Schizophrenia Bulletin vol. 33 no. 4 pp.

912–920, 2007
doi:10.1093/schbul/sbm046
Advance Access publication on June 13, 2007

How Should DSM-V Criteria for Schizophrenia Include Cognitive Impairment?

Richard S. E. Keefe1,2 and Wayne S. Fenton3 Key words: schizophrenia/cognition/diagnosis

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at H.E.C. Bibliotheque Myriam & J. Robert Ouimet on June 7, 2015
2
Department of Psychiatry and Behavioral Sciences, Duke
University Medical Center, Box 3270, Durham, NC 27710;
3
Division of Adult Translational Research, National Institute of
Mental Health Introduction
Neurocognitive deficits of schizophrenia are profound
and clinically relevant. Patients with schizophrenia per-
Neurocognitive impairment is considered a core component form 1.5 to 2.0 standard deviation below healthy controls
of schizophrenia and is increasingly under investigation as on a variety of neurocognitive tasks. The most prominent
a potential treatment target. On average, cognitive impair- of these deficits are memory, attention, working memory,
ment is severe to moderately severe compared with healthy problem solving, processing speed, and social cognition.1
controls, and almost all patients with schizophrenia demon- These impairments exist prior to the initiation of antipsy-
strate cognitive decrements compared with their expected chotic treatment2 and are not caused by psychotic symp-
level if they had not developed the illness. Compared with toms in patients who are able to complete cognitive
patients with affective disorders, cognitive impairment in testing, which includes the overwhelming majority of
schizophrenia appears earlier, is more severe, and tends patients.3 The various cognitive deficits in schizophrenia
to be more independent of clinical symptoms. While the Di- have all been shown to be associated with functional out-
agnostic and Statistical Manual of Mental Disorders, comes such as difficulty with community functioning, dif-
Fourth Edition, Text Revision, description of schizophrenia ficulty with instrumental and problem-solving skills,
includes several references to cognitive impairment, neither reduced success in psychosocial rehabilitation programs,4
the diagnostic criteria nor the subtypology of schizophrenia and the inability to maintain successful employment.5 In
include a requirement of cognitive impairment. We forward fact, cognitive deficits are better able to explain important
for consideration a proposal that the Diagnostic and Sta- functional outcomes, such as work performance and in-
tistical Manual of Mental Disorders, Fifth Edition, criteria dependent living,6 than positive or negative symptoms.
include a specific criterion of ‘‘a level of cognitive function- The importance of cognitive deficits in schizophrenia
ing suggesting a consistent severe impairment and/or a sig- goes beyond their severity and relation to functional out-
nificant decline from premorbid levels considering the comes. Cognitive deficits appear to be present in some
patient’s educational, familial, and socioeconomic back- patients with schizophrenia prior to the onset of psycho-
ground.’’ The inclusion of this criterion may increase the sis and are correlated with measurable brain dysfunction
‘‘point of rarity’’ with affective psychoses and may increase more than any other aspect of the illness. While the num-
clinicians’ awareness of cognitive impairment, potentially ber of studies associating negative or positive symptoms
leading to more accurate prognosis and better treatment with abnormal brain imaging results is small, the imaging
outcomes. Future research will need to address the validity literature in schizophrenia is filled with associations be-
of these possibilities. The reliable determination of cogni- tween cognitive deficits and structural and functional im-
tive impairment as part of a standard diagnostic evaluation aging results that differ from healthy controls. Perhaps
may present challenges to diagnosticians with limited re- most importantly, cognition is increasingly considered
sources or insufficient expertise. Various cognitive assess- as a primary target for treatment.7–10
ment methods for clinicians, including brief assessments Despite the relevance of cognitive impairment to biol-
and interview-based assessments, are discussed. Given ogy, function, and treatment in schizophrenia, it is not
the current emphasis on the development of cognitive treat- included in the Diagnostic and Statistical Manual of Men-
ments, the evaluation of cognition in schizophrenia is an tal Disorders, Fourth Edition (DSM-IV), criteria. It is
essential component of mental health education. noteworthy, however, that the first sentence of the de-
scription of schizophrenia in DSM-IV includes 4 referen-
ces to cognitive disturbances: ‘‘the characteristics of
1
To whom correspondence should be addressed; tel: 919-684- schizophrenia involve a range of cognitive and emotional
4306, fax: 919-684-2632, e-mail: [email protected]. dysfunctions that include perception, inferential thinking,
Ó The Author 2007. Published by Oxford University Press on behalf of the Maryland Psychiatric Research Center. All rights reserved.
For permissions, please email: [email protected].
912
 Cognitive Impairment in Schizophrenia

language and communication, behavioral monitoring, af- Schizophrenia Major Depressive Disorder Euthymic Bipolar Disorder

fect, fluency, and production of thought and speech, he- 0.5

Z-Score (SD Units)

donic capacity, volition and drive, and attention.’’11 0
Thus, it is clear that cognition is deemed important by
diagnostic experts; however, a method for including -0.5

this fundamental aspect of the illness in the diagnostic cri- -1

teria for schizophrenia has not been determined. The cur- -1.5
rent review will emphasize the importance of cognition in
-2
schizophrenia and forward a proposal for consideration

