Chapter 7. Depression and cognitive impairment in old age: what comes first?

Transcription

1 Chapter 7 Depression and cognitive impairment in old age: what comes first? Vinkers DJ,Gussekloo J,StekML,W estendorp RGJ,van der Mast RC. Depression and cognitive impairment in old age: what comes first? Outcomes of the prospective population-based Leiden 85-plus Study. BMJ 2004;329: 881.

2 Chapter 7 60

3 Depression and cognitive impairment in old age: what comes first? Abstract Objectives To examine the temporal relation between depression and cognitive impairment in old age. DesignProspective population-based study. SettingCity of Leiden, the Netherlands. Participants 500 subjects from age 85 years through 89 years. M ain outcome measures Annual assessments of depressive symptoms (15-item Geriatric Depression Scale), global cognitive function (Mini-Mental State Examination), attention (Stroop Test), processing speed (Letter Digit Coding Test), immediate recall memory (Word Learning Test Immediate Recall), and delayed recall memory (Word Learning Test Delayed Recall). Results At age 85 years, depressive symptoms and cognitive impairment were highly significant correlated (all domains, p < 0.001). During follow-up, impaired attention (0.08 points, 95 % CI ), immediate recall memory (0.17 points, 95 % CI ), and delayed recall memory (0.10 points, 95 % CI )at baseline were all associated with an accelerated annual increase of depressive symptoms during follow-up. In contrast, depressive symptoms at baseline were not related to an accelerated cognitive decline during follow-up (all domains, p >.05). Conclusion Caregivers should be aware of the development of depressive symptoms when cognitive impairment is present. The presence of depression only does not mean that cognitive decline is yet to come. 61

4 Chapter 7 Introduction Depression and cognitive impairment rank among the most significant mental health problems in the elderly. The consequences of both depression and cognitive impairment in the elderly are known to be severe, including diminished quality of life, functional decline, increased service utilisation and high mortality 1. Frequently, late-onset depression and cognitive impairment co-occur, suggesting a close relationship between them It is not known, however, if depression leads to cognitive decline or whether cognitive impairment leads to depression 4 5. Clinical practice and research evidence suggest that depression precedes cognitive decline in old age Inference of the data so far is hampered by the fact that most most studies on this topic examined only the association between depression and the subsequent development of cognitive impairment As depression may be an early sign instead of an independent risk factor for cognitive impairment, the temporal relation between depression and cognitive impairment in old age remains unclear. Hitherto, only one study has investigated the temporal relationship between depression and cognitive impairment in subjects aged 65 years and older 15. The study is limited, however, by the use of the clinical diagnosis of depression and dementia as dichotomous endpoint. Therefore, we followed 500 community-dwelling elderly people from age 85 through 89 years with annual assessments of depressive symptoms and various cognitive functions and determined the temporal relation. Methods The Leiden 85-plus Study The Leiden 85-plus Study is a prospective population-based study of all 85-year old inhabitants of Leiden, The Netherlands. Between 1 September 1997 and 1 September 1999, 599 subjects were enrolled (response rate 87 %). Subjects were visited annually from age 85 through 89 years at their home for face-to-face interviews. The Medical Ethical Committee of the Leiden University Medical Center approved the study and all subjects gave informed consent to participate 16. In the present analysis, all 500 subjects (83 %) without severe cognitive impairment at baseline (MMSE score 19 points) were included. Figure 7.1 shows the annual follow-up visits of these subjects from age 85 through 89 years within the Leiden 85-plus Study. During 1459 personyears of follow-up (mean per person, 2.9 years), 39 subjects (8 %) declined to participate in the repeated measurements, most of them at the first follow-up visit. From the 334 subjects without significant depressive symptoms at baseline (GDS-15 score 2 points), 97 subjects (29 %) died and 28 subjects (8 %) declined to participate during follow-up. From the 415 non-demented subjects at baseline (MMSE score 24 points), 124 subjects (30 %) died and 32 subjects (8 %) declined to participate during follow-up. 62

