Schizophrenia Research 133 (2011) 223–231

Contents lists available at SciVerse ScienceDirect

Schizophrenia Research
j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c h r e s

Effectiveness of different modalities of cognitive remediation on symptomatological,

neuropsychological, and functional outcome domains in schizophrenia: A
prospective study in a real-world setting
Antonio Vita a, b,⁎, Luca De Peri a, Stefano Barlati b, Paolo Cacciani b, Giacomo Deste a, Roberto Poli c,
Emilia Agrimi c, Bruno M. Cesana a, d, Emilio Sacchetti a, b, e
a
University of Brescia, School of Medicine, Brescia, Italy
b
Department of Psychiatry, Spedali Civili Hospital, Brescia, Italy
c
Hospital Istituti Ospitalieri of Cremona, Italy
d
Medical Statistics Division, Department of Biomedical Sciences and Biotechnologies, University of Brescia, Italy
e
Center for Neurodegenerative Disorders and EULO, University of Brescia, Italy

a r t i c l e i n f o a b s t r a c t

Article history: Objectives: The efficacy of cognitive remediation interventions in schizophrenia has been demonstrated in
Received 11 May 2011 several experimental studies. However, the effectiveness of such treatments in the usual setting of care of
Received in revised form 30 July 2011 schizophrenia and a direct comparison of different modalities of interventions have not been systematically
Accepted 14 August 2011 analyzed. The aim of the study was to assess the effectiveness of the cognitive subprograms of Integrated
Available online 9 September 2011
Psychological Therapy (IPT-cog) and of a computer-assisted cognitive remediation (CACR) method on
symptomatological, neuropsychological and functional outcome measures in schizophrenia.
Keywords:
Schizophrenia
Methods: Ninety patients with schizophrenia were assigned to IPT-cog, CACR or usual rehabilitative
Cognitive dysfunctions interventions (REHAB) in a naturalistic setting of care. Clinical, neuropsychological, and functional outcome
Cognitive remediation variables were assessed at baseline and after 24 weeks of treatment.
Functional outcome Results: Both the IPT-cog and CACR groups improved more than the comparison group with respect to all
outcome variables. The more responsive cognitive domains were speed of processing and working memory.
The effectiveness of the 2 remediation methods on the outcome dimensions considered was comparable.
However, IPT-cog, but not CACR, was more effective than REHAB on speed of processing, and the CACR group
had better outcome than both the REHAB and the IPT-cog groups when the Health of the Nation Outcome
Scale was considered. Few correlations between neurocognitive and functional outcome changes were found.
Conclusions: The study demonstrates the effectiveness, although nongeneralized, of IPT-cog and CACR in
schizophrenia when applied within a psychiatric and psychosocial treatment regimen representative of the usual
setting and modality of care, with no evident superiority of any of the methods, and indicates that the changes in
functional outcome during treatment are modestly mediated by improvement in specific cognitive domains.
© 2011 Elsevier B.V. All rights reserved.

1. Introduction (Simon et al., 2007), and such impairment persists with the
progression of the illness (Heaton et al., 2001; Morrison et al.,
The presence of cognitive deficits in patients with schizophrenia 2006). Almost all patients with schizophrenia demonstrate some
has been extensively documented since Kraepelin's first description decline in measures of neurocognitive functioning and the global
of the illness as dementia praecox. Cognitive impairment may be cognitive deficit in schizophrenia averages between 1 and 2 standard
considered a core feature of schizophrenia, detectable at the onset of deviations (SDs) below the mean of healthy control subjects (Green
the illness (Gopal and Variend, 2005), in neuroleptic-naive patients et al., 2004; Keefe et al., 2006), with deficits in psychomotor speed,
(Saykin et al., 1994) and even in the prodromal phase of the disease attention, verbal and working memory, and executive functions (Lee
and Park, 2005; Dickinson et al., 2007; Knowles et al., 2010). These
are believed to underlie part of the functional disability associated
with the disorder (Jaeger et al., 2006; Niendam et al., 2007). Several
studies have also shown that cognitive deficits are related to poorer
⁎ Corresponding author at: University of Brescia, School of Medicine, Brescia, Italy.
outcomes in different functional domains (Bowie et al., 2006, 2008),
Tel.: + 39 030 2184856; fax: + 39 030 2184871. quality of life and rehabilitation outcomes (Evans et al., 2004; Green
E-mail address: [email protected] (A. Vita). et al., 2004; Milev et al., 2005). With this more detailed knowledge of

0920-9964/$ – see front matter © 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.schres.2011.08.010
224 A. Vita et al. / Schizophrenia Research 133 (2011) 223–231

