- Email: [email protected]
Prevalence and risk factors for depression in non-demented primary care attenders aged 75 years and older
Journal of Affective Disorders 111 (2008) 153 – 163 www.elsevier.com/locate/jad
Research report
Prevalence and risk factors for depression in non-demented primary care attenders aged 75 years and older Siegfried Weyerer a,⁎, Sandra Eifflaender-Gorfer a , Leonore Köhler a , Frank Jessen b , Wolfgang Maier b , Angela Fuchs c , Michael Pentzek c , Hanna Kaduszkiewicz d , Cadja Bachmann d , Matthias C. Angermeyer e , Melanie Luppa e , Birgitt Wiese f , Edelgard Mösch g , Horst Bickel g for the German AgeCoDe Study group (German Study on Ageing, Cognition and Dementia in Primary Care Patients) 1 a
Central Institute of Mental Health, Mannheim, Germany Department of Psychiatry, University of Bonn, Germany c Department of General Practice, University Medical Center Düsseldorf, Germany Institute of Primary Medical Care, University Medical Center Hamburg-Eppendorf, Germany e Department of Psychiatry, University of Leipzig, Germany f Institute for Biometrics, Hannover Medical School, Germany g Department of Psychiatry, Technical University of Munich, Germany b
d
Received 12 December 2007; received in revised form 12 February 2008; accepted 12 February 2008 Available online 26 March 2008
Abstract Background: Depression among the elderly is an important public health issue. The aims of this study were to report the prevalence of depression and to determine the impact of socio-demographic variables, functional impairment and medical diagnoses, lifestyle factors, and mild cognitive impairment on depression as part of the German Study on Ageing, Cognition and Dementia in Primary Care Patients (AgeCoDe Study). Methods: Included in the cross-sectional survey were 3327 non-demented subjects aged 75 and over attending general practitioners (GPs) (n = 138) in an urban area of Germany. The GDS-15 Geriatric Depression Scale was used to measure depression with a threshold of b6/6+. Associations with social and clinical risk factors were assessed by means of multiple logistic regression models. Results: The prevalence of depression was 9.7% (95% confidence interval 8.7–10.7). In a univariate analysis, the following variables were significantly associated with depression: female gender, increasing age, living alone, divorce, lower educational status, functional impairment, comorbid somatic disorder, mild cognitive impairment, smoking, and abstinence from alcohol. After full adjustment for confounding variables, odds ratios for depression were significantly higher only for functional impairment, smoking, and multi-domain mild cognitive impairment.
⁎ Corresponding author. Tel.: +49 621 1703 6351; fax: +49 621 1703 1305. E-mail address: [email protected] (S. Weyerer). 1 Further members of the German AgeCoDe Study group: Steffi G. Riedel-Heller, Tobias Luck, Heike Kölsch, Hendrik van den Bussche, HeinzHarald Abholz, Gabriela Cvetanovska, Franziska Haller, Michael Wagner, Thomas Zimmermann, Teresa Kaufeler, Jochen Werle, Manfred Mayer, Heinz-Peter Romberg, Hagen Sandholzer, Anja Frenzen, Anja Wollny. 0165-0327/$ - see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jad.2008.02.008
154
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
Limitations: Recruitment procedures might have led to an underestimation of current prevalence. The cross-sectional data did not allow us to analyze the temporal relationship between risk factors and depression. Conclusions: The prevalence of depression in the elderly is high and remains high into old age. In designing prevention programs, it is important to call more attention to the impact of functional and cognitive impairment on depression. © 2008 Elsevier B.V. All rights reserved. Keywords: Depression; Epidemiology; Risk factors; Old age; General practice
1. Introduction Depression in old age is common. Prevalence rates among elderly people seen in primary care range from 6.5 to 9% for major depression (Lyness et al., 2002) and from 10 to 25% for clinically relevant depressive symptoms (Speer and Schneider, 2003). In meta-analyses (e.g., Cole and Dendukuri, 2003), a number of risk factors for the occurrence of depression in old age have been isolated: Socio-demographic factors such as gender and marital status are associated with depression, showing higher rates for women and lower rates for the married. Depression in old age is also associated with more physical disorders and increased functional impairment (e.g., Braam et al., 2005). In addition to poor functioning, depression increases the utilization of medical services and health care costs (Luppa et al., 2008). Although the association between age and depression received considerable attention, very little is known about the prevalence and course of depression among those 75 years of age and older. Studies that treat the group 65+ as one entity are often heavily weighted towards the age group 65–75 (Riedel-Heller et al., 2006). Therefore, the prevalence of depression in the very old is uncertain since many community-based studies lack adequate samples over the age of 75. The findings from the few studies which report prevalence rates for the older old are inconsistent, some showing a rise over the age 75, some a fall (Palsson and Skoog, 1997; McDougall et al., 2007). To study the older old is also important since some crucial risk factors such as bereavement, social isolation, somatic diseases, and functional impairment become more common with increasing age. These factors may exert different effects in the younger old compared to the older old. Knowledge of risk factors is a prerequisite to designing tailored interventions, either to tackle the factors itself or to define high risk groups, since depression is treatable in many cases. Based on a large sample of non-demented primary care attenders (age 75+), the objectives of this study are: • to report age and gender-specific prevalence rates of depression and