5 Depression and cognitive impairment in old age: what comes first? 500 subjects 42 died 430 subjects 28 refused 41 died 385 subjects 4 refused 36 died 346 subjects 3 refused 44 died 298 subjects 4 refused Figure 7.1 Annual follow-up visits of 500 participants without severe cognitive impairment at baseline (MMSE score 19 points) from age 85 through 89 years within the Leiden 85-plus. Cognitive function Global cognitive function was assessed with the Mini-Mental State Examination (MMSE) 17. Possible scores on the MMSE range from 0 to 30 points, with lower scores indicating an impaired global cognitive functioning. Attention was measured using the third chart of the 40- item Stroop Test. Outcome is the time needed to name the ink colour of incongruously printed names of colours. Higher scores indicate a lesser attention. Processing speed was assessed with the Letter Digit Coding Test. Outcome is the total number of correct digits according to a code key in 60 seconds. Lower scores indicate a slower speed. Immediate and delayed recall memories were determined with the 12 Word Learning Test. In this test, 12 pictures are presented, and the subject is asked to recall the presented pictures. This procedure is carried out three times. Outcome for the immediate recall memory is the total number of correct recalled words immediately after each procedure. Outcome for the delayed recall memory is the total number of correct recalled words after 20 minutes. Lower scores indicate an impaired memory. Parallel versions for the coding and memory test, using identical procedures, but with different items were available to prevent learning effects. These neuropsychological tests are sensitive to small differences in elderly subjects 18. Attention, processing speed, immediate and delayed recall memory were not assessed in subjects with a MMSE score of 18 points or lower, because these tests lack reliability and validity in subjects with severe cognitive impairment. 63

6 Chapter 7 Depressive symptoms Depressive symptoms were annually assessed with the 15-item Geriatric Depression Scale (GDS-15), a questionnaire especially developed as a screening instrument for depressive symptoms in elderly populations 19. The use of the GDS-15 to detect depression has been validated within the Leiden 85-plus Study 20, whereas we have shown earlier that the GDS-15 is able to ascertain longitudinal alterations in depressive symptoms 21. The GDS-15 was not administered in subjects with a MMSE score of 18 points or lower, because this test lacks reliability and validity in subjects with severe cognitive impairment. Statistical analysis Continuous data at baseline are presented as medians and interquartile ranges. The crosssectional correlation between depressive symptoms and cognitive function was estimated with Pearson s correlation coefficient. The temporal relation between depression and cognitive function was assessed using separate linear mixed models adjusted for sex and level of education. Mixed models use all available data during follow-up, can properly account for correlation between repeated measurements, and can handle missing data more appropriately than traditional models. They allow for the use of time-independent and timedependent covariates 22. First, we examined the impact of cognitive function at baseline on the course of depressive symptoms. This analysis was restricted to subjects without significant depressive symptoms at baseline (GDS-15 score 2 points). Models included cognitive function at baseline, time, and the interaction of cognitive function at baseline and time. The estimate for "cognitive function at baseline"reflects the cross-sectional impact of cognitive function on depressive symptoms avaraged across all time-points. The estimate for "time"reflects the annual change of depressive symptoms. The estimate for "interaction"of cognitive function at baseline and time reflects the additional annual effect per standard deviation of the mean of cognitive function at baseline on the course of depressive symptoms. The impact of cognitive function at baseline on the course of depressive symptoms is reflected by the estimate for "interaction" of cognitive function and time. Second, we examined the impact of depressive symptoms at baseline on the course of cognitive function. This analysis was restricted to subjects without serious cognitive impairment at baseline (MMSE score 24 points). Models included depressive symptoms at baseline, time, and the interaction of depressive symptoms at baseline and time. The estimate for "depressive symptoms at baseline" reflects the cross-sectional impact of depressive symptoms on cognitive function averaged across all time-points. The estimate for "time" reflects the annual change of cognitive function. The estimate for "interaction"of depressive symptoms at baseline and time reflects the additional annual effect per standard deviation of the mean of depressive symptoms at baseline on the course of cognitive function. The impact of depressive symptoms at baseline on the course of cognitive function is reflected by the estimate for "interaction"of depressive symptoms and time. 64

