Research report
Identifying a cognitive impairment subgroup in adults with mood disorders

https://doi.org/10.1016/j.jad.2011.03.001Get rights and content

Abstract

Background

We hypothesized that only a minority of patients with mood disorders have measurable cognitive impairment, and this minority drives the small-to-medium effect sizes detected in group studies. Removal of this minority from group statistical analyses will illustrate that the majority appear to have broadly normal cognitive functioning.

Methods

Participants were adults between the ages of 20 and 54, including 659 healthy control subjects, 84 unmedicated outpatients diagnosed with depression, 59 outpatients diagnosed with depression who were on medications at the time of the evaluation, and 43 outpatients with bipolar disorder. All completed the CNS Vital Signs computerized cognitive screening battery.

Results

The prevalence rates of low cognitive test scores were calculated for the healthy control subjects and the patients with mood disorders. Having two scores at or below the 5th percentile occurred in 31.2% of the patients and only 8.2% of the control subjects [χ2(1) = 66.67, p < .0001; Odds Ratio = 5.1, 95% CI = 3.4–7.7]. For the control subjects, this low false positive rate for cognitive impairment was maintained across age groups, sexes, and education levels. A larger proportion of patients with bipolar disorder (41.9%) than patients with depression (27.1–28.6%) met this criterion for cognitive impairment.

Conclusions

This study suggests that cognitive impairment associated with mood disorders is limited to a minority of patients with the majority being broadly cognitively normal. Future research should determine if this identified subgroup has neuroanatomical, neurophysiological, or neuroendocrine abnormalities. Cognitive screening tools of this type might be useful in selecting participants for studies.

Introduction

It is well established that mood disorders are associated with cognitive impairment (Robinson et al., 2006, Zakzanis et al., 1998). The nature and extent to which depression causes objective cognitive impairment, however, is not fully understood. Some studies suggest that cognitive impairment associated with depression is quite limited (Grant et al., 2001, Rohling et al., 2002), difficult to detect, and is more likely to occur in those who are more seriously ill (Fossati et al., 2002, MacQueen et al., 2003, McDermott and Ebmeier, 2009, Tarbuck and Paykel, 1995). There is evidence that neurocognitive functioning improves following treatment (e.g., Bayless et al., 2010, Deuschle et al., 2004, Doraiswamy et al., 2003, Hviid et al., 2010, Neu et al., 2005, O'Connor et al., 2005, Rocca et al., 2005, Vythilingam et al., 2004, Wroolie et al., 2006), although this is not always the case (Frasch et al., 2009, Paelecke-Habermann et al., 2005, Reppermund et al., 2009, Weiland-Fiedler et al., 2004) — especially in older adults (Bhalla et al., 2009, Culang et al., 2009). Indeed, there is some suggestion that cognitive or executive functioning deficits may be a trait risk factor for depression (Douglas and Porter, 2009, Frasch et al., 2009, Micco et al., 2009, Reppermund et al., 2009). Furthermore, worse neuropsychological test performance at baseline is associated with poorer response to treatment (Dunkin et al., 2000, Kampf-Sherf et al., 2004, Mohlman and Gorman, 2005), and cognitive deficits are more pronounced in patients who are unemployed (Baune et al., 2010). It is possible that treatment refractory depression is a subtype characterized in part by cognitive impairment. Likewise, cognitive impairment is pronounced in patients with bipolar disorder and it persists when the patients are euthymic (Langenecker et al., 2007, Langenecker et al., 2010, Robinson et al., 2006, Torres et al., 2007), with lesser extent and severity of cognitive difficulties relative to schizophrenia (Schretlen et al., 2007). Cognitive impairment is associated with poor treatment adherence in bipolar disorder, but the direction of the relationship is unknown (Martinez-Aran et al., 2009). Mood disorders with psychosis are associated with a large adverse effect on neurocognitive functioning (Bora et al., 2009).

The accurate identification and quantification of neurocognitive impairment are important for research relating to neurobiological underpinnings, treatment, and functional outcome in patients with mood disorders. It is essential, methodologically, that we have accurate methods for identifying those patients who are objectively cognitively impaired and separate them from patients who have the subjective experience of poor thinking skills or thinking that is easily perturbed by negative affect, but perform normally on cognitive testing in controlled conditions. The treatments and outcomes for these two groups may differ markedly, as well as the prognosis. Based on group statistics, in individual studies or in meta-analyses, mood disorders are associated with a small-to-medium adverse effect on cognitive functioning (Robinson et al., 2006, Torres et al., 2007, Zakzanis et al., 1998). However, group statistics can obscure individual and subgroup differences. If present, these individual or subgroup differences in cognition might have important implications for research and clinical practice.