• to determine the impact of putative risk factors: socio-demographic variables (marital status, education), functional impairment and medical diagnoses, lifestyle factors (consumption of alcohol and nicotine), and mild cognitive impairment. 2. Methods 2.1. Study design and sample The sample consists of all subjects participating in the baseline assessment of a prospective longitudinal study on the early detection of mild cognitive impairment and dementia in primary care. The study was conducted in six centers (Bonn, Düsseldorf, Hamburg, Leipzig, Mannheim, and Munich) representing an urban area of cities with a total population ranging between about 300,000 (Mannheim) and almost 1,8 million (Hamburg). The subjects were recruited between January 1, 2003 and November 30, 2004. In each center, 19 to 29 GPs participated in the recruitment process: all in all, 138 GPs. Inclusion criteria for GP patients were an age of 75 years and over, the absence of dementia in the GP's view, and at least one contact with the GP within the last 12 months. Exclusion criteria were consultations only by home visits, residence in a nursing home, a severe illness which the GP would deem fatal within three months, an insufficient knowledge of the German language, deafness or blindness, inability to consent, and not being a regular patient of the participating GP. On average, each practice comprised 24 patients. Information on sampling frame, eligible subjects, and respondents is given in Fig. 1. Finally, 3327 selected GP patients were assessed by means of structured clinical interviews. Eighty-five (2.6%) of the 3327 interviewed subjects were excluded from the following analyses: 41 (48.2%) were classified as demented, 39 (45.9%) fell short of the age limit of 75 years, and 5 (5.9%) had incomplete neuropsychological test data. The calculation of the prevalence of depression is based on the remaining 3242 subjects. In order to analyze possible non-response bias, data on age and gender was collected for 1770 (99.7%) of the 1775 subjects refusing participation. The
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
155
psychodiagnostic interview (Mini-DIPS; Margraf, 1994), a cut-off score of 6 yielded the best sensitivity (84.0%) and specificity (88.9%) (Gauggel and Birkner, 1999). Therefore, this cut-off was used in the present study. Neuropsychological assessment was based on the Structured Interview for Diagnosis of Dementia of Alzheimer type, Multi-infarct Dementia and Dementia of other Etiology according to DSM-III-R, DSM-IV and ICD-10 (SIDAM; Zaudig and Hiller, 1996). In order to evaluate impairment of cognitive functioning, age- and education-specific norms for the cognitive domains were applied (Luck et al., 2007b). Dementia according to DSM-IV was excluded by means of the SIDAM. Mild Cognitive Impairment (MCI) was diagnosed according to new consensus criteria proposed by the International Working Group on Mild Cognitive Impairment. According to Winblad et al. (2004), four subtypes for MCI were examined based solely on differences in the criterion of objective cognitive decline (Luck et al., 2007a):
Fig. 1. Sampling frame and sample.
mean age of the included subjects was 80.2 years (SD = 3.6) vs. 80.8 years (SD = 3.8) for those refusing participation (t = - 6.104, p = 0.000). 2127 (65.6%) of all participants were female and 1115 (34.4%) were male, among non-participants 1219 (68.9%) were female and 551 (31.1%) were male (χ2 = 5.594, df = 1, p = 0.018). Thus, group of participants was significantly younger and included more males than did the group of nonparticipants. 2.2. Instruments Structured clinical interviews were conducted by trained physicians and psychologists during visits to the participants' homes. Depressive symptoms were assessed by means of the 15-item short version of the Geriatric Depression Scale (GDS; Sheikh and Yesavage, 1986; D'Ath et al., 1994). Due to the simplified yes/no response format and the exclusion of questions on somatic symptoms, it is especially suitable for elderly people. This instrument has good psychometric properties also for Germanspeaking populations (Wancata et al., 2006). Comparing the German version of the GDS-15 with a
• Single-domain amnestic MCI was used for subjects having an objective deficit in memory but not in any other area of cognitive functioning. • Single nonmemory MCI was diagnosed only if a single domain other than memory was impaired. • Multi-domain MCI amnestic was diagnosed if memory and at least one other cognitive domain were impaired. • Multi-domain MCI nonamnestic was diagnosed if at least two cognitive domains other than memory showed an objective impairment. In this study modified versions of the MCI (see Luck et al., 2007a) were used. With the exception of subjective cognitive complaints, each of these modifications was defined by the same subtypes according to Winblad et al. (2004). Lifestyle factors assessed include self-reported alcohol consumption and smoking, and it was differentiated between current consumers and nonconsumers. With regard to alcohol consumption, current consumers were distinguished according to their level of consumption: “normal” or “at risk”. Following widely accepted international guidelines (e.g., British Medical Association, 1995), at-risk drinking was defined as the daily consumption of N 20 g alcohol for women and N 30 g alcohol for men. The measurement of functional impairment (mobility, vision, and hearing) was based on the SIDAMADL-Scale (Zaudig and Hiller, 1996).
