen, N.C. (1984) Scale for the assessment of po-
sitive symptoms (SAPS). University of Iowa, Iowa City.
[33] Andreasen, N.C. (1983) Scale for the assessment of neg-
ative symptoms (SANS) University of Iowa, Iowa City.
[34] Psarros, C., Theleritis, C.G., Paparrigopoulos, T.J., Politis,
A.M. and Papadimitriou, G.N. (2009) Amisulpride for the
treatment of very-late-onset schizophrenia-like psychosis.
International Journal of Geriatric Psychiatry, 24, 518-
522. doi:10.1002/gps.2146
[35] Scott, J., Greenwald, B.S., Kramer, E. and Shuwall, M.
(2011) Atypical (second generation) antipsychotic treat-
ment response in very late-onset schizophrenia-like psy-
chosis. International Psychogeriatrics, 23, 742-748.
doi:10.1017/S1041610210002188
[36] Brodaty, H., Sachdev, P., Koschera, A., Monk, D. and
Cullen, B. (2003) Long-term outcome of late-onset schi-
zophrenia: 5-year follow-up study. British Journal of Psy-
chiatry, 183, 213-219. doi:10.1192/bjp.183.3.213
[37] Casanova, M.F., Stevens, J.R., Brown, R., Royston, C.
and Bruton, C. (2002) Disentangling the pathology of schi-
zophrenia and paraphrenia. Acta Neuropathologica, 103,
313-320. doi:10.1007/s00401-001-0468-6
[38] Tsujino, N., et al. (2011) Cerebral blood flow changes in
very-late-onset schizophrenia-like psychosis with cata-
tonia before and after successful treatment. Psychiatry
and Clinical Neurosciences, 65, 600-603.
doi:10.1111/ j.1440-1819.2011.02257.x
[39] Girard, G., et al. (2011) Late-onset-psychosis: Cognition.
International Psychogeriatrics, 23, 1301-1306.
doi:10.1017/S1041610211000238
[40] Cohen, J. (1988) Statistical power analysis for the behav-
ioral sciences. 2nd Edition, Lawrence Erlbaum Associates,

C. Girard, M. Simard / Open Journal of Psychiatry 2 (2012) 305-316 316
Hillsdale.
[41] Jockers-Scherübl, M.C., Bauer, A., Godemann, F., Reis-
chies, F.M., Selig, F. and Schlattman, P. (2005) Negative
symptoms of schizophrenia are improved by the addition
of paroxetine to neuroleptics: A double-blind placebo con-
trolled study. International Clinical Psychopharmacol-
ogy, 20, 27-31. doi:10.1097/00004850-200501000-00006
[42] Gareri, P., De Fazio, P., De Fazio, S., Marigliano, N.,
Ferreri Ibbadu, G. and Sarro, G. (2006) Adverse effects
of atypical antipsychotics in the elderly: A review. Drugs
and Aging, 23, 937-956.
doi:10.2165/00002512-200623120-00002
[43] Lee, S., Saito, T., Takahashi, M. and Kai, I. (2008) Vol-
unteer participation among older adults in Japan: An
analysis of participation and reasons for non-participation.
Archives of Gerontology and Geriatry, 47, 173-187.
doi:10.1016/j.archger.2007.08.004
[44] Wiederkehr, S., Simard, M., Fortin, C. and van Reekum,
R. (2008) Validity of the clinical diagnostic criteria for
vascular dementia: A critical review. Part II. The Journal
of Neuropsychiatry and Clinical Neurosciences, 20, 162-
177. doi:10.1176/appi.neuropsych.20.2.162
[45] Sachdev, P., Brodaty, H., Rose, N. and Cathcart, S. (1999)
Schizophrenia with onset after age 50 years. 2: Neuro-
logical, neuropsychological and MRI investigation. Brit-
ish Journal of Psychiatry, 175, 416-421.
doi:10.1192/bjp.175.5.416
[46] Ostling, S., Borjesson-Hanson, A. and Skoog, I. (2007)
Psychotic symptoms and paranoid ideation in a popula-
tion-based sample of 95-year-olds. American Journal of
Geriatric Psychiatry, 15, 999-1004.
doi:10.1097/JGP.0b013e31814622b9
[47] Almeida, O., Levy, R., Howard, R.J. and David, A.S.
(1996) Insight and paranoid disorders in late life (late
paraphrenia). International Journal of Geriatric Psychi-
atry, 11, 653-658.
doi:10.1002/(SICI)1099-1166(199607)11:7<653::AID-G
PS380>3.0.CO;2-9
[48] Howard, R. (2001) Late-onset schizophrenia and very
late-onset schizophrenia-like psychosis. Review in Clini-
cal Gerontology, 11, 337-352.
doi:10.1017/S0959259801011455