7 Depression and cognitive impairment in old age: what comes first? Results Table 7.1 presents the demographic and clinical characteristics of all 500 participants at baseline. There were 184 men (37 %) and 303 subjects (61 %) with a low level of education defined as a maximum of six years of schooling. There were 334 (67 %) subjects without significant depressive symptoms (GDS-15 score 2 points) and 415 (83 %) subjects without serious cognitive impairment (MMSE score 24 points). Table 7.1 Demographic and clinical characteristics of 500 community-dwelling subjects aged 85 years. Men 184 (37 %) Low level of education 303 (61 %) Independent living 444 (89 %) Depressive symptoms 15-item Geriatric Depression Scale (points) a 2 (1 3) GDS-15 score 2 points 334 (67 %) Cognitive function Mini-Mental State Examination (points) a 27 (24-28) MMSE score 24 points 415 (83 %) Stroop Test (seconds) a 74 (60-97) Letter Digit Coding Test (digits) a 16 (12-21) Word Learning Test Immediate Recall (words) a 25 (21-28) Word Learning Test Delayed Recall (words) a 9 (7-11) a Continous data are presented as medians with interquartile ranges. In table 7.2 the cross-sectional correlation between depressive symptoms and cognitive function at baseline is shown. Depressive symptoms were significantly correlated with lower test scores for global cognitive function, attention, processing speed, immediate and delayed episodic memory, and with higher test scores for attention indicating lesser attention (all domains, p <.001). Table 7.2 Cross-sectional correlation of depressive symptoms with various domains of cognitive function in 500 community-dwelling subjects aged 85 years. R P value Global cognitive function <.001 Attention <.001 Processing speed <.001 Immediate recall memory <.001 Delayed recall memory <.001 Correlation was calculated with Pearson`s correlation coefficient. Depressive symptoms were assessed with the 15-item Geriatric Depression Scale (GDS-15). 65

8 Chapter 7 Table 7.3 demonstrates the impact of cognitive function at baseline on the course of depressive symptoms in 334 non-depressed subjects (67 %) with a GDS-15 score of 2 points or lower at baseline. Impairment of attention (0.08 points, 95 % CI ), immediate recall memory (0.17 points, 95 % CI ), as well as delayed recall memory (0.10 points, 95 % CI ) at baseline were associated with an accelerated annual increase of depressive symptoms during follow-up. These estimates remained similar when subjects with a GDS-15 score of higher than 2 points were included (data not shown). Table 7.3 The impact of cognitive function at baseline on depressive symptoms from age 85 through 89 years in 334 community-dwelling subjects without depressive symptoms at baseline (GDS-15 score 2 points). Additional annual increase of depressive P value symptoms (points ± SE) Global cognitive function at baseline (per SD) ± Attention at baseline (per SD) 0.08 ± Processing speed at baseline (per SD) ± Immediate episodic memory at baseline (per SD) ± Delayed episodic memory at baseline (per SD) ± The impact of cognitive function is presented per SD of the cognitive testscores at baseline (respectively 2.65 points, seconds, 7.08 digits, 5.72 words, and 2.74 words). Associations were assessed by separate linear mixed models adjusted for sex and educational level. There was a highly significant increase of the GDS-15 score over time in all models (all, p < 0.001). Table 7.4 shows the impact of depressive symptoms at baseline on the course of cognitive function in 415 non-demented subjects (83 %) with a MMSE score of 24 points or more at baseline. Depressive symptoms at baseline were not associated with an accelerated cognitive decline during follow-up (all domains, p > 0.05). These estimates remained similar when subjects with a MMSE score lower than 24 points were included (data not shown). Table 7.4 The impact of depressive symptoms at baseline on cognitive decline from age 85 through 89 years in 415 community-dwelling subjects without cognitive impairment at baseline (MMSE score 24 points). Additional annual impact on the course of P value cognitive function (estimate ± SE) Global cognitive function (points extra per year) ± Attention (seconds extra per year) ± Processing speed (digits extra per year) ± Immediate recall memory (words extra per year) ± Delayed recall memory (words extra per year) ± Depressive symptoms were assessed with the 15-item Geriatric Depression Scale (GDS-15). The impact of depressive symptoms at baseline is presented per standard deviation of the mean of the GDS-15 score at baseline (2.11 points). Associations were assessed by separate linear mixed models adjusted for sex and educational level. There was a highly significant decline of all cognitive domains over time (all, p < 0.001). 66

