1
Journal of Alzheimer’s Disease xx (20xx) xx–xx DOI 10.3233/JAD-142178 IOS Press
3
roo f
2
Depression and Mild Cognitive Impairment in the General Population: Results of the Heinz Nixdorf Recall Study
or P
1
6
Martha Dlugaja,∗ , Angela Winklera , Nico Draganob , Susanne Moebusc , Karl-Heinz J¨ockelb , Raimund Erbeld and Christian Weimara on behalf of the Heinz Nixdorf Recall Study Investigative Group
7
a Department
8
b Institute
4 5
14
Handling Associate Editor: Roberto Monastero
12
dA
11
cte
9 10
uth
13
of Neurology, University Hospital of Essen, University of Duisburg-Essen, Germany for Medical Sociology, Centre for Health and Society, University of D¨usseldorf, Medical Faculty, University of D¨usseldorf, Germany c Institute for Medical Informatics, Biometry and Epidemiology, University Hospital of Essen, University Duisburg-Essen, Germany d Clinic of Cardiology, West German Heart Centre, University Hospital of Essen, University Duisburg-Essen, Germany
Accepted 17 November 2014
32
Keywords: Aging, depression, epidemiology, mild cognitive impairment
18 19 20 21 22 23 24 25 26 27 28 29 30
co
17
Un
16
rre
31
Abstract. Background: The literature suggests an association between depression and mild cognitive impairment (MCI) and dementia, but not all studies have examined this association with regard to MCI subtypes reflecting different dementia etiologies. Objective: To examine if there is a cross-sectional relationship of depression and MCI and to examine if the relationship differs depending on the type of depression (currently elevated depressive symptoms or a positive history of lifetime depression or both) and on the MCI subtype (amnestic versus non-amnestic MCI (aMCI/naMCI)). Methods: From the second examination of the population-based Heinz Nixdorf Recall study (50% men, 50–80 years), 583 participants with MCI (aMCI n = 304; naMCI n = 279) and 1,446 cognitively normal participants were included in the analyses. Currently elevated depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale (CES-D; score ≥18). Furthermore, participants were asked if they have ever received a previous diagnosis of depression. Log-Poisson regression models (adjusted for sociodemographic/cardiovascular risk factors) were calculated to determine the association of MCI and its subtypes with all depression variables. Results: The fully adjusted prevalence rate ratios for MCI, aMCI, and naMCI in depressed versus non-depressed participants were 2.06 (95% confidence interval, 1.60–2.64), 3.06 (2.21–4.23), and 1.93 (1.46–2.57). A positive history of lifetime depression without current depressive symptoms was solely associated with naMCI (1.31 (0.99–1.73)). Conclusion: These results suggest that the relationship of depression/depressive symptoms and MCI might differ depending on the timing of depression and on the MCI subtype. Our longitudinal follow-up will allow us to further elucidate this relationship.
15
∗ Correspondence to: Martha Dlugaj, PhD, University Hospital Essen, Department of Neurology, Hufelandstr. 55, 45147 Essen, Germany. Tel.: +49 201 723 2588; Fax: +49 201 723 5901; E-mail: [email protected].
ISSN 1387-2877/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved
40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83
Participants
roo f
39
METHODS
Participants of the Heinz Nixdorf Recall (Risk Factors, Evaluation of Coronary Calcium and Lifestyle; HNR) study were randomly sampled from mandatory city registries in the Ruhr area in Germany. The major aim of the HNR study is to evaluate the predictive value of coronary artery calcification using electronbeam computed tomography for myocardial infarction and cardiac death in comparison to other cardiovascular risk factors. Study methods have been previously described in detail [29, 30]. Briefly, 4,814 participants (45 to 75 years of age) were enrolled between 2000 and 2003. After five years, participants were invited for a second examination (response rate: 90.2%), which additionally included a standardized cognitive performance assessment. Data regarding cognitive status were incomplete or missing in 71 participants (2%). Data about depression (measured on the Center for Epidemiologic Studies Depression Scale (CES-D); see ‘Measurement of depression’ below) were not assessed in 40 participants (1%). Twenty-one participants were excluded due to a physician’s diagnosis of dementia or AD, with intake of cholinesterase inhibitors (ATC, anatomic-therapeutic-chemical classification issued by the World Health Organization (WHO) [31], code: N06DA) or other antidementia drugs (N06DX), or fulfilling the DSM-IV dementia diagnosis. Participants who did not meet the MCI criteria or the criteria for ‘cognitively normal’ (see below) were excluded as well (n = 1,996). Thus, the final analysis sample consisted of 2,029 participants (Fig. 1). All participants provided written informed consent. The study was approved by the institutional review board of the University Duisburg-Essen and followed established guidelines of good epidemiological practice. The study has been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki.