156
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
Table 1 Univariate association with depression N (% of total 3242)
Cases on GDS 6+ (%)
Univariate odds ratio (95%CI)
Gender Female 2127 (65.6) 237 (11.1) 1.7 (1.28–2.20) Male 1115 (34.4) 76 (6.8) 1 Age 75–79 1725 (53.2) 149 (8.6) 1 80–84 1210 (37.3) 122 (10.1) 1.2 (0.92–1.53) 85+ 307 (9.5) 42 (13.7) 1.7 (1.16–2.42) Living alone Yes 1661 (51.2) 186 (11.2) 1.4 (1.14–1.83) No 1581 (48.8) 127 (8.0) 1 Marital status Single 203 (6.3) 23 (11.3) 1 Married 1377 (42.5) 99 (7.2) 0.6 (0.38–0.98) Divorced 193 (6.0) 36 (18.7) 1.7 (1.02–3.16) Widowed 1469 (45.3) 155 (10.6) 0.9 (0.58–1.47) Level of education (Based on the revised version of the international CASMIN educational classification; Brauns and Steinmann, 1999) Low 2013 (62.1) 210 (10.4) 1 Middle 886 (27.3) 80 (9.0) 0.9 (0.64–1.10) High 343 (10.6) 23 (6.7) 0.6 (0.36–0.90) Mobility impairment Yes 1146 (35.3) 199 (17.4) 3.7 (2.87–4.66) No 2096 (64.7) 114 (5.4) 1 Vision impairment Yes 469 (14.5) 81 (17.3) 2.3 (1.74–3.01) No 2773 (85.5) 232 (8.4) 1 Hearing impairment Yes 993 (30.6) 126 (12.7) 1.6 (1.26–2.04) No 2249 (69.4) 187 (8.3) 1 Somatic comorbidity No somatic comorbidity 143 (4.3) 9 (6.5) 1 1–4 somatic diagnoses 2233 (68.5) 192 (8.6) 1.4 (0.68–2.73) 5 and more somatic diagnoses 888 (27.2) 112 (12.7) 2.1 (1.04–4.25) Smoking (at present) Yes 244 (7.5) 34 (13.9) 1.6 (1.08–2.31) No 2994 (92.5) 279 (9.3) 1 Alcohol consumption (at present) Completely abstinent 1613 (50.1) 192 (11.9) 1.7 (1.33–2.17) Moderate 1478 (45.9) 109 (7.4) 1 At risk (men: 30 g+ per day; 128 (4.0) 9 (7.0) 1.0 (0.47–1.92) women: 20 g+ per day) Mild Cognitive Impairment (MCI)-modified a Unimpaired 2422 (74.8) 209 (8.6) 1 Single-domain amnestic MCI 97 (3.0) 15 (15.5) 1.9 (1.10–3.42) Single nonmemory MCI 486 (15.0) 44 (9.1) 1.1 (0.75–1.48) Multi-domain MCI amnestic 134 (4.1) 26 (19.4) 2.5 (1.62–4.00) Multi-domain MCI nonamnestic 100 (3.1) 18 (18.0) 2.3 (1.37–3.95) a
p-value b0.001 – – 0.184 0.006 0.002 – – 0.041 0.043 0.736 – 0.247 0.034 b0.001 – b0.001 – b0.001 – – 0.377 0.038 0.020 – b0.001 – 0.886
– 0.023 0.762 b0.001 0.002
Modified subtypes (exclusion of the criterion of a subjective cognitive complaint) of MCI according to Winblad et al. (2004).
For each study participant, his or her GP filled in a questionnaire to assess comorbidity. The following clinical diagnoses were predetermined by the questionnaire (answer: yes/no/I don't know): diabetes mellitus, high blood pressure, cardiac arrhythmia, coronary heart disease, myocardial infarction, hyperlipidemia, hyper-
cholesterolemia, peripheral arterial obstructive disease, Parkinson's disease, epilepsy, carotic artery stenosis (N 80%), transient ischemic attack (TIA), stroke, hypothyroidism, hyperthyroidism, renal insufficiency, and liver disease. Based on this information, somatic comorbidity was defined: no comorbidity/1–4 diagnoses/5+ diagnoses.
