- Top of page
Aims: Although cognitive deficits are a common and potentially debilitating feature of major depressive disorder (MDD), such subjective declines in cognitive function are seldom validated by objective methods as a clinical routine. The aim of this study was to validate the Taiwanese Depression Questionnaire (TDQ) for detecting cognitive deficits in a sample of drug-free patients with MDD.
Methods: The subjects consisted of 40 well-characterized medication-free patients with MDD and 40 healthy controls. Clinical and neuropsychological assessments, including the Wisconsin Card Sorting Test, the Wechsler Memory Scale–Revised, the Continuous Performance Test, and the Finger-Tapping Test, were administered at the time of recruitment.
Results: Factor analyses of the TDQ yielded three factors. Memory, attention and psychomotor performance were significantly poorer in patients with MDD. The performances of verbal and delayed memory of the Wechsler Memory Scale–Revised were correlated with the cognitive domains of the TDQ. Generalization of our results must be undertaken with caution considering the relatively small sample size, which could lead to increased β-error.
Conclusion: Cognitive subdomains might be considered important for including in patient-administered questionnaires used to measure symptoms of MDD when developing a new scale.
THE INCLUSION OF cognitive symptoms in the DSM-IV criteria for major depressive episodes highlights the importance of cognition. For example, the criteria for the diagnosis of this condition include a diminished ability to concentrate, indecisiveness, and psychomotor retardation or agitation. There is a growing body of evidence of a significant reduction in neurocognitive functions in major depressive disorder (MDD),1 and depression-related disturbances of cognitive function have been demonstrated in a range of domains, including attention,2–4 memory,5,6 executive functions,7–10 and psychomotor functions.11 However, findings to date are varied,12 with many factors potentially contributing to inconsistencies between studies, including age, hospitalization duration, severity and subtype of depression, and the effect of psychotropic medication.13,14
Although cognitive deficits are a common and potentially debilitating feature of MDD, subjective cognitive decline is seldom validated by objective methods. Only a limited number of items pertain to neurocognitive decline in the most widely used depression rating scales, for example, the Hamilton Depression Rating Scale (HDRS)15 and the Beck Depression Inventory (BDI).16 However, in a recently developed patient-administered depression questionnaire, more comprehensive incorporation of cognitive items was attempted,17,18 although the validity concerning the neurocognitive domain of depression has not yet been confirmed.
The aims of this study were to: (i) validate the Taiwanese Depression Questionnaire (TDQ) for cognitive deficits in a sample of patients with MDD; and (ii) compare the cognitive function of depressed patients with that of their controls.
- Top of page
Three subdomains of the TDQ, the cognitive, affective, and somatic domains, were indeed confirmed from factor analysis (Table 2). The reliability of the overall TDQ was good (Cronbach's α = 0.93), and the reliabilities of the TDQ subdomains (the somatic, affective and cognitive domains) are shown in Table 2. A significant correlation between TDQ score and HDRS score was noted in this study (r = 0.56, P < 0.0001). Cognitive declines in patients with MDD were observed from their performance in the WMS-R, CPT and FTT (Table 1). In the patients group, there was a significant negative correlation between the TDQ cognition subscale and the WCST index of the number of categories completed (r = −0.33, P = 0.038); significant negative correlations between the TDQ cognition subscale and the WMS-R indices of verbal memory (r = −0.36, P = 0.04) and delayed recall memory (r = −0.37, P = 0.035) were also observed. In addition, the correlation between the TDQ cognition subscale score and the general memory index was marginally negative (r = −0.34, P = 0.057). The unmasked d' obtained from the CPT was also correlated with the cognitive domain of the TDQ (Table 3).
Table 2. Cronbach's α for the Taiwanese Depression Questionnaire subdomains (n = 140)
| ||Factor loading|
|Factor 1: Cognitive subdomain Cronbach's α = 0.87|
| 10. I had poor memory||0.71||0.17||0.34|
| 11. I could not concentrate when doing things||0.77||0.23||0.36|
| 12. I was slower in thinking and doing things than before||0.79||0.25||0.16|
|Factor 2: Affective subdomain Cronbach's α = 0.95|
| 1. I often felt like crying||0.06||0.54||0.32|
| 3. I felt more agitated than before||0.20||0.41||0.39|
| 13. I felt less confident than before||0.48||0.61||0.26|
| 14. I tended to look at the dark side of everything||0.15||0.78||0.24|
| 15. I felt miserable and even wanted to die||0.25||0.74||0.19|
| 16. I lost interest in everything||0.34||0.56||0.43|
| 18. I felt worthless||0.29||0.68||0.25|
|Factor 3: Somatic subdomain Cronbach's α = 0.96|
| 2. I felt blue and depressed||0.14||0.56||0.64|
| 4. I had trouble sleeping||0.23||0.21||0.52|
| 5. I had poor appetite||0.25||0.43||0.47|
| 6. I frequently had chest tightness ('sim-guan-tau-bang-bang')||0.26||0.35||0.64|
| 7. I felt uneasy, uncomfortable||0.27||0.37||0.67|
| 8. I felt tired and weak (‘Xu’, ‘mo wan qi’)||0.35||0.36||0.62|
| 9. I felt upset||0.28||0.46||0.64|