the role and meaning of cognitive deficits in schizophrenia, 2. Materials and methods
improvement in cognitive functions by means of cognitive remedi-
ation interventions has become a relevant target in the care of 2.1. Subjects
schizophrenia (Twamley et al., 2003; McGurk et al., 2007). In recent
decades, several cognitive remediation techniques, computerized All patients enrolled in the study were recruited and followed in
and non-computerized, designed for individual or group settings, the setting of 3 rehabilitative centres at the University Department of
have been developed and adopted in multimodal treatment Mental Health of the Spedali Civili Hospital in Brescia and the Istituti
approaches to the disorder, and studies analyzing their efficacy Ospitalieri Hospital in Cremona (Italy). They fulfilled DSM-IV-TR
have recently been reviewed quantitatively (McGurk et al., 2007; criteria (American Psychiatric Association, 2000) for schizophrenia
Grynszpan et al., 2010; Wykes et al., 2011). On the one hand, and were aged between 18 and 50 years. Diagnosis was made by
computer-assisted cognitive remediation (CACR) techniques, which expert clinicians through chart review and a clinical interview using a
enable selective treatment of different cognitive domains, have been symptom checklist allowing DSM-IV TR criteria to be applied.
shown to improve neurocognitive functions in schizophrenia in a Exclusion criteria were: (a) a concomitant diagnosis of mental
wide range of cognitive domains (Grynszpan et al., 2010). Moreover, retardation (as revealed by a Wechsler Adult Intelligence Scale-
a number of studies have demonstrated that CACR may also affect Revised (WAIS-R) (Wechsler, 1981) total IQ lower than 70) or of
psychotic symptoms and the psychosocial functioning of patients substance use disorder; (b) severe positive symptoms or impulsive
with schizophrenia (Bellucci et al., 2003; Wexler and Bell, 2005). On behaviour requiring a higher security setting; (c) significant changes
the other hand, several studies examining non-computerized, in psychopathologic status (requiring hospitalization or major change
individual- or group-based cognitive remediation found a significant in pharmacologic treatment) in the last 3 months. All patients not
improvement in cognitive performance associated with an improve- presenting these exclusion criteria entered the recruitment phase
ment in psychosocial functioning of schizophrenia (for reviews see independent from other clinical or anamnestic variables. These are the
Roder et al., 2006; McGurk et al., 2007; Wykes et al., 2011). Among same criteria for exclusion of patients to the rehabilitative centres
the latter interventions, a large body of research suggests that where the study was conducted, according to the admission/discharge
Integrated Psychological Therapy (IPT) (Brenner et al., 1994) has criteria for day centres and rehabilitation centres in the Lombardia
positive effects on the outcome of schizophrenia. In a recent review of Region, Italy, so that no further selection of patients among those
30 studies performed by Roder et al. (2006), positive mean effect treated in these centres was done.
sizes favouring IPT over usual treatment were found in the domains Patients referred consecutively to each of the rehabilitative centres of
of symptoms, psychosocial functioning and neurocognition. We the 3 psychiatric units were randomly assigned to the IPT-cog, CACR, or
confirmed the effectiveness of the cognitive subprograms of IPT on comparison groups. The allocation of patients followed a centralized
symptomatological, neuropsychological and functional outcome stratified randomization generated by the project coordinators outside
variables and found that some of the changes in functional outcome the centres. Six were subsequently excluded before treatment assign-
may be mediated by improvement in specific cognitive domains (Vita ment, all for mental retardation. The remaining 84 were included in the
et al., 2011). rehabilitation interventions. The patients in the comparison group
Despite these promising results, some key issues on the efficacy and entered non-cognitive group rehabilitation (REHAB). In this study,
limits of cognitive remediation interventions in schizophrenia have not completion was considered as attending at least 70% of the planned
yet been investigated. Most of the studies performed with cognitive sessions of treatment. At the time of baseline assessment, 52 patients
remediation methods have analyzed single measures in each outcome were taking an atypical antipsychotic, 4 a typical and some were on more
domain, and were quite heterogeneous as to the treatment modalities complex pharmacologic regimen (2 atypicals n =12, an atypical and a
used, duration of treatment, and the characteristics of the clinical typical n = 16). Other concomitant pharmacologic treatment was
treatment setting. In particular, McGurk et al. (2007) suggest that there allowed. All patients were then maintained on pharmacologic treatment
is a need for studies analyzing the effects of cognitive remediation on during the study but the drug dose or regimen could be modified as
most of the outcomes of interest and examining its action when needed. Doses of antipsychotics taken were calculated as chlorpromazine
embedded within usual psychiatric rehabilitation programs. Moreover, equivalents using the conversion proposed by Woods (2003) at baseline
the comparative efficacy of different modalities of intervention (e.g. and throughout the treatment period. This allowed us to obtain the mean
individual vs group, computerized vs non-computerized) is largely cumulative dose of antipsychotics taken during the study period and the
unknown. The IPT and CACR programs have been studied by different mean doses taken at baseline and at the end point.
authors and reported to be effective in experimental conditions, and Written informed consent to treatment was obtained from all
may be implemented in usual clinical settings. They represent 2 rather participants after the nature of the intervention procedures had been
different modalities of delivering cognitive remediation in clinical fully explained. The project was approved by the Board for Innovation
practice, one as group treatment, the other as an individual computer- in Psychiatry of the Health Authority of the Lombardia Region, Italy.
ized intervention, and may promote different strategies of remediation, The work has been carried out in accordance with the Code of Ethics of
including different degrees of compensatory strategies or action on the World Medical Association.
social cognition. Despite the obvious clinical and heuristic interest in this
issue, cognitive remediation interventions have not yet been directly 2.2. Assessment
compared. The aim of the present study was to compare the
effectiveness of 2 different cognitive remediation modalities (i.e. the The following evaluations were completed at baseline and after
cognitive subprograms of IPT (IPT-cog) and a CACR program) with 24 weeks of treatment by raters who were independent from those
respect to a similar amount of standard, non-cognitive-targeted, involved in the cognitive remediation treatment of the patients.
rehabilitative interventions in patients with schizophrenia. The 2
modalities were assessed by means of a comprehensive evaluation 2.2.1. Clinical
battery of clinical, cognitive and functional outcome domains, embedded Psychopathologic assessment was made by means of the Positive
within a psychiatric and psychosocial treatment regimen representative and Negative Syndrome Scale (PANSS) (Kay et al., 1987). The
of the usual setting and modality of care of Italian psychiatric symptoms were assessed following a semi-structured interview
rehabilitative centres. The impact of cognitive rehabilitation on referring to the month before the evaluation by raters who used the
psychosocial functioning and possible mediators of such effects were scale routinely. Inter-rater reliability was measured and found to be
also investigated. high (intraclass correlation coefficient N0.9 for PANSS total scores).
 A. Vita et al. / Schizophrenia Research 133 (2011) 223–231 225

Also, the Clinical Global Impression-Severity (CGI-S) (Guy, 1976) functional outcomes has been shown in several studies (McClelland et al.,
scores were rated. 2000; Parker et al., 2002; Pirkis et al., 2005; Salvi et al., 2005; Bech et al.,
2006). HoNOS has also been found to perform well against other
2.2.2. Neurocognitive established clinician-rated instruments that measure constructs related
Patients received a comprehensive battery of neuropsychological to psychosocial outcome (Parker et al., 2002). Reliability for this scale was
measures chosen to assess the principal cognitive domains found to measured and was found to be satisfactory (intraclass correlation
be impaired in schizophrenia, namely speed of processing, working coefficient=0.78).
memory, verbal memory and executive functions (Flashman and
Green, 2004; Nuechterlein et al., 2004) and for which cognitive 2.2.4. Time and modalities of assessment
remediation interventions have proved to be useful (McGurk et al., All scales were administered at baseline and at the end of the
2007; Wykes et al., 2011). The battery comprised the WAIS-R, adopted study, after 24 weeks. Neuropsychological tests were administered by
at study entry as an inclusion criterion measure; the Trail Making Test trained personnel, external to the treatment teams, who did not know
A (TMT-A), Trail Making Test B (TMT-B) (Reitan, 1979); the individual patients and their treatment allocation. On the other hand,
Wisconsin Card Sorting Test (WCST), number and percentage of the functional outcome measures were completed by the multi-
perseverative errors (Heaton, 1981); the Self-Ordered Pointing Task professional team taking care of the patients in the rehabilitative
(SOPT), number of wrong answers (Petrides and Milner, 1982); and centres, who were trained in the use of these instruments and used
the California Verbal Learning Test (CVLT), number of correct them routinely. This team, which did not include any personnel
responses at immediate free recall, short- and long-delay free recall involved in the experimental treatment for the individual patients,
and short- and long-delay cued recall (Benton and Hamsher, 1989). completed the evaluations with team consensus, using all the useful
The single tests were chosen among those frequently used in informative sources (clinical records, patient interview, information
neuropsychological studies of schizophrenia, as being an acceptable obtained from close relatives and the case manager). The symptom-
compromise between completeness in relation to cognitive domains atological evaluations were completed by the psychiatrists who
found to be affected in schizophrenia and their ease of use, or because followed each patient in the psychiatric outpatient units and were
they were already usually applied in our centres (Sacchetti et al., not directly involved in the study or aware of the rehabilitative
2008). The neuropsychological tests used have validated Italian treatments followed by their patients.
versions and Italian normative data (Giovagnoli et al., 1996; Heaton,
1981) or have been analyzed in the general Italian population in 2.3. Rehabilitative therapy programs
published (Sacchetti et al., 2008) or unpublished studies (further
psychometric data available on request from the authors). The IPT is a group-based structured cognitive behavioural program
Four cognitive constructs were used to represent specific domains for schizophrenia in which neurocognitive and social cognitive
of cognitive functioning: processing speed, working memory, verbal remediation are integrated with psychosocial rehabilitation. It is
memory, and executive functions. The Z score for each cognitive made up of 5 modules, applied in the following order: cognitive
construct was calculated by averaging the z scores of the different differentiation, social perception, verbal communication, social skills,
tests for evaluating the specific construct as follows: processing speed, and interpersonal problem solving (Brenner et al., 1994). The original
TMT-A; memory, all items of the CVLT; executive functions, TMT-B manual was translated into Italian and adapted for Italian language
minus TMT-A (used as a flexibility index; Reitan and Wolfson, 1993), and culture by one of the authors (AV) (Brenner et al., 1997). In this
mean of WCST percent of perseverative and total errors; working study, patients were administered the first 2 subprograms (cognitive
memory, TMT-B and SOPT number of errors. Z scores for each differentiation and social perception) of IPT (IPT-cog). The cognitive
neuropsychological test were obtained using Italian normative data components of IPT represent a practical compromise between a
for TMT and WCST (Giovagnoli et al., 1996; Heaton, 1981) or using rigorously targeted cognitive approach, as commonly used in
control data published in previous studies for SOPT (Sacchetti et al., laboratory studies, and a more comprehensive and ecologically
2008) or obtained in healthy subjects recruited by our group (n = 109, meaningful clinical approach.
see Sacchetti et al., 2008; Vita et al., 2011) for CVLT. The Z scores of the The IPT-cog groups, composed of 8–10 patients, attended therapy
4 cognitive constructs (composite Z scores) and a mean global sessions twice a week, 45 min each session, for 24 weeks. They were
cognitive score (mean of the 4 cognitive dimensions Z scores) were conducted by one trained psychiatrist or psychologist and another
calculated for 2 conditions, pre- and post-treatment. The pre- professional who administered the 2 cognitive subprograms following
treatment score was then subtracted from the post-treatment score the IPT manual.
for each of the 5 variables (4 cognitive dimensions and global The CACR used the Cogpack (Marker Software®) programme. The
cognitive score). Cogpack includes different neurocognitive exercises that can be divided
into domain-specific exercises, aimed at training specific cognitive areas
2.2.3. Psychosocial functioning among those known to be impaired in schizophrenia (verbal memory,
The following scales were adopted to evaluate psychosocial function- verbal fluency, psychomotor speed and coordination, executive function,
ing: the Global Assessment of Functioning (GAF) scale (American working memory, attention) and non–domain-specific exercises that
Psychiatric Association, 2000) and the Health of the Nation Outcome require the use of various functions at the same time and engage culture,
Scale (HoNOS) (Wing et al., 1998; Lora et al., 2001). The GAF scale in DSM- language and calculation skills. Most exercises are adaptive and the
IV is designed to measure patients' functioning and is an easy-to-use computer sets the level of difficulty, based on the patient's performance
instrument in common use. However, since it has been argued that its during the course of the session. The programme records the
ratings are more strongly correlated with ratings of clinical symptoms performance of each patient for every session, giving patients the chance
than functioning, reliance on the GAF scale as the only tool to assess to receive feedback on the session and on their progress. CACR was
patient functioning may be problematic (Gigantesco et al., 2006). HoNOS administered individually twice a week, in 45-min sessions, for
is a scale that assesses different aspects of psychosocial functioning. Its 24 weeks.
global score is determined by the severity of problems caused by The 2 methods target similar cognitive functions, but in different
difficulties in behaviour, clinical symptoms, cognition, and social and ways. The Cogpack addresses different cognitive functions through the
living conditions; a manual with well-described anchor points is practice of specific tasks; although engagement of these exercises is
provided, and the definition of specific symptoms is given. It is widely adaptive, no strategies are suggested by the software, and each function
adopted in Italian rehabilitative settings and its ability to predict global can be trained individually. Since little is known about the direct effects
226 A. Vita et al. / Schizophrenia Research 133 (2011) 223–231

