Wk 5 Assignment 1

Original Article

Five-year follow-up of cognitive impairment in older adults with bipolar disorder

Schouws SNTM, Comijs HC, Dols A, Beekman ATF, Stek ML. Five-year follow-up of cognitive impairment in older adults with bipolar disorder. Bipolar Disord 2016: 18: 148–154. © 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.

Objective: To date, cognitive impairment has been thought to be an integral part of bipolar disorder. In clinical staging models, cognitive impairment is one of the hallmarks to define the clinical stage and it plays an important role in identifying the risk factors for progression to later stages of the illness. It is important to examine neurocognitive performance over longer periods to test the hypothesis of neuroprogression of bipolar disorder.

Methods: A comprehensive neuropsychological test battery was applied at baseline and five years later to 56 euthymic older outpatients with bipolar disorder (mean age = 68.35 years, range: 60–90 years) and to a demographically matched sample of 44 healthy subjects. A group-by- time repeated measures multivariate analysis of variance was performed to measure changes over time for the two groups. The impact of baseline illness characteristics on the intra-individual change in neurocognitive performance within the bipolar disorder group was studied by using logistic regression analysis.

Results: At baseline and at follow-up, patients with bipolar disorder performed worse on all neurocognitive measures compared to the matched healthy subjects. However, there was no significant group-by- time interaction between the patients with bipolar disorder and the comparison group.

Conclusions: Although older patients with bipolar disorder had worse cognitive function than healthy subjects, they did not have greater cognitive decline over a five-year period. The change in acquired cognitive impairment of patients with bipolar disorder might parallel the cognitive development as seen in normal aging.

Sigfried NTM Schouwsa, Hannie C Comijsa,b, Annemieke Dolsa, Aartjan TF Beekmanb and Max L Steka

aDepartment of Old Age Psychiatry, GGZ

inGeest, bDepartment of Psychiatry, EMGO

Institute of Care and Health Research VU

University Medical Center, Amsterdam, The

Netherlands

doi: 10.1111/bdi.12374

Key words: bipolar disorder – cognition – old

age

Received 19 June 2015, revised 11 November

2015, revised and accepted for publication 16

November 2015

Corresponding author:

Sigfried N.T.M. Schouws, Ph.D.

Department of Old Age Psychiatry

GGZ ingest

Amstelveenseweg 589

1081 JC Amsterdam

The Netherlands

Fax: +31-20-3016926

E-mail: [email protected]

Bipolar disorder is a psychiatric illness character- ized by periods of elevated mood and of depres- sion. It is a chronic illness that is associated with functional impairment, including impairment of cognitive functioning. A large number of cross-sec- tional studies have revealed cognitive impairment in patients with bipolar disorder of any age, even in the euthymic state (1–3). Although some studies suggest that there is a progressive increase in cogni- tive impairment in older patients with bipolar dis- order, even exposing them to an increased risk of dementia (4), the debate continues as to whether

the illness is related to (progressive) neurodegener- ation (5). New models are emerging to understand the long-term development of severe mental ill- nesses such as bipolar disorder (6). In these clinical staging models, earlier and milder symptoms are distinguished from later, more impairing factors. Later stages of the illness might mimic accelerated aging – that is, oxidative damage to cells due to chronic stress associated with mood disorders. The process of a lifetime accumulation of toxic pro- cesses and cerebrovascular brain changes may lead to more cognitive impairment later in life (7). It is

148

Bipolar Disorders 2016: 18: 148–154 © 2016 John Wiley & Sons A/S Published by John Wiley & Sons Ltd.

BIPOLAR DISORDERS

important to identify the risk factors that underlie these changes, as an implicit aim of staging is to prevent progression to advanced stages of the ill- ness. On the one hand, this knowledge can enable personalized interventions for individual patients in clinical psychiatry, and on the other hand it can shed light on specific neurophysiological pathways involved in the progression of the bipolar disorder. Cognition can thus be seen both as an indicator of their functional impairment and as a profiler of psychopathological progression of the disease.

