Archives of Clinical Neuropsychology 20 (2005) 547553

Brief Report

The relation of self-report of mood and anxiety to CVLT-C, CVLT, and CVLT-2 in a psychiatric sample
Judith R. OJile , Gregory W. Schrimsher, Sid E. OBryant
Department of Psychiatry and Human Behavior, 2500 North State St., Jackson, MS 39216-4505, USA Accepted 6 December 2004

Abstract The relation between mood and cognitive status has been examined extensively over the years suggesting a signicant impact of mood and potentially anxiety on memory. This relation is of particular interest to practitioners conducting evaluations in settings that regularly treat individuals diagnosed with psychiatric disorders. With this in mind, the present study sought to evaluate the impact of self-report of depression and anxiety on the California Verbal Learning Testchildrens version (CVLT-C), the California Verbal Learning Test (CVLT), and the California Verbal Learning Test2nd edition (CVLT-2) in a mixed psychiatric sample. Records from 107 patients aged 678 evaluated in an outpatient psychiatry unit were examined. Results indicated minimal predictive utility was provided from self-report symptoms of anxiety or depression on CVLT-C, CVLT, or CVLT-2: Trial 1, Trial 5, total score, Short Delay Free Recall, Short Delay Cued Recall, long delay free recall, or long delay cued recall performance above the variance predicted by age, gender, and Full Scale IQ. Additional variance predicted by depression as measured by the Beck Depression Inventory (BDI) and the Child Depression Inventory (CDI) or anxiety as measured by the State-Trait Anxiety Inventory (STAI) was less than 3.0% over that accounted for by the covariates for the great majority of measurements from the various CVLT versions. Exceptions included the CDI that tended to account for approximately 5.0% of the variance on all of the CVLT-C measures and the STAI that accounted for approximately 5.010.0% additional variance on the short and long delay measures of the CVLT-2. The present results suggest that performance on the various forms of the CVLT are minimally predicted by self-reported depression and anxiety in a psychiatric setting. 2004 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved.
Keywords: CVLT; Memory; Depression; Anxiety; BDI; CDI; STAI

Corresponding author. Tel.: +1 601 984 5804. E-mail address: [email protected] (J.R. OJile).

0887-6177/$ see front matter 2004 National Academy of Neuropsychology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.acn.2004.12.001

548

J.R. OJile et al. / Archives of Clinical Neuropsychology 20 (2005) 547553

There is a large body of research reporting a relation between mood and cognitive functioning. In particular, mood has been shown to signicantly impact scores on measures of memory (Humphreys & Revelle, 1984; Kizilbash, Vanderploeg, & Curtiss, 2002; Porter, Gallagher, Thomson, & Young, 2003). This association is of particular concern for practitioners in psychiatric settings who are regularly asked to evaluate the cognitive status of patients who have mood or anxiety disorders in addition to other comorbid psychiatric and/or neurological conditions that may lead to memory impairment or decline. The deleterious effect of depression on memory has been reported on both short-term and long-term memory (Colby & Gotlib, 1988; Krames & McDonald, 1985; Richards & Ruff, 1989), and these ndings have been further supported by meta-analytic studies (e.g., Kindermann & Brown, 1997), but such ndings are not always consistent (e.g., Miller, Faustman, Moses, & Csernansky, 1991). The effect of anxiety on memory is less well established. For example, Kizilbash et al. (2002) found that, while anxiety alone had no impact on memory, combined anxiety and depression had a more deleterious effect on memory than depression alone. Because of the frequency with which the various versions of the CVLT are administered, it is important to understand the impact that depression and anxiety may exert on scores in different populations of interest. Recently Foldi, Brickman, Schaefer, and Knutelska (2003) found that depressed elderly patients performed more poorly on a number of CVLT measures than did controls. Kizilbash et al. (2002) found that, while depression impacted immediate recall and total acquisition of new information on the CVLT in a sample of Vietnam era veterans, anxiety alone exerted no impact on CVLT scores. Horan, Pogge, Borgaor, Stokes, and Harvey (1997) found that depressed adolescents performed more poorly than the normative standards on all CVLT indices. Otto et al. (1994) went further and presented norms for depressed patients. To our knowledge, no published studies have directly examined the impact of depression or anxiety on the CVLT-C or the CVLT-2, nor has any study examined the impact of depression and anxiety on any CVLT version from within a mixed psychiatric sample. Consequently, the aim of the present study was to examine the degree to which selfreport of depression or anxiety would predict memory performance as measured by the California Verbal Learning Test (CVLT) in a clinically relevant heterogeneous psychiatric sample.