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
2.3. Statistical analyses The data were collected in the centers via an internetbased Remote-Data-Entry-System into a central ORACLE version 9 database. The statistical analyses were performed with SPSS for Windows, version 12.01. According to the defined cut-off (6+), prevalence rates of depression were estimated as the percentage of the completely assessed non-demented subjects aged 75 years and over. In addition to prevalence, 95% confidence intervals (95% CI) were calculated. Associations between risk factors and depression (scoring 6+ on the GDS-15) were analyzed using multiple logistic regression. Univariate Odds Ratios (OR) were generated for each variable and the independence of any association was initially controlled for sex and age, then additionally by entering all variables into the model. A p-value less than 0.05 was considered statistically significant. 2.4. Ethical approval The ethics committees of the participating centers approved the study. Written informed consent was obtained from all participants. 3. Results 3.1. Demographic characteristics Of the 3242 patients interviewed, 2127 were women (65.6%) and 1115 men (34.4%). Their average age was 80.2 years (SD = 3.6; range: 75–99). Based on the revised CASMIN classification according to Brauns and Steinmann (1999), 2013 (62.1%) of the subjects exhibited a low level of education, 886 (27.3%) a middle level of education, and 343 (10.6%) a high level of education. The majority of the patients were widowed (45.3%) or married (42.5%); only 203 (6.3%) of the subjects were never married, while a further 193 subjects (6.0%) were divorced.
157
the rate among female subjects was 11.1% (95% CI 9.76–12.44). The prevalence of depression increased with age: In the group of 75–79 year-olds, 8.6% (95% CI 7.28– 9.92), among the 80–84 year-olds already 10.1% (95% CI 8.40–11.80), and in the group of those over 85 years of age and older, 13.7% (95% CI 9.85–17.55) were affected by a depressive disorder. Above and beyond this, Table 2 contains the Odds Ratios for each potential risk factor, adjusted first of all for age and gender, and in a further step, for all the variables studied. In order to better visualize the adjustment effects, the univariate correlations were entered into a second column. 3.3. Socio-demographic factors Independent of age, women suffer from a depressive disorder significantly more often than men (OR = 1.7, 95% CI 1.27–2.18). Independent of their gender, the subjects 85 years of age and older (OR = 1.7, 95% CI 1.16–2.42) ran a 70% higher risk of depressive disorder compared to those in the group aged 75–79 years. With regard to marital status, a greater number of depressive subjects is found in the group of divorced persons (18.7%), as well as in the unmarried group (11.3%), while in the widowed group (10.6%), and especially in the married group (7.2%), the respective number of depressive subjects is much lower (OR = 0.6, 95% CI 0.38–0.98). The prevalence of depression was significantly higher among those living alone (11.2%) compared to those not living alone (8.0%). Adjusted for all variables, age, gender, marital status, and living alone are not significantly associated with depression. Univariate analysis revealed a significant correlation between depression and the level of education, with a low prevalence (6.7%) in subjects with a high level of education (OR = 0.6, 95% CI 0.36–0.90) compared to the group with a middle level of education (9.0%), respectively a low level of education (10.4%). After controlling for age and gender, this correlation was no longer significant (OR = 0.7, 95% CI 0.46–1.15).
3.2. Age- and gender-specific prevalence rates of depressive disorders
3.4. Functional impairment and comorbidity
The numbers of depressive subjects, including their proportional distribution within the categories of the various independent variables, are depicted in Table 1. According to the case definition, at baseline 313/3242 subjects (9.7%; 95% CI 8.68–10.72) could be identified as “cases”. While only 6.8% (95% CI 5.32–8.28) of the male subjects suffered from depressive symptoms,
Substantial differences in prevalence rates of depression are found in the area of functional limitations. The depression rate is 17.4% among those subjects impaired in their mobility, compared to a rate of 5.4% among those who are mobile. The association was also highly significant in the fully adjusted model (OR = 2.9, 95% CI 2.26–3.78).