Recent longitudinal studies in older patients with bipolar disorder, including one from our group, have not found a faster cognitive decline when compared to matched, healthy subjects, despite cognitive impairment at baseline (8, 9). However, in these studies, a two-year interval was used, which may be too short to detect declines in cognitive functioning. Although it is anticipated that persons with bipolar disorder would exhibit more cognitive deficits than a healthy comparison group, and that they would exhibit a faster decline at follow-up, it is not clear from the existing litera- ture what the outcomes might be. There is some evidence that individuals with bipolar disorder develop dementia at a higher rate than expected compared to healthy subjects (10), while other studies suggest that lithium treatment could reduce the risk of Alzheimer’s disease in patients with bipolar disorder (11). The association between risk factors such as course of the illness and cognitive deterioration is not clear (12).

The main object of the present study was to examine the course of cognitive functioning during a five-year period in a sample of older patients with bipolar disorder meeting strict criteria for euthy- mia as compared to a healthy cohort. We con- trolled for residual affective symptoms at both time points because even small degrees of residual manic or depressive symptoms may have a nega- tive impact on cognition. Mortality, dropout, and cognitive status were compared with outcomes for healthy comparison subjects. In addition, we stud- ied whether clinical characteristics predicted the course of cognitive function.

Methods

Subjects

As previously described, patients (>60 years) with bipolar I and bipolar II disorder, who were cur- rently euthymic, were recruited from outpatient clinics in four regions in the Netherlands (13). With help from the Dutch bipolar patient association (VMDB, Association for Manic Depressives and

their Relatives), additional patients were recruited. Extensive care was taken to exclude patients with current mood symptoms, which could influence cognition. Eligible subjects were reported by their psychiatrist to have been euthymic for at least three weeks. Excluded were patients with a primary diag- nosis of alcohol dependence or substance abuse and those with dementia. The control group con- sisted of persons who had no current or lifetime history of psychiatric or addiction disorders or recent memory complaints. They were recruited from community centers for older people and through advertisements in local newspapers. In Figure 1, a flow diagram shows inclusion and attri- tion in the study. The study was approved by the institutional review board, and written informed consent was obtained from all subjects.

The original sample consisted of 119 patients with bipolar disorder and 78 healthy controls. At follow-up after five years, all subjects were invited to participate in reassessment of their neuropsycho- logical functioning. Patients with current mood symptoms that could influence their cognitive func- tioning were excluded. From the original sample, 56 patients (47%) and 44 healthy persons (56%) par- ticipated in the follow-up assessment. The main rea- sons for loss to follow-up were: refusal (25 patients and six controls), death (seven patients and four controls), not euthymic (13 patients and three con- trols), not able because of physical illness (eight patients and five controls), and could not be reached (ten patients and 16 controls). Compared to persons who were lost to follow-up, the longitu- dinal sample (completers) did not differ with respect to age or gender. However, subjects from the longi- tudinal sample had a significantly higher premorbid intelligence quotient (IQ) than subjects who were lost to follow-up (p < 0.05). In addition, the com- pleters performed significantly better at baseline on the learning condition of the 10-word Test and Occupation Naming Test compared to subjects who were lost to follow-up (both p < 0.05).

Clinical measures

At baseline (T1), the clinical diagnosis of bipolar disorder was obtained using the Structured Clinical Interview for DSM-IV (SCID) (14). Illness charac- teristics such as age at onset and medication were derived from patient interviews and hospital medi- cal records. Premorbid intelligence was estimated using the Dutch Reading Test for Adults (NLV) and the Dutch version of the New Adult Reading Test (15). At baseline (T1) and at the five-year fol- low-up (T2), current depressive symptoms were assessed using the Centre for Epidemiologic

149

Five-year follow-up older adults with BD

Studies Depression Scale (CES-D) (16), and cur- rent mania symptoms using the Young Mania Rat- ing Scale (YMRS) (17). The Mini Mental State Examination (MMSE) (18) was used to provide an overall assessment of cognitive functioning. For the present study, a cerebrovascular risk score [range: 0–2; 0 = no cerebrovascular diseases (CDs), 1 = 1 CD, and 2 = more than one CD] was determined by asking respondents whether they currently or had ever had any of the following chronic diseases or disease events: disease of the circulatory system (myocardial infarction, angina pectoris, heart failure, cardiac arrhythmia), hyper- tension, a history of ischemic attack or stroke, or diabetes. Answers were coded as either yes or no for each of these diseases.