1. Method 1.1. Participants Archival data was gathered on 107 referred patients to the neuropsychology clinic located within an outpatient psychiatry unit at a large southern medical center hospital that had data for all variables of interest. The sample was subdivided into three groups based on the version of the CVLT administered. The primary DSM-IV diagnoses of patients in all three groups was determined by referring psychiatrist as well as results of psychological testing.

J.R. OJile et al. / Archives of Clinical Neuropsychology 20 (2005) 547553 Table 1 Descriptive information and mean test scores for the CVLT and CVLT-2 groups CVLT Mean (SD) Age Education Gender Ethnicity BDI STAI State STAI Trait WAIS-III IQ CVLT score Trial 1 Trial 5 Total SDFR SDCR LDFR LDCR 37.9 (15.3) 13.3 (2.9) 24/24 36/10/1 22.5 (12.8) 47.0 (15.1) 51.3 (13.6) 90.6 (18.8) 6.0 (2.4) 10.2 (3.4) 43.9 (13.2) 8.4 (4.0) 10.0 (3.6) 8.3 (4.2) 9.7 (3.8) Range 1778 620 049 2078 2177 56128 213 416 1570 016 316 016 216 CVLT-2 Mean (SD) 33.4 (12.4) 13.4 (4.2) 19/6 22/3/0 21.0 (14.5) 44.2 (11.2) 49.4 (14.9) 96.2 (20.2) 5.9 (1.8) 11.1 (3.2) 46.2 (12.1) 9.0 (3.9) 10.4 (3.8) 10.2 (4.3) 10.7 (4.0) Range 1759 723 052 2465 2375 64147 39 416 2464 016 216 016 016 CVLT-C Mean (SD) 10.7 (2.3) 10/24 18/12/2 12.4 (7.9)1 30.5 (7.1)2 37.8 (9.4)3 81.2 (15.8)4 6.0 (1.9) 9.3 (2.8) 41.6 (10.9) 7.9 (3.5) 8.3 (3.1) 8.3 (3.1) 8.7 (3.2)

549

Range 615 029 2048 2258 55112 210 214 1561 014 015 215 215

Note. 1 Childrens Depression Inventory (CDI) score, 2 State-Trait Anxiety Inventorychilds version (STAI-C) State score, 3 State-Trait Anxiety Inventorychilds version (STAI-C) Trait score, 4 Wechsler Intelligence Scale for Children-III (WISC-III) Full Scale IQ score; all scores, except Wechsler Adult Intelligence Scale-Full Scale (WAIS-III) IQ and WISC-III Full Scale IQ, are raw scores; gender = female/male; ethnicity = Caucasian/AfricanAmerican/other.

1.2. CVLT-C group Demographics of the CVLT-C group are presented in Table 1. Records from children diagnosed with a formal memory disorder were not included in the study, nor were children with a Full Scale IQ below 55 (a 3 standard deviation cut-off was implemented to remove any extreme outliers). The primary DSM-IV diagnoses of the sample were: anxiety disorder NOS (n = 1), attention decit hyperactivity disorder (n = 17), cognitive disorder NOS (n = 10), conduct disorder (n = 1), general anxiety disorder (n = 1), intermittent explosive disorder (n = 2), learning disability (n = 3), major depression, single episode (n = 1), oppositional deant disorder (n = 3), schizophrenia (n = 1), and no diagnosis (n = 3). 1.3. CVLT group Descriptive demographics of the CVLT group are presented in Table 1. Any patient records that included a diagnosis of dementia (regardless of etiology) or that had a Full Scale IQ less than 55 were excluded from both the CVLT and CVLT-2 group. The primary DSM-IV diagnoses in the CVLT group were: alcohol dependence (n = 4), anxiety disorder NOS (n = 4), Aspergers Syndrome (n = 1), attention decit hyperactivity disorder (n = 7), bipolar disorder (n = 1), borderline personality disorder (n = 2), depression NOS (n = 3), intermittent explosive