of individual training tasks on specific cognitive functions, we designed of cognitive dimensions scores) and treatment group (IPT-cog, CACR
a broad range training battery, aimed at targeting the same cognitive or REHAB) as independent variables.
functions targeted by IPT-cog. Furthermore, we decided to standardize
the training protocol for all participants to guarantee uniformity of 3. Results
remediation modalities, to train all cognitive functions, and to allow
direct comparison with the IPT-cog results. This is also compatible with All but 2 of the treated patients completed the study according to
the demonstration of the substantial comparability of the effects of the criterion of at least 70% attendance at treatment sessions. A mean
specific vs generalized cognitive training with computerized modalities of 42 (range 37–48) sessions were completed, with no differences
(see Grynszpan et al., 2010). On the other hand, the IPT-cog is a more between treatment conditions (p N 0.8). Two patients in the IPT-cog
integrated and organized program, which requires the acquisition of group left the study early. Since it was not possible to obtain a retest
specific skills before advancing to the next step; in this program the battery for these patients, the analysis was carried out on the last
interaction of every single patient with the group is also essential. observation carried forward (LOCF) according to an intention-to-treat
Compensatory strategies are encouraged in both methods, but they are approach. The same approach was followed in the case of any tests
more openly and systematically included in the learning process for the missed at the last visit. This happened in an additional 5 cases: 3 cases
CACR method. Social cognitive skills are more directly addressed by the missed a symptomatological evaluation (2 in the CACR and 1 in the
IPT-cog (social perception subprogram) intervention. IPT-cog group), and 2 missed a single cognitive test (in the CACR
The comparison group participated in non–cognitive-oriented group).
group psychosocial interventions, consisting of different types of The characteristics of the 3 treatment groups at baseline are reported
activities (such as art therapy, physical training, or occupational in Table 1. The 3 groups were comparable with the exception of age
therapies), administered for the same amount of time as IPT-cog and (p b 0.05). No significant differences in clinical, cognitive and psychosocial
CACR. The REHAB modality was designed to match IPT-cog and CACR functioning variables at baseline were detected, with few exceptions
with respect to length and frequency of therapy sessions and, in (negative symptoms higher in IPT-cog than both CACR and REHAB
general, the amount of treatment received. groups: p b 0.01, and HoNOS scores higher in the IPT-cog vs REHAB group:
In addition to participating in the IPT-cog, CACR or REHAB programs, p= 0.017). For this reason, and to avoid the baseline severity of a given
the patients continued with standard psychiatric care provided by a variable affecting the degree of its response to treatment, we included age
multidisciplinary mental health team, including other rehabilitative and the baseline value of the variable as a covariate in the specific
interventions (not cognitive-oriented) aimed at improving social skills, analyses. On the other hand, there were no differences between groups
social relationships or work abilities, individually tailored in relation to for the mean drug dosage (chlorpromazine equivalents) taken at baseline
clinical needs and personal preferences and delivered in a balanced way or at the end point or for the cumulative antipsychotic drug doses taken
between the 3 groups (IPT-cog, CACR and REHAB). This allowed good during the treatment period (mean chlorpromazine equivalent doses:
comparability of the groups on treatments received and time spent in IPT-cog, 710.72 ±437.63, CACR, 572.53±280.65 and REHAB, 641.91±
the outpatient services and rehabilitative centres of the psychiatric units 370.33 mg, F =1.1, p =0.31), or the proportion of patients taking a single
involved in their care. atypical or typical antipsychotic or a combination of a typical and an
atypical or of 2 atypicals at baseline or end point. During the treatment
2.4. Statistical analysis period, patients had no significant deterioration of their psychopatho-
logic status requiring hospitalization, interruption of the rehabilitative
Data were analyzed using SAS version 9.13 software package. The programs or major changes in their pharmacologic treatment regimen.
univariate analysis of variance (ANOVA) or chi-squared test was used
to assess between-group differences in demographic data at baseline. 3.1. Clinical variables
Differences in pre- and post-treatment scores were compared
between the IPT-cog, CACR and comparison groups on each of the The direct comparison of PANSS score changes in the 3 groups by
clinical, neuropsychological and functional outcome measures with means of the ANCOVA with age and PANSS baseline scores as
an analysis of covariance (ANCOVA) with baseline scores of the covariates revealed that the 3 groups differed significantly for
specific variable and age as covariates. This was necessary, given the positive, negative and total PANSS score changes (Table 2). Similarly,
differences at baseline of some demographic or clinical measures (see the 3 groups differed in change in CGI-S scores during treatment. Post-
below) and the fact that baseline severity of clinical, neuropsycho- hoc analyses (with Bonferroni's correction) showed that both groups
logical performance and psychosocial functioning could potentially treated with cognitive remediation had greater improvement in
affect the comparison of effects between treatments. The ANCOVA of
scale score changes were adopted if the linearity and parallelism
assumptions were fulfilled. When such assumptions were not Table 1
Demographic and clinical characteristics of the treatment groups at baseline.
fulfilled, a repeated measures (pre- and post-treatment) ANOVA
was used to compare the score changes between groups and the Total IPT-Cog CACR REHAB p
time × group effect was calculated. In all cases, data were treated sample
according to an intention-to-treat model. N 84 26 30 28
Post-hoc analyses (with Bonferroni's correction) were completed M:F 58:26 21:5 19:11 18:10 0.297
in the case of significant results, in order to compare each of the 3 Age (years) 39.00 ± 37.15 ± 36.87 ± 43.00 ± 0.030
9.90 9.10* 11.40* 7.76
treatments directly (significance set at the level of p ≤ 0.017).
Age at onset (years) 23.50 ± 21.92 ± 22.87 ± 25.64 ± 0.119
In order to analyze the relationship between cognitive change and 6.95 5.44 6.01 8.63
psychosocial functioning change over the follow-up period, a Duration of illness (years) 15.89 ± 14.94 ± 14.80 ± 17.93 ± 0.401
correlation analysis (Pearson regression coefficient) was carried out 9.73 9.76 9.78 9.68
Education (years) 10.51 ± 10.00 ± 10.83 ± 10.64 ± 0.568
between these differences.
3.00 2.99 2.84 3.22
Two multiple regression analyses, according to a backward WAIS-R full scale IQ 86.67 ± 83.62 ± 88.63 ± 87.39 ± 0.334
procedure, were performed using the psychosocial scale scores (GAF 12.96 12.61 12.53 13.67
and HoNOS) as dependent variables and demographic data (age, Chlorpromazine 660.99 ± 674.08 ± 600.17 ± 714.00 ± 0.384
duration of illness, and sex), symptomatological data (change of equivalents at baseline 406.70 417.97 362.59 445.32

