DEPRESSION AND ANXIETY 30:515–527 (2013)

ReviewCOGNITIVE DEFICITS AND FUNCTIONAL OUTCOMESIN MAJOR DEPRESSIVE DISORDER: DETERMINANTS,SUBSTRATES, AND TREATMENT INTERVENTIONS

Roger S. McIntyre, MD, FRCPC,1,2,3,4∗ Danielle S. Cha, HBSc,3,4 Joanna K. Soczynska, HBSc,3,4

Hanna O. Woldeyohannes, HBSc,2 Laura Ashley Gallaugher, HBSc,2 Paul Kudlow, BSc, MD,2 MohammadAlsuwaidan, MD, MPH, FRCPC, Dip ABPN,3

and Anusha Baskaran, HBSc, MSc2,5

Background: Few reports have aimed to describe the mediational effect of cog-nitive deficits on functional outcomes in major depressive disorder (MDD), andrelatively few interventions are demonstrated to mitigate cognitive deficits inMDD. Methods: Studies enrolling subjects between the ages of 18–65 were se-lected for review. Bibliographies from identified articles were reviewed to identifyadditional original reports aligned with our objectives. Results: Cognitive deficitsin MDD are consistent, replicable, nonspecific, and clinically significant. The ag-gregated estimated effect size of cognitive deficits in MDD is small to medium.Pronounced deficits in executive function (≥1 SD below the normative mean) areevident in ∼20–30% of individuals with MDD). Other replicated abnormalitiesare in the domains of working memory, attention, and psychomotor processingspeed. Mediational studies indicate that cognitive deficits may account for thelargest percentage of variance with respect to the link between psychosocial dys-function (notably workforce performance) and MDD. No conventional antide-pressant has been sufficiently studied and/or demonstrated robust procognitiveeffects in MDD. Conclusions: Cognitive deficits in MDD are a principal me-diator of psychosocial impairment, notably workforce performance. The hazardsposed by cognitive deficits in MDD underscore the need to identify a consensus-based neurocognitive battery for research and clinical purposes. Interventions(pharmacological, behavioral, neuromodulatory) that engage multiple physio-logical systems implicated in cognitive deficits hold promise to reduce, reverse,and prevent cognitive deficits. Depression and Anxiety 30:515–527, 2013.C© 2013 Wiley Periodicals, Inc.

Key words: major depressive disorder; cognitive dysfunction; cognitive deficits;dementia; duloxetine; vortioxetine; functional outcome

1Institute of Medical Science, University of Toronto, Toronto,Canada2Mood Disorders Psychopharmacology Unit, University HealthNetwork, Toronto, Canada3Department of Psychiatry, University of Toronto, Toronto,Canada4Department of Pharmacology, University of Toronto, Toronto,Canada5Centre for Neuroscience Studies, Queen’s University,Kingston Canada

Contract grant sponsor: Takeda Pharmaceutical Company, Ltd.

∗Correspondence to: Roger S. McIntyre, MD, FRCPC, Mood Dis-orders Psychopharmacology Unit, University Health Network, Uni-versity of Toronto, 399 Bathurst Street, Toronto, ON, Canada M5T2S8. E-mail: roger.mcintyre@uhn.caReceived for publication 5 November 2012; Revised 4 January2013; Accepted 4 January 2013

DOI 10.1002/da.22063Published online 6 May 2013 in Wiley Online Library(wileyonlinelibrary.com).

C© 2013 Wiley Periodicals, Inc.

516 McIntyre et al.

INTRODUCTIONMajor depressive disorder (MDD) is a multidimen-sional mental disorder affecting approximately one inseven individuals at some time in their life.[1] MDD isassociated with a high rate of nonrecovery and recur-rence, with chronicity rates estimated at approximately20%.[2] The estimated annual costs attributable to MDDare approximately $83 billion, with indirect costs due todecreased psychosocial function (notably workforce per-formance) being a major contributor.[3] For example, it isestimated that MDD is associated with an annual loss of27.2 workdays per ill worker.[4] The human capital costattributable to MDD provides the impetus for identi-fying determinants of functional impairment. Availableevidence indicates that cognitive dysfunction is a crit-ical mediator of adverse psychosocial outcomes in thispopulation.[5–7]

The Diagnostic and Statistical Manual of Mental Dis-orders – Fourth Edition – Text Revision (DSM-IV-TR)identifies cognitive impairment (i.e., poor concentra-tion or indecisiveness) as a criterion item of a majordepressive episode (MDE). Cognitive complaints dur-ing the symptomatic and remitted phases are commonlyreported by individuals with MDD.[3, 4, 6, 8] When com-pared to other severe mental disorders (e.g., schizophre-nia, bipolar disorder), relatively fewer studies havecritically reviewed the affected cognitive domains andestimated magnitude of cognitive impairment in MDD.In addition, there have been fewer original reportsthat have primarily aimed to parse out the neurobi-ological substrate(s) of cognitive dysfunction in thispopulation, to determine the contribution of cognitivedysfunction to psychosocial impairment, and to primar-ily evaluate the procognitive effects of treatment (regard-less of modality).

Replicated evidence indicates that cognitive dysfunc-tion causes and maintains psychosocial impairment insevere mental disorders.[5–7] Moreover, deficits in cog-nitive function in adults under the age of 65 with MDDcannot be sufficiently explained by age-related cogni-tive changes. Available evidence suggests that cognitivedysfunction is a critical determinant of functional out-come in MDD.[7] It also remains to be determined ifthe neurobiological substrate(s) that subserve cognitivedysfunction in MDD are discrete and/or if they overlapwith substrate(s) implicated in other mental disorders.Questions regarding which, if any, treatment modality ismore effective in mitigating cognitive deficits, enhancingcognitive function and/or preventing their occurrence inMDD remain unanswered. A general impression, albeitbased on relatively few empirical studies, is that cogni-tive deficits are suboptimally treated with conventionaltreatment approaches [e.g., selective serotonin reuptakeinhibitors (SSRIs)].[9–11]

The persistence of cognitive dysfunction in MDDbeyond resolution of the acute episode suggests thatit may represent a trait or a residual phenomenon inmany individuals. Moreover, it provides the impetus to

fundamentally re-think and perhaps redefine how “re-mission” should be conceptualized.[6] Studies assessingtemporality of onset provide supporting evidence thatin some individuals, cognitive dysfunction may predatethe onset of the first MDE.[12] Cognitive dysfunctioncan be disaggregated into two broad interrelated cate-gories: (1) deficits in one or more cognitive domain(s)or (2) cognitive bias: representing attentional alloca-tion toward negatively valenced stimuli and/or aberrantinterpretation of social cues. It could be further sur-mised that cognitive bias represents a trait characteristicin individuals who are temperamentally predisposed toMDD (e.g., neuroticism). The preponderance of evi-dence indicates that cognitive deficits are a consequenceand/or residual phenomenon of MDD and in relativelyfewer cases may be an antecedent to MDD. (The no-tion that cognitive deficits predate the onset of MDDis more frequently reported in older individuals withMDD).

Relatively few studies have evaluated the effectof conventional antidepressants on cognitive perfor-mance in nongeriatric, adult MDD samples. Selectiveserotonin reuptake inhibitors (SSRIs), selective nore-pinephrine inhibitors (SNRIs), dopamine modulators(bupropion), and norepinephrine inhibitors (reboxe-tine) have been reported to improve cognitive per-formance in adults with MDD.[13–18] There is alsoa paucity of evidence comparing the effects of dif-ferent classes of antidepressants.[16–18] Questions per-taining to the degree to which the improvement incognitive deficits can be dissociated from the ef-fect of other MDD psychopathology domains (e.g.,chronicity, subtype of MDD, number and duration ofMDEs) and/or comorbidities have been insufficientlyaddressed.

The overarching aim of this review is to describe therole of cognitive dysfunction as a determinant of func-tional outcome in MDD. Toward this aim, we succinctlyreview (1) cognitive deficits and their determinants inadults with MDD 65 years old or younger (i.e., 18–65years) and (2) the effect of treatment on cognitive per-formance. The secondary objectives of this review are toidentify underlying substrate(s) that subserve cognitivedeficits and to propose treatment approaches that pri-marily aim to mitigate, reverse, and prevent deficits incognitive function.