The purpose of this study was to examine cognitive functioning in mood disorders at the level of the individual. We hypothesized, based on our preliminary research, that (a) only a minority of patients with mood disorders have measurable cognitive impairment, (b) this minority is driving the small-to-medium effect sizes detected in group statistics (see also Bora et al., 2009, Burt et al., 1995), and (c) if you remove this minority from the group statistical analyses, the significant effect sizes will virtually disappear. The alternative hypothesis would be that the entire distribution of performance in depressed subjects is shifted about one half standard deviation lower relative to the control group without a bimodal distribution. If our hypotheses are true, the effect sizes reported in the literature seriously under-estimate the degree to which cognitive impairment is associated with mood disorders in a subset of patients. They are diluted by the majority of patients who have no measurable cognitive impairment. Using a large healthy normative sample and archival clinical groups, we will (a) develop and evaluate psychometric criteria for identifying cognitive impairment in adults with mood disorders, and (b) evaluate the above-mentioned hypotheses.

Section snippets

Participants

A healthy normative sample and three clinical groups were used for this study. Ethical approval for the use of this large, de-identified, archival database was granted by the University of British Columbia. Older adults were excluded. Participants were adults between the ages of 20 and 54, including 659 healthy control subjects, 84 unmedicated outpatients diagnosed with depression, 59 outpatients diagnosed with depression who were on medications at the time of the evaluation, and 43 outpatients

Results

Calculations for the base rates of low scores involve simultaneously examining the five domain scores, rather than performance on each domain in isolation. The base rates of low domain scores were calculated by using four cutoff scores that might be routinely used in clinical practice, including: (a) more than 1 standard deviation (SD) below the mean (i.e., <85 SS), (b) below the 10th percentile (i.e., <81 SS), (c) at or below the 5th percentile (i.e., ≤76 SS), and (d) more than 2 SDs below the

Discussion

The purpose of this study was to examine cognitive functioning in mood disorders at the individual and subgroup levels. As hypothesized (a) only a minority of patients with mood disorders have measurable cognitive impairment, (b) this minority appears to be driving the effect sizes detected in group statistics, and (c) if you remove this substantial minority from the group statistical analyses, the significant effect sizes disappear. Obviously, if you identify a subgroup of patients who perform

Role of funding source

This study was funded by AstraZeneca Canada, Inc. AstraZeneca Canada had no further role in study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the paper for publication. Dr. Langenecker's effort was supported by an NIH K23 Career Development Award (MH074459, SAL).

Conflicts of interest

Dr. Iverson has received research funding from the CNS Vital Signs Company. No funding from that company was provided for this study, however. Dr. Young has received honoraria and grant funding from AstraZeneca.

Acknowledgements

This study was presented at the International Neuropsychological Society, February, 2010, Acapulco, Mexico. The authors thank Dr. Gualtieri and the CNS Vital Signs company for providing all of the data used in this study.

References (42)

  • L.M. McDermott et al.

    A meta-analysis of depression severity and cognitive function

    J. Affect. Disord.

    (2009)
  • J. Mohlman et al.

    The role of executive functioning in CBT: a pilot study with anxious older adults

    Behav. Res. Ther.

    (2005)
  • P. Neu et al.

    Cognitive function over the treatment course of depression in middle-aged patients: correlation with brain MRI signal hyperintensities

    J. Psychiatr. Res.

    (2005)
  • Y. Paelecke-Habermann et al.

    Attention and executive functions in remitted major depression patients

    J. Affect. Disord.

    (2005)
  • L.J. Robinson et al.

    A meta-analysis of cognitive deficits in euthymic patients with bipolar disorder

    J. Affect. Disord.

    (2006)
  • M.L. Rohling et al.

    Depressive symptoms and neurocognitive test scores in patients passing symptom validity tests

    Arch. Clin. Neuropsychol

    (2002)
  • D.J. Schretlen et al.

    Neuropsychological functioning in bipolar disorder and schizophrenia

    Biol. Psychiatry

    (2007)
  • M. Vythilingam et al.

    Hippocampal volume, memory, and cortisol status in major depressive disorder: effects of treatment

    Biol. Psychiatry

    (2004)
  • P. Weiland-Fiedler et al.

    Evidence for continuing neuropsychological impairments in depression

    J. Affect. Disord.

    (2004)
  • J.D. Bayless et al.

    Pre- and post-electroconvulsive therapy multidomain cognitive assessment in psychotic depression: relationship to premorbid abilities and symptom improvement

    J. ECT

    (2010)
  • E. Bora et al.

    Cognitive impairment in affective psychoses: a meta-analysis

    Schizophr. Bull.

    (2009)
  • Cited by (107)

    • Features of cognitive impairment and related risk factors in patients with major depressive disorder: A case-control study

      2022, Journal of Affective Disorders
      Citation Excerpt :

      In addition to the T-scores of each test, composite scores for each of the five cognitive domains were also obtained by calculating the average T-scores of the corresponding tests. CI in a specific domain was defined as a score 1.5 standard deviations (SDs) below the mean score of the HCs (Iverson et al., 2011). General CI was defined as having impairments in two or more cognitive domains (Gualtieri and Morgan, 2008).

    View all citing articles on Scopus
    View full text