Neuropsychological test battery

at T1 and T2, all subjects completed a comprehen- sive battery of neuropsychological tests, grouped into four cognitive domains:

(i) Attention: Digit Span subtest of the Wechsler Adult Intelligence Scale (19), Trail Making Test– Part A (20), and The Amsterdam Short-Term Memory Test (ASTM) (21); (ii) Learning and memory: The 10-Word Test, a modified version of the Auditory Verbal Learning Test (22); (iii) Executive functioning: Trail Making Test–Part B (17), modified version of the Stroop Color Word Test (23), and Mazes (1–4) subtest of the Wechsler Intelligence Scale for Children (24); and (iv) Verbal fluency: Control Oral Word Association Test (COWAT) (25) and Animal and Occupation

Naming subtest of the Groningen Intelligence Test (26).

Statistical analyses

We compared patients with bipolar disorder and comparison subjects on each of the demographic and clinical variables, using t-tests for continuous measures and chi-square tests for categorical ones at baseline and follow-up. Distributions of the con- tinuous measures were examined prior to analyses. Differences between mood symptoms (CES-D and YMRS) and MMSE scores of patients with bipolar disorder and comparison subjects at baseline and follow-up were compared using multivariate analy- ses of variance.

Composite scores for each neurocognitive domain were calculated by converting the subject’s raw scores to standardized z-scores. These were then summed in each domain to provide a single score. A higher score indicated better neurocogni- tive performance.

Repeated measures analyses of covariance (mancova) were employed to analyze the longitu- dinal changes on neuropsychological testing of the two groups. Analyses were adjusted for age, pre- morbid intelligence, and manic symptoms. Effects of group, time, and group-by-time interactions were examined. A significant group-by-time inter- action would indicate that the change in cognitive functions over time differed between the groups. Effect sizes were calculated by means of eta- squared (with g < 0.15 indicating small to med- ium effects and g > 0.14 indicating large effects) (27).

Bipolar disorder

(n = 119)

Healthy controls

(n = 78)

• Refused (n = 25) • Died (n = 7) • Could not be

contacted (n = 10) • Not euthymic (n = 13) • Medical illness (n = 8)

• Refused (n = 6) • Died (n = 4) • Could not be

contacted (n = 16) • Not euthymic (n = 3) • Medical illness (n = 5)

(n = 56) (n = 44)

Baseline

After five years

Fig. 1. Flow chart follow-up of patients with bipolar disorder and controls after five years.

150

Schouws et al.

In order to examine inter-individual differences in patients with bipolar disorder with respect to cogni- tive decline over a five-year period, we dichotomized the difference scores from the domain scores in 0 = highest two tertiles representing no cognitive decline and 1 = lowest tertile representing relevant cognitive decline. The categorization into tertiles was chosen to distinguish persons with relevant cog- nitive decline from those with no or minor decline, given the sample size. To test whether demographic and illness characteristics were associated with cog- nitive decline, we firstly carried out a univariate logistic regression analysis. We then performed mul- tivariate logistic regression analyses on the variables that influence cognitive changes, adjusted for age and premorbid intelligence.

Two-tailed tests were used for all analyses and statistical significance was defined as p < 0.05. Statistics were processed using spss version 20.0 for Windows (SPSS Inc., Chicago, IL, USA).

Results

Demographic and baseline clinical and

neuropsychological comparisons

The baseline characteristics of the 56 participants with bipolar disorder and 44 control subjects are displayed in Table 1. There were no significant differences in premorbid IQ, education, or vascu- lar risk factors. Subjects with bipolar disorder

were significantly younger than controls and included more males (v2 = 7.61, p < 0.01). Lithium (71% of the patients), valproic acid, and carbamazepine were the typical medications for remitted patients in the present study. Among the 56 patients, only three subjects (5%) did not use any medication.

The scores on the YMRS and CES-D were below the cut-off at each time point, confirming a euthymic state both in patients with bipolar disor- der and healthy controls. However, the bipolar dis- order group had significantly higher YMRS scores than the comparison group on both measurements. The bipolar disorder group did not differ from the comparison group on CES-D scores, either at baseline or after five years, but both the patient group and the comparison group had higher CES- D scores after five years (F = 21.81, p < 0.01).