Vis Mem

Fluency

Trails B

WCST

BD

Voc
Verb Mem

Verb Mem
that severe cognitive impairment should be part of the

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at H.E.C. Bibliotheque Myriam & J. Robert Ouimet on June 7, 2015
(D)

(I)
criteria for schizophrenia in Diagnostic and Statistical
Manual of Mental Disorders, Fifth Edition (DSM-V). A
research agenda for determining the validity and useful- Fig. 1. Cognitive Profiles in Schizophrenia, Major Depression, and
ness of including cognitive impairment as part of the cri- Euthymic Bipolar Disorder From Published Meta-analyses.16 Data
teria for schizophrenia will be discussed. from Heinrichs and Zakzanis,44 Zakzanis et al,14 and van Gorp
et al15 Healthy group mean 5 0. Verb Mem (D), delayed verbal
memory; Verb Mem (I), immediate verbal memory; Vis Mem, visual
Will Cognitive Impairment Help Distinguish the Diagnosis memory; Trails B, Trail Making Test, B; WCST, Wisconsin Card
of Schizophrenia From Affective Disorders? Sorting Test; BD, Wechsler Adult Intelligence Scale (WAIS) block
design test; Voc, WAIS vocabulary. Reprinted with permission from
The first question that will be considered is whether add- Buchanan et al.10.
ing some definition of cognitive impairment or cognitive
decline to the criteria for schizophrenia will help define ecutive control, visual memory, mental speed, and verbal
a ‘‘point of rarity’’ with affective psychoses.12 The ability memory.13 Even when patients with schizophrenia and
of a diagnostic refinement to improve the distinction be- patients with bipolar disorder were matched on the sever-
tween 2 entities and thus create an increased nonoverlap ity of their clinical symptoms, the deficits of schizophrenia
between them is considered to be a crucial determinant surpass those of patients with bipolar disorder by 0.5 SD.13
for inclusion. Studies of patients with first-episode schizophrenia
and affective disorder appear to support the meta-
Diagnostic Differences in Severity of Cognitive analyses completed on more chronic patients. In an epide-
Impairment miological study of all first-admission psychotic disorders
The conclusions from cognitive experts in the Measure- in Suffolk County, NY, patients who received a diagnosis
ment And Treatment Research to Improve Cognition in of schizophrenia at 24 months of follow-up (n = 148) were
Schizophrenia (MATRICS) project were that ‘‘schizo- found to have significantly greater cognitive deficits com-
phrenia and schizoaffective disorder share a similar pat- pared with those first-episode psychotic patients who
tern of cognitive impairments, which is distinct from were diagnosed with bipolar disorder (n = 87) and de-
patterns in major depression, bipolar disorder, and Alz- pression (n = 56) 24 months later. Again, the differenti-
heimer’s dementia.’’10 This group of experts came to this ation between schizophrenia and affective psychoses were
conclusion based upon a series of studies indicating that particularly profound with regard to memory, executive
patients with schizophrenia have a pattern of deficits that functions, and mental speed tasks (A. Reichenberg, PhD,
is more profound than those in major depression and bi- unpublished data, 2007). These data suggest that cogni-
polar disorder, more stable over the course of illness, and tive information at first episode may aid in the determi-
more related to clinical state. Meta-analyses of the cog- nation of whether an individual’s later diagnosis will be in
nitive profiles of patients with schizophrenia, major de- the affective or schizophrenia spectrum.
pression, and bipolar disorder are described in figure 1.
Patients with schizophrenia have more cognitive im- Diagnostic Differences Regarding Relation of Cognitive
pairment on all the cognitive tests that were measured Impairment to Clinical State
in each of the diagnostic groups. While the pattern of def- While patients with affective psychoses also have cogni-
icits among these groups may not differ dramatically, it is tive impairment, it appears as though these cognitive def-
well accepted that the deficits of schizophrenia are more icits are more strongly associated with clinical symptoms
profound than those in affective disorders.10,13 A recent and state-related factors than in patients with schizophre-
meta-analysis comparing the performances of patients nia.14,15 In a study of patients with schizophrenia or bi-
with schizophrenia and bipolar disorder concluded that polar disorder who were assessed when psychotic at
patients with schizophrenia have cognitive deficits that baseline and then 8 months later, patients who were psy-
are about 0.5 SD larger than those in patients with bipo- chotic at follow-up in both diagnostic groups had no dif-
lar disorder. These deficits were found to be particularly ference in their cognitive impairment 8 months later.
profound on tests of verbal fluency, working memory, ex- Among those patients whose psychosis had remitted
913
R. S. E. Keefe & W. S. Fenton

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at H.E.C. Bibliotheque Myriam & J. Robert Ouimet on June 7, 2015
Fig. 2. Distributions of Total Scores on the Repeatable Battery for the Assessment of Neuropsychological Status in Patients With
Schizophrenia and Healthy Controls From Published Norms. Reprinted with permission from Randolph et al22 and Wilk et al.23