550

J.R. OJile et al. / Archives of Clinical Neuropsychology 20 (2005) 547553

disorder (n = 3), major depressive disorder, recurrent (n = 7), major depressive disorder, single episode (n = 6), polysubstance dependence (n = 1), post-traumatic stress disorder (n = 3), psychosis NOS (n = 1), schizophrenia (n = 1), no diagnosis (n = 4). 1.4. CVLT-2 group Demographics of the CVLT-2 group are presented in Table 1. The primary DSM-IV diagnoses for this group were: attention decit hyperactivity disorder (n = 4), anxiety disorder NOS (n = 1), bipolar disorder (n = 2), cognitive disorder NOS (n = 4), delusional disorder (n = 1), depression NOS (n = 3), major depressive disorder, recurrent (n = 2), major depressive disorder, single episode (n = 1), multiple sclerosis (n = 1), post-traumatic stress disorder (n = 2), schizophrenia (n = 1), undifferentiated somatoform disorder (n = 1), and no diagnosis (n = 2).

2. Materials and procedures Patients were administered the California Verbal Learning Testchildrens version (CVLTC), the California Verbal Learning Test (CVLT), or the California Verbal Learning Test2nd edition (CVLT-2) as part of a larger xed neuropsychological battery. The format of each version of the CVLT is similar, and therefore a general description can be given. The CVLT is a list learning task that assesses ability to learn words over ve successive learning trials, and then, following a single attempt to learn a novel list of words, ability to immediately recall the previously learned words (with and without being cued). After a 20 min delay during which unrelated nonverbal tasks are performed, ability to recall (with and without being cued) and recognize the words is again tested (Delis, Kramer, Kaplan, & Ober, 1987, 1994, 2000). Additional measures administered as part of the xed neuropsychological battery included self-report measures of depression: the Beck Depression Inventory (BDI) for adults and the Child Depression Inventory (CDI) for children (Beck, Steer, & Brown, 1996; Kovacs, 1992). Both of these instruments are self-report rating measures of various affective, cognitive, and physiological manifestations of depression. Self-report measures of anxiety included the StateTrait Anxiety Inventory (STAI) for adults and the State-Trait Anxiety Inventory for Children (STAI-C) (Spielberger, 1973, 1983). Both instruments are self-report rating measures of the affective, cognitive, and physiological manifestations of anxiety in terms of current experience (i.e., state anxiety) and long-standing patterns (i.e., trait anxiety). Intellectual assessment was based on the Wechsler Intelligence Scale for Children3rd edition (WISC-III Full Scale IQ) and the Wechsler Adult Intelligence Scale3rd edition (WAIS-III Full Scale IQ). The Mattis Dementia Rating Scale (DRS) was also administered as part of the adult battery and was used to screen out patients thought to have dementia. All tests were administered according to their standardized protocols. 2.1. Statistical analyses To examine the relationship between CVLT outcome variables and self-reports of depression and anxiety, multiple regression analyses were performed to determine the amount of variance

J.R. OJile et al. / Archives of Clinical Neuropsychology 20 (2005) 547553

551

the CDI, STAI-C, BDI, and STAI individually predicted above the covariates of age, gender, and IQ. The following dependent variables were analyzed for the CVLT-C, CVLT, and CVLT2: Trial 1, Trial 5, total Trials 15, short delay free and cued recall and long delay free and cued recall. Tests of normality for the dependent variables indicated all were normally distributed. Regression diagnostics indicated no problems with variance ination or tolerance values in any regression model for any version of the CVLT. Effect sizes in terms of additional variance accounted for beyond that predicted by the three covariates are reported.