Table 2 shows the mean [standard deviation (SD)] of the raw neurocognitive scores at baseline and after five years. As in our earlier study with the original, larger sample (13), the subset of patients with bipolar disorder in the present study did less well on all cognitive tests when compared to the healthy subjects (all tests p < 0.05) at base- line and after five years.

Rate of cognitive decline in cognitive domains

Table 2 shows the mean (SD) of the z-scores on the different cognitive domains at baseline and at

Table 1. Sociodemographic and clinical characteristics in patients with bipolar disorder and healthy comparison subjects

Bipolar disorder (n = 56)

Healthy subjects (n = 44)

t-value p-value

Age, years, mean (SD) 68.16 (7.0) 70.16 (7.6) 1.36 0.24 Premorbid IQ, mean (SD) 111.21 (11.4) 114.3 (8.7) 1.48 0.11 Education, years, mean (SD) 5.18 (1.5) 5.32 (1.3) 0.48 0.27 Gender, male, n (%)a 28 (50) 11 (25) Vascular risk factors, mean (SD) 0.41 (0.6) 0.50 (0.7) 0.66 0.11 Hospital admissions, mean (SD) 3.84 (3.5) Depressive episodes, mean (SD) 4.25 (3.0) Manic episodes, mean (SD) Psychotic episode, n (%) 37 (66) Months euthymic, mean (SD) 55.63 (63.6) Illness duration, mean (SD) 31.31 (12.8) Age at onset, years, mean (SD) 38.52 (14.4) Age at first psychiatric treatment, years, mean (SD) 43.25 (15.4) Taking lithium, n (%) 40 (71)

Baseline After 5 years Baseline After 5 years

MMSE score, mean (SD) 28.75 (1.2) 27.80 (2.6) 28.36 (4.7) 28.98 (1.6) 0.92b 0.34 YMRS score, mean (SD) 0.96 (1.7) 1.18 (2.2) 0.05 (0.3) 0.09 (0.4) 16.42b <0.01 CES-D score, mean (SD) 10.00 (7.1) 15.1 (4.9) 8.98 (6.4) 14.79 (4.9) 0.50b 0.48

CES-D = Centre for Epidemiologic Studies Depression Scale; IQ = intelligence quotient; MMSE = Mini Mental State Examination; SD = standard deviation; YMRS = Young Mania Rating Scale. av2 = 7.61, p < 0.01. bF repeated-measures multivariate analyses of variance of the main effects of groups (tests of between-subject effects).

151

Five-year follow-up older adults with BD

the five-year follow-up. In order to compare changes over time for each cognitive domain, a group-by-time repeated-measures mancova was performed, controlling for the covariates (premor- bid IQ and age). The results show the main effects of groups for all cognitive domains (attention, memory, executive function, and language func- tion) but no significant interactions with time were found. This suggests that the rate of decline over the five-year follow-up was the same for patients with bipolar disorder and the mentally healthy comparison group.

Predicting cognitive decline

To test whether demographic and illness character- istics were associated with cognitive decline within the group of patients with bipolar disorder, uni- variate logistic regression analysis was performed. This showed that the severity of manic symptoms at baseline predicted a decline in the domain of memory [odds ratio (OR) = 0.73, 95% confidence interval (CI): 0.52–0.98], and the number of hospi- talizations predicted a decline in language function (OR = 0.85, 95% CI: 0.73–0.97). Multivariate logistic regression analysis controlling for age and premorbid intelligence showed that more manic

symptoms were significantly associated with a greater decline in memory (OR = 0.72, 95% CI: 0.53–0.99). The association between number of hospitalizations and language function was no longer significant.

Discussion

Older adults with bipolar disorder exhibit worse cognitive functioning compared to healthy older subjects but no greater cognitive decline at a five- year follow-up. None of the demographic or illness characteristics were associated with cognitive decline in the domains of attention, and executive and language function. Only subclinical manic symptoms at baseline predicted a decline in mem- ory. This is in agreement with earlier suggestions (28) that the recurrence of mania may have a long- term neuropsychological impact.