or P
38
or both) and depending on the MCI subtype (amnestic versus non-amnestic MCI) in the general population aged 50–80 years.
dA
37
cte
36
rre
35
Although depression and dementia are distinct clinical entities, they share some common features [1]. For instance, depressed participants often complain about problems in attention and subsequent memory deficits as well as altered sleep patterns [1]. This is supported by studies showing that depressive symptoms occur in about 30–40% of individuals with Alzheimer’s disease (AD) [2–4] and in up to 50% of individuals with vascular dementia (VaD) [5]. Case-control and longitudinal studies have shown a 2-fold increased risk for AD and VaD in patients with a positive history of depression [6–8], whereas, other longitudinal studies reported no increased dementia risk [9–11]. It still remains controversial whether depression is an independent risk factor for dementia, a prodromal dementia symptom, a reaction to cognitive and functional impairment, or a symptom of a related disease such as cerebrovascular disorders [12–14]. In addition, the association of depression and cognitive functioning might vary as a result of the underlying dementia etiology [15]. As AD and VaD are representing the end stage of pathological changes in the brain [29], researchers focus on early stages of cognitive impairment. Mild cognitive impairment (MCI) represents an intermediate stage in the trajectory from normal cognition to dementia [16, 17]. MCI can be classified as amnestic MCI (aMCI), most likely reflecting the prodromal AD stage, and non-amnestic MCI (naMCI), most likely reflecting the prodromal stage of VaD (but also dementia with Lewy bodies or frontotemporal dementia) [17, 18]. Neuropsychiatric symptoms in individuals with MCI have been reported in several studies [4, 19]. However, an increased risk of progression to dementia in depressed individuals with MCI has only been only reported in some [19–21], but not all studies [22–25]. This might be due to the fact that some studies did not differentiate between current depressive symptoms, a positive history of lifetime depression or both [26, 27]. Growing evidence implies that timing of depression (early life or late life depression) may be important for the development of different dementia types (AD or VaD) [28]. Identifying the association of different depression patterns and MCI has important implications as depression could serve as a marker in the early detection of AD and other dementias. We, therefore, examined (1) if there is a crosssectional relationship of depression and MCI in our cohort and (2) if the relationship differs depending on the type of depression (currently elevated depressive symptoms or a positive history of lifetime depression
co
34
INTRODUCTION
Un
33
M. Dlugaj et al. / Depression and Mild Cognitive Impairment in the General Population
uth
2
Cognitive measures The cognitive performance assessment has been previously described in detail [32, 33]. Briefly, the assessment was conceptualized as a multidimensional test using established measures of immediate and
84 85 86
87
88
89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125
126
127 128 129 130
3
co
rre
cte
dA
uth
or P
roo f
M. Dlugaj et al. / Depression and Mild Cognitive Impairment in the General Population
131 132 133 134 135 136 137 138 139 140
Un
Fig. 1. Flowchart of study participants in the Heinz Nixdorf Recall study. CES-D, Center for Epidemiologic Studies Depression Scale; MCI, mild cognitive impairment; SD, standard deviation.