METHODWe conducted a review of computerized databases (i.e., PubMed,

Google Scholar) from 1980 to 2012. “MDD” was cross-referencedwith the following terms: cognitive dysfunction, cognitive deficits, de-mentia, and functional outcome. Studies enrolling subjects betweenthe ages of 18–65 were primarily selected for review: select studies inlate life depression in the absence of studies in adults 65 years of age orless were also reviewed were also reviewed. Bibliographies from iden-tified articles were also reviewed in order to identify any other originalreports that were aligned with the objectives of this paper.

Depression and Anxiety

Review: Cognitive Deficits and Functional Outcomes in MDD 517

RESULTSNEUROCOGNITIVE TESTS

Table 1 contains a list of conventional instru-ments and batteries frequently administered for as-sessing cognitive function. The available tests measureintellectual functioning, working memory, verbal learn-ing and memory, attention and information processingspeed, executive function, psychomotor performance,and global cognitive function. None of the foregoingcognitive measures have been specifically developed toevaluate cognitive performance and/or its functional im-plications in the workforce.

COGNITIVE DEFICITS IN MDDThe interpretation of study outcomes as they pertain

to cognitive deficits in MDD begins with a review ofmethodological factors that may affect the interpreta-tion of the data. First, available studies are often inclu-sive of heterogeneous sample compositions that have in-cluded mixed ages, disparate definitions of MDD, andillness characteristics (e.g., age at onset, duration, num-ber of episodes, depression subtype, presence or absenceof psychotic features) as well as treatments. Additionally,the co-occurrence of medical and/or psychiatric condi-tions has not always been controlled. The relevance ofthis latter observation is underscored by evidence in-dicating that comorbid conditions in MDD may exertdirect effects on cognitive performance (e.g., attentiondeficit/hyperactivity disorder, obesity).[19, 20] For exam-ples of determinants of cognitive deficits in MDD, seeTable 2.

Most studies evaluating cognitive dysfunction inMDD have enrolled symptomatic and treated individ-uals whereas relatively fewer studies enrolled “asymp-tomatic” and/or individuals not receiving medication.Most studies have evaluated cognitive function follow-ing the onset of MDD. However, there is little docu-mentation reporting on cognitive performance prior tothe onset of an index MDE. A separate understudied is-sue relates to the impact of early childhood adversity oncognitive measures in adult samples with MDD and itsmoderational effect.[21]

A notable factor affecting interpretation of study re-sults is the heterogeneity of neurocognitive tests and bat-teries. Most available studies include tests that evaluateperformance across several principal cognitive domains(see Table 1). There is, however, no accepted “goldstandard” psychometric of cognitive function in MDDthat has been embraced by the research/clinical com-munity. In contradistinction, there have been efforts toagree upon a common set of tests for individuals withprimary psychotic disorders for both descriptive andinterventional research [e.g., measurement and treat-ment in research to improve cognition in schizophrenia(MATRICS)]. In the clinical setting, subjectivecomplaints of cognitive deficits are common.[22–24]

Notwithstanding the ubiquity of cognitive complaints,

TABLE 1. Common neurocognitive tests

Cognitive domain Neurocognitive tests

Executive function(conceptformation,abstraction, setshifting, setmaintenance,planning,self-monitoring,divided attention)

• Wisconsin Card Sorting Test (WCST)• Trail-Making Test Part B (TMT B)• Stroop Colour-Word Interference Test

(SCWT)• Categories Test• Block Design (WAIS-R)• Picture Completion (WAIS-R)• Concept Shifting Task (CST)• Tower of London (TOL; CANTAB)• Stockings of Cambridge (SOC; CANTAB)• Intra/Extradimensional Shift Test

(CANTAB; IED)• Spatial Span (SSP; CANTAB)• Ruff Figural Fluency Test (RFFT)• Verbal Fluency–Letter Fluency & Category

Fluency• Controlled Oral Word Association Test

(COWAT)• The Delis-Kaplan Executive Function

System (D-KEFS)Attention and

processing speed• Digit Symbol Substitution Test (DSST;

WAIS-R)• Digit Span Forwards and Backwards

(WAIS-R)• Continuous Performance Task (CPT)• Reaction Time (RTI; CANTAB)• Choice Reaction Time (CRT; CANTAB)• Simple Reaction Time (SRT; CANTAB)• Trail-Making Test Part A (TMT A)• Paced Auditory Serial Addition Test

(PASAT)• Serial Sevens Subtraction Test (SSST)

Working memory • Arithmetic (WAIS-R)• Digit Span Forwards and Backwards

(WAIS-R)• Delayed Recognition Span Test (DRST)• Spatial Working Memory (SWM;

CANTAB)• Letter-Number Sequencing (LNS; WMS-R)• Logical Memory (WMS-R)• n-Back Test

Verbal learning andmemory

• California Verbal Learning Test (CVLT)• Rey Auditory Verbal Learning Test

(RAVLT)• Rivermead Behavioral Memory Test

(RBMT)• Hopkins Verbal Learning Test Revised

(HVLT-R)• Logical Memory (WMS-R)• Verbal Paired Associates (VPA; WMS-R)• Visual Verbal Learning Test (VVLT)• Digit Span Forwards and Backwards

(WAIS-R)• Luria Verbal Learning Test (LVLT)• Serial Sevens Subtraction Test (SSST)• Verbal Recognition Memory Test (VRM;

CANTAB)Visual learning and

memory• Visual Reproduction (WMS-R)• Benton Visual Retention Test (VRT)• Benton Visual Form Discrimination (VFD)

Depression and Anxiety

518 McIntyre et al.

TABLE 1. Continued

Cognitive domain Neurocognitive tests

• Rey-Osterrieth Complex Figure Test(ROCF)

• Kimura’s Recurring Figures Test (RFT)• Visual Verbal Learning Test (VVLT)• Pattern Recognition Memory (PRM;

CANTAB)• Spatial Recognition Memory (SRM;

CANTAB)• Delayed Matching to Sample (DMS;

CANTAB)• Paired Associates Learning (PAL;

CANTAB)• Matching Familiar Figures Test 2

(MFFT-20)Language and verbal

comprehension• Controlled Oral Word Association Test

(COWAT)• Verbal Fluency–Category Fluency & Letter

Fluency• Similarities (WAIS-R)• Vocabulary (WAIS-R)• Information (WAIS-R)• Comprehension (WAIS-R)• Token Test

Visuospatial/perceptualprocessing

• Judgement of Line Orientation (JOLO)• Benton Visual Form Discrimination (VFD)• Block Design (WAIS-R)• Visuospatial Span Forwards and Backwards

(WMS-R)Brief mental status • Mini Mental State Exam (MMSE)General intelligence • Raven’s Progressive Matrices

• Wechsler Adult Intelligence Scale—Revised(WAIS-R)

• National Adult Reading Test (NART)• Wechsler Test of Adult Reading (WTAR)• Test of Nonverbal Intelligence-3 (TONI-3)

Cognitive battery • Cambridge Neuropsychological TestAutomated Battery (CANTAB)

• Wechsler Memory Scale—Revised(WMS-R)

• Wechsler Adult Intelligence Scale—Revised(WAIS-R)

• Victoria Symptom Validity Test (VSVT)• California Computerised Assessment

Package (CalCAP)• CNS Vital Signs• Massachusetts General Hospital Cognitive

and Physical Functioning Questionnaire(CPFQ)

• The Delis-Kaplan Executive FunctionSystem (D-KEFS)

Psychomotorperformance

• Finger Tapping• Grooved Pegboard Test• Purdue Pegs

Decision makingand responsecontrol

• The Go/No-go Association Task (GNAT;CANTAB)

• Information Sampling Task (IST;CANTAB)

• Cambridge Gambling Task (CGT;CANTAB)

Induction • Big/Little Circle (BLC; CANTAB)

TABLE 2. Determinants of cognitive deficits in majordepressive disorder (MDD)

Age[86]

Age at onset[87]

Educational attainment[88]

Baseline depression severity[89]