Our findings are in line with those of Depp et al. (29), Santos et al. (12), and our previous follow-up study over a two-year period (8), despite the differ- ences in study design. Depp et al. did not establish the euthymic state of patients, which may have influenced the results, and the patients studied by Santos et al. were much younger – the mean age was 44.7 years.

Table 2. Neuropsychological performance (tests and domain) at baseline (T1) and five-year endpoint (T2) for all participantsa

Measures

Bipolar disorder (n = 56) Healthy subjects (n = 44)

F-value p-value 2Baseline After 5 years Baseline After 5 years

Attention 0.96 (2.5) 0.70 (2.8) 1.49 (2.3) 1.04 (2.3) 25.91 <0.01 0.22 Time group 2.09 0.15 0.02 Digits forward 5.20 (0.9) 5.34 (1.1) 5.82 (0.9) 5.82 (1.3) Digits backward 4.05 (0.9) 4.07 (1.1) 4.82 (1.1) 4.70 (1.0) TMT-Part A, sec 55.24 (22.4) 59.35 (32.1) 42.16 (15.6) 47.30 (16.9) ASTM (Items 1–10) 28.35 (1.5) 27.80 (2.0) 29.14 (0.9) 28.53 (1.4) Declarative memory 1.01 (2.5) 0.76 (2.2) 1.40 (2.1) 0.97 (2.5) 25.62 <0.01 0.21 Time group 1.18 0.28 0.01 10-word Test Learning (Trials 1–5) 30.56 (6.1) 29.87 (6.9) 37.95 (5.4) 36.79 (7.6) Retention 4.64 (2.1) 4.56 (2.1) 6.89 (2.2) 6.30 (2.5) Recognition 18.48 (1.6) 18.63 (1.8) 19.14 (1.0) 18.93 (1.8) Executive function 0.43 (2.4) 0.64 (2.5) 1.05 (1.3) 1.02 (1.5) 28.27 <0.01 0.24 Time group 0.96 0.33 <0.01 Stroop Color Word Test,b sec 56.38 (27.0) 60.49 (24.6) 44.67 (13.4) 46.49 (15.2) TMT-Part B, sec 153.98 (105.9) 192.35 (123.9) 96.93 (38.9) 124.67 (87.0) Mazes (Items 1–4) Planning time 132.89 (66.5) 177.96 (136.1) 99.19 (50.8) 110.40 (57.1) Language function 1.07 (2.1) 0.84 (2.3) 1.37 (2.3) 1.07 (2.6) 42.48 <0.01 0.31 Time group 0.57 0.45 0.01 Letters D-A-T 27.34 (9.9) 28.46 (12.8) 38.48 (10.4) 35.20 (11.8) Animal naming 19.66 (4.8) 18.95 (5.4) 24.57 (6.3) 22.30 (6.3) Occupation naming 15.30 (4.7) 13.63 (4.6) 18.59 (4.8) 15.65 (5.5)

Values are expressed as mean (standard deviation). ASTM = Amsterdam Short Term Memory Test; TMT = Trail Making Test. aRepeated-measures analyses of covariance of the main effects of groups (tests of between-subject effects). bThis is a short form, consisting of reading the first four lines.

152

Schouws et al.

The patients with bipolar disorder studied by Dhingra and Rabins (10) and Gildengers et al. (30) exhibited significant cognitive decline in contrast to our results. In the former study, no control group was included, so it is difficult to detect the effect of other variables such as aging. In the rela- tively small group studied by Gildengers et al., almost one-third of patients had a history of alco- hol use disorder, which may have influenced the results negatively. Furthermore, these authors did not control for the effect of mood symptoms, to check whether a change in cognitive impairment corresponded to changes in mood symptoms. It seems likely that the group studied by Gildengers et al. had a greater impairment and was more prone to deterioration.