delayed verbal memory [34] (eight word list, performance measured as number of words recalled in each trial), problem solving/speed of processing [34] (labyrinth test, time in seconds needed to complete the task), verbal fluency [35] (semantic category “animals”, number of recalled words within one minute), and abstraction (as an executive function)/visual-spatial organization (clock-drawing test [36], performance was rated from 1 (perfect clock) to 6 (poor performance; was not reached in our sample
(maximum value was 5)). The short cognitive performance assessment reached a good accuracy (area under the curve (AUC) = 0.82, 95% confidence interval (CI) = 0.78–0.85) against a detailed neuropsychological and neurological examination to assess MCI in a previous study [32]. Besides using the raw cognitive data, we used adjusted data by stratifying into three age (50–59 years, 60–69 years, and 70–80 years) and three education groups (≤10 years, 11–13 years, ≥14 years of education).
141 142 143 144 145 146 147 148 149 150
185
Measurement of depression
157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183
186 187 188 189 190 191 192 193 194 195 196 197 198 199
cte
156
rre
155
co
154
We assessed current depressive symptoms using the German 15-item short form of the Center for Epidemiologic Studies Depression Scale (CES-D [38, 39]). The CES-D was developed by The National Institute of Mental Health for epidemiological research. It showed good sensitivity and specificity and high internal consistency across wide age ranges [40, 41] and in participants with cognitive impairment [40]. Participants are asked to rate how often over the past week they have experienced symptoms associated with depression, such as depressed mood, restless sleep, and feeling lonely. The answers are rated on a Likert scale with four possible categories: “0” is “rarely or none of the time (less than 1 day)”, “1” is “sometimes (1–2
Un
153
dA
184
MCI was diagnosed according to the International Working Group on MCI criteria [16, 18]. The diagnosis requires a self and/or informant reported cognitive complaint. Participants were asked if their cognitive performance changed during the past two years. A complaint was considered present if the participant reported a decline in cognitive performance over time. We do not have any informant information. Furthermore, the diagnosis requires a cognitive impairment that is insufficient to fulfill criteria for dementia (Diagnostic and Statistical Manual of Mental Disorders, DSM-IV [37]) and reflects generally intact activities of daily living. A cognitive subtest (see cognitive measures) was rated as impaired if the performance was more than one standard deviation (SD) below the age and education adjusted mean or a score of ≥3 in the clock drawing test. We distinguished two subtypes of MCI [16, 18]: Participants presenting an objective impairment in memory (immediate and/or delayed verbal memory subtest) with or without impairment in any other cognitive domain received a diagnosis of aMCI (n = 304). If a non-memory domain was impaired, the participants received a diagnosis of naMCI (n = 279). Participants who presented neither a subjective cognitive complaint nor objective impairment were defined as ‘cognitively normal’ (n = 1,446). As mentioned above, participants who reported either a subjective cognitive complaint (without objective impairment, n = 543) or showed objective impairment (without subjective cognitive complaint, n = 1,440) were excluded from the regression analyses as well as participants with missing information regarding subjective cognitive complaint (n = 13, Fig. 1).
152
days), “2” is often (3–4 days), and “4” is “almost or all of the time (5–7 days)”. Scores range from 0 to 45 with higher scores reflecting greater levels of depressive symptoms. Participants with more than three missing items on the CES-D (n = 40, 1%, Fig. 1) were excluded from the analyses. In 1,946 of 2,029 (96%) participants the answers of all 15 items were complete (‘depression score complete sample’). In 83 (4%) participants, the answers of one to three questions were missing. To build a depression score for those participants, values of missing items were imputed based on mean values of answered items (‘imputed sample’). We calculated our regression models in the ‘depression score complete sample’ (n = 1,946) and in the ‘depression score complete+imputed sample’ (n = 2,027 (1,946 +83)). As the results did not differ, we included participants with up to three missing CES-D items in our final models. The cut-off point for ‘elevated depressive symptoms’ was ≥18 as recommended in the validation studies of the instrument [39]. To assess the ‘history of lifetime depression’, participants were also asked if they have ever received a depression diagnosis by a physician or a psychologist/psychotherapist (missing data for n = 9 (0.4%) participants). Medication use was recorded according to ATC codes, including antidepressants (ATC code N06A) [31]. Participants were asked to bring their medication in the original packing or the package insert to the study visit. To avoid any missing information regarding current regular medication use, drug intake was also recorded by using a computer-assisted interview performed by a physician in the study center (‘physician-based questionnaire’) [29].