positive, negative and total symptoms), neurocognitive data (change *p b 0.05 vs REHAB (ANOVA).
 A. Vita et al. / Schizophrenia Research 133 (2011) 223–231 227

Table 2 Table 3
Psychopathologic variables before and after 24 weeks of treatment. Neurocognitive variables before and after 24 weeks of treatment.

IPT-Cog (n = 26) CACR (n = 30) REHAB (n = 28) p IPT-Cog CACR REHAB p

(n = 26) (n = 30) (n = 28)
CGI-S T0 5.0 ± 0.63 4.67 ± 0.75 4.71 ± 0.93 0.025
T6 4.04 ± 0.87 3.87 ± 0.81 4.36 ± 0.91 Mean processing T0 − 0.49 ± − 0.08 ± − 0.15 ± 0.028
Changea − 0.96 ± 0.72 − 0.80 ± 0.76 − 0.36 ± 0.78 speed 1.05 1.24 0.72
PANSS T0 19.0 ± 4.45 18.97 ± 5.91 19.68 ± 6.67 b 0.001 T6 0.019 ± 0.09 ± − 0.18 ±
Pos T6 14.0 ± 3.12 13.50 ± 4.18 17.89 ± 6.47 0.57 0.91 0.94
Changeb − 5.0 ± 3.40 − 5.47 ± 4.92 − 1.79 ± 4.33 Changea 0.51 ± 0.18 ± − 0.02 ±
PANSS T0 28.73 ± 6.65 22.27 ± 7.95 21.18 ± 7.33 b 0.001 0.81 0.66 0.73
Neg T6 21.77 ± 5.20 17.37 ± 6.70 21.21 ± 6.22 Mean working T0 − 1.34 ± − 0.89 ± − 0.79 ± 0.020
Changec − 6.96 ± 5.67 − 4.90 ± 6.31 0.04 ± 4.13 memory 1.08 1.07 1.0
PANSS T0 91.96 ± 12.74 88.07 ± 20.16 84.61 ± 19.94 b 0.001 T6 − 0.82 ± − 0.56 ± − 1.12 ±
Tot T6 70.31 ± 13.17 67.27 ± 18.09 80.89 ± 19.89 1.08 1.18 1.32
Changed − 21.65 ± 15.40 − 20.80 ± 18.35 − 3.71 ± 14.68 Changeb 0.51 ± 0.32 ± − 0.33 ±
a 0.81 0.73 1.03
IPT-cog vs REHAB, p = 0.011 (Effect Size (ES) = − 0.80); CACR vs REHAB, p = 0.031
Mean memory T0 − 2.64 ± − 2.29 ± − 2.20 ± 0.632
(ES = − 0.57); IPT-cog vs CACR, p = 0.642. Test for parallelism, p = 0.313; for linearity,
1.27 1.58 1.43
p b 0.001.
b T6 − 1.74 ± − 1.12 ± − 0.98 ±
IPT-cog vs REHAB, p b 0.001 (ES = − 0.83); CACR vs REHAB, p b 0.001 (ES = − 0.79);
1.45 1.78 1.93
IPT-cog vs CACR, p = 0.975. Test for parallelism, p = 0.436; for linearity, p b 0.001.
c Change 0.89 ± 1.17 ± 1.21 ±
IPT-cog vs REHAB, p b 0.001 (ES = − 1.29); CACR vs REHAB, p b 0.001 (ES = − 0.94);
0.95 1.23 1.32
IPT-cog vs CACR, p = 0.645. Test for parallelism, p = 0.836; for linearity, p b 0.001.
d Man executive T0 − 0.73 ± − 0.52 ± − 0.70 ± 0.702
IPT-cog vs REHAB, p b 0.001 (ES = − 1.19); CACR vs REHAB, p b 0.001 (ES = − 1.03);
functions .93 0.92 0.74
IPT-cog vs CACR, p = 0.477. Test for parallelism, p = 0.447; for linearity, p b 0.001.
T6 − 0.66 ± − 0.31 ± − 0.64 ±
1.02 0.93 1.23
Change 0.06 ± 0.21 ± 0.05 ±
0.48 0.58 0.96
PANSS positive, negative and total scores, but not CGI-S scores than Mean Global Cognitive T0 − 1.30 ± − 0.95 ± − 0.96 ± 0.281
Composite score 0.85 0.96 0.73
the REHAB group (Table 2). No differences emerged between the IPT- T6 − 0.80 ± − 0.47 ± − 0.73 ±
cog and the CACR group for improvement of such measures of clinical 0.82 1.02 1.12
severity. Change 0.49 ± 0.47 ± 0.22 ±
0.43 0.45 0.74
a
3.2. Cognitive variables IPT-cog vs REHAB, p = 0.008 (ES = 0.69); CACR vs REHAB, p = 0.086 (ES = 0.28);
IPT-cog vs CACR, p = 0.252. Test for parallelism: p = 0.052; for linearity p b 0.001.
b
IPT-cog vs REHAB, p = 0.010 (ES = 0.91); CACR vs REHAB, p b 0.021 (ES = 0.73);
Comparison of changes in neuropsychological test scores during IPT-cog vs CACR, p = 0.653. Test for parallelism: p = 0.391; for linearity p b 0.001.
treatment revealed a significant difference between the 3 groups for
the domains of mean processing speed and mean working memory
(Table 3). Post-hoc analyses (with Bonferroni's correction) showed
with significant effects of treatment group, patient's age, change in
that, compared with the REHAB group, the IPT-cog group had greater
positive symptoms and in mean memory scores (trend) during
improvement in mean processing speed and mean working memory
treatment (Table 6).
scores, while the CACR group had non-significantly greater improve-
ment in mean working memory (Table 3). No significant difference
4. Discussion
emerged, however, for improvement in neurocognitive measures
when the IPT-cog and the CACR groups were compared directly
After 6 months of treatment, patients undergoing cognitive remedi-
(Table 3).
ation interventions demonstrated significantly greater improvements in
psychopathologic, cognitive and functional outcome measures with
3.3. Functional outcome variables respect to patients following the usual setting of psychiatric and
psychosocial treatment. To our knowledge, this is the first time a
Change in HoNOS scores between the 3 groups was found to differ controlled comparison between 2 cognitive remediation interventions
significantly and a non-significant trend in the same direction was with different modalities of administration has been performed. Both the
found with GAF (Table 4). Post-hoc analyses showed that the CACR cognitive subprograms of IPT and the CACR interventions had a
group had significantly better outcome than both the REHAB and the significantly larger effect on negative, positive and total symptom
IPT-cog groups when the HoNOS scale was considered (Table 4). severity than usual care. IPT-cog showed a significantly better effect on