It has been suggested that lithium may have neu- roprotective abilities and may reduce the risk of developing dementia (11), and that longer lithium exposure is associated with better white matter integrity in older adults with bipolar disorder (31). In the present study, more than two-thirds of the patient group used lithium, while in the study by Gildengers et al. (31) only one-third of the patients used lithium. This could mean that in the present study the use of lithium may have compensated for cognitive decline, although the very large number of patients in our study using lithium led to a restricted of range of scores and thus made it diffi- cult to detect an effect of lithium on cognition. The finding that patients with bipolar disorder do not have a faster rate of cognitive decline may have been due to the shorter interval between baseline and follow-up than in the study of Dhingra and Rabins (10). A more appropriate follow-up inter- val for detecting cognitive changes might be between seven years (as used by Dhingra and Rabins) and ten years.

Although our naturalistic study had several strong points, such as the inclusion of a rela- tively large group of patients, using a compre- hensive battery of neuropsychological tests, and including a careful selection of patients who were in a euthymic state at both time points, it also had some limitations. As a result of the nat- ure of the study design, and due to old age, not all subjects completed the follow-up. Ten patients were excluded from the follow-up because they were not euthymic. There was no association between age of onset and decline in cognitive function for the group-as-a-whole. However, we observed that all of the five patients who had a more rapid cognitive decline or converted to a dementia (MMSE <25) had a late onset. These persons may have been the most severely ill, and may have had greater cog-

nitive decline. Finally, although efforts were made to recruit a representative bipolar disorder sample, this was not a catchment area study, so our euthymic outpatients and comparison sub- jects comprised a convenience sample. They may have belonged to a better outcome subgroup (survivors), in which case the results would only have applied to the subset of patients with bipo- lar disorder.

Conclusions

In spite of the limitations cited above, cognitive impairment is part of bipolar disorder in older people, and seems to be fairly consistent over time for at least a subgroup of patients. We speculate that patients with bipolar disorder may have reached their clinical stage in adulthood and do not decline in later life faster than healthy older subjects, or progress to more advanced clinical stages of bipolar disorder because of aging. The acquired cognitive impairment of patients with bipolar disorder might parallel the cognitive devel- opment seen in normal aging once patients reach 60–65 years of age. Prospective studies, following heterogeneous bipolar disorder samples, collecting a detailed description of past and current drug treatment and clinical features, are awaited. An important issue to be solved is whether cognitive impairment in bipolar disorder might be associated with relapse and poor clinical outcome. An impor- tant implication of the present study is that mood stabilizing seems important for preventing a decline in cognitive functioning as even subclinical symptoms may be associated with cognitive decline. Furthermore, late-life bipolar disorder patients may provide a perfect sample for studying clinical staging and profiling as these patients may have reached a stable clinical stage.

Acknowledgements

This research was supported by the Fonds voor Psychische Gezondheid.

Disclosures

The authors of this paper do not have any commercial associa- tions that might pose a conflict of interest in connection with this manuscript.

References

1. Quraishi S, Frangou S. Neuropsychology of bipolar disor- der: a review. J Affect Disord 2002; 72: 209–226.

2. Robinson LJ, Ferrier IN. Evolution of cognitive impair- ment in bipolar disorder: a systematic review of cross-sec- tional evidence. Bipolar Disord 2006; 8: 103–116.

153

Five-year follow-up older adults with BD

3. Samam�e C, Martino DJ, Strejilevich SA. A quantitative review of neurocognition in eythymic late-life bipolar dis- order. Bipolar Disord 2013; 15: 633–644.

4. Kessing LV, Andersen PK. Does the risk of developing dementia increase with the number of episodes in patients with depressive disorder and in patients with bipolar dis- order? J Neurol Neurosurg Psychiatry 2004; 75: 1662– 1666.

5. Gildengers AG, Chung K-H, Huang S-H, Begley A, Aizenstein HJ, Tsai S-Y. Neuroprogressive effects of lifetime illness duration in older adults with bipolar disorder. Bipolar Disord 2014; 16: 617–623.

6. Scott J, Lobyer M, Hichic I et al. Clinical staging in psy- chiatry: a cross-cutting model of diagnosis with heuristic and practical value. Br J Psychiatry 2013; 202: 243–245.

7. Rizzo LB, Costa LG, Mansur RB et al. The theory of bipolar disorder as an illness of accelerated aging: implica- tions for clinical care and research. Neurosci Biobehav Rev 2014; 42C: 157–169.