9 Depression and cognitive impairment in old age: what comes first? Discussion Depressive symptoms and cognitive impairment were highly significantly correlated crosssectionally showing that indeed they co-occur in old age. Critical is our finding that cognitive impairment at baseline was associated with an accelerated increase of depressive symptoms. In contrast, depressive symptoms at baseline were not related to an accelerated cognitive decline. Taken together, the prospective analysis of our data shows that cognitive impairment precedes the onset of depressive symptoms and not the other way round. We found specifically that impairment of attention and memory precedes the development of depressive symptoms. How does cognitive impairment lead to depression in old age? The awareness of cognitive decline may cause depressive symptoms as a psychological reaction to the loss of cognitive functioning. Indeed, memory complaints in old age may be an early sign of dementia and as such upset elderly people 23. Thus, especially adequate functioning of attention and memory may be important to elderly subjects, explaining why their loss is associated with an accelerated increase of depressive symptoms. A common etiology or sharing of risk factors is less likely to explain the association between depression and cognitive impairment. In that case, we would have expected that depressive symptoms precede cognitive decline also. Causal inference of the relation between depression and cognitive impairment in old age has been hampered by the fact that most studies examined only one direction of this relation. Some studies found that depression is a risk factor for the development of cognitive decline , while other studies could not confirm this finding Examination of both directions of the relationship between depression and cognitive impairment shows that depression in old age is a concomitant phenomenon of already existing cognitive impairment, instead of an independent risk factor. Our findings, using various domains of cognitive function instead of a dichotomous endpoint, are in line with a large population-based study in subjects of 65 years and older showing that depression is an early manifestation rather than a predictor of Alzheimer s disease 15. Thus, in the general population of the oldest old the presence of depressive symptoms does not mean that cognitive decline is yet to come. Major strengths of this study are the annual assessments of depressive symptoms and various cognitive functions, the use of linear mixed models, and the population-based sample of elderly subjects. A possible weakness is that depression is not formally diagnosed. It becomes however increasingly clear that depressive symptoms in old age have similar serious consequences as compared with depressive disorders 24. Hence, depression may better be interpreted as a continuum of depressive symptoms than the mere presence or absence of a depressive disorder. In conclusion, cognitive decline precedes depression in old age. We found that specifically impairment of attention or memory precedes the development of depressive symptoms. Depression appears to be a concomitant symptom of cognitive impairment instead of an independent risk factor. Therefore, caregivers should pay special attention to early detection and treatment of depressive phenomena in elderly subjects with cognitive impairment. 67