roo f
Diagnosis of mild cognitive impairment
or P
151
M. Dlugaj et al. / Depression and Mild Cognitive Impairment in the General Population
uth
4
Covariates Computer-assisted interviews were conducted, at both time-points, to collect information about medical/family history, coronary heart disease, smoking, and socio-economic status [29, 30]. The variables for model adjustment were selected based on the current literature describing their effects on depression and cognitive functioning: It is known that depressive symptoms differ in dependence of age [42], gender [42], or education level [43], and those three factors have also an influence on cognition [44]. Furthermore, several studies have reported a negative influence of coronary heart disease [45], high or low systolic and diastolic blood pressure [46], stroke [47], and smoking [8] on cognition, and some studies have also reported higher rates of depression in patients with coronary
200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233
234
235 236 237 238 239 240 241 242 243 244 245 246 247 248 249
M. Dlugaj et al. / Depression and Mild Cognitive Impairment in the General Population
258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301
roo f
257
or P
256
Statistical analyses
Continuous variables were compared between the MCI and the cognitively normal participants using the Mann-Whitney U test. Pearson χ2 test was used for categorical variables. The same methods were used for drop-out analyses. To estimate prevalence rate ratios (PRR) and their 95% confidence intervals (CI), we used Log-Poisson regression models (overall MCI versus cognitively normal participants; aMCI versus cognitively normal participants; naMCI versus cognitively normal participants). To examine the association of currently elevated depressive symptoms and MCI, we used “CES-D
Journal of Alzheimer’s Disease xx (20xx) xx–xx DOI 10.3233/JAD-142178 IOS Press
3
roo f
2
Depression and Mild Cognitive Impairment in the General Population: Results of the Heinz Nixdorf Recall Study
or P
1
6
Martha Dlugaja,∗ , Angela Winklera , Nico Draganob , Susanne Moebusc , Karl-Heinz J¨ockelb , Raimund Erbeld and Christian Weimara on behalf of the Heinz Nixdorf Recall Study Investigative Group
7
a Department
8
b Institute
4 5
14
Handling Associate Editor: Roberto Monastero
12
dA
11
cte
9 10
uth
13
of Neurology, University Hospital of Essen, University of Duisburg-Essen, Germany for Medical Sociology, Centre for Health and Society, University of D¨usseldorf, Medical Faculty, University of D¨usseldorf, Germany c Institute for Medical Informatics, Biometry and Epidemiology, University Hospital of Essen, University Duisburg-Essen, Germany d Clinic of Cardiology, West German Heart Centre, University Hospital of Essen, University Duisburg-Essen, Germany
Accepted 17 November 2014
32
Keywords: Aging, depression, epidemiology, mild cognitive impairment
18 19 20 21 22 23 24 25 26 27 28 29 30
co
17
Un
16
rre
31
Abstract. Background: The literature suggests an association between depression and mild cognitive impairment (MCI) and dementia, but not all studies have examined this association with regard to MCI subtypes reflecting different dementia etiologies. Objective: To examine if there is a cross-sectional relationship of depression and MCI and to examine if the relationship differs depending on the type of depression (currently elevated depressive symptoms or a positive history of lifetime depression or both) and on the MCI subtype (amnestic versus non-amnestic MCI (aMCI/naMCI)). Methods: From the second examination of the population-based Heinz Nixdorf Recall study (50% men, 50–80 years), 583 participants with MCI (aMCI n = 304; naMCI n = 279) and 1,446 cognitively normal participants were included in the analyses. Currently elevated depressive symptoms were assessed using the Center for Epidemiologic Studies Depression Scale (CES-D; score ≥18). Furthermore, participants were asked if they have ever received a previous diagnosis of depression. Log-Poisson regression models (adjusted for sociodemographic/cardiovascular risk factors) were calculated to determine the association of MCI and its subtypes with all depression variables. Results: The fully adjusted prevalence rate ratios for MCI, aMCI, and naMCI in depressed versus non-depressed participants were 2.06 (95% confidence interval, 1.60–2.64), 3.06 (2.21–4.23), and 1.93 (1.46–2.57). A positive history of lifetime depression without current depressive symptoms was solely associated with naMCI (1.31 (0.99–1.73)). Conclusion: These results suggest that the relationship of depression/depressive symptoms and MCI might differ depending on the timing of depression and on the MCI subtype. Our longitudinal follow-up will allow us to further elucidate this relationship.