MDD subtype[90]

Symptomatic status (i.e., remission vs. nonremission)[14,30,91]

Psychiatric comorbidity[92]

Medical comorbidity[93,94]

Illness duration[9]

Episode frequency[32]

Treatment[66]

Childhood adversity[95,96]

the congruence between subjectively reported and ob-jectively measured cognitive deficits is not consistent inMDD.[25]

A common convention among extant studies is thereporting and comparing of group means rather thanan emphasis on individuals who fall below a pre-specified cut score for cognitive dysfunction [i.e., 1–2 standard deviations (SD) below the norm].[26] Thisconvention may inadvertently decrease the “assay sensi-tivity” and not provide a sufficient estimate of the mag-nitude of deficits, and/or the functional implications, inMDD. Moreover, the use of mean scores in parametricstatistics assumes normal distribution of the dependentvariable. As the mean scores influence extreme values,the emphasis on mean scores rather than the SD fromthe normative population mean could underestimate themagnitude of deficits that affect a subgroup of individu-als with MDD.[26] For example, individuals with MDDfall, on average, 0.5–1 SD below the normal populationmean.[14]

Notwithstanding it is reported that approximately 25–50% of patients with MDD exhibit deficits that aremore than 1 SD below the mean on at least one cog-nitive domain and as many as 48% score more than 2SD below the mean.[26] A summary statement regardingcognitive deficits in MDD is that cognitive deficits areconsistent, replicable, nonspecific, clinically significant,and of small to medium in effect size. The most repli-cated cognitive deficits are in the domains of executivefunction, working memory, attention, as well as generaland psychomotor processing speed (e.g., sensorimotor,cognitive).[27] A limitation of the approach that definesdeficits based on deviations from the standard popula-tion means is that it is not inclusive of individuals whoremain above the mean, possibly due to cognitive re-serve (e.g., education, occupation, intelligence),[28] butreport deterioration compared with baseline. It could behypothesized that these “uncaptured” deficits likewisecontribute significantly to relative functional decline inthese individuals.

The conventional assumption has been that cognitivedeficits in MDD are a consequence or a residual fea-ture of an acute MDE. For example, individuals with

Depression and Anxiety

Review: Cognitive Deficits and Functional Outcomes in MDD 519

remitted MDD often exhibit persisting cognitive deficitsin measures of attention, executive function, and verbalmemory.[29, 30] There have been relatively few studiesevaluating the temporality of onset, wherein cognitivedeficits are documented prior to the onset of MDD.Notwithstanding the paucity of data, a longitudinal,population-based study has reported a significant associ-ation between episodic memory function and emergentdepressive symptoms during a 2-year follow-up.[31]

A separate study that broadly aimed to identify pre-morbid markers of depression in a population-basedsample of nondepressed individuals (N = 708) evalu-ated prospectively over a 3-year period reported thatfemale gender, lower educational attainment, low so-cial support, financial strain, and poor episodic memory(but not verbal fluency, psychomotor speed, or men-tal speed) were associated with an increased risk ofdepression.[12] These results suggest that hippocampal-dependent memory may be abnormal prior to the clinicalpresentation of depression.[12] The persistence of cog-nitive deficits in asymptomatic individuals as well as theidentification of cognitive deficits prior to the presenceof symptomatic depression indicates that in many in-dividuals, cognitive deficits can be conceptualized as acore deficit rather than an epiphenomenon of a MDE(i.e., secondary to diminished initiative and effortfulprocessing).

Replicated evidence indicates that episode frequencyand illness duration are significantly associated withthe presence of cognitive deficits.[32, 33] For example,a large study (N = 8,229) of outpatients with MDDreported that declarative memory (measured with thedelayed paragraph recall index from the WechslerMemory Scale—Revised), a surrogate marker of hip-pocampal function, decreases approximately 2–3% witheach MDE, up to four episodes.[32] This decrementin memory parallels results from morphometric stud-ies that have documented volumetric reduction in thehippocampus in MDD with a more pronounced de-crease observed during the first several episodes.[32, 34]

Current illness severity was the major determinant ofperformance, as opposed to the intensity of their pre-vious depressive history (i.e., the number and lengthof past episodes). However, following clinical responseat the second visit, the length of previous depres-sive history became more significant than currentsymptoms.[32, 34]

Not all reports identify episode frequency as a de-terminant of cognitive deficit. For example, a meta-analysis of cognitive deficits in first episode MDDreported small-to-moderate effect size decreases in psy-chomotor speed, attention, visual and learning memoryas well as all aspects of executive function.[27] The pres-ence of cognitive deficits during the first MDE as partof MDD indicates that cognitive performance repre-sents a target for early identification, measurement, andintervention.[27, 35]

Taken together, there are multiple determinants ofcognitive outcome in MDD: heterogeneity of sample

composition, treatment regimens, patterns of comorbid-ity, definitions of depression and cognition measures, aswell as insufficient adjustment for early childhood ad-versity and colinearity, are major methodological lim-itations. Notwithstanding cognitive deficits in MDDare a consistent observation exhibiting trait, state, andprogressive features.

COGNITIVE DEFICITS IN MDD: LINK TOFUNCTIONAL OUTCOME

Results from the Global Alliance and Chronic DiseaseReport extend and replicate the observation that MDDis associated with the highest level of disability-adjustedlife years among all mental, neurological, and substanceuse disorders.[36] A significant component of the overalldisability and cost associated with depression relates toimpaired workplace performance.[4]

The effect of neurocognitive deficits as a mediatorof disability in individuals with MDD has been insuffi-ciently studied. Jaeger et al. reported that measures of at-tention, ideational fluency, nonverbal (i.e., visuo-spatial)and learning domains were highly associated with dis-ability 6 months following hospitalization for a MDE ina relatively small cohort (N = 48) of adults with MDD.[5]

Naismith et al. sought to determine whether sub-jective and/or objective measures of cognitive per-formance were related to measures of psychomotorperformance.[25] A small sample (N = 21) of adultstreated for MDD exhibited a moderate relationshipbetween objectively measured psychomotor speed andphysical disability.[25] Functional disability was mod-erately correlated to objective measures of memoryretention.[25] Measures of mental health disability (i.e.,Brief Disability Questionnaire Mental Health and SF-12Well-Being) were not associated with any neuropsycho-logical measure.[25] These results replicate and extendobservations of an association between cognitive deficitsand functional outcomes in a separate small sample ofadults with MDD as well as individuals with late-lifeMDD.[29, 37, 38]

Buist-Bouwman et al. conducted a mediational anal-ysis of the European Study of the Epidemiology ofMental Disorders (ESEMeD), a cross-sectional sur-vey representative of the adult population in Belgium,France, Germany, Italy, Netherlands, and Spain(N = 21,425, age ≥ 18).[7] The mediating effects ofsix activity limitations (Mobility, Self Care, Cogni-tion/Concentration/Attention/Memory, Social Interac-tion, Discrimination, Embarrassment) articulated in theWHO International Classification of Functioning, Dis-ability, and Health were evaluated.[7] Cognition and em-barrassment were the only determinants significantlyassociated with both MDEs and work functioning.[7]

Moreover, cognition and embarrassment accounted forapproximately half of the association between a MDEand work loss, underscoring the salience of cognitivefunction in mediating and reducing role functioning inMDEs.[7]

Depression and Anxiety

520 McIntyre et al.

Taken together, cognitive deficits in MDD are notonly characterized by poor performance on objective andsubjective measures, but also appear to be a principal me-diator of functional impairment.[39, 40] This observationunderscores the prioritization of cognitive deficits in theevaluation and management of individuals with MDD.