158
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
Table 2 Univariate and adjusted associations with depression Univariate odds ratio
Odds ratio adjusted for age and sex
(95%CI)
(95%CI)
Gender Female 1.7 (1.28–2.20) Male 1 Age 75–79 1 80–84 1.2 (0.92–1.53) 85+ 1.7 (1.16–2.42) Living alone Yes 1.4 (1.14–1.83) No 1 Marital status Single 1 Married 0.6 (0.38–0.98) Divorced 1.7 (1.02–3.16) Widowed 0.9 (0.58–1.47) Level of education (CASMIN) Low 1 Middle 0.9 (0.64–1.10) High 0.6 (0.36–0.90) Mobility impairment Yes 3.7 (2.87–4.66) No 1 Vision impairment Yes 2.3 (1.74–3.01) No 1 Hearing impairment Yes 1.6 (1.26–2.04) No 1 Somatic comorbidity No somatic comorbidity 1 1–4 somatic diagnoses 1.4 (0.68–2.73) 5 and more somatic diagnoses 2.1 (1.04–4.25) Smoking (at present) Yes 1.6 (1.08–2.31) No 1 Alcohol consumption (at present) Completely abstinent 1.7 (1.33–2.17) Moderate 1 At risk (men: 30 g+ per day; 1.0 (0.47–1.92) women: 20 g+ per day) Mild Cognitive Impairment (MCI)-modified a Unimpaired 1 Single-domain amnestic MCI 1.9 (1.10–3.42) Single nonmemory MCI 1.1 (0.75–1.48) Multi-domain MCI amnestic 2.5 (1.62–4.00) Multi-domain MCI nonamnestic 2.3 (1.37–3.95) a
p-value
Odds ratio fully adjusted for all variables
p-value
(95%CI)
1.7 (1.27–2.18) 1
b0.001 –
1.3 (0.90–1.76) 1
0.178 –
1 1.2 (0.92–1.52) 1.7 (1.16–2.42)
0.196 0.006
1 0.9 (0.70–1.21) 1.2 (0.79–1.75)
– 0.540 0.419
1.2 (0.89–1.52) 1
0.266 –
0.8 (0.57–1.26) 1
0.411 –
1 0.7 (0.44–1.19) 1.7 (0.99–3.10) 0.9 (0.55–1.40)
– 0.199 0.055 0.576
1 0.6 (0.33–1.10) 1.4 (0.79–2.61) 0.8 (0.48–1.26)
– 0.097 0.240 0.300
1 0.8 (0.61–1.05) 0.7 (0.46–1.15)
– 0.109 0.173
1 0.8 (0.61–1.09) 0.7 (0.41–1.10)
– 0.168 0.112
3.5 (2.71–4.45) 1
b0.001 –
2.9 (2.26–3.78) 1
b0.001 –
2.1 (1.61–2.82) 1
b0.001 –
1.7 (1.25–2.26) 1
b0.001 –
1.6 (1.26–2.07) 1
b0.001 –
1.4 (1.10–1.83) 1
0.008 –
1 1.4 (0.68–2.73) 2.2 (1.08–4.46)
– 0.379 0.029
1 1.4 (0.64–3.16) 1.8 (0.80–4.09)
– 0.387 0.154
1.6 (1.10–2.39) 1
0.016 –
1.6 (1.03–2.36) 1
0.038 –
1.5 (1.15–1.93) 1 0.9 (0.46–1.87)
0.003 – 0.822
1.3 (0.95–1.64) 1 0.9 (0.43–1.84)
0.106 – 0.760
1 2.0 1.0 2.3 2.0
– 0.017 0.876 b0.001 0.014
1 1.6 (0.86–2.86) 1.0 (0.73–1.50) 2.1 (1.30–3.43) 1.5 (0.88–2.81)
– 0.140 0.798 0.003 0.125
(1.13–3.56) (0.73–1.45) (1.46–3.68) (1.15–3.41)
Modified subtypes (exclusion of the criterion of a subjective cognitive complaint) of MCI according to Winblad et al. (2004).