8 months later, schizophrenia patients also showed the there is considerable overlap between these 2 distribu-
same level of cognitive impairment. Only the bipolar pa- tions, it is noteworthy that there are very few healthy con-
tients whose psychosis had remitted at follow-up had im- trols at the lower ends of this distribution and very few
proved in their cognitive performance.16 Similar data have schizophrenia patients at the upper ends of this distribu-
been reported in first-episode samples. While first- tion. Traditional neuropsychological criteria for cognitive
episode patients with affective psychoses performed simi- impairment would identify those individuals who per-
larly to those with first-episode schizophrenia in 1 study, formed better than 1 SD below the healthy control
patients with nonpsychotic affective disorders performed mean as ‘‘unimpaired’’.24–26 By these criteria, about 20%
significantly better than both psychotic groups.17 Thus, of the patients in this study would be considered to have
while the cognitive deficits of affective disorders may normal cognitive functions. However, it is possible that
be profound in some cases, these cognitive deficits appear many of the individuals at the upper end of this schizophre-
to be related to clinical symptoms. In contrast, cognitive nia distribution have demonstrated cognitive decline com-
impairment in schizophrenia patients has been repeatedly pared with what their cognitive functions would have been
demonstrated to be uncorrelated with psychotic symp- if they had never developed the illness. While this conjec-
toms.3,18–20 Part of the explanation for these low correla- ture can never be proved, it is useful to investigate the re-
tions between cognitive deficits and symptoms in patients lationship between antecedent factors, such as parental
with schizophrenia is that while the symptoms of schizo- education and reading scores, and their relationship to cur-
phrenia wax and wane in almost all patients, leading to rent cognitive functions in patients with schizophrenia.
low stability coefficients over time,21 the stability of cog- In healthy controls, current cognitive ability is strongly
nitive deficits in all domains is very high, with test-retest predicted by antecedent factors such as maternal educa-
coefficients ranging between 0.7 and 0.85 even in patients tion and reading score.27 As demonstrated in figure 3, the
tested 1 year apart following their initial treatment for healthy controls whose mothers had greater education
psychosis.21 Thus, while there are cognitive deficits in af- clearly had higher cognitive functions. The regression
fective disorders, they fluctuate in parallel with clinical line in this figure describes the relationship between these
symptom changes. In schizophrenia, however, they 2 factors. Because of the natural variability of cognitive
may be the most stable aspect of the disorder. performance among healthy controls, about half of this
distribution performs above expectation while half per-
Prevalence of Cognitive Impairment in Schizophrenia forms below expectations. However, an overwhelming
If cognitive impairment is to be considered as part of the majority of the patients with schizophrenia in this sample
diagnosis of schizophrenia, it will be important to dem- performed below the expectations established by anteced-
onstrate that its prevalence among patients with schizo- ent factors, in this case, maternal education. Thus, it is
phrenia is high. Patients with schizophrenia and healthy likely that almost all schizophrenic patients have some
controls both show normal distributions of scores on measure of cognitive deficit compared with what their
cognitive batteries such as the Repeatable Battery for level of cognitive function would have been if they had
Assessment of Neuropsychological Status22 (see figure 2). never developed the illness.
However, as has been frequently demonstrated, the dis-
tribution of a large number of patients with schizophre- Early Cognitive Decline in Schizophrenia?
nia (n = 575) is shifted about 2 SDs below the 540 healthy While most patients with schizophrenia appear to show
controls from the standardization sample.22,23 While some cognitive decline based upon what would have been
914
 Cognitive Impairment in Schizophrenia

2 nitive performance that is slightly worse than their peers.

Group As childhood progresses, cognitive performance tends to
Cognitive Composite Score

1 Above Expectation
Patients (n = 150) worsen in those children who will eventually develop
0 Normal control schizophrenia. By the time psychosis develops in late ad-
regression line
Controls (n = 50)
olescence or early adulthood, patients perform substan-
-1
tially worse than their healthy peers. While patients with
-2 affective disorders also demonstrate cognitive impairment
in adulthood, it appears as though these individuals do not
-3
show impairment until the adult onset of their disorders.28
The literature review above supports the notion that

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at H.E.C. Bibliotheque Myriam & J. Robert Ouimet on June 7, 2015
-4 Below Expectation
the severity and longitudinal course of cognitive impair-
-5
0 5 10 15 20
ment in schizophrenia differ substantially from that
Maternal Education (years)* found in patients with affective psychosis. Yet, the ques-
tion of whether including a criterion of cognitive impair-
* Maternal education values jittered for clarify.
ment or cognitive decline from healthy premorbid levels
in the diagnosis of schizophrenia will help define a ‘‘point
Fig. 3. Expected Neurocognitive Performance Based on Maternal
Education of Healthy Controls. Reprinted with permission from
of rarity’’ with affective psychoses remains unanswered.
Keefe et al.27 Research studies and analyses of existing databases are
needed that address this question in large numbers of psy-
predicted by antecedent factors, it is also important to note chotic individuals with schizophrenia and affective disor-
that on average patients with schizophrenia start out at ders assessed on measures of cognition and symptoms.
a lower baseline prior to the onset of the illness. Children These studies will help determine whether the inclusion
who will eventually develop schizophrenia show cognitive of a criterion for cognitive impairment in the diagnosis
impairment compared with healthy controls and children of schizophrenia will increase the point of rarity between
who later develop affective disorders.28,29 However, indi- these diagnostic entities.
viduals who will eventually develop schizophrenia also
appear to show decline on scholastic measures between
early childhood and late adolescence.29 The presence of How Can Cognitive Impairment Be Assessed for
cognitive deficits or cognitive decline during adolescence Diagnostic Purposes?
has been found to predict the conversion to schizophrenia While formal cognitive testing appears to be very sensi-
in various samples.29–34 Thus, as depicted in figure 4, tive to the cognitive impairment in schizophrenia, the re-
patients with schizophrenia appear to begin life with cog- sources required to complete neuropsychological evaluations