3.4. Cognitive domains and functional outcome score changes over time

Table 4
Change in neuropsychological variables, considered as composite
Functional outcome variables before and after 24 weeks of treatment.
scores of each cognitive domain, were not generally correlated with
changes in the functional outcome measures, both in the IPT-cog and IPT-Cog (n = 26) CACR (n = 30) REHAB (n = 28) p
CACR groups, with the exception of a significant correlation found GAF T0 45.62 ± 9.28 47.40 ± 11.71 50.11 ± 11.06 0.079
between improvement in mean global cognitive score and change in T6 54.19 ± 9.56 55.63 ± 8.26 52.71 ± 10.16
GAF scores during CACR treatment (r = 0.39; p = 0.03). Changea 8.58 ± 5.78 8.23 ± 9.02 2.61 ± 7.33
HoNOS T0 18.85 ± 5.27 17.53 ± 5.34 14.71 ± 5.42 b0.001
Multiple regression analyses conducted with functional scale scores
T6 14.35 ± 6.23 8.20 ± 5.58 11.64 ± 6.05
as dependent variables and demographic, clinical, neuropsychological Changeab − 4.50 ± 3.15 − 9.33 ± 5.14 − 3.07 ± 2.86
and treatment variables as independent variables revealed that changes a
IPT-cog vs REHAB, p = 0.055 (ES = 0.91); CACR vs REHAB, p = 0.043 (ES = 0.68);
in GAF scores were significantly predicted just by change in positive and IPT-cog vs CACR, p = ns. Test for parallelism, p = 0.120; for linearity, p b 0.001.
negative symptoms during treatment (Table 5); on the other hand, a b
IPT-cog vs REHAB, p = 0.919; CACR vs REHAB, p b 0.001 (ES = − 1.56); IPT-cog vs
more complex model of prediction of HoNOS score changes emerged, CACR, p = 0.001 (ES = 1.16). Test for parallelism, p = 0.130; for linearity, p b 0.001.
228 A. Vita et al. / Schizophrenia Research 133 (2011) 223–231

Table 5 psychosocial therapies on the favourable symptomatological outcome of

Multiple regression analysis of factors affecting GAF score changes after treatment: this group of patients (Gigantesco et al., 2006; Patterson and
dependent variable GAF T6-T0.
Leewenkamp, 2008). The differential effect vs usual rehabilitation
Source DF Sum of squares Mean square F value Pr N F interventions may have been influenced in part by the unblinded nature
Model 2 3307.49804 1653.74902 68.02 b 0.001 of the study, which did not completely exclude the possibility that the
Error 81 1969.39482 24.31352 treating psychiatrist could be informed of the type of rehabilitative
Corrected total 83 5276.89286 treatment undertaken. However, none of the psychiatrists were directly
Variable DF Parameter estimate Standard error t value Pr N − t
involved in the study design and performance, so other possible
Intercept 1 0.87335 0.74899 1.17 0.247
explanations may be hypothesized. One explanation, although specula-
Positive symptoms tive, could be that improvement in cognitive functions may positively
T6–T0 1 − 0.61202 0.13995 − 4.37 b 0.001 affect the patients' insight into illness, which has been reported to be
correlated with the severity of positive symptoms (Buchy et al., 2009).
Negative symptoms
Moreover, improvement in cognitive flexibility has been suggested to
T6–T0 1 − 0.72906 0.10382 − 7.02 b 0.001
reduce positive symptom severity (Lecardeur et al., 2009), in line with
the hypothesis that dysfunction of lower-level functions could be
implicated in the emergence of positive psychotic symptoms (Aleman
processing speed and working memory variables, on which CACR had et al., 2003).
smaller, trend effects; both cognitive treatments had somewhat larger The improvement in neurocognition observed in both the IPT-cog
effects on some functional (GAF) measures, but only the CACR group and CACR groups was not generalized to all cognitive domains, but
showed a better effect on psychosocial functioning as revealed by the was detected only for the domains of processing speed and working
HoNOS scale. Even if partial and somewhat contradictory, the treatment memory. Moreover, it was more limited than that reported by other
effects were compared with an actively treated control condition, and authors (for reviews see Grynszpan et al., 2010; McGurk et al., 2007;
this may justify the small effects found and accentuate the positive Wykes et al., 2011). Our results are only partially consistent with
findings. These results are in good agreement with those reported by previous evidence of the positive effect of the Cogpack program on the
other authors using different cognitive remediation modalities as domains of processing speed (Sartory et al., 2005; Lindenmayer et al.,
quantitatively reviewed by McGurk et al. (2007), Grynszpan et al. 2008), cognitive flexibility (Cavallaro et al., 2009) and verbal memory
(2010) and, most recently, by Wykes et al. (2011), who provided (Sartory et al., 2005), since the differential effects detected in the
evidence of favourable effects of cognitive remediation interventions on CACR group on speed of processing was only tendential and that on
symptomatological, neuropsychological and functional outcome vari- working memory was no longer significant after the restrictive
ables of schizophrenia. They are also in line with those of a previous study correction for multiple comparisons was applied. On the other hand,
of ours that evaluated the effectiveness of the cognitive subprograms of they suggest a more selective effect of the IPT-cog treatment than that
IPT in schizophrenia (Vita et al., 2011) and of a recently published study reported in the meta-analysis of IPT studies performed by Roder et al.
on CACR from the Italian population (Cavallaro et al., 2009). (2006). We administered only the first 2 subprograms of IPT (the
The interpretation of the significant effect detected in the present cognitive subprograms) and thus relating our findings to those
study on symptom severity is not obvious. Previous results (Vita et al., obtained with the complete IPT program is questionable. When
2011) suggest that the cognitive component of IPT may be beneficial applied selectively, the cognitive remediation component of IPT
specifically for negative and, to a lesser extent, total symptom severity. seems to be more specific to the domains of attention (Spaulding
Moreover, the effect of cognitive remediation on negative symptom- et al., 1999), verbal and working memory (Vita et al., 2011). The role
atology has also been recognized by other authors using a CACR method of the usual clinical setting of care, especially when intensive and
(Bellucci et al., 2003). In the latter case, the authors speculated that CACR highly structured as in the case of the Italian system of community
could influence negative symptoms by stimulating brain reward systems care, may explain such diluted effect when the remediation
by the feedback and success experiences that CACR provides (Bellucci et intervention is translated from the laboratory to the clinic. On the
al., 2003). On the other hand, the possibility of a significant effect on other hand, comparison with an active treatment within a structured
positive symptoms and global psychopathology severity of cognitive framework of interventions strengthens the value of positive findings,
rehabilitation remains uncertain, even though some evidence of CACR even if modest.
efficacy on positive symptoms already exists (Bark et al., 2003; Lecardeur Few significant effects were found for variables indicating patients'
et al., 2009). The effect detected in this study on psychopathology, with global psychosocial functioning. All the scales used showed favourable
robust effect sizes, in both active treatment groups could be attributable changes with treatments, with improvements somewhat more
to a non-specific incremental benefit of the application of structured pronounced for the remediation groups; however, only the difference
for the HoNOS scale was significant, while that for GAF was only
tendential. Both CACR and IPT-cog had a modestly good effect on
Table 6
Multiple regression analysis of factors affecting HoNOS score changes after treatment: psychosocial functioning, but change in the HoNOS scale was more
dependent variable HoNOS T6-T0. evident under the CACR treatment. The HoNOS is a multidimensional
scale that includes behavioural, social and relational items, and also an
Source DF Sum of squares Mean square F value Pr N F
item specifically relative to cognitive functioning. This may explain
Model 4 610.84338 152.71084 9.61 b 0.001
the differences detected with different functioning scales and for
Error 79 1254.90662 15.88489
Corrected total 83 1865.75000
different treatment modalities and raises the question of possible
Variable DF Parameter estimate Standard error t value Pr N − t differential effects of different cognitive remediation interventions on
Intercept 1 − 12.24601 2.42987 − 5.04 b 0.001 the heterogeneous outcome domains of schizophrenia. On the other
Age 1 0.13694 0.04729 2.90 0.005 hand, the larger effect on functioning found in the CACR group may be
Treatment group 1 2.34228 1.01899 2.30 0.024
due to the fact that compensatory strategies, although present in both
Positive symptoms modalities, are more openly included in the learning process of the
T6–T0 1 0.27033 0.11260 2.40 0.019 CACR method. It is also possible to hypothesize that the differential
effects on functional outcome found in our study may have been
Memory score
influenced by the prevailing role of the naturalistic setting in which
T6–T0 1 − 0.75203 0.38767 − 1.94 0.056
the cognitive remediation interventions were offered. There is
 A. Vita et al. / Schizophrenia Research 133 (2011) 223–231 229