10 Chapter 7 References 1 Macdonald AJD, Mental health in old age. BMJ 1997; 315: Migliorelli R, Teson A, Sabe L, Petracchi M, Leiguarda R, Starkstein SE. Prevalence and correlates of dysthymia and major depression among patients with Alzheimer's disease. Am J Psychiatry 1995; 152: Zubenko GS, Zubenko WN, McPherson S, et al. A collaborative study of the emergence and clinical features of the major depressive syndrome of Alzheimer`s disease. Am J Psychiatry 2003; 160: Jorm AF. History of depression as a risk factor for dementia: an updated review. Aust N Z J Psychiatry 2001; 35: Schweitzer I, Tuckwell V, O'Brien J, Ames D. Is late onset depression a prodrome to dementia? Int J Geriatr Psychiatry 2002;17: Devanand DP, Sano M, Tang M-X, Taylor S, Gurland BJ, Wilder D, et al. Depressed mood and the incidence of Alzheimer's disease in the elderly living in the community. Arch Gen Psychiatry 1996; 53: Yaffe K, Blackwell T, Gore R, Sands L, Reus V, Browner WS. Depressive symptoms and cognitive decline in nondemented elderly women. Arch Gen Psychiatry 1999; 56: Paterniti S, Verider-Taillerfer M-H, Dufouil C, Alpérovitch A. Depressive symptoms and cognitive decline in elderly people. Br J Psychiatry 2002; 181: Green RC, Cupples LA, Kurz A, et al. Depression as a risk factor for Alzheimer disease.arch Neurol 2003; 60: Wilson RS, Mendes de Leon CF, Bennett DA, Bienias JL, Evans DA. Depressive symptoms and cognitive decline in a community population of older persons. J Neurol Neurosurg Psychiatry 2004; 75: Dufouil C, Fuhrer R, Dartigues J-F, Alpérovitch A. Longitudinal analysis of the association between depressive symptomatology and cognitive deterioration. Am J Epidemiol 1996; 144: Henderson AS, Korten AE, Jacomb PA, Jorm AF, Rodgers B, Jacomb P, et al. The course of depression in the elderly: a longitudinal community-based study in Australia. Psychol Med 1997;27: Cervilla JA, Prince M, Joels S, Mann A. Does depression predict cognitive outcome 9 to 12 years later? Evidence from a prospective study of elderly hypertensives. Psychol Med 2000; 30: Broday H, Luscombe G, Anstey KJ, Cramsie J, Andrews G, Peisah C. Neuropsychological performance and dementia in depressed patients after 25-year follow-up: a controlled study. Psychol Med 2003; 33: Chen P, Ganguli M, Mulsant BH, DeKosky ST. The temporal relationship between depressive symptoms and dementia. Arch Gen Psychiatry 1999; 56: Bootsma - van der Wiel A, van Exel E, de Craen AJM, et al. A high response is not essential to prevent selection bias: results from the Leiden 85-plus Study. J Clin Epidemiol 2002; 55: Folstein MF, Folstein SE, McHugh PR. "Mini-Mental State": a practical method for grading the cognitive state of patients for the clinician.j Psychiatr Res 1975; 12: Houx PJ, Shepherd J, Blauw GJ, et al. Testing cognitive function in elderly populations: the PROSPER study. J Neurol Neurosurg Psychiatry 2002; 73: Yesavage JA, Brink TL, Rose TL, et al. Development and validation of a geriatric depression screening scale: a preliminary report. J Psychiatr Res 1982; 1: De Craen AJM, Heeren TJ, Gussekloo J. Accuracy of the 15-item Geriatric Depression Scale (GDS- 15) in a community sample of the oldest old. Int J Geriatr Psychiatry 2003; 18:

11 Depression and cognitive impairment in old age: what comes first? 21 Vinkers DJ, Gussekloo J, Stek ML, Westendorp RGJ, van der Mast RC. The 15-item Geriatric Depression Scale (GDS-15) detects changes in depressive symptoms after a major negative life event. The Leiden 85-plus Study. Int J Geriatr Psychiatry 2004; 19: Gueorguieva R, Krystal JH. Move over ANOVA: progress in analyzing repeated-measures data and its reflection in papers published in the Archives of General Psychiatry. Arch Gen Psychiatry 2004; 61: Jonker C, Geerlings MI, Schmand B. Are memory complaints predictive for dementia? A review of clinical and population-based studies. Int J Geriatr Psychiatry 2000; 15: Geerlings SW, Beekman AT, Deeg DJ, et al. Duration and severity of depression predict mortality in older adults in the community. Psychol Med 2002; 32:

12