The prevalence of depression among subjects with impaired vision and impaired hearing was significantly higher than among subjects without such deficits. The extent of the relationship was influenced by the variables age and gender; however, a significant correlation
with depression remained for subjects with impaired vision (OR = 2.1, 95% CI 1.61–2.82), as well as for those with impaired hearing (OR = 1.6, 95% CI 1.26– 2.07). After controlling for potential risk factors, the relationship decreases further, but there still remains a
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
significantly higher risk of depression among subjects with impaired vision (OR = 1.7, 95% CI 1.25–2.26) and among those with impaired hearing (OR = 1.4, 95% CI 1.10–1.83). Univariate analysis (results not shown in table) revealed that in 5 out of 17 somatic diseases assessed by the primary care physician, depression rates are statistically significant higher: for Parkinson's disease (OR = 3.6, 95% CI 1.99–6.66), transient ischemic attack (OR = 1.7, 95% CI 1.15–2.40), coronary heart disease (OR = 1.5, 95% CI 1.16–1.86), cardiac arrhythmia (OR = 1.4, 1.06–1.73), and diabetes mellitus (OR = 1.3, 95% CI 1.04–1.73). Compared to subjects without any somatic comorbidity, the prevalence of depression among the group of multimorbid subjects (5 or more somatic diagnoses) is almost twice as high (12.7% vs. 6.5%). The relationship between depression and multimorbidity remains significant even when age and gender are taken into account (OR = 2.2, 95% CI 1.08–4.46), but falls below the level of significance after controlling for all potential risk factors (OR = 1.8, 95% CI 0.80–4.09). 3.5. Lifestyle factors While the rate of depression is higher among current smokers than among non-smokers (13.9% vs. 9.3%), the opposite is true in terms of alcohol consumption. Here, the number of depressive subjects is higher in the group of abstinent subjects (11.9%) than in the groups whose alcohol consumption is either “normal” (7.4%) or “at risk” (7.0%). The risk of depression is significantly higher for smokers, both after controlling for age and gender (OR = 1.6, 95% CI 1.10–2.39) and in the fully adjusted model (OR = 1.6, 95% CI 1.03–2.36). With regard to alcohol consumption, the correlation with depression was no longer significant once all variables had been adjusted. 3.6. Mild Cognitive Impairment (MCI) Compared to the cognitively unimpaired subjects, subjects with “single nonmemory MCI” do not exhibit an increased risk of depression (OR = 1.1, 95% CI 0.75– 1.48). In comparison thereto, subjects with “singledomain amnestic MCI” (OR = 1.9, 95% CI 1.10–3.42), “multi-domain MCI nonamnestic” (OR = 2.3, 95% CI 1.37–3.95) and “multi-domain MCI amnestic” (OR = 2.5, 95% CI 1.62–4.00) have a significantly higher risk of depression. While age and gender influence these correlations only slightly, after controlling for all
159
potential determinants only those subjects with “multidomain MCI amnestic” (OR = 2.1, 95% CI 1.30–3.43) exhibited a significantly increased risk of depression ( p = 0.003). 4. Discussion The current study is one of the largest studies worldwide to have been conducted on depression in very old patients in primary care. We found that depression in urban areas of Germany is a common condition: One in ten subjects 75 years of age and older is affected. A direct comparison with other studies is difficult, however, due to the varying operationalization of depressive disorders and the heterogeneous age groups. In a large-scale study of general practices carried out by Osborn et al. (2003) in the United Kingdom, the same age group (75+) was studied, excluding, as in our study, people in long-term care, with terminal diseases, or with moderate/severe cognitive impairment. In both studies, the 15-item Geriatric Depression Scale was administered. At an identical cut-off (6+), the prevalence rate of 9.7% which we determined was slightly higher than that in the English study (7.7%). With regard to the total population of the elderly in both studies, depression rates are probably underestimated, since residents in nursing homes were excluded. It is well known that not only rates of dementia but also rates of depression are particularly high among the institutionalized elderly. For instance, Jongenelis et al. (2004) reported that the prevalence rates of depression found among nursing home residents were three to four times higher than in the communitydwelling elderly. Numerous community-based studies indicate that also in old age depression rates are significantly higher among women compared to men. This was also confirmed in our study. However, similar to the results reported by Osborn et al. (2003), after controlling for potential risk factors there was no significant correlation between female gender and depressive disorder. This finding corroborates the hypothesis of Jorm (1987) that the overly proportional rate of depressive disorders among middle-aged women is less markedly so with increasing age. The age effect reported by Osborn et al. (2003), with significantly higher depression rates in the older age groups (85+), could also be confirmed. However, controlling for all variables this association was no longer significant. Univariate analysis revealed significant associations between depression and marital status and education
160
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
that at first led one to presume a protective effect for subjects with a high level of education and for married subjects. However, the association was no longer significant after adjustment for sex and age. The results of the Berlin Aging Study (Linden et al., 1998) show a similar trend with regard to marital status, with a lesser number of depressive disorders found among married subjects. In the Gospel Oak Project (Prince et al., 1997) only married men ran less risk of a depressive disorder, while the risk was noticeably higher among married women. In correspondence to the results of the Berlin Aging Study (Linden et al., 1998), we found increased depression rates among subjects with impaired mobility, vision, or hearing. The extent of the relationship was modified by the variables age and gender, and in the further course also by the other potential risk factors, but remained significantly correlated with a depressive disorder for all functionally impaired groups. With regard to the relationship between physical illnesses and depressive disorders, Osborn et al. (2003) discovered that the subjects with only one severe illness showed no greater risk than those without such an illness. However, those subjects with two or more illnesses ran a 60% greater risk of depression (GDS). Different from our study, this particular study did not consider functional impairments as potential risk factors. In our study