15
∗ Correspondence to: Martha Dlugaj, PhD, University Hospital Essen, Department of Neurology, Hufelandstr. 55, 45147 Essen, Germany. Tel.: +49 201 723 2588; Fax: +49 201 723 5901; E-mail: [email protected].
ISSN 1387-2877/14/$27.50 © 2014 – IOS Press and the authors. All rights reserved
40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83
Participants
roo f
39
METHODS
Participants of the Heinz Nixdorf Recall (Risk Factors, Evaluation of Coronary Calcium and Lifestyle; HNR) study were randomly sampled from mandatory city registries in the Ruhr area in Germany. The major aim of the HNR study is to evaluate the predictive value of coronary artery calcification using electronbeam computed tomography for myocardial infarction and cardiac death in comparison to other cardiovascular risk factors. Study methods have been previously described in detail [29, 30]. Briefly, 4,814 participants (45 to 75 years of age) were enrolled between 2000 and 2003. After five years, participants were invited for a second examination (response rate: 90.2%), which additionally included a standardized cognitive performance assessment. Data regarding cognitive status were incomplete or missing in 71 participants (2%). Data about depression (measured on the Center for Epidemiologic Studies Depression Scale (CES-D); see ‘Measurement of depression’ below) were not assessed in 40 participants (1%). Twenty-one participants were excluded due to a physician’s diagnosis of dementia or AD, with intake of cholinesterase inhibitors (ATC, anatomic-therapeutic-chemical classification issued by the World Health Organization (WHO) [31], code: N06DA) or other antidementia drugs (N06DX), or fulfilling the DSM-IV dementia diagnosis. Participants who did not meet the MCI criteria or the criteria for ‘cognitively normal’ (see below) were excluded as well (n = 1,996). Thus, the final analysis sample consisted of 2,029 participants (Fig. 1). All participants provided written informed consent. The study was approved by the institutional review board of the University Duisburg-Essen and followed established guidelines of good epidemiological practice. The study has been performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki.
or P
38
or both) and depending on the MCI subtype (amnestic versus non-amnestic MCI) in the general population aged 50–80 years.