evidence that cognitive remediation has somewhat different effects only to maintaining independence of the treatment delivery and rating
on outcome when embedded within a more general rehabilitation processes (including blind neuropsychological assessment and psycho-
treatment framework (McGurk et al., 2007; Wykes et al., 2011). This social outcome ratings obtained with staff consensus) but also to
could sometimes strengthen the effects of treatment, but may also maintaining a naturalistic setting of care, very close to the usual delivery
dilute them in other cases, especially when other highly structured of treatment in Italian psychiatric services. This reflected routine clinical
interventions are administered intensively. practice and improved the trial's acceptability for patients, psychiatrists,
Our finding of similar results obtained by different cognitive and the treatment team, facilitating the recruitment of a group of
remediation techniques, targeting cognitive functions in substantially patients with schizophrenia representative of those usually treated in
different modalities, poses an intriguing question about the pathways rehabilitative centres and so increasing the trial's external validity. The
underlying improvement in cognitive functions with cognitive remedia- neuropsychological tests selected for the study were not always the
tion interventions. It could be speculated that the observed improvement most representative of the cognitive domain under investigation.
with both modalities of remediation is consistent with a common However, they were chosen because they were related to the cognitive
pathway that, starting from activation of single cognitive domains, may domains found to be affected in schizophrenia and for their ease of use
extend to more general neurocognitive and psychosocial outcome or because they were already usually applied in our centres. Many other
improvements, even though at present it is unclear how the cognitive possible variables, including those related to pharmacologic treatment,
changes are causally linked to changes in behaviour (Green, 1999; could have affected some of the results, even though we controlled the
Hogarty et al., 2006). Moreover, a greater understanding of the sample for the pharmacologic regimen. We found no significant
relationship between improved cognitive function and better psychoso- differences among groups both in terms of the amount of medication
cial functioning is needed to identify appropriate cognitive targets for (expressed as chlorpromazine equivalents) or percentage of patients
treatment (Lindenmayer et al., 2008). An alternative explanation of this among groups taking typical, atypical antipsychotics or both. This
finding may be that other non-specific factors related to participation in maximized the likelihood that the differences in outcome variables
cognitive rehabilitation, independent from the modalities adopted, may detected in the patient groups may be relevant to the different
contribute to better clinical outcome. Among these potentially therapeutic rehabilitation modalities delivered.
factors are improved self-efficacy, self-esteem and self-confidence in Even with these limitations, the results of this study indicate that
completing cognitive training exercises (McGurk et al., 2009). In our cognitive remediation may be effective on clinical, neuropsychologic
study, we also addressed the question of the relationship between and functional outcome variables in schizophrenia, and that some of
improved cognitive function and greater psychosocial functioning by the functional outcome changes may be modestly mediated by
analyzing whether functional improvement was correlated with im- improvement in specific cognitive domains. The transfer of the efficacy
provement in specific cognitive domains (mediator variable) and whether of cognitive remediation interventions obtained in the laboratory
these correlations were different in relation to the type of rehabilitation under experimental conditions to the real world may justify some
treatment (moderator variable). The results suggest that, in the present dilution of the effects of treatment or explain the more complex
study, improvement in cognition was a weak mediator of psychosocial and less consistent results obtained. However, the confirmation of
outcome improvement in patients undergoing cognitive remediation their effectiveness in a highly structured and integrated system of
treatment. The modest level of functional improvement specifically care, when compared with an actively treated control condition,
associated with the cognitive remediation interventions probably reduced emphasizes the relevance of even limited positive findings. The direct
the possibility of obtaining significant results from mediator analyses, and comparison of 2 different cognitive remediation approaches (an
likely explains why little association between cognitive and functional individual computerized method versus a group interactive method)
change was found. On the other hand, the moderator role of such a did not reveal systematic differences in effectiveness on several
relationship emerged just for CACR with the GAF scale, although the small outcome measures of schizophrenia. Thus no inference can be drawn
number of patients in the different treatment groups reduces the from the present study about the opportunity of a priori choice of
statistical power of the analyses. The lack of consistent correlations specific remediation interventions in schizophrenia.
between cognitive and psychosocial functioning changes in this study Additional studies are needed on the comparative efficacy of
may suggest that cognitive improvement may be not a direct or consistent different cognitive remediation interventions, their effect on patient
mediator of functional outcome improvement in schizophrenia, as functioning and their long-term efficacy in the naturalistic settings of
suggested by other authors (Bowie et al., 2006, 2008; Aubin et al., 2009; care of schizophrenia. Specific effort should be put into studies
González-Blanch et al., 2010), at least in naturalistic treatment conditions, designed to better define the clinical, demographic and biological
where several factors within and outside the system of care may interfere predictors of effectiveness of different modalities of cognitive
with and dilute not only the effects of cognitive remediation but also those remediation in schizophrenia.
of cognitive improvement on social functioning. Although with the same
problem of poor statistical power, the multivariate analysis of factors
Role of funding source
affecting change in psychosocial functioning revealed a more robust effect Funding for this study was partially provided by an unrestricted grant from the
of symptomatological change than of cognitive change on outcome at Lombardia Region (project TR11 and project 195) and by a 60% grant from the
least for the GAF scale, while the change in the HoNOS scale may capture a University of Brescia (School of Medicine).
more complex interplay of patient, illness or treatment characteristics.
This may put into question the use of GAF as the sole scale for measuring Contributors
psychosocial functioning and functional outcome of schizophrenia and the AV designed the project, reviewed and discussed the data and statistical analyses, and
need to better define and measure the dimensions of functional outcome the final version of the paper; LDP participated in the design of the study, in the choice of
of the disease (Naber and Vita, 2004; Nasrallah et al., 2005) before assessment scales and tests, reviewed the methodology of remediation interventions and
of scale completion by the centres and wrote the first draft of the paper; GD administered
drawing definitive conclusions on the strength and nature of the
and scored neuropsychological tests, prepared the database and participated in the
relationships between cognitive and functional improvement in analyses; SB, PC, RP and EA followed patients in the rehabilitative interventions; BMC
schizophrenia. designed and undertook the statistical analyses; ES participated in the design of the
This study suffers from several limitations. In particular, it was not project, and discussion of the data and manuscript. All authors contributed to and
possible to make the study a fully blinded design, a limitation common approved the final manuscript.

to many studies on psychosocial treatments. In our study, the treating

psychiatrists and the team were not fully blinded to the assigned Conflict of interest
rehabilitative treatment. However, particular attention was given not There are no conflicts of interest to report in relation to the content of this study.
230 A. Vita et al. / Schizophrenia Research 133 (2011) 223–231