COGNITIVE DYSFUNCTION IN MDD:NEUROBIOLOGICAL SUBSTRATE

Contemporary models of disease pathoetiology inMDD have been extensively reviewed elsewhere.[41]

Progress in molecular and cellular biology as well asresults from neuroimaging investigations have servedto refine the hypothesis that disparate domains ofpsychopathology in MDD are subserved by abnor-malities in the structure, function, and chemicalcomposition of fronto-subcortical circuitry.[42, 43] A pre-liminary study also suggests that brain activation pat-terns suggestive of functional disconnectivity in MDDmay be different in individuals who are on long-term sickleave when compared with those who are able to main-tain workplace performance.[44] Moreover, several nodalstructures (e.g., hippocampus, amygdala, anterior cin-gulate cortex) are susceptible to volumetric/functionalchanges as a consequence of illness duration, episodefrequency, and severity.[34, 43, 45] The neurochemical ab-normalities that are well documented in individuals withMDD can be conceptualized as a consequence of aber-rant cellular functioning within the implicated neuralcircuits.[46–48] For example, disturbances in monoaminesystems, notably catecholamine systems, have been doc-umented to play a role in attentional and executivefunction deficits.[49]

A derivative of this hypothesis is that the neurobiolog-ical correlates (i.e., neural circuits) of cognitive functionand deficits, overlap with those implicated in other com-ponents of MDD.[47] The most replicated volumetricabnormality in MDD is bilateral hippocampal reduction,which is a consequence of loss of neuropil, decreaseddendritic density, and reduced neuronal soma size.[34, 50]

Hypofrontality of the prefrontal cortex and overactivityof the anterior cingulate cortex are hypothesized to bethe correlate of a functional disconnection between cor-tical and subcortical structures.[43, 51] These alterationsmay mediate deficits in measures of executive function,attention, learning and memory, as well as informationprocessing speed.

The mediators of aberrant neural circuitry, structureand function are hypothesized to include alterationsin regulatory–counterregulatory hormone balance(i.e., glucocorticoid-signaling abnormalities, insulinresistance), immunoinflammatory activation, neu-rotrophins [e.g., Brain-Derived Neurotrophic Factor(BDNF)], and oxidative stress.[46, 52–54] A separate, andpossibly related, observation is that recurrent MDDis associated with an increased hazard for Alzheimer’sdisease (AD).[44] For example, results from postmortemstudies indicate that a history of MDD is associated

with greater neuritic plaque density in neuropatho-logically confirmed Alzheimer’s brain.[55] Moreover,cognitively intact adults with MDD manifest lowercerebrospinal fluid concentration of beta-amyloid,providing indirect evidence for parenchymal depositionof beta-amyloid-42.[56]

A bidirectional relationship between MDD andmetabolic disorders has been amply documented.[57] Ac-cumulating evidence supports the hypothesis that al-terations in insulin signaling may be relevant to neu-rocognitive decline in subpopulations of individuals withMDD. For example, insulin resistance and diabetes mel-litus type 2 are associated with cognitive deficits inyounger adults as well as those with mild cognitiveimpairment (MCI) and AD.[58] An imbalance betweeninsulin and counterregulatory neurohormonal systems(i.e., glucocorticoids) may alter pro-apoptotic intracellu-lar signaling cascades thereby resulting in neuronal/glialloss and neurocognitive decline.[59, 60]

The default mode network (DMN), a proposed net-work of neural circuits that connects cortical and subcor-tical structures, may be relevant to cognitive function.[61]

The DMN comprises several nodal structures includingthe ventral–medial–prefrontal cortex, posterior cingu-late/retrosplenial cortex, and bilateral inferior parietallobe.[61] The DMN is usually more active during the“resting state” than it is during a cognitive, emotional,and/or motor task.[62] Abnormalities in the DMN havebeen implicated in pathological disease states of whichcognitive deficits are a defining feature (e.g., AD).[63]

An integrated translational research report indicated thatabnormalities in insulin resistance are accompanied bydisturbances in the DMN in individuals at risk for AD.Moreover, modafinil, an agent with established procog-nitive effects by enhancing catecholamine neurotrans-mission, augments deactivation in the major nodal struc-tures of the DMN.[61]

In summary, the pathoetiological model in MDDposits structural and functional disturbances in inter-connected neural circuits and distributed networks. Thesubstrates that subserve cognitive performance in MDDhave overlapping and discrete components from thesubstrates subserving affective processing; refining themodel subserving domains of psychopathology in MDDhold promise to inform interventional strategies for cog-nitive deficits in MDD.

THE EFFECT OF INTERVENTION ONCOGNITIVE DEFICITS IN MDD

All commercially available and approved conventionalantidepressants augment central indolamine and/or cat-echolamine neurotransmission.[64, 65] This pharmacody-namic profile provides the basis for hypothesizing thatantidepressants, in addition to mitigating both the moodand vegetative symptoms of an MDE, would also be ex-pected to ameliorate cognitive deficits. There are, how-ever, relatively few studies that have primarily aimedto evaluate the effect of conventional antidepressants

Depression and Anxiety

Review: Cognitive Deficits and Functional Outcomes in MDD 521

on cognitive performance in nongeriatric, adult MDDsamples.[13–15] Available evidence indicates that SSRIs,SNRIs, dopamine modulators (bupropion), and nore-pinephrine inhibitors (reboxetine) improve cognitiveperformance in adults with MDD.[16–18] There are alsorelatively few studies that have compared the cognitiveeffects of different classes of antidepressants on cogni-tive function.[16–18] Insufficiently answered questions in-clude the degree to which the improvement in cognitivedeficits can be dissociated from the effect on other MDDpsychopathology domains.

Herrera-Guzman et al. reported on the cognitive ef-fects of escitalopram and duloxetine in adults (aged 20–50) with MDD.[16] They observed that both treatmentsimproved working memory, attention, and executivefunction as well as mental processing speed and mo-tor performance.[16, 66] Duloxetine was superior to es-citalopram on episodic and working memory whereasno significant differences between the antidepressantswere observed on measures of attention and executivefunction.[16, 66]

Raskin et al. sought to determine the effect of dulox-etine versus placebo on several measures of cognitiveperformance.[10] A study in older adults (62–90 years ofage) is reviewed as, to our knowledge, it is the only studythat primarily aimed to evaluate a conventional antide-pressant’s ability to mitigate cognitive deficits in a cohortof adults with MDD.[10] They reported that when com-pared to placebo, 8 weeks of treatment with duloxetinesignificantly improved the cognitive composite score.[10]

The improvement in the duloxetine-treated group waslargely accounted for by improvement in measures ofverbal learning and recall with trending between-groupdifferences noted on attention [i.e., Digit Symbol Sub-stitution Test (DSST)], visual attention and executivefunction (i.e., Two-Digit Cancellation Test), and work-ing memory/executive function (i.e., Letter-Number Se-quencing Test).[10] Path analyses showed that for theimprovement of the cognitive composite score, therewas a 90.9% direct effect and 90.1% indirect effectthrough improvement in a Geriatric Depression ScaleTotal Score.[10]

Vortioxetine (Lu AA21004) is a novel multi-modalantidepressant that exhibits 5-HT3 and 5-HT7 re-ceptor antagonism, 5-HT1B receptor partial agonism,5-HT1A receptor agonism, and inhibition of the 5-HTtransporter. In addition to its effect on indolamine sig-naling, vortioxetine also increased central neurotrans-mission of noradrenaline, dopamine, acetylcholine, andhistamine.[67] Vortioxetine has demonstrated antide-pressant efficacy in MDD in adult and elderly pop-ulations as well as in adults with generalized anxietydisorder.[68, 69] In addition to antidepressant proper-ties, vortioxetine has demonstrated cognitive-enhancingproperties in animal and human studies.[70] For ex-ample, vortioxetine showed statistically significant im-provement on acquisition and delayed recall (as mea-sured by the Rey Auditory Verbal Learning Test), aswell as information processing speed (as measured by

the DSST) in a group of elderly (N = 453) nonde-mented individuals with MDD.[68] In this study, duloxe-tine, the active control, also demonstrated improvementon RAVLT, but not DSST.[68] A path analysis concludedthat 83% of the improvement on DSST with vortioxe-tine was a direct effect (duloxetine 26%), while the directeffect on acquisition and delayed recall (i.e., RAVLT)was 71% and 72%, respectively (for duloxetine 65%and 66%, respectively).[68] The beneficial effects of vor-tioxetine in this study on cognitive performance were asecondary outcome, providing an empirical basis forevaluating the effect of this agent on cognition and con-sequently functional outcomes as a primary outcome inyounger adults with MDD.[68]

Notwithstanding the paucity of studies, a signal thatemerges from the foregoing studies is that treatment in-terventions that engage multiple neurochemical systemssimultaneously may be more likely to improve cogni-tive performance when compared to interventions thatprincipally target a single system (e.g., SSRIs). In keep-ing with this view, it is not uncommon for practition-ers to coprescribe mechanistically diverse agents (e.g.,modafinil, psychostimulants) and/or suggest behavioralinterventions (e.g., aerobic exercise) in addition to con-ventional antidepressant treatment with an aim to pri-marily target the cognitive domain. Hitherto, no Foodand Drug Administration (FDA) approved agent for de-pression has been purposefully evaluated to mitigate cog-nitive deficits and improve psychosocial functioning inMDD.