depression was also associated with somatic comorbidity (1–4 diagnoses: OR = 1.4; 5+ diagnoses: OR = 2.4). However, even for the multimorbid subjects (5 and more illnesses), the association in the fully adjusted model failed to reach the level of significance. McDougall et al. (2007) have arrived at similar results after controlling for potential risk factors, finding no significant correlation between depression and comorbid disorders. In this study as well, the strongest associations resulted between functional impairments and depression. Our results indicate that impairments of mobility, vision and hearing exhibit a stronger association with depression than do individual somatic illnesses, respectively the number of medical diagnoses. Similar results are reported by the EURODEP Study (Braam et al., 2005), a consortium of 14 research groups in 11 European States, according to which the association of depressive symptoms with functional disability was stronger than that with chronic physical conditions. With regard to the relationship between lifestyle factors (consumption of nicotine and alcohol) and depression, our results supported the findings reported by Osborn et al. (2003): No significant correlations between consumption of alcohol and depressive symp-
toms were found in either study; however, the correlations between current smoking and depressive symptoms were highly significant, even after controlling for potential risk factors. In a cross-sectional study of older adults attending a representative sample of general practitioners in Western Australia, Almeida and Pfaff (2005) also found that current smoking is associated with increased frequency and severity of depression. There is general agreement on the cross-sectional correlation between depressive and cognitive symptoms (Vinkers et al., 2004; Ganguli et al., 2006; Geda et al., 2006). Thus, cognitive impairments frequently occur in the course of a depression, which can take the shape of attention and concentration disorders, slower cognitive processing, and weaknesses in the storage of new information (Reischies and Neu, 2000). The significant correlation between depressive disorders and the “multidomain MCI amnestic” confirms in this context that the concurrent manifestation of depressive and cognitive disorders is characterized particularly by the occurrence of mild impairments in various functional areas of cognition. Still unclear, however, is whether or not depressive symptoms lead to cognitive deterioration over the course of time, respectively, whether or not these represent an independent risk factor for a subsequent dementia disorder. Several studies (e.g., Wilson et al., 2002; Green et al., 2003; Modrego and Ferrández, 2004) have shown an association between depressive symptoms and an increased risk of developing Alzheimer's dementia. Modrego and Ferrández (2004) report that particularly those subjects with mild cognitive impairments (MCI according to Petersen et al., 1999) and depression as compared to non-depressives, run a more than twofold greater risk of developing an Alzheimer dementia within the course of three years. The analysis of the longitudinal data of our study will help us to clarify the direction of the relationship between depression and cognitive impairment. 5. Limitations A number of factors may have influenced the findings and should be considered when interpreting the results: The general practices taking part were not randomly selected. Since only about 50% of the randomly selected patients consented to participate in the study, selection bias might also play a role. The survey excluded institutionalized elderly people, subjects who were unable to attend a primary care physician, and those who suffered from dementia. Since depressive disorders occur more frequently among those in
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
161
residential care and those with functional impairment and dementia, prevalence rates in our study are probably underestimated. Cases were defined according to scores on the Geriatric Depression Scale (GDS-15), not according to clinical diagnostic criteria. However, this widely used instrument has shown good psychometric properties and allows numerous comparisons with epidemiological studies in different countries. Finally, the results are based on a cross-sectional analysis and therefore provide no insight into the direction of effects.
Role of the funding source This study is part of the German Competence Network Dementia and was funded by the German Federal Ministry of Education and Research (BMBF, grant O1 GI 0420). The funding source had no involvement.
6. Conclusions
This study was funded by the German Ministry of Education and Research within the framework of the Competence Network Dementia (BMBF, grant O1 GI 0420). Our thanks go to all participating patients and their general practitioners for their good collaboration:
With the demographic change in the forthcoming decades, more emphasis should be put on large-scale epidemiological studies of the older old since in many countries the increase in this age group will be particularly high. Our study revealed that the prevalence of depression among elderly patients in general practices is high and remains high into old age. Short scales such as the GDS-15 may be useful in helping GPs and practice staff to identify elderly patients with depressive symptoms (D'Ath et al., 1994). In agreement with previous studies, we found that depression is more prevalent among women compared to men. However, after controlling for confounding variables, gender-specific differences are no longer significant. We found high levels of functional disability, mild cognitive impairment, and smoking to be the most important risk factors for depression. The strong association between functional impairment and depression might stem from a reciprocally causal relationship between the two conditions. Either condition might precipitate the other or prolong its duration, thus raising the prevalence rate of their concurrence. Primary care practice offers ample opportunity to treat mental health problems such as depression that occur in relation to physical disability. If functional impairment accounts for higher rates of depression, the modification of this risk factor could well lower rates of depression among the elderly. Functional impairment can be modified by a large array of measures: maintaining physical health, change in personal health habits, exercise, correcting or compensating functional deficits by means of medical and surgical treatments, walking aids, as well as environmental modifications. Intervention strategies that focus on maintaining health and preventing functional impairment may reduce the prevalence and impact of depression. There is an obvious need for intervention studies of this type.