dA
37
cte
36
rre
35
Although depression and dementia are distinct clinical entities, they share some common features [1]. For instance, depressed participants often complain about problems in attention and subsequent memory deficits as well as altered sleep patterns [1]. This is supported by studies showing that depressive symptoms occur in about 30–40% of individuals with Alzheimer’s disease (AD) [2–4] and in up to 50% of individuals with vascular dementia (VaD) [5]. Case-control and longitudinal studies have shown a 2-fold increased risk for AD and VaD in patients with a positive history of depression [6–8], whereas, other longitudinal studies reported no increased dementia risk [9–11]. It still remains controversial whether depression is an independent risk factor for dementia, a prodromal dementia symptom, a reaction to cognitive and functional impairment, or a symptom of a related disease such as cerebrovascular disorders [12–14]. In addition, the association of depression and cognitive functioning might vary as a result of the underlying dementia etiology [15]. As AD and VaD are representing the end stage of pathological changes in the brain [29], researchers focus on early stages of cognitive impairment. Mild cognitive impairment (MCI) represents an intermediate stage in the trajectory from normal cognition to dementia [16, 17]. MCI can be classified as amnestic MCI (aMCI), most likely reflecting the prodromal AD stage, and non-amnestic MCI (naMCI), most likely reflecting the prodromal stage of VaD (but also dementia with Lewy bodies or frontotemporal dementia) [17, 18]. Neuropsychiatric symptoms in individuals with MCI have been reported in several studies [4, 19]. However, an increased risk of progression to dementia in depressed individuals with MCI has only been only reported in some [19–21], but not all studies [22–25]. This might be due to the fact that some studies did not differentiate between current depressive symptoms, a positive history of lifetime depression or both [26, 27]. Growing evidence implies that timing of depression (early life or late life depression) may be important for the development of different dementia types (AD or VaD) [28]. Identifying the association of different depression patterns and MCI has important implications as depression could serve as a marker in the early detection of AD and other dementias. We, therefore, examined (1) if there is a crosssectional relationship of depression and MCI in our cohort and (2) if the relationship differs depending on the type of depression (currently elevated depressive symptoms or a positive history of lifetime depression
co
34
INTRODUCTION
Un
33
M. Dlugaj et al. / Depression and Mild Cognitive Impairment in the General Population
uth
2
Cognitive measures The cognitive performance assessment has been previously described in detail [32, 33]. Briefly, the assessment was conceptualized as a multidimensional test using established measures of immediate and
84 85 86
87
88
89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125
126
127 128 129 130
3
co
rre
cte
dA
uth
or P
roo f
M. Dlugaj et al. / Depression and Mild Cognitive Impairment in the General Population
131 132 133 134 135 136 137 138 139 140
Un
Fig. 1. Flowchart of study participants in the Heinz Nixdorf Recall study. CES-D, Center for Epidemiologic Studies Depression Scale; MCI, mild cognitive impairment; SD, standard deviation.
delayed verbal memory [34] (eight word list, performance measured as number of words recalled in each trial), problem solving/speed of processing [34] (labyrinth test, time in seconds needed to complete the task), verbal fluency [35] (semantic category “animals”, number of recalled words within one minute), and abstraction (as an executive function)/visual-spatial organization (clock-drawing test [36], performance was rated from 1 (perfect clock) to 6 (poor performance; was not reached in our sample
(maximum value was 5)). The short cognitive performance assessment reached a good accuracy (area under the curve (AUC) = 0.82, 95% confidence interval (CI) = 0.78–0.85) against a detailed neuropsychological and neurological examination to assess MCI in a previous study [32]. Besides using the raw cognitive data, we used adjusted data by stratifying into three age (50–59 years, 60–69 years, and 70–80 years) and three education groups (≤10 years, 11–13 years, ≥14 years of education).
141 142 143 144 145 146 147 148 149 150
185
Measurement of depression
157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183
186 187 188 189 190 191 192 193 194 195 196 197 198 199
cte
156
rre
155
co
154
We assessed current depressive symptoms using the German 15-item short form of the Center for Epidemiologic Studies Depression Scale (CES-D [38, 39]). The CES-D was developed by The National Institute of Mental Health for epidemiological research. It showed good sensitivity and specificity and high internal consistency across wide age ranges [40, 41] and in participants with cognitive impairment [40]. Participants are asked to rate how often over the past week they have experienced symptoms associated with depression, such as depressed mood, restless sleep, and feeling lonely. The answers are rated on a Likert scale with four possible categories: “0” is “rarely or none of the time (less than 1 day)”, “1” is “sometimes (1–2
Un
153
dA
184
MCI was diagnosed according to the International Working Group on MCI criteria [16, 18]. The diagnosis requires a self and/or informant reported cognitive complaint. Participants were asked if their cognitive performance changed during the past two years. A complaint was considered present if the participant reported a decline in cognitive performance over time. We do not have any informant information. Furthermore, the diagnosis requires a cognitive impairment that is insufficient to fulfill criteria for dementia (Diagnostic and Statistical Manual of Mental Disorders, DSM-IV [37]) and reflects generally intact activities of daily living. A cognitive subtest (see cognitive measures) was rated as impaired if the performance was more than one standard deviation (SD) below the age and education adjusted mean or a score of ≥3 in the clock drawing test. We distinguished two subtypes of MCI [16, 18]: Participants presenting an objective impairment in memory (immediate and/or delayed verbal memory subtest) with or without impairment in any other cognitive domain received a diagnosis of aMCI (n = 304). If a non-memory domain was impaired, the participants received a diagnosis of naMCI (n = 279). Participants who presented neither a subjective cognitive complaint nor objective impairment were defined as ‘cognitively normal’ (n = 1,446). As mentioned above, participants who reported either a subjective cognitive complaint (without objective impairment, n = 543) or showed objective impairment (without subjective cognitive complaint, n = 1,440) were excluded from the regression analyses as well as participants with missing information regarding subjective cognitive complaint (n = 13, Fig. 1).