DISCUSSIONSeveral observations emanate from the synthesis of the

data reviewed herein.[3, 36, 71, 72] The functional impair-ments (notably workforce impairment) associated withMDD disproportionately account for the overall costsattributable to MDD.[3, 73] This observation providesthe basis for elucidating determinants of psychosocialimpairment in MDD.[3, 73] A global shift in the work-force away from primary industry toward a skilled hu-man capital economy will only amplify the importance ofdiseases/disorders of the central nervous system (CNS)and their actuarial costs.[3, 72, 74]

Cognitive deficts in MDD are prevalent and mayrepresent a core dimension of psychopathology inMDD.[31, 32, 75] Cognitive deficits are often a persist-ing abnormality in individuals with MDD and in somecases may predate the onset of clinically significantdepressive symptoms.[12, 27] The observation that cog-nitive deficits mediate psychosocial impairment (i.e.,accounting for a significant percentage of variance) andintervention.[7, 26, 74] When compared with severe men-tal disorders (e.g., schizophrenia, bipolar disorder), rela-tively less emphasis has been given to the hazards posedby cognitive deficits in MDD. Along with the method-ological limitations that affect inferences and interpre-tations that may be drawn from the extant data per-taining to cognitive deficits in MDD, there is no “gold

Depression and Anxiety

522 McIntyre et al.

TABLE 3. Effects of pharmacological treatment on cognitive deficits in major depressive disorder

Author Year Population Treatment Cognitive testsCognitive

deficits/improvements

Ferguson et al. 2003 74 MDD (age range =18–65)

Reboxetine (8–10 mg/day;N = 25)

Paroxetine (20–40mg/day; N = 23)

Placebo (N = 26)

SRT, Digit Vigilance,CRT, Numeric WM,Word Recognition, andCritical FlickerFrequency

Reboxetine significantlyimproved sustainedattention and speed ofcognitive functioningcompared with baseline

No significant changes ortrends in this directionwere observed inindividuals receivingeither paroxetine orplacebo

Constant et al. 2005 20 MDD (mean age =47.67)

26 healthy controls (meanage = 48.85)

Sertraline (50–75 mg/day) Phasic Alertness Task,Classic SCWT, and theSupraliminal andSubliminal EmotionalStroop test

MDDs had psychomotorslowing associated withattentional andexecutive disturbance

Following first weeks oftreatment, a beneficialeffect on psychomotorslowing on attentionaland executive functionswas observed

Gualtieri et al. 2006 38 drug-free MDD (meanage = 38.11, SD = 9.95)

31 MDD antidepressantmonotherapyresponders (mean age =43.55, SD = 10.68)

69 healthy controls (meanage = 41.30, SD =11.40)

Citalopram (N = 1)Fluoxetine (N = 3)Escitalopram (N = 8)Paroxetine (N = 3)Mirtazapine (N = 1)Trazodone (N = 1)Venlafaxine (N = 5)Bupropion (N = 6)Sertraline (N = 3)

CNS Vital Signs Untreated MDDs hadglobalneuropsychologicalimpairment.Successfully treatedMDD`s performanceimproved, but notnormalized

Specificdepression-relateddeficits were observedin executive functionand processing speed,but not in memory,psychomotor speed, orreaction time

Gualtieri et al. 2007 81 MDD (mean age =43.85)

27 healthy controls (meanage = 43.85)

Bupropion (N = 27)Venlafaxine (N = 27)SSRI (N = 27)

CNS Vital Signs SSRI group scoredsignificantly belowcontrols in tests ofpsychomotor speed,cognitive flexibility, andreaction time

Venlafaxine group scoredworse than controls inreaction time

Bupropion group did notdiffer from controls inany of the cognitivedomains

Raskin et al. 2007 311 MDDDuloxetine (N = 207;

mean age = 72.6, SD =5.7)

Placebo (N = 104; meanage = 73.3, SD = 5.7)

Duloxetine (60 mg/day; N= 207)

Placebo (N = 104)

Verbal Learning andRecall Test, DSST,Cancellation Test, LNS

Duloxetine demonstratedsignificantly greaterimprovement versusplacebo. MDD withduloxetine treatmentimproved verballearning and memory

Depression and Anxiety

Review: Cognitive Deficits and Functional Outcomes in MDD 523

TABLE 3. Continued

Author Year Population Treatment Cognitive testsCognitive

deficits/improvements

Herrera-Guzmanet al.

2010 73 MDDSSRI (N = 36; mean age

= 32.91, SD = 8.73)SNRI (N = 37; mean age

= 33.21, SD = 8.61)37 healthy controls (mean

age = 33.05, SD = 8.04)

Escitalopram (10 mg/day;N = 36)

Duloxetine (60 mg/day; N= 37)

Vocabulary, Digit Spanand SWM, RapidVisual InformationProcessing, DMS,SCWT, IED, SOC

SSRI and SNRItreatments bothimproved WM,attention, and allexecutive functions.However, MDD`scognitive function didnot improve enough toreach levels of controlsubjects

Herrera-Guzmanet al.

2010 73 MDDSSRI (N = 36; mean age

= 32.91, SD = 8.73)SNRI (N = 37; mean age

= 33.21, SD = 8.61)37 healthy controls (mean

age = 33.05, SD = 8.04)

Escitalopram (10 mg/day;N = 36)

Duloxetine (60 mg/day; N= 37)

Vocabulary, Digit Span,AVLT, PRM, PAL,DMS, SRM, RTI,SCWT

SSRI and SNRItreatments bothimproved episodicmemory and, to a lesserextent, WM, mentalprocessing speed andmotor performance.SNRI was superior toSSRI at improvingepisodic memory andWM

Herrera-Guzmanet al.

2010 73 MDDSSRI (N = 36; mean age

= 32.91, SD = 8.73)SNRI (N = 37; mean age

= 33.21, SD = 8.61)37 healthy controls (mean

age = 33.05, SD = 8.04)

Escitalopram (10 mg/day;N = 36)

Duloxetine (60 mg/day; N= 37)

Vocabulary, SWM,RAVLT, PAL, SOC,Rapid VisualInformation Processing

MDDs in remissionshowed deficits in verbaland visual episodicmemory, sustainedattention, mnemonicand strategic aspects ofWM, and planning.MDDs in recoveryshowed the sameneuropsychologicaldeficit pattern. MDDstreated with SSRIshowed moreimpairment in episodicvisual and verbalmemory than thosetreated with SNRI

Hinkelmann etal.