| 17. I felt sick (headache, dizziness, palpitations, or abdominal distress)||0.18||0.10||0.72|
Table 3. Relationship between TDQ subscale scores and scores for various cognitive tests in patients with major depressive disorder
|WCST (n = 40)|| || || || |
| Preservative errors||r = −0.05||r = −0.01||r = 0.12||r = −0.01|
|P = 0.76||P = 0.93||P = 0.47||P = 0.96|
|Number of categories completed||r = 0.01||r = −0.10||r = −0.33||r = −0.10|
|P = 0.94||P = 0.55||P = 0.04*||P = 0.54|
|WMS-R (n = 33)|| || || || |
| Verbal memory index||r = −0.16||r = −0.24||r = −0.36||r = −0.26|
|P = 0.39||P = 0.18||P = 0.04*||P = 0.15|
| Visual memory index||r = 0.19||r = 0.08||r = −0.24||r = −0.09|
|P = 0.28||P = 0.66||P = 0.19||P = 0.63|
| General memory index||r = −0.03||r = −0.13||r = −0.34||r = − 0.14|
|P = 0.88||P = 0.46||P = 0.057||P = 0.44|
| Attention/concentration index||r = −0.09||r = −0.03||r = −0.22||r = −0.10|
|P = 0.60||P = 0.89||P = 0.21||P = 0.58|
| Delayed recall memory index||r = −0.13||r = −0.29||r = −0.37||r = −0.27|
|P = 0.46||P = 0.10||P = 0.04*||P = 0.13|
|CPT (n = 28)|| || || || |
| CPT unmasked d'||R = −0.35||R = −0.39||r = −0.45||r = −0.44|
|P = 0.06||P = 0.04*||P = 0.02*||P = 0.02*|
| CPT masked d'||R = −0.57||R = −0.43||r = −0.42||r = −0.54|
|P = 0.01**||P = 0.05||P = 0.06||P = 0.01*|
|FTT (n = 27)|| || || || |
| Dominant hand||R = −0.01||r = 0.02||r = −0.28||r = −0.04|
|P = 0.95||P = 0.90||P = 0.16||P = 0.83|
- Top of page
The results of poorer cognitive performance in drug-free patients with MDD in this study were similar to those of previous studies.26 The findings of significant memory decline in patients with MDD supported the review by Buchanan et al.,27 which reported that memory decline was more profound than executive function decline in patients with MDD. Additionally, it is known that the hippocampus plays a key role in the regulation of mood and cognition. Hickie et al. showed that reduction in hippocampal volumes was associated with declines in visual and verbal memory functions in older people with depression,28 while Taylor Tavares et al. also found that unmedicated MDD patients experienced a significant reduction in executive function as compared with healthy subjects and bipolar II depressed patients.29
In this study, neuropsychological measures in depressed patients were found to be correlated with the cognitive domain of a depression questionnaire; in other words, cognitive symptoms perceived by depressed patients are correlated with objective measures of neurocognitive performance. Antikainen et al. reported that patients complaining of memory disturbance had higher BDI and HDRS scores than patients who did not, and that they also performed poorly in objective memory tests but not in tests of other cognitive functions.30 Even in the context of suspected cognitive disorder, the discrepancy between memory complaints and memory performance in patients is a subject of debate.31,32 A few studies have examined the association between subjective cognitive symptoms and objective neurocognitive measurements in patients. As such, Crane et al. found that depressive symptoms were independently associated with subjective memory complaints in non-demented older adults.33 Another example is the study of Bruce et al.,34 which demonstrated that BDI scores were positively correlated with mini-mental status and memory impairment. However, Bruce et al. did not identify which of the subdomains of the BDI are correlated with objective neuropsychological assessments.34 In contrast to these results, some studies have shown that memory complaints are associated with depressive symptoms35–37 but not with all domains of neuropsychological measurements, particularly psychomotor speed measures.38 These inconsistent results may be due to the use of different symptom checklists or different subdomains in the different studies.
In this study it was demonstrated that the cognitive subdomain of the validated depression questionnaire, the TDQ, is correlated with objectively measured cognitive functions in patients with MDD. Lee et al.39 also demonstrated that mild cognitive decline is highly prevalent among geriatric depressed patients in the course of the acute phase. It might therefore be important to include cognitive symptoms of major depression when developing a depression questionnaire.
No significant correlation was found in this study between self-rated cognitive domain scores of the TDQ and psychomotor speed, as measured by the FTT, as with the study of Naismith et al.;38 however, self-rating of cognitive deficits by depressed patients is still important. The discrepancy between objective and subjective measurements of cognitive performance still exists,38 as demonstrated by the large unexplained variance, and clinicians cannot rely solely on subjective reports of cognitive decline to assess cognitive function in depressed patients.
The results of the current study need to be interpreted with caution due to the following limitations. First, this study only recruited a small sample. Second, no significant correlations were observed between TDQ cognitive subdomain scores and some indexes of cognitive performance. Third, MDD patients can be subgrouped into non-psychotic or psychotic, single episode or multiple episodes, and other grouping variations; however, only non-psychotic and unipolar MDD patients were recruited to this study. Finally, the lack of inter-rater reliability of MDD diagnosis, detailed clinical assessment by psychiatrists and the short disease duration could confound the results.