8. Schouws SNTM, Stek ML, Comijs HC, Dols A, Beek- man ATF. Cognitive decline in elderly bipolar disorder patients: a follow-up study. Bipolar Disord 2012; 14: 749–755.

9. Gildengers AG, Chisholm D, Butters MA et al. Two- year course of cognitive function and instrumental activ- ities of daily living in older adults with bipolar disorder: evidence for neuroprogression? Psychol Med 2013; 43: 801–811.

10. Dhingra U, Rabins PV. Mania in the elderly: a 5-7 year follow-up. J Am Geriatr Soc 1991; 39: 581–583.

11. Kessing LV, Forman JL, Andersen PK. Does lithium pro- tect against dementia? Bipolar Disord 2010; 12: 87–94.

12. Santos JL, Aparicio A, Bagney A et al. A five-year follow- up study of neurocognitive functioning in bipolar disorder. Bipolar Disord 2014; 16: 722–731.

13. Schouws SNTM, Comijs HC, Stek ML et al. Cognitive impairment in early and late bipolar disorder. Am J Geri- atr Psychiatry 2009; 17: 508–515.

14. First MB, Spitzer RL, Gibbon M, Williams JBW. Struc- tured Clinical Interview for DSM-IV Axis Disorders SCID-I/P (Version 2.0). New York: Biometrics Research Department, 1996.

15. Nelson HE, O’Connell A. Dementia: the estimation of pre- morbid intelligence levels using the New Adult Reading Test. Cortex 1978; 14: 234–244.

16. Radloff CS. The CES-D Scale: a self report depression scale for research in the general population. Appl Psychol Meas 1977; 1: 385–401.

17. Young RC, Biggs JT, Ziegler VE, Meyer DA. A rating scale for mania: reliability, validity and sensitivity. Br J Psychiatry 1978; 133: 429–435.

18. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state” a practical method for grading the cognitive state of patient for the clinician. J Psychiatr Res 1975; 12: 189–198.

19. Wechsler D. WAIS-R Manual, Wechsler Adult Intelli- gence Scale- Revised. Cleveland: Psychological Corpora- tion, 1981.

20. Reitan RM. Validity of the Trailmaking Test as an indica- tion of organic brain damage. Percept Mot Skills 1958; 8: 271–276.

21. Schmand B, Lindeboom J. The Amsterdam Short Term Memory Test. The Netherlands: PITS Leiden, 2005.

22. Rey A. L’examen clinique en psychologie. Paris: Presses Universitaires de France, 1964.

23. Golden CJ. Stroop Color Word Test. Chigago: Stroeting, 1978.

24. Wechsler D. Wechsler Intelligence Scale for Children – Revised UK Edition. Windsor: NFER- Nelson, 1976.

25. Benton AL, Hamsher K. Multilangual Aphasia Examina- tion. Iowa City: University of Iowa, 1976.

26. Luteijn F, van der Ploeg FAE. Handleiding Groninger Intelligentietest (GIT) [Manual Groningen Intelligence Test]. Lisse, The Netherlands: Swets & Zeitlinger, 1983.

27. Cohen J. Statistical Power Analysis for the Behavorial Sciences, 2nd edn. Erlbaum: Hillsdale, 1988.

28. Lopez-Jaramillo C, Lopera-V�asquez J, Gallo A et al. Effects of recurrence on the cognitive performance of patients with bipolar disorder: implications for relapse prevention and treatment adherence. Bipolar Disord 2010; 12: 557–567.

29. Depp CA, Savla GN, Moore DJ et al. Short-term course of neuropsychological abilities in middle-aged and older adults with bipolar disorder. Bipolar Disord 2008; 10: 684– 690.

30. Gildengers AG, Mulsant BH, Begley A et al. The longitu- dinal course of cognition in older adults with bipolar disor- der. Bipolar Disord 2009; 11: 744–752.

31. Gildengers AG, Butters MA, Aizenstein HJ et al. Longer lithium exposure is associated with better white matter integrity in older adults with bipolar disorder. Bipolar Dis- ord 2015; 17: 248–256.

154

Schouws et al.

Copyright of Bipolar Disorders is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.