Fig. 4. Standardized Scholastic Test Performance in Grades 4, 8, and 11 Relative to State Norms for 70 Subjects Who Later Developed
Schizophrenia. Reprinted with permission from Fuller et al.29

915
R. S. E. Keefe & W. S. Fenton

are prohibitive in various treatment settings. In fact, the methodology that assesses cognition with interviews of
diagnoses of Alzheimer disease and attention deficit/hy- patients and caregivers, such as relatives or caseworkers,
peractivity disorder (ADHD), while being clearly cogni- may have improved reliability and validity. For example,
tive disorders, do not require formal cognitive testing. the Schizophrenia Cognition Rating Scale (SCoRS) has
Thus, it may not be realistic to expect that the diagnosis been found to have excellent reliability and substantial
of schizophrenia would depend upon cognitive perfor- correlations with cognitive performance and functional
mance testing. It needs to be established how cognitive outcomes36 (M. F. Green, PhD, unpublished data,
impairment will be assessed by clinicians who will diag- 2007). In fact, SCoRS global outcome measures have
nose schizophrenia so that the additional criterion will met several of the criteria for coprimary measures out-
contribute sufficiently to diagnostic validity and treat- lined by the MATRICS meeting for optimal designs

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at H.E.C. Bibliotheque Myriam & J. Robert Ouimet on June 7, 2015
ment success. Recent work suggests that almost all the for cognitive enhancement trials.10 One of the potential
variance in cognitive composite scores can be accounted weaknesses of this methodology however, is that reports
for by a small number of tests.3 Thus, clinicians may be from patients have been found to have reduced reliability
able to develop the capacity to assess cognitive impair- if patients are the only source of information. The rele-
ment in schizophrenia without overwhelming resource vance of this weakness is particularly important in the
requirements. However, education and training on the assessment of patients with schizophrenia because a sub-
use of standardized cognitive tests for clinicians will be stantial percentage of patients do not have an available
essential to assure that the assessment procedures are informant who can provide information about the pa-
completed in a manner that maintains test standardiza- tient’s cognitive deficits and how these deficits affect the
tion. This aspect of training is usually included in the cur- patient’s daily behavior. For example, in the MATRICS
riculum of clinical psychologists and neuropsychologists Psychometric and Standardization Study (PASS), test-
but is rarely a component of education for physicians, retest reliability data over the course of 1 month was
social workers, and nurses. A program for training in high when ratings were based upon a patient and infor-
cognitive testing will be an essential step to increase mant as a source of information (intra class correlation
the capacity for clinicians to assess the cognitive impair- [ICC] = 0 .82, goodness of fit [GF] = 123; P = .001);
ments of schizophrenia. It will take time to work this (M. F. Green, PhD, unpublished data, 2007). However,
training into the traditional education of psychiatrists when patients were the only source of information, the
and other nonpsychologists. This training does not need test-retest reliability coefficient (ICC) was 0.60, (GF
to be limited to formal neuropsychological tests but = 148, P < .001). While reliability is enhanced when in-
may be better aimed toward the assessment of patients’ formation can be obtained from both sources, a consider-
ability on practical cognitive tasks, which may have stron- able number of patients do not have an available
ger direct correlations with outcome.35 informant. A more extensive series of questions, as found
in the Clinical Global Impression of Cognition in Schizo-
phrenia (CGI-CoGS),37 appears to improve the reliabil-
Interview-Based Assessments of Cognition ity of patient reports up to ICC = 0.80, but patients
While the availability of formal test methods and trained describe these longer interviews, which require up to
testers may prohibit testing in many clinical environ- 45 minutes per interview, as burdensome (M. F. Green,
ments, recent methodological advances have included PhD, unpublished data, 2007). A shorter, less burden-
the assessment of cognition in patients with schizophre- some instrument that would not require an informant
nia with interview-based measures. Similar to ADHD as- and could be completed on almost all patients would
sessment methods, which do not involve formal testing, be ideal, although is not currently available. Future stud-
these measures involve a series of questions directed to- ies should focus on this methodology and must also
ward the patient with schizophrenia and his or her rela- determine whether interview-based assessments of cogni-
tives or caregivers. These questions address whether tion can contribute to the diagnostic separation between
people with schizophrenia have cognitive deficits that schizophrenia and affective psychoses.
impair fundamental aspects of their daily lives. For
instance, some of the questions ask whether patients
Is the Clinical Importance of Cognition in Schizophrenia
have difficulty remembering names, concentrating well
Sufficient to Include It in the Formal Criteria?
enough to read a newspaper or book, being able to follow
group conversations, and handling changes in daily rou- Even if including a definition of ‘‘cognitive impairment’’
tines.36,37 Interview-based assessments of cognition have in the criteria for schizophrenia does not increase the
historically been unreliable and have demonstrated low point of rarity between schizophrenia and other psy-
correlations with cognitive performance. However, these chotic disorders, an additional consideration is whether
measures have generally relied upon the reports from it would be able to ‘‘provide useful information not con-
patients and their treating clinicians, which have been tained in the definition of the disorder that helps in deci-
notoriously unreliable and potentially invalid.38–41 A sions about management and treatment.’’42 The inclusion
916
 Cognitive Impairment in Schizophrenia