2012 N = 52 MDD with SSRItreatment and add-ontreatment modulatingthe mineralocorticoidreceptor

N = 50 healthy subjects

BaselineEscitalopram (10–20

mg/day; N = 52)Add-on treatmentMR-agonist

fludrocortisone (0.2mg/day; N = 19)

MR-antagonistspironolactone (100mg/day; N = 22)

Placebo (N = 11)

RAVLT, Digit Span F andB, ROCF, Cancellationtest, TMT A and B

MDD performed worsecompared with healthycontrols (Digit SpanForward, ROCF,Cancellation Test),indicating ongoingrelative cognitivedeficits in thesedomains. No differencesbetween the threetreatment groups wereobserved over time

standard” to measure cognitive deficits in researchsettings and/or to succinctly and comprehensively mea-sure cognitive deficits in the clinical ecosystem. More-over, as interest in patient-reported outcomes in psy-chiatric and medical disorders increases, there will bea need to identify not only objective, but also sub-jective measures of cognitive functions that are valid,

reliable, and have utility for personalizing treatmentinterventions.[76]

MDD is a heterogeneous phenotype in phenomenol-ogy and pathoetiology. There is no single disease modelthat will sufficiently explain the panoply of abnormal-ities observed across disparate populations. Admittedlycomplex, a conceptual framework of disease modeling in

Depression and Anxiety

524 McIntyre et al.

MDD that would be comprehensive and coherent wouldposit that there are multiple (i.e., hundreds, if not thou-sands) disease pathways in MDD. Succinct neurobio-logical substrate(s) of neurocognitive dysfunction is notcurrently available. Notwithstanding available evidenceimplicates altered cellular, synaptic, and circuitry abnor-malities in MDD manifesting as disturbances in centralneurotransmitter signaling.[46, 52–54, 77]

Robust therapeutic interventions specifically tar-geting cognitive deficits in MDD are not currentlyavailable, largely due to insufficient empirical obser-vation. Notwithstanding available evidence indicatesthat targeting multiple monoamine systems mitigatescognitive deficits independent of its effect on otherdepressive symptoms.[10, 16, 66] The complex pathoeti-ology of cognitive deficits provides the basis for hy-pothesizing that interventions capable of engagingmultiple physiological systems may be differentially ef-fective for this domain of psychopathology. Moreover,it remains a testable hypothesis that mitigating cog-nitive deficits in MDD would result in an improve-ment in overall psychosocial performance, notably in theworkforce.

Refining disease models in MDD will provide a plat-form for identifying genuinely novel, pharmacologicaltreatment approaches to treat, reverse, and prevent cog-nitive deficits in MDD (Table 3). Extrapolating from re-sults reported in schizophrenia and bipolar disorder, it isreasonable to expect that some adults with MDD wouldbe candidates for cognitive remediation and neurore-habilitative activities/endeavors.[78–82] Moreover, novelneuromodulatory approaches [e.g., repetitive Transcra-nial Magnetic Stimulation (rTMS)], complementary al-ternative medicine (e.g., N-acetyl-cysteine), and behav-ioral interventions (e.g., aerobic exercise) may also havea place in the armamentarium against cognitive deficitsin MDD beyond resolution of the acute episode.[13, 83–85]

Acknowledgments. This review was sponsored bythe Takeda Pharmaceutical Company, Ltd., as part ofa joint clinical development program with H. Lund-beck A/S. Dr. McIntyre drafted and reviewed successiveversions of the manuscript. Editorial support, includingstyling and editing for journal submission, was providedby The Medicine Group, New Hope, Pennsylvania.

Conflict of interest. The authors have no conflicts ofinterest. Joanna K. Soczynska is a recipient of the EliLilly Canada Fellowship Award.

REFERENCES1. Kessler RC, Petukhova M, Sampson NA, Zaslavsky AM, Wittchen

HU. Twelve-month and lifetime prevalence and lifetime morbidrisk of anxiety and mood disorders in the United States. Int JMethods Psychiatr Res 2012.

2. van RA, Huffmeier J, Victor D, Klocke K, Borlinghaus J, PawelzikM. Contrasting chronic with episodic depression: an analysis of

distorted socio-emotional information processing in chronic de-pression. J Affect Disord 2012.

3. Greenberg PE, Kessler RC, Birnbaum HG, et al. The eco-nomic burden of depression in the United States: how did itchange between 1990 and 2000? J Clin Psychiatry 2003;64(12):1465–1475.

4. Kessler RC, Akiskal HS, Ames M, et al. Prevalence and effectsof mood disorders on work performance in a nationally repre-sentative sample of U.S. workers. Am J Psychiatry 2006;163(9):1561–1568.

5. Jaeger J, Berns S, Uzelac S, Davis-Conway S. Neurocognitivedeficits and disability in major depressive disorder. Psychiatry Res2006;145(1):39–48.

6. Conradi HJ, Ormel J, de JP. Presence of individual (residual)symptoms during depressive episodes and periods of remission:a 3-year prospective study. Psychol Med 2010 Oct. 8:1–10.

7. Buist-Bouwman MA, Ormel J, de GR, et al. Mediators of the as-sociation between depression and role functioning. Acta PsychiatrScand 2008;118(6):451–458.

8. Reppermund S, Zihl J, Lucae S, et al. Persistent cognitive im-pairment in depression: the role of psychopathology and alteredhypothalamic-pituitary-adrenocortical (HPA) system regulation.Biol Psychiatry 2007;62(5):400–406.

9. Herrera-Guzman I, Gudayol-Ferre E, Herrera-Abarca JE, et al.Major depressive disorder in recovery and neuropsychologicalfunctioning: effects of selective serotonin reuptake inhibitor anddual inhibitor depression treatments on residual cognitive deficitsin patients with major depressive disorder in recovery. J AffectDisord 2010;123(1–3):341–350.

10. Raskin J, Wiltse CG, Siegal A, et al. Efficacy of duloxetine on cog-nition, depression, and pain in elderly patients with major depres-sive disorder: an 8-week, double-blind, placebo-controlled trial.Am J Psychiatry 2007;164(6):900–909.

11. Millan MJ, Agid Y, Brune M, et al. Cognitive dysfunction in psy-chiatric disorders: characteristics, causes and the quest for im-proved therapy. Nat Rev Drug Discov 2012;11(2):141–168.

12. Airaksinen E, Wahlin A, Forsell Y, Larsson M. Low episodicmemory performance as a premorbid marker of depression: evi-dence from a 3-year follow-up. Acta Psychiatr Scand 2007;115(6):458–465.

13. Furtado CP, Hoy KE, Maller JJ, Savage G, Daskalakis ZJ, Fitzger-ald PB. An investigation of medial temporal lobe changes andcognition following antidepressant response: a prospective rTMSstudy. Brain Stimul 2012 Jul 6;1e9.

14. Wagner S, Doering B, Helmreich I, Lieb K, Tadic A. A meta-analysis of executive dysfunctions in unipolar major depressivedisorder without psychotic symptoms and their changes duringantidepressant treatment. Acta Psychiatr Scand 2012;125(4):281–292.

15. McLennan SN, Mathias JL. The depression-executive dys-function (DED) syndrome and response to antidepressants:a meta-analytic review. Int J Geriatr Psychiatry 2010;25(10):933–944.

16. Herrera-Guzman I, Herrera-Abarca JE, Gudayol-Ferre E, et al.Effects of selective serotonin reuptake and dual serotonergic-noradrenergic reuptake treatments on attention and executivefunctions in patients with major depressive disorder. PsychiatryRes 2010;177(3):323–329.

17. Herrera-Guzman I, Gudayol-Ferre E, Lira-Mandujano J, et al.Cognitive predictors of treatment response to bupropion and cog-nitive effects of bupropion in patients with major depressive dis-order. Psychiatry Res 2008;160(1):72–82.

18. Ferguson JM, Wesnes KA, Schwartz GE. Reboxetine versusparoxetine versus placebo: effects on cognitive functioning in de-pressed patients. Int Clin Psychopharmacol 2003;18(1):9–14.

Depression and Anxiety

Review: Cognitive Deficits and Functional Outcomes in MDD 525

19. Larochette AC, Harrison AG, Rosenblum Y, Bowie CR.Additive neurocognitive deficits in adults with attention-deficit/hyperactivity disorder and depressive symptoms. Arch ClinNeuropsychol 2011;26(5):385–395.

20. Cserjesi R, Luminet O, Poncelet AS, Lenard L. Altered executivefunction in obesity. Exploration of the role of affective states oncognitive abilities. Appetite 2009;52(2):535–539.

21. Aas M, Steen NE, Agartz I, et al. Is cognitive impairment fol-lowing early life stress in severe mental disorders based on spe-cific or general cognitive functioning? Psychiatry Res 2012 Apr 1;198(3):495–500.

22. Martinez-Aran A, Vieta E, Colom F, et al. Cognitive dysfunctionsin bipolar disorder: evidence of neuropsychological disturbances.Psychother Psychosom 2000;69(1):2–18.