3. Results 3.1. CVLT-C group Desriptives and mean test scores for the CVLT-C, CVLT and CVLT-2 groups can be found in Table 1. Variance predicted by CDI ranged from .035 (Long Delay Free Recall) to .078 (Trial 1). Variance predicted by STAI-C State scores ranged from .005 (Trial 5) to .033 (Short and Long Delay Cued Recall trials), and variance predicted by STAI-C Trait scores ranged from .000 (Long Delay Cued Recall) to .023 (Long Delay Free Recall). 3.2. CVLT group Variance accounted for beyond that by the BDI ranged from .013 (Short Delay Cued Recall) to .029 (Short Delay Free Recall). Variance accounted for by the STAI State and Trait scores ranged from .000 (Trial 1, Short Delay Cued Recall, and Long Delay Free Recall) to .003 (Trial 5) and .000 (Trial 1) to .029 (Short Delay Free Recall), respectively. 3.3. CVLT-2 group Variance predicted by the BDI ranged from .000 (Long Delay Cued Recall) to .018 (Trial 5). Additional variance predicted by STAI State and Trait scores ranged from .001 (Trial 1) to .098 (Short Delay Cued Recall) and .000 (Trials 1 and 5) to .034 (Short Delay Free Recall).

4. Discussion The potential relationship between mood and anxiety and memory performance is of obvious importance to practitioners functioning within psychiatric settings as noted by several recent studies (Foldi et al., 2003; Horan et al., 1997; Kizilbash et al., 2002). However, the current study is the rst to explicitly examine this relationship on all three versions of the CVLT within a mixed psychiatric sample. The current study found that neither anxiety nor depression accounted for a signicant portion of variance in CVLT scores once the covariates of age, gender, and Full Scale IQ were accounted for. In fact, our ndings are in line with those of Kizilbash and colleagues (2002). That is, even though the impact of depression was statistically signicant in their study, the amount of variance accounted for by either depression or

552

J.R. OJile et al. / Archives of Clinical Neuropsychology 20 (2005) 547553

anxiety was very small. Depression only accounted for up to 2% of the variance in any CVLT score and anxiety only 1%. Therefore, our ndings are congruent with these authors ndings that the effect sizes of depression and anxiety on CVLT scores are small. These results suggest that when practitioners are screening psychiatric patients for neuropsychological functioning, frank memory decits observed on the various versions of the CVLT are likely representative of actual memory disturbance rather than epiphenomena associated with anxious or depressive symptomatology. For example, if a patient is being evaluated for memory decits and also presents with concomitant depression and/or anxiety, other tests known to be sensitive to depression and/or anxiety should be examined (such as processing speed) to support a conceptualization of negative affect being an etiological factor in poor test performance. Some potential limitations of the current study are noteworthy. First was the use of a single method of assessing depressive and anxious symptomatology (i.e., self-report measures). However, these assessment devices are frequently utilized in neuropsychological evaluations. It should be noted that no diagnoses were made based on these self-report measures, rather, the present study sought to examine if self-report of depressive and anxious symptomatology could be shown to appreciably predict CVLT performance. Secondly is the relatively small sample size per CVLT group. However, for the purposes of the current study the sample sizes should be sufcient, since the main goal was to examine the effect size as indicated by variance predicted rather than simple signicance testing. Obviously, with larger sample sizes the small amount of variance predicted for most of the outcome variables would become signicant as seen in Kizilbash et al. (2002). A larger sample size would address the issue of limited power in the present study, although it is likely that effect sizes would not change substantially, as noted by the consistency between the effect sizes found in the current study and those of Kizilbash and colleagues. Thirdly, the generally poor performance on the CVLT in the current samples may potentially limit the generalizability of the current ndings to lower functioning psychiatric populations. A fourth issue is the exclusion of subjects with memory disorders. This was done in order to provide a clearer picture of the effects of affective distress on verbal memory functions, but may have also reduced the generalizability of these ndings. Lastly, the range of IQs in the current samples was large and the mean was relatively low. It is possible that a higher functioning group (regarding both intelligence and memory) might demonstrate a different relation between CVLT scores and self-report measures of depressive and anxious symptoms. The current results suggest that when neuropsychological practitioners are assessing functioning in psychiatric settings, memory decits observed on the various versions of the CVLT may be minimally associated with self-report of anxious and depressive symptomatology. Obviously, a thorough clinical interview conducted (both with the patient and family members or other caregive) would be an essential feature of exploring such a conceptualization as well. These ndings await replication in a larger similar sample.