of cognitive impairment in the criteria for schizophrenia This change in DSM will potentially increase the point
may increase psychiatrists’ attention toward a neglected of rarity with other psychoses. It is likely that some
aspect of the core components of schizophrenia.7 Because patients diagnosed with schizophrenia who have little
cognition is rarely considered among psychiatrists as an or no cognitive impairment have treatment responses
important treatment target, inclusion of cognitive impair- and courses of illness that are more consistent with a di-
ment in the criteria for schizophrenia may help to educate agnosis of affective disorder. If this is the case, it will ben-
clinicians about the importance of cognition in their efit clinicians to change their expectations based upon
treatment options. Furthermore, representatives from this revised diagnosis. On the other hand, some patients
the US Food and Drug Administration have indicated diagnosed with affective disorders and severe cognitive
that the recognition of cognitive impairment in the diag- impairment may follow the longitudinal course and treat-

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at H.E.C. Bibliotheque Myriam & J. Robert Ouimet on June 7, 2015
nostic nomenclature would be an important step in ap- ment response of patients with schizophrenia. One of the
proving a drug for a cognitive improvement indication.10 important research questions that will need to be ad-
Many government agencies and pharmaceutical compa- dressed is whether patients whose diagnosis changes
nies are currently involved in intense work to try to develop based upon the new criteria are more likely to have ge-
compounds that may improve cognition with schizophre- netic and other biological indicators consistent with
nia. If successful, these compounds have the potential to the new diagnosis.
alter the way that schizophrenia is currently treated. Changing the DSM criteria for schizophrenia to include
However, if the pathway is not established to allow these cognitive impairment will also force clinicians to consider
medications to be approved by FDA, and if clinicians the cognitive impairment of their patients, which has been
are not trained to recognize cognitive improvement, largely ignored among clinical psychiatrists. This change
this potential area of great benefit to patients may be would thus direct clinicians’ attention toward the aspect
missed. Inclusion of cognitive impairment in the diagnos- of the disorder that is the largest determinant of long-
tic criteria for schizophrenia may be one of the steps that term functioning. It may also help develop the pathway
could be taken to help clinicians target and potentially for new treatments to improve this fundamental compo-
improve cognition in patients with schizophrenia. nent of the illness and force educational systems to teach
clinicians how to recognize cognitive impairment and
improvement.
How Should Cognitive Impairment in Schizophrenia Be
However, the implementation of this change in DSM
Defined?
will present several challenges. If this criterion is included
We have presented an argument for the importance of in the criteria for schizophrenia, it will be crucial to con-
cognitive impairment in schizophrenia and have sug- sider how cognition will be measured by clinicians and
gested that cognitive impairment should be represented researchers making a diagnosis. It is unrealistic to expect
in the diagnosis for schizophrenia. However, there are that all patients with schizophrenia would receive formal
several considerations regarding how it should be in- neuropsychological testing by psychologists, which is
cluded. We propose that the following criterion should time consuming and expensive. In most treatment set-
be considered for the diagnosis of schizophrenia: ‘‘a level tings, these costs are prohibitive. However, if cognitive
of cognitive functioning suggesting a consistent severe paradigms were developed that were able definitively
impairment and/or a significant decline from premorbid to separate diagnostic entities, a case could be made
levels considering the patient’s educational, familial, and that this testing is essential to patient diagnosis and treat-
socioeconomic background.’’ Diagnosticians should con- ment planning. Unfortunately, as discussed above, we are
sider all aspects of cognitive impairment in this definition not yet at this stage.
but should be alerted that, in general, schizophrenia A second consideration regarding the use of cogni-
patients may have particularly severe deficits in the cog- tive impairment as a criterion for schizophrenia is that
nitive domains of memory, attention, working memory, current cognitive performance is affected by factors un-
reasoning and problem solving, processing speed, and related to cognitive decline in patients with schizophrenia
social cognition.1 It is not uncommon for some aspects such as level of education and environments that are var-
of cognition to be unimpaired in the context of severe iably conducive to normal learning.23 Some patients may
impairments in other areas, with an overall level of im- have very poor cognitive functioning due to factors un-
pairment in the severe range. A statement that the assess- related to schizophrenia while other patients may have
ment of cognitive function must consider the patients’ cognitive performance that is in the ‘‘normal range’’ de-
background was included to avoid overdiagnosing schizo- spite significant decline from premorbid levels. How will
phrenia in individuals whose environments deprive them diagnosticians determine how schizophrenia may interact
of their ability to develop cognitive abilities. If in the with these factors to result in a patient’s current cognitive
event that DSM-V changes to a completely dimensional levels? Since not all patients are defined as ‘‘impaired’’ on
approach to the symptoms of psychosis,43 cognitive im- cognitive tests, it is important to emphasize that the
pairment should be one of the key dimensions. criterion will be met if a patient’s current cognitive
917
R. S. E. Keefe & W. S. Fenton