23. Martinez-Aran A, Vieta E, Reinares M, et al. Cognitive functionacross manic or hypomanic, depressed, and euthymic states inbipolar disorder. Am J Psychiatry 2004;161(2):262–270.

24. Grut M, Jorm AF, Fratiglioni L, Forsell Y, Viitanen M, WinbladB. Memory complaints of elderly people in a population survey:variation according to dementia stage and depression. J Am Geri-atr Soc 1993;41(12):1295–1300.

25. Naismith SL, Longley WA, Scott EM, Hickie IB. Disability inmajor depression related to self-rated and objectively-measuredcognitive deficits: a preliminary study. BMC Psychiatry 2007;7:32.

26. Gualtieri CT, Morgan DW. The frequency of cognitive impair-ment in patients with anxiety, depression, and bipolar disorder: anunaccounted source of variance in clinical trials. J Clin Psychiatry2008;69(7):1122–1130.

27. Lee RS, Hermens DF, Porter MA, Redoblado-Hodge MA. Ameta-analysis of cognitive deficits in first-episode major depressivedisorder. J Affect Disord 2012;140(2):113–124.

28. Jones RN, Manly J, Glymour MM, Rentz DM, Jefferson AL, SternY. Conceptual and measurement challenges in research on cogni-tive reserve. J Int Neuropsychol Soc 2011 Mar 17; 17(4):593–601.

29. Godard J, Baruch P, Grondin S, Lafleur MF. Psychosocial andneurocognitive functioning in unipolar and bipolar depression:a 12-month prospective study. Psychiatry Res 2012;196(1):145–153.

30. Preiss M, Kucerova H, Lukavsky J, Stepankova H, Sos P,Kawaciukova R. Cognitive deficits in the euthymic phase of unipo-lar depression. Psychiatry Res 2009;169(3):235–239.

31. Simons CJ, Jacobs N, Derom C, et al. Cognition as predictor ofcurrent and follow-up depressive symptoms in the general popu-lation. Acta Psychiatr Scand 2009;120(1):45–52.

32. Gorwood P, Corruble E, Falissard B, Goodwin GM. Toxic ef-fects of depression on brain function: impairment of delayed re-call and the cumulative length of depressive disorder in a largesample of depressed outpatients. Am J Psychiatry 2008;165(6):731–739.

33. Singh-Manoux A, Akbaraly TN, Marmot M, et al. Persistentdepressive symptoms and cognitive function in late midlife: theWhitehall II study. J Clin Psychiatry 2010;71(10):1379–1385.

34. MacQueen GM, Campbell S, McEwen BS, et al. Course of ill-ness, hippocampal function, and hippocampal volume in majordepression. Proc Natl Acad Sci U S A 2003;100(3):1387–1392.

35. Snyder HR. Major depressive disorder is associated with broad im-pairments on neuropsychological measures of executive function:a meta-analysis and review. Psychol Bull 2012 May 28; 139(1):81–132.

36. Collins PY, Patel V, Joestl SS, et al. Grand challenges in globalmental health. Nature 2011;475(7354):27–30.

37. Kiosses DN, Klimstra S, Murphy C, Alexopoulos GS. Executivedysfunction and disability in elderly patients with major depres-sion. Am J Geriatr Psychiatry 2001;9(3):269–274.

38. Kim JM, Stewart R, Glozier N, et al. Physical health, depressionand cognitive function as correlates of disability in an older Koreanpopulation. Int J Geriatr Psychiatry 2005;20(2):160–167.

39. McCall WV, Dunn AG. Cognitive deficits are associated withfunctional impairment in severely depressed patients. PsychiatryRes 2003;121(2):179–184.

40. Sweeney JA, Kmiec JA, Kupfer DJ. Neuropsychologic im-pairments in bipolar and unipolar mood disorders on theCANTAB neurocognitive battery. Biol Psychiatry 2000;48(7):674–684.

41. Krishnan V, Nestler EJ. Linking molecules to mood: newinsight into the biology of depression. Am J Psychiatry2010;167(11):1305–1320.

42. Jiao Q, Ding J, Lu G, et al. Increased activity imbalance in fronto-subcortical circuits in adolescents with major depression. PLoSOne 2011;6(9):e25159.

43. Pizzagalli DA. Frontocingulate dysfunction in depression: towardbiomarkers of treatment response. Neuropsychopharmacology2011;36(1):183–206.

44. Sandstrom A, Sall R, Peterson J, et al. Brain activation patterns inmajor depressive disorder and work stress-related long-term sickleave among Swedish females. Stress 2012;15(5):503–513.

45. Malykhin NV, Carter R, Hegadoren KM, Seres P, Coupland NJ.Fronto-limbic volumetric changes in major depressive disorder. JAffect Disord 2012;136(3):1104–1113.

46. Andreazza AC. Combining redox-proteomics and epigenomics toexplain the involvement of oxidative stress in psychiatric disorders.Mol Biosyst 2012 Jun 18; 8(10):2503–2512.

47. Drevets WC, Price JL, Furey ML. Brain structural and func-tional abnormalities in mood disorders: implications for neurocir-cuitry models of depression. Brain Struct Funct 2008;213(1–2):93–118.

48. Yucel K, Nazarov A, Taylor VH, Macdonald K, Hall GB, Mac-Queen GM. Cerebellar vermis volume in major depressive disor-der. Brain Struct Funct 2012; Jun 14.

49. Stahl SM, Zhang L, Damatarca C, Grady M. Brain circuits deter-mine destiny in depression: a novel approach to the psychophar-macology of wakefulness, fatigue, and executive dysfunction inmajor depressive disorder. J Clin Psychiatry 2003;64(Suppl 14):6–17.

50. Malykhin NV, Carter R, Seres P, Coupland NJ. Structuralchanges in the hippocampus in major depressive disorder: con-tributions of disease and treatment. J Psychiatry Neurosci2010;35(5):337–343.

51. Zeng LL, Shen H, Liu L, et al. Identifying major depression usingwhole-brain functional connectivity: a multivariate pattern analy-sis. Brain 2012;135(Pt 5):1498–1507.

52. Li M, Soczynska JK, Kennedy SH. Inflammatory biomarkers indepression: an opportunity for novel therapeutic interventions.Curr Psychiatry Rep 2011;13(5):316–320.

53. Ryan JP, Sheu LK, Critchley HD, Gianaros PJ. A neural cir-cuitry linking insulin resistance to depressed mood. PsychosomMed 2012;74(5):476–482.

54. Kessing LV, Jorgensen OS. Apolipoprotein E-epsilon 4 fre-quency in affective disorder. Biol Psychiatry 1999;45(4):430–434.

55. Rapp MA, Schnaider-Beeri M, Grossman HT, et al. Increased hip-pocampal plaques and tangles in patients with Alzheimer diseasewith a lifetime history of major depression. Arch Gen Psychiatry2006;63(2):161–167.

56. Pomara N, Bruno D, Sarreal AS, et al. Lower CSF amyloidbeta peptides and higher F2-isoprostanes in cognitively intactelderly individuals with major depressive disorder. Am J Psychia-try 2012;169(5):523–530.

Depression and Anxiety

526 McIntyre et al.

57. McIntyre RS, Rasgon NL, Kemp DE, et al. Metabolic syndromeand major depressive disorder: co-occurrence and pathophysio-logic overlap. Curr Diab Rep 2009;9(1):51–59.

58. Baker LD, Cross DJ, Minoshima S, Belongia D, Watson GS, CraftS. Insulin resistance and Alzheimer-like reductions in regionalcerebral glucose metabolism for cognitively normal adults withprediabetes or early type 2 diabetes. Arch Neurol 2011;68(1):51–57.

59. McIntyre RS, Vagic D, Swartz SA, et al. Insulin, insulin-like growth factors and incretins: neural homeostatic regula-tors and treatment opportunities. CNS Drugs 2008;22(6):443–453.

60. Craft S, Watson GS. Insulin and neurodegenerative disease:shared and specific mechanisms. Lancet Neurol 2004;3(3):169–178.

61. Minzenberg MJ, Yoon JH, Carter CS. Modafinil modulationof the default mode network. Psychopharmacology (Berlin)2011;215(1):23–31.