Acknowledgements Hogarty, G.E., Greenwald, D.P., Eack, S.M., 2006. Durability and mechanism of effects of
The Authors wish to thank S. Baresi, M. Bonomi, R. Briganti, L. Cadeddu, M. Goiza, E. cognitive enhancement therapy. Psychiatr. Serv. 57 (12), 1751–1757.
Jaeger, J., Tatsuoka, C., Berns, S., Varadi, F., Czobor, P., Uzelac, S., 2006. Associating
Lodi, A. Mainardi, D. Massussi , G. Peviani, A. Puerari, G. Roselli, M. Tinti, for their
functional recovery with neurocognitive profiles identified using partially ordered
participation in the clinical work with patients included in the study, and Lorna O’Brian
classification models. Schizophr. Res. 85 (1–3), 40–48.
for the revision of english language. Kay, S.R., Fiszbein, A., Opler, L.A., 1987. The Positive and Negative Syndrome Scale
(PANSS) for schizophrenia. Schizophr. Bull. 13, 261–276.
Keefe, R.S., Bilder, R.M., Harvey, P.D., Davis, S.M., Palmer, B.W., Gold, J.M., Meltzer, H.Y.,
References Green, M.F., Miller, D.D., Canive, J.M., Adler, L.W., Manschreck, T.C., Swartz, M.,
Rosenheck, R., Perkins, D.O., Walker, T.M., Stroup, T.S., McEvoy, J.P., Lieberman, J.A.,
Aleman, A., Bocker, K.B., Hijman, R., de Haan, E.H., Kahn, R.S., 2003. Cognitive basis of 2006. Baseline neurocognitive deficits in the CATIE schizophrenia trial. Neuropsy-
hallucinations in schizophrenia: role of top-down information processing. chopharmacology 31 (9), 2033–2046.
Schizophr. Res. 64, 175–185. Knowles, E.E., David, A.S., Reichenberg, A., 2010. Processing speed deficits in
American Psychiatric Association, 2000. Diagnostic and Statistical Manual of Mental schizophrenia: reexamining the evidence. Am. J. Psychiatry 167 (7), 828–835.
Disorders, Fourth Edition, Text Revision. American Psychiatric Association, Lecardeur, L., Stip, E., Giguere, M., Blouin, G., Rodriguez, J.P., Champagne-Lavau, M.,
Washington, DC. 2009. Effects of cognitive remediation therapies on psychotic symptoms and
Aubin, G., Stip, E., Gélinas, I., Rainville, C., Chapparo, C., 2009. Daily activities, cognition and cognitive complaints in patients with schizophrenia and related disorders: a
community functioning in persons with schizophrenia. Schizophr. Res. 107 (2–3), randomized study. Schizophr. Res. 111, 153–158.
313–318. Lee, J., Park, S., 2005. Working memory impairments in schizophrenia: a meta-analysis.
Bark, M., Revheim, N., Huq, F., Kaldherov, V., Ganz, Z.V., Medalia, A., 2003. The impact of J. Abnorm. Psychol. 114 (4), 599–611.
cognitive remediation on psychiatric symptoms in schizophrenia. Schizophr. Res. Lindenmayer, J.P., McGurk, S.R., Mueser, T., Khan, A., Wance, D., Hoffman, L., Wolfe, R.,
63 (3), 229–235. Xie, H., 2008. Randomized controlled trial of cognitive remediation among
Bech, P., Bille, J., Waarst, S., Wiese, M., Borberg, L., Treufeldt, P., Kessing, L., 2006. Validity inpatients with persistent mental illness. Psychiatr. Serv. 59, 241–247.
of HoNOS in identifying frequently hospitalized patients with ICD-10 mental Lora, A., Bai, G., Bianchi, S., Bolongaro, G., Civenti, G., Erlicher, A., 2001. La versione
disorders. Acta Psychiatr. Scand. 113, 485–491. italiana della HoNOS (Health of the Nation Outcome Scales), una scala per la
Bellucci, D.M., Glaberman, K., Haslam, N., 2003. Computer-assisted cognitive rehabil- valutazione della gravità e dell'esito nei servizi di salute mentale. Epidemiol.
itation reduces negative symptoms in the severely mentally ill. Schizophr. Res. 59 Psichiatr. Soc. 10, 198–212.
(2–3), 225–232. McClelland, R., Trimble, P., Fox, M.L., Stevenson, M.R., Bell, B., 2000. Validation of an
Benton, A.L., Hamsher, K.D., 1989. Multilingual Aphasia Examination. AJA Associates, outcome scale for use in adult psychiatric practice. Qual. Health Care 9, 98–105.
Iowa City, IA. McGurk, S.R., Twamley, E.W., Sitzer, D.I., McHugo, G.J., Mueser, K.T., 2007. A meta-
Bowie, C.R., Reichenberg, A., Patterson, T.L., Heaton, R.K., Harvey, P.D., 2006. De- analysis of cognitive remediation in schizophrenia. Am. J. Psychiatry 164 (12),
terminants of real-world functional performance in schizophrenia subjects: 1791–1802.
correlations with cognition, functional capacity, and symptoms. Am. J. Psychiatry McGurk, S.R., Mueser, K.T., DeRosa, T.J., Wolfe, R., 2009. Work, recovery, and
163 (3), 418–425. comorbidity in schizophrenia: a randomized controlled trial of cognitive
Bowie, C.R., Leung, W.W., Reichenberg, A., McClure, M.M., Patterson, T.L., Heaton, R.K., remediation. Schizophr. Bull. 35 (2), 319–335.
Harvey, P.D., 2008. Predicting schizophrenia patients' real-world behavior with Milev, P., Ho, B.C., Arndt, S., Andreasen, N.C., 2005. Predictive values of neurocognition
specific neuropsychological and functional capacity measures. Biol. Psychiatry 63 and negative symptoms on functional outcome in schizophrenia: a longitudinal
(5), 505–511. first-episode study with 7-year follow-up. Am. J. Psychiatry 162 (3), 495–506.
Brenner, H., Roder, V., Hodel, B., Kienzie, N., Reed, D., Liberman, R., 1994. Integrated Morrison, G., O'Carroll, R., McCreadie, R., 2006. Long-term course of cognitive
Psychological Therapy for Schizophrenic Patients. Hogrefe & Huber, Seattle, WA. impairment in schizophrenia. Br. J. Psychiatry 189, 556–557.
Brenner, V., Roder, V., Hodel, B., Kienzle, N., Invernizzi, G., Vita, A., 1997. Terapia Naber, D., Vita, A., 2004. Tools for measuring clinical effectiveness. Eur. Neuropsycho-
Psicologica Integrata (ITP). Programma per la Riabilitazione del Paziente Schizo- pharmacol. 14 (Suppl. 4), 435–444.
frenico. McGraw-Hill, Milan. (Italian edition edited by Vita, A., Invernizzi, G.). Nasrallah, H.A., Targum, S.D., Tandon, R., McCombs, J.S., Ross, R., 2005. Defining and
Buchy, L., Torres, I.J., Liddle, P.F., Woodward, T.S., 2009. Symptomatic determinants of measuring clinical effectiveness in the treatment of schizophrenia. Psychiatr. Serv.
insight in schizophrenia spectrum disorders. Compr. Psychiatry 50 (6), 578–583. 56 (3), 273–282.
Cavallaro, R., Anselmetti, S., Poletti, S., Bechi, M., Ermoli, E., Cocchi, F., Stratta, P., Vita, A., Niendam, T.A., Bearden, C.E., Zinberg, J., Johnson, J.K., O'Brien, M., Cannon, T.D., 2007.
Rossi, A., Smeraldi, E., 2009. Computer-aided neurocognitive remediation as an The course of neurocognition and social functioning in individuals at ultra high risk
enhancing strategy for schizophrenia rehabilitation. Psychiatry Res. 169 (3), 191–196. for psychosis. Schizophr. Bull. 33 (3), 772–781.