performance represents a ‘‘decline from premorbid cogni- be very similar in patients with schizophrenia and those
tive functioning’’. On average, the longitudinal course of with affective psychoses.10,14,15 Thus, for the patient to
cognitive function in patients with schizophrenia appears meet the criterion of cognitive impairment, it will be im-
to decline at least one full SD from childhood. During portant for the cognitive deficits to be stable throughout
childhood and adolescence, patients who will eventually a long period of illness. This would help to differentiate
develop schizophrenia perform about 0.5 SDs below the cognitive impairment found in schizophrenia from
their peers who will not develop schizophrenia.28,29,31 Im- those in affective psychoses. However, it will also result
mediately prior to the onset of psychosis, patients who in delays in definitive diagnoses in cases where cog-
are about to develop schizophrenia demonstrate a wors- nitive impairment is present in the context of symptom
ened cognitive function, such that the average person at exacerbation.

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at H.E.C. Bibliotheque Myriam & J. Robert Ouimet on June 7, 2015
ultra high risk for schizophrenia disorders who will even- In sum, we have recommended for consideration that
tually convert to psychosis performs about 1 SD below a criterion for consistent severe cognitive impairment be
healthy controls.30,34 It will be important in these cases added to the DSM diagnosis of schizophrenia. There
for diagnosticians to determine whether there has been are several challenges for this suggestion to meet accep-
a decline in cognitive functions from expected cognitive tance by the research and clinical communities. Research
levels based upon antecedent factors such as parental ed- is needed to determine: if such a criterion will increase the
ucation, early school performance, and reading level. It point of rarity between schizophrenia and other diagnostic
will be essential for diagnosticians to collect a complete entities; if clinicians are able assess cognition reliably with
history on the cognitive performance of each patient, in- brief formal assessment instruments or interview-based
cluding how the patient’s current cognitive performance methods; and if the inclusion of such a criterion will im-
compares to early school performance and any academic, prove the value of the diagnosis of schizophrenia for prog-
intelligence, or cognitive testing that was performed dur- nosis and treatment outcomes.
ing premorbid and prodromal periods. Further, a pa-
tient’s level of cognitive performance will need to be
Acknowledgments
compared with other members of the patient’s family
and sociocultural background, if available. In some cases, This article was generated from a meeting on
testing would benefit this assessment. In other cases, the ‘‘Deconstructing Psychosis’’ at the offices of the
amount of cognitive impairment in a patient would be American Psychiatric Association in Alexandria,
clearly obvious and in direct contrast to early cognitive Virginia, on February 16–17, 2006. In that meeting, Dr
competence in an individual. Finally, because cognitive Keefe presented many of the ideas discussed in this
impairment in affective disorders in the context of clinical article, and they were commented on formally by Dr
exacerbation may be difficult to distinguish from schizo- Fenton and informally by other panel participants.
phrenia cross-sectionally, longitudinal assessment will be While Dr Fenton agreed to coauthor this article, he
important for an accurate diagnosis. While this historical was not able to make comments on the manuscript
and longitudinal data collection may initially appear to before his tragic death on September 2, 2006. Dr
add burden, if indeed the level and course of cognitive Keefe discloses that he has received research funding
deficit is crucial not only to diagnosis but also to progno- from Eli Lilly and Pfizer through Duke University. He
sis and treatment planning, it is likely that this ‘‘front- also owns a company that trains and certifies cognitive
loading’’ of clinical care may actually reduce clinical testers for clinical trials, and he consults and has
burden in the form of improved treatment response received funding from various pharmaceutical
and long-term functioning. companies, universities, and government agencies to
Third, while clinician judgment will be an important carry out this work. He receives royalties through
component of assessing cognition in schizophrenia, Duke University for cognitive measures developed in
recent data suggest that clinicians cannot be the sole his laboratory, including the Brief Assessment of
source of information for making this determination. Cognition in Schizophrenia (BACS) and the BACS
The challenge that arises here is that many patients symbol coding subtest of the MATRICS Consensus
with schizophrenia will not have enough contact with Cognitive Battery, some of which are discussed here.
other people for someone to be able to report reliably Currently, there is no fee charged for the use of the
on their usual level of cognitive functioning. Patients SCoRS, copyrighted by Duke University, discussed in
without available informants will need to have additional this article; however, in the future, a fee may be
assessments such as more extensive interviews or an ac- charged. Dr Keefe has devoted much of his career to
tual cognitive assessment, which is the most informative research on cognitive impairment in schizophrenia.
method for collecting cognitive information about a Thus, if the suggestions of this article are carried out,
patient. he stands potentially to benefit academically and
As discussed above, if a patient is assessed during a pe- financially. Courtney Kennel and Cathy Lefebvre
riod of clinical exacerbation, cognitive impairment may provided editorial assistance for this article.
918
 Cognitive Impairment in Schizophrenia