152
days), “2” is often (3–4 days), and “4” is “almost or all of the time (5–7 days)”. Scores range from 0 to 45 with higher scores reflecting greater levels of depressive symptoms. Participants with more than three missing items on the CES-D (n = 40, 1%, Fig. 1) were excluded from the analyses. In 1,946 of 2,029 (96%) participants the answers of all 15 items were complete (‘depression score complete sample’). In 83 (4%) participants, the answers of one to three questions were missing. To build a depression score for those participants, values of missing items were imputed based on mean values of answered items (‘imputed sample’). We calculated our regression models in the ‘depression score complete sample’ (n = 1,946) and in the ‘depression score complete+imputed sample’ (n = 2,027 (1,946 +83)). As the results did not differ, we included participants with up to three missing CES-D items in our final models. The cut-off point for ‘elevated depressive symptoms’ was ≥18 as recommended in the validation studies of the instrument [39]. To assess the ‘history of lifetime depression’, participants were also asked if they have ever received a depression diagnosis by a physician or a psychologist/psychotherapist (missing data for n = 9 (0.4%) participants). Medication use was recorded according to ATC codes, including antidepressants (ATC code N06A) [31]. Participants were asked to bring their medication in the original packing or the package insert to the study visit. To avoid any missing information regarding current regular medication use, drug intake was also recorded by using a computer-assisted interview performed by a physician in the study center (‘physician-based questionnaire’) [29].
roo f
Diagnosis of mild cognitive impairment
or P
151
M. Dlugaj et al. / Depression and Mild Cognitive Impairment in the General Population
uth
4
Covariates Computer-assisted interviews were conducted, at both time-points, to collect information about medical/family history, coronary heart disease, smoking, and socio-economic status [29, 30]. The variables for model adjustment were selected based on the current literature describing their effects on depression and cognitive functioning: It is known that depressive symptoms differ in dependence of age [42], gender [42], or education level [43], and those three factors have also an influence on cognition [44]. Furthermore, several studies have reported a negative influence of coronary heart disease [45], high or low systolic and diastolic blood pressure [46], stroke [47], and smoking [8] on cognition, and some studies have also reported higher rates of depression in patients with coronary
200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233
234
235 236 237 238 239 240 241 242 243 244 245 246 247 248 249
M. Dlugaj et al. / Depression and Mild Cognitive Impairment in the General Population
258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301
roo f
257
or P
256
Statistical analyses
Continuous variables were compared between the MCI and the cognitively normal participants using the Mann-Whitney U test. Pearson χ2 test was used for categorical variables. The same methods were used for drop-out analyses. To estimate prevalence rate ratios (PRR) and their 95% confidence intervals (CI), we used Log-Poisson regression models (overall MCI versus cognitively normal participants; aMCI versus cognitively normal participants; naMCI versus cognitively normal participants). To examine the association of currently elevated depressive symptoms and MCI, we used “CES-D