References
Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory (2nd ed.). San Antonio: The Psychological Corporation.

J.R. OJile et al. / Archives of Clinical Neuropsychology 20 (2005) 547553

553

Colby, C. A., & Gothlib, I. H. (1988). Memory decits in depression. Cognitive Therapy and Research, 12, 611627. Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (1987). California verbal learning: Manual. San Antonio: The Psychological Corporation. Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (1994). California Verbal Learning Testchildrens version: Manual. San Antonio: The Psychological Corporation. Delis, D. C., Kramer, J. H., Kaplan, E., & Ober, B. A. (2000). California Verbal Learning Test2nd edition: Manual. San Antonio: The Psychological Corporation. Foldi, N. S., Brickman, A. M., Schaefer, L. A., & Knutelska, M. E. (2003). Distinct serial position proles and neuropsychological measures differentiate late life depression from normal aging and Alzheimers disease. Psychiatry Research, 120, 7184. Horan, W. P., Pogge, D. L., Borgaro, S. R., Stokes, J. M., & Harvey, P. D. (1997). Learning and memory in adolescent psychiatric inpatients with major depression: A normative study of the California Verbal Learning Test. Archives of Clinical Neuropsychology, 12, 578584. Humphreys, M. S., & Revelle, W. (1984). Personality, motivation, and performance: A theory of the relationship between individual differences and information processing. Psychological Review, 91, 153184. Kindermann, S. S., & Brown, G. G. (1997). Depression and memory in the elderly: A meta-analysis. Journal of Clinical and Experimental Neuropsychology, 19, 625642. Kizilbash, A. H., Vanderploeg, R. D., & Curtiss, G. (2002). The effects of depression and anxiety on memory performance. Archives of Clinical Neuropsychology, 17, 5767. Kovacs, M. (1992). Manual for the childrens depression inventory manual. New York: Multi-Health Systems, Inc. Krames, L., & McDonald, M. (1985). Distraction and depressive cognitions. Cognitive Therapy and Research, 9, 561573. Miller, L. S., Faustman, W. O., Moses, J. A., Jr., & Csernansky, J. G. (1991). Evaluating cognitive impairment in depression with the Luria-Nebraska Neuropsychological Battery: Severity correlates and comparisons with nonpsychiatric controls. Psychiatry Research, 37, 219227. Otto, M. W., Bruder, G. E., Fava, M., Delis, D. C., Quitkin, F. M., & Rosenbaum, J. F. (1994). Norms for depressed patients for the California Verbal Learning Test: Associations with depression severity and self-report of cognitive difculties. Archives of Clinical Neuropsychology, 9, 8188. Porter, R. J., Gallagher, P., Thompson, J. M., & Young, A. H. (2003). Neurocognitive impairment in drug-free patients with major depressive disorder. British Journal of Psychiatry, 182, 214220. Richards, P. M., & Ruff, R. M. (1989). Motivational effects on neuropsychological functioning: Comparison of depressed and nondepressed individuals. Journal of Consulting and Clnical psychology, 57, 396402. Spielberger, C. D. (1973). Manual for the State-Trait Anxiety Inventory for children. Palo Alto, CA: Consulting Psychologists Press. Spielberger, C. D. (1983). Manual for the State-Trait Anxiety Inventory: STAI. Palo Alto, CA: Consulting Psychologists Press.