Richard Keefe: Potential financial conflicts for myself in 16. Harvey PD, Docherty NM, Serper MR, Rasmussen M. Cog-
this paper include: I receive research funding from Eli Lilly nitive deficits and thought disorder: II. An 8-month followup
study. Schizophr Bull. 1990;16:147–156.
and Pfizer through Duke University. I also own a company
17. Albus M, Hubmann W, Wahlheim C, Sobizack N, Franz U,
that trains and certifies cognitive testers for clinical trials. I
Mohr F. Contrasts in neuropsychological test profile be-
consult and have received funding from various tween patients with first-episode schizophrenia and first-
pharmaceutical companies, universities and government episode affective disorders. Acta Psychiatr Scand. 1996;94:
agencies to carry out this work. I receive royalties 87–93.
through Duke University for cognitive measures 18. Addington J, Addington D. Neurocognitive and social func-
developed in my lab, including the Brief Assessment. tioning in schizophrenia: a 2.5 year follow-up study. Schiz-
ophr Res. 2000;44:47–56.

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at H.E.C. Bibliotheque Myriam & J. Robert Ouimet on June 7, 2015
19. Hughes C, Kumari V, Soni W, et al. Longitudinal study of
symptoms and cognitive function in chronic schizophrenia.
Schizophr Res. 2003;59:137–146.
References
20. Strauss ME. Relations of symptoms to cognitive deficits in
1. Nuechterlein KH, Barch DM, Gold JM, Goldberg TE, Green schizophrenia. Schizophr Bull. 1993;19:215–231.
MF, Heaton RK. Identification of separable cognitive factors 21. Bilder RM, Goldman RS, Robinson D, et al. Neuropsychol-
in schizophrenia. Schizophr Res. 2004;72:29–39. ogy of first-episode schizophrenia: initial characterization and
2. Saykin AJ, Shtasel DL, Gur RE, et al. Neuropsychological clinical correlates. Am J Psychiatry. 2000;157:549–559.
deficits in neuroleptic naive patients with first-episode schizo- 22. Randolph C. RBANS Manual-Repeatable Battery for the Assess-
phrenia. Arch Gen Psychiatry. 1994;51:124–131. ment of Neuropsychological Status. PAR, Inc., Lutz, FL. 1998.
3. Keefe RSE, Bilder RM, Harvey PD, et al. Baseline neurocog- 23. Wilk CM, Gold JM, Humber K, Dickerson F, Fenton WS,
nitive deficits in the CATIE schizophrenia trial. Neuropsycho- Buchanan RW. Brief cognitive assessment in schizophrenia:
pharmacology. 2006;31:2033–2046. normative data for the repeatable battery for the assessment
4. Green MF, Kern RS, Braff DL, Mintz J. Neurocognitive def- of neuropsychological status. Schizophr Res. 2004;70:
icits and functional outcome in schizophrenia: are we measur- 175–186.
ing the ‘‘right stuff’’? Schizophr Bull. 2000;26:119–136. 24. Bryson GJ, Silverstein ML, Nathan A, Stephen L. Differen-
5. Bryson G, Bell MD. Initial and final work performance in tial rate of neuropsychological dysfunction in psychiatric
schizophrenia: cognitive and symptom predictors. J Nerv disorders: comparison between the Halstead-Reitan and Luria-
Ment Dis. 2003;191:87–92. Nebraska batteries. Percept Mot Skills. 1993;76:305–306.
25. Heinrichs RW, Awad AG. Neurocognitive subtypes of
6. Harvey PD, Howanitz E, Parrella M, et al. Symptoms, cogni-
chronic schizophrenia. Schizophr Res. 1993;9:49–58.
tive functioning, and adaptive skills in geriatric patients with
lifelong schizophrenia: a comparison across treatment sites. 26. Palmer BW, Heaton RK, Paulsen JS, et al. Is it possible to be
Am J Psychiatry. 1998;155:1080–1086. schizophrenic yet neuropsychologically normal? Neuropsy-
chology. 1997;11:437–446.
7. Hyman SE, Fenton WS. Medicine. What are the right targets
for psychopharmacology? Science. 2003;299:350–351. 27. Keefe RS, Eesley CE, Poe MP. Defining a cognitive func-
tion decrement in schizophrenia. Biol Psychiatry. 2005;57:
8. Gold JM. Cognitive deficits as treatment targets in schizo-
688–691.
phrenia. Schizophr Res. 2004;72:21–28.
28. Cannon M, Caspi A, Moffitt TE, et al. Evidence for early-
9. Davidson M, Keefe RSE. Cognitive impairment as a target childhood, pan-developmental impairment specific to schizo-
for pharmacological treatment in schizophrenia. Schizophr phreniform disorder: Results from a longitudinal birth
Res. 1995;17:123–129. cohort. Arch Gen Psychiatr. 2002;59:449–456.
10. Buchanan RW, Davis M, Goff D, et al. A summary of the 29. Fuller R, Nopoulos P, Arndt S, O’Leary D, Ho BC, Andrea-
FDA-NIMH-MATRICS workshop on clinical trial designs sen NC. Longitudinal assessment of premorbid cognitive
for neurocognitive drugs for schizophrenia. Schizophr Bull. functioning in patients with schizophrenia through examina-
2005;31:5–21. tion of standardized scholastic test performance. Am J Psy-
11. Diagnostic and Statistical Manual of Mental Disorders, 4th chiatry. 2002;159:1183–1189.
Edition, Text Revision. American Psychiatric Association, 30. Reichenberg A, Weiser M, Rapp MA, et al. Elaboration on
Inc., Arlington, VA 2000. premorbid intellectual performance in schizophrenia: premor-
12. Kendell R, Jablensky A. Distinguishing between the validity bid intellectual decline and risk for schizophrenia. Arch Gen
and utility of psychiatric diagnoses. Am J Psychiatry. 2003; Psychiatry. 2005;62:1297–1304.
160:4–12. 31. Davidson M, Reichenberg A, Rabinowitz J, Weiser M,
13. Krabbendam L, Arts B, van Os J, Aleman A. Cognitive func- Kaplan Z, Mark M. Behavioral and intellectual markers for
tioning in patients with schizophrenia and bipolar disorder: schizophrenia in apparently healthy male adolescents. Am J
a quantitative review. Schizophr Res. 2005;80:137–149. Psychiatry. 1999;156:1328–1335.
14. Zakzanis KK, Leach L, Kaplan E. On the nature and pattern 32. Brewer WJ, Wood SJ, McGorry PD, et al. Impairment of ol-
of neurocognitive function in major depressive disorders. factory identification ability in individuals at ultra-high risk
Neuropsychiatry Neuropsychol Behav Neurol. 1998;11: for psychosis who later develop schizophrenia. Am J Psychi-
111–119. atry. 2003;160:1790–1794.
15. van Gorp WG, Altshuler L, Theberge DC, Wilkins J, Dixon 33. Brewer WJ, Francey SM, Wood SJ, et al. Memory impair-
W. Cognitive impairment in euthymic bipolar patients with ments identified in people at ultra-high risk for psychosis
and without prior alcohol dependence. Arch Gen Psychiatry. who later develop first-episode psychosis. Am J Psychiatry.
1998;55:41–46. 2005;162:71–78.