Dickinson, D., Ramsey, M.E., Gold, J.M., 2007. Overlooking the obvious: a meta-analytic Nuechterlein, K.H., Barch, D.M., Gold, J.M., Goldberg, T.E., Green, M.F., Heaton, R.K.,
comparison of digit symbol coding tasks and other cognitive measures in 2004. Identification of separable cognitive factors in schizophrenia. Schizophr. Res.
schizophrenia. Arch. Gen. Psychiatry 64 (5), 532–542. 72, 29–39.
Evans, J.D., Bond, G.R., Meyer, P.S., Kim, H.W., Lysaker, P.H., Gibson, P.J., Tunis, S., 2004. Parker, G., O'Donnell, M., Hadzi-Pavlovic, D., Proberts, M., 2002. Assessing outcome in
Cognitive and clinical predictors of success in vocational rehabilitation in community mental health patients: a comparative analysis of measures. Int. J. Soc.
schizophrenia. Schizophr. Res. 70 (2–3), 331–342. Psychiatry 48, 11–19.
Flashman, L.A., Green, M.F., 2004. Review of cognition and brain structure in Patterson, T.L., Leewenkamp, O.R., 2008. Adjunctive psychosocial therapies for the
schizophrenia: profiles, longitudinal course, and effects of treatment. Psychiatr. treatment of schizophrenia. Schizophr. Res. 100, 108–119.
Clin. North Am. 27, 1–18. Petrides, M., Milner, B., 1982. Deficits on subject-ordered tasks after frontal- and
Gigantesco, A., Vittorielli, M., Pioli, R., Falloon, I.R.H., Rossi, G., Morosini, P., 2006. The temporal-lobe lesions in man. Neuropsychologia 20, 249–262.
VADO approach in psychiatric rehabilitation: a randomized controlled trial. Pirkis, J., Burgess, M., Kirk, K., Dodson, S., Coombs, T.J., Williamson, M.K., 2005. A review
Psychiatr. Serv. 57, 1778–1783. of the psychometric properties of the Health of the Nation Outcome Scales
Giovagnoli, A.R., Del Pesce, M., Mascheroni, S., Simoncelli, M., Laiacona, M., Capitani, E., (HoNOS) family of measures. Health Qual. Life Outcomes 3, 76.
1996. Trail making test: normative values from 287 normal adult controls. Ital. J. Reitan, R.M., 1979. Trail Making Test: Manual for Administration and Scoring. Reitan
Neurol. Sci. 17, 305–309. Neuropsychology Laboratory, South Tucson, AZ.
González-Blanch, C., Perez-Iglesias, R., Pardo-García, G., Rodríguez-Sánchez, J.M., Reitan, R.M., Wolfson, D., 1993. The Halstead–Reitan Neuropsychological Test Battery:
Martínez-García, O., Vázquez-Barquero, J.L., Crespo-Facorro, B., 2010. Prognostic Theory and Clinical Interpretation, 2nd ed. Neuropsychology Press, Tucson, AZ.
value of cognitive functioning for global functional recovery in first-episode Roder, V., Mueller, D.R., Mueser, K.T., Brenner, H.D., 2006. Integrated psychological
schizophrenia. Psychol. Med. 40 (6), 935–944. therapy (IPT) for schizophrenia: is it effective? Schizophr. Bull. 32 (Suppl. 1),
Gopal, Y., Variend, H., 2005. First-episode schizophrenia: review of cognitive deficits S81–S93.
and cognitive remediation. Adv. Psychiatr. Treat. 11, 38–44. Sacchetti, E., Galluzzo, A., Panariello, A., Parrinello, G., Cappa, S.F., 2008. Self-ordered
Green, M.F., 1999. Interventions for neurocognitive deficits: editor's introduction. pointing and visual conditional associative learning tasks in drug-free schizophre-
Schizophr. Bull. 25 (2), 197–200. nia spectrum disorder patients. BMC Psychiatry 8, 6.
Green, M.F., Kern, R.S., Heaton, R.K., 2004. Longitudinal studies of cognition and Salvi, G., Leese, M., Slade, M., 2005. Routine use of mental health outcome assessments:
functional outcome in schizophrenia: implications for MATRICS. Schizophr. Res. 72 choosing the measure. Br. J. Psychiatry 186, 146–152.
(1), 41–51. Sartory, G., Zorn, C., Groetzinger, G., Windgassen, K., 2005. Computerized cognitive
Grynszpan, O., Perbal, S., Pelissolo, A., Fossati, P., Jouvent, R., Dubal, S., Perez-Diaz, F., remediation improves verbal learning and processing speed in schizophrenia.
2010. Efficacy and specificity of computer-assisted cognitive remediation in Schizophr. Res. 75, 219–223.
schizophrenia: a meta-analytical study. Psychol. Med. 41 (1), 163–173. Saykin, A.J., Shtasel, D.L., Gur, R.E., Kester, D.B., Mozley, L.H., Stafiniak, P., Gur, R.C., 1994.
Guy, W., 1976. ECDEU Assessment Manual for Psychopharmacology. US Dept of Neuropsychological deficits in neuroleptic naive patients with first-episode
Health, Education and Welfare. National Institute of Mental Health, Washington, schizophrenia. Arch. Gen. Psychiatry 51, 124–131.
DC, pp. 76–338. Simon, A.E., Cattapan-Ludewig, K., Zmilacher, S., Arbach, D., Gruber, K., Dvorsky, D.N.,
Heaton, R.K., 1981. The Wisconsin Card Sorting Test Manual. Psychological Assessment Roth, B., Isler, E., Zimmer, A., Umbricht, D., 2007. Cognitive functioning in the
Resources, Odessa, FL. (Italian translation and validation edited by Giunti schizophrenia prodrome. Schizophr. Bull. 33 (3), 761–771.
Organizzazioni Speciali, Florence, 2000). Spaulding, W.D., Reed, D., Sullivan, M., Richardson, C., Weiler, M., 1999. Effects of
Heaton, R.K., Gladsjo, J.A., Palmer, B.W., Kuck, J., Marcotte, T.D., Jeste, D.V., 2001. cognitive treatment in psychiatric rehabilitation. Schizophr. Bull. 25, 657–676.
Stability and course of neuropsychological deficits in schizophrenia. Arch. Gen. Twamley, E.W., Jeste, D.V., Bellack, A.S., 2003. A review of cognitive training in
Psychiatry 58 (1), 24–32. schizophrenia. Schizophr. Bull. 29 (2), 359–382.
 A. Vita et al. / Schizophrenia Research 133 (2011) 223–231 231

Vita, A., De Peri, L., Barlati, S., Cacciani, P., Cisima, M., Deste, G., Sacchetti, E., 2011. Wing, J.K., Beevor, A.S., Curtis, R.H., Park, S.B., Hadden, S., Burns, A., 1998. Health of the
Psychopathologic, neuropsychological and functional outcome measures during Nation Outcome Scales (HoNOS). Research and development. Br. J. Psychiatry 172,
cognitive rehabilitation in schizophrenia: a prospective controlled study in a real 11–18.
world setting. Eur. Psychiatry 26, 276–283. Woods, S.W., 2003. Chlorpromazine equivalent doses for the newer atypical
Wechsler, D., 1981. Wechsler Adult Intelligence Scale–Revised. Harcourt, Brace, antipsychotics. J. Clin. Psychiatry 64, 663–667.
Jovanovich, New York. Wykes, T., Huddy, V., Cellard, C., McGurk, S.R., Czobor, P., 2011. A meta-analysis of
Wexler, B.E., Bell, M.D., 2005. Cognitive remediation and vocational rehabilitation for cognitive remediation for schizophrenia: methodology and effect sizes. Am. J.
schizophrenia. Schizophr. Bull. 31 (4), 931–941. Psychiatry 168 (5), 472–485.