62. Mars RB, Neubert FX, Noonan MP, Sallet J, Toni I, RushworthMF. On the relationship between the “default mode network” andthe “social brain.” Front Hum Neurosci 2012;6:189.

63. Kenna H, Hoeft F, Kelley R, et al. Fasting plasma insulin and thedefault mode network in women at risk for Alzheimer’s disease.Neurobiol Aging 2012; Jul 5; 34(3):641–649.

64. Muller N, Schwarz MJ. Immunological aspects of depressive dis-orders. Nervenarzt 2007;78(11):1261–1273.

65. Xu Z, Zhang Z, Shi Y, et al. Influence and interaction of geneticpolymorphisms in catecholamine neurotransmitter systems andearly life stress on antidepressant drug response. J Affect Disord2011;133(1–2):165–173.

66. Herrera-Guzman I, Gudayol-Ferre E, Herrera-Guzman D,Guardia-Olmos J, Hinojosa-Calvo E, Herrera-Abarca JE. Ef-fects of selective serotonin reuptake and dual serotonergic-noradrenergic reuptake treatments on memory and mentalprocessing speed in patients with major depressive disorder. J Psy-chiatr Res 2009;43(9):855–863.

67. Bang-Andersen B, Ruhland T, Jorgensen M, et al. Discovery of 1-[2-(2,4-dimethylphenylsulfanyl)phenyl]piperazine (Lu AA21004):a novel multimodal compound for the treatment of major depres-sive disorder. J Med Chem 2011;54(9):3206–3221.

68. Katona C, Hansen T, Olsen CK. A randomized, double-blind,placebo-controlled, duloxetine-referenced, fixed-dose study com-paring the efficacy and safety of Lu AA21004 in elderly pa-tients with major depressive disorder. Int Clin Psychopharmacol2012;27(4):215–223.

69. Alvarez E, Perez V, Dragheim M, Loft H, Artigas F. A double-blind, randomized, placebo-controlled, active reference study ofLu AA21004 in patients with major depressive disorder. Int J Neu-ropsychopharmacol 2012;15(5):589–600.

70. Mork A, Pehrson A, Brennum LT, et al. Pharmacological ef-fects of Lu AA21004: a novel multimodal compound for thetreatment of major depressive disorder. J Pharmacol Exp Ther2012;340(3):666–675.

71. Kessler RC, Barber C, Birnbaum HG, et al. Depression in theworkplace: effects on short-term disability. Health Aff (Millwood)1999;18(5):163–171.

72. Kessler RC, Berglund P, Demler O, et al. The epidemiology ofmajor depressive disorder: results from the National ComorbiditySurvey Replication (NCS-R). JAMA 2003;289(23):3095–3105.

73. Kessler RC, Merikangas KR, Wang PS. The prevalence and cor-relates of workplace depression in the national comorbidity surveyreplication. J Occup Environ Med 2008;50(4):381–390.

74. Birnbaum HG, Kessler RC, Kelley D, Ben-Hamadi R, Joish VN,Greenberg PE. Employer burden of mild, moderate, and severe

major depressive disorder: mental health services utilization andcosts, and work performance. Depress Anxiety 2010;27(1):78–89.

75. van EP, van WG, Fernandez G, et al. Amygdala responsiv-ity related to memory of emotionally neutral stimuli consti-tutes a trait factor for depression. Neuroimage 2011;54(2):1677–1684.

76. Fava M, Graves LM, Benazzi F, et al. A cross-sectional study of theprevalence of cognitive and physical symptoms during long-termantidepressant treatment. J Clin Psychiatry 2006;67(11):1754–1759.

77. Blank T, Prinz M. Microglia as modulators of cognition and neu-ropsychiatric disorders. Glia 2012; Jun 27; 61(1):62–70.

78. Bowie CR, Harvey PD. Cognition in schizophrenia: impairments,determinants, and functional importance. Psychiatr Clin NorthAm 2005;28(3):613–633, 626.

79. Bowie CR, McGurk SR, Mausbach B, Patterson TL, Har-vey PD. Combined cognitive remediation and functional skillstraining for schizophrenia: effects on cognition, functionalcompetence, and real-world behavior. Am J Psychiatry 2012;May 11.

80. Bowie CR, Depp C, McGrath JA, et al. Prediction of real-world functional disability in chronic mental disorders: a com-parison of schizophrenia and bipolar disorder. Am J Psychiatry2010;167(9):1116–1124.

81. Goodwin GM, Martinez-Aran A, Glahn DC, Vieta E. Cognitiveimpairment in bipolar disorder: neurodevelopment or neurode-generation? An ECNP expert meeting report. Eur Neuropsy-chopharmacol 2008;18(11):787–793.

82. Bora E, Yucel M, Pantelis C. Cognitive endophenotypes of bipo-lar disorder: a meta-analysis of neuropsychological deficits in eu-thymic patients and their first-degree relatives. J Affect Disord2009;113(1–2):1–20.

83. Chan A, Shea TB. Effects of dietary supplementation with N-acetyl cysteine, acetyl-L-carnitine and S-adenosyl methionine oncognitive performance and aggression in normal mice and miceexpressing human ApoE4. Neuromolecular Med 2007;9(3):264–269.

84. Vasques PE, Moraes H, Silveira H, Deslandes AC, Laks J. Acuteexercise improves cognition in the depressed elderly: the effect ofdual-tasks. Clinics (Sao Paulo) 2011;66(9):1553–1557.

85. Knochel C, Oertel-Knochel V, O’Dwyer L, et al. Cognitive andbehavioural effects of physical exercise in psychiatric patients. ProgNeurobiol 2012;96(1):46–68.

86. Thomas AJ, Gallagher P, Robinson LJ, et al. A comparison ofneurocognitive impairment in younger and older adults with majordepression. Psychol Med 2009;39(5):725–733.

87. Herrmann LL, Le MM, Ebmeier KP. White matter hyperintensi-ties in late life depression: a systematic review. J Neurol NeurosurgPsychiatry 2008;79(6):619–624.

88. Gildengers AG, Butters MA, Chisholm D, et al. Cognition inolder adults with bipolar disorder versus major depressive disor-der. Bipolar Disord 2012;14(2):198–205.

89. McDermott LM, Ebmeier KP. A meta-analysis of depressionseverity and cognitive function. J Affect Disord 2009;119(1–3):1–8.

90. Withall A, Harris LM, Cumming SR. A longitudinal study ofcognitive function in melancholic and non-melancholic subtypesof major depressive disorder. J Affect Disord 2010;123(1–3):150–157.

91. Gallagher P, Robinson LJ, Gray JM, Porter RJ, Young AH.Neurocognitive function following remission in major depressivedisorder: potential objective marker of response? Aust N Z J Psy-chiatry 2007;41(1):54–61.

Depression and Anxiety

Review: Cognitive Deficits and Functional Outcomes in MDD 527

92. Liu IC, Chiu CH, Yang TT. The effects of gender anda co-occurring depressive disorder on neurocognitive func-tioning in patients with alcohol dependence. Alcohol Alcohol2010;45(3):231–236.

93. McIntyre RS, Kenna HA, Nguyen HT, et al. Brain volume abnor-malities and neurocognitive deficits in diabetes mellitus: points ofpathophysiological commonality with mood disorders? Adv Ther2010; Apr 13; 27(2):63–80.

94. McIntyre RS, Soczynska JK, Konarski JZ, et al. Should depressive

syndromes be reclassified as “metabolic syndrome type II”? AnnClin Psychiatry 2007;19(4):257–264.

95. Lemos-Miller A, Kearney CA. Depression and ethnicity as in-termediary variables among dissociation, trauma-related cogni-tions, and PTSD symptomatology in youths. J Nerv Ment Dis2006;194(8):584–590.

96. Riso LP, Miyatake RK, Thase ME. The search for determinantsof chronic depression: a review of six factors. J Affect Disord2002;70(2):103–115.

Depression and Anxiety

Copyright of Depression & Anxiety (1091-4269) is the property of John Wiley & Sons, Inc.and its content may not be copied or emailed to multiple sites or posted to a listserv withoutthe copyright holder's express written permission. However, users may print, download, oremail articles for individual use.