919
R. S. E. Keefe & W. S. Fenton

34. Keefe RSE, Perkins DO, Gu H, Zipursky RB, Christensen BK, 39. van den Bosch RJ, Rombouts RP. Causal mechanisms of
Lieberman JA. A longitudinal study of neurocognitive function subjective cognitive dysfunction in schizophrenic and de-
in individuals at-risk for psychosis. Schizophr Res. In press. pressed patients. J Nerv Ment Dis. 1998;186:364–368.
35. Bowie CR, Reichenberg A, Patterson TL, Heaton RK, 40. Harvey PD, Serper MR, White L, et al. The convergence of
Harvey PD. Determinants of real-world functional perfor- neuropsychological testing and clinical ratings of cognitive
mance in schizophrenia subjects: correlations with cognition, impairment in patients with schizophrenia. Compr Psychiatry.
function, functional capacity, and symptoms. Am J Psychia- 2001;42:306–313.
try. 2006;163:418–425. 41. Stip E, Caron J, Renaud S, Pampoulova T, Lecomte Y. Ex-
36. Keefe RSE, Poe M, Walker TM, Kang JW, Harvey PD. The ploring cognitive complaints in schizophrenia: the subjective
Schizophrenia Cognition Rating Scale (SCoRS): interview- scale to investigate cognition in schizophrenia. Compr Psychi-
based assessment and its relationship to cognition, real-world atry. 2003;44:331–340.

Downloaded from http://schizophreniabulletin.oxfordjournals.org/ at H.E.C. Bibliotheque Myriam & J. Robert Ouimet on June 7, 2015
functioning and functional capacity. Am J Psychiatry. 2006; 42. Spitzer RL. Values and assumptions in the development of
163:426–432. DSM-III and DSM-III-R: an insider’s perspective and a be-
37. Ventura J, Bilder R, Cienfuegos A, Boxer O. Interview based lated response to Sadler, Hulgus, and Agich’s ‘‘On values in
measures of cognition in schizophrenia. Biol Psychiatry. 2006; recent American psychiatric classification’’. J Nerv Ment
59:1715. Dis. 2001;189:351–359.
38. Mortiz S, Ferahli S, Naber D. Memory and attention perfor- 43. Carpenter WT. Schizophrenia: diagnostic class or domains of
mance in psychiatric patients: lack of correspondence be- pathology. Schizophr Bull. 2007;33:203.
tween clinician-rated and patient-rated functioning with 44. Heinrichs RW, Zakzanis KK. Neurocognitive deficit in
neuropsychological test results. J Int Neuropsychol Soc. In schizophrenia: a quantitative review of the evidence. Neuro-
press. psychology. 1998;12:426–445.

920