Conflict of interest All authors declare that there are not any actual or potential conflicts of interest.
Acknowledgements
Hamburg: Gundula Bormann, Winfried Bouché, Doris Fischer-Radizi, Michael Funke, Heike Gatermann, Wolfgang Herzog, Petra Hütter, Stefanie Kavka-Ziegenhagen, Günther Klötzl, Bernd-Uwe Krug, Dietrich Lau, Ursula Linn, Andrea Moritz, Karl-Christian Münter, Detlef Niemann, Klaus Richard-Klein, Walter Schreiber, Ursula SchröderHöch, Gerhard Schulze, Klaus Stelter, Carl-Otto Stolzenbach, Ljudmila Titova, Klaus Weidner, OttoPeter Witt, Eckehard Zeigert; Mannheim: Gerhard Arnold, Veit-Harold Bauer, Werner Besnier, Hanna Böttcher-Schmidt, Hartmut Grella, Gernot Kunzendorf, Ingrid Ludwig, Manfred Mayer, Hubert Mühlig, Arnt Müller, Adolf Noky, Helmut Perleberg, Carsten Rieder, Michael Rosen, Georg Scheer, Michael Schilp, Matthias Schneider, Jürgen Wachter, Brigitte Weingärtner, Hans-Georg Willhauck; Bonn: Jörg Eimers-Kleene, Klaus Fischer, Maria Goebel-Schlatholt, Peter Gülle, Wolf-Dietrich Honig, Hans Jürgen Kaschell, Hanna Liese, Manfred Marx, Eberhard Prechtel, Heinz-Peter Romberg, Heribert Schützendorf, Annemarie Straimer, Martin Tschoke, Karl-Michael Werner; Halstenbek: Herrmut Mayen; Königswinter: Theodor Alfen; Bad Honnef: Klaus Weckbecker; Niederkassel: Inge Bürfent; Alfter-Oedekoven: Johann von Aswege; Erfstadt-Liblar: Arndt Uhlenbrock; Windeck-Herchen: Wolf-Rüdiger Weisbach; Leipzig: Martina Amm, Heinz-Michael Assmann, Horst Bauer, Barbara Bräutigam, Jochen Ebert, Angelika Gabriel, Eva Hager, Gunter Kässner, Ina
162
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
Lipp, Thomas Lipp, Ute Mühlmann, Gabi Müller, Thomas Paschke, Gabriele Rauchmaul, Ina Schmalbruch, Holger Schmidt, Hans-Christian Taut, Ute Voβ, Bettina Winkler, Sabine Ziehbold; München: Eugen Allwein, Guntram Bloβ, Peter Dick, Johann Eiber, Lutz-Ingo Fischer, Peter Friedrich, Helga Herbst, Andreas Hofmann, Günther Holthausen, Karl-Friedrich Holtz, Ulf Kahmann, Elke Kirchner, Hans Georg Kirchner, Luitpold Knauer, Andreas Koeppel, Heinz Koschine, Walter Krebs, Franz Kreuzer, Karl Ludwig Maier, Christoph Mohr, Elmar Schmid, Gabriel Schmidt, Johann Thaller; Haar: Richard Ellersdorfer, Michael Speth;Düsseldorf: Angela Ackermann, Pauline Berger, Florinela Cupsa, Barbara Damanakis, Klaus-Wolfgang Ebeling, Tim Oliver Flettner, Michael Frenkel, Friederike Ganßauge, Kurt Gillhausen, Hans-Christian Heede, Uwe Hellmessen, Benjamin Hodgson, Bernhard Hoff, Helga Hümmerich, Boguslaw-Marian Korman, Dieter Lüttringhaus, Dirk Matzies, Vladimir Miasnikov, Wolfgang Josef Peters, Birgitt Richter-Polynice, Gerhard Erich Richard Schiller, Ulrich Schott, Andre Schumacher, Harald Siegmund, Winfried Thraen, Roland Matthias Unkelbach, Clemens Wirtz.
References Almeida, O.P., Pfaff, J.J., 2005. Depression and smoking amongst older general practice patients. Journal of Affective Disorders 86, 317–321. Braam, A.W., Prince, M.J., Beekman, A.T.F., Delespaul, P., Dewey, M.E., Geerlings, S.W., Kivelä, S.-L., Lawlor, B.A., Magnússon, H., Meller, I., Pérès, K., Reischies, F.M., Roelands, M., Schoevers, R.A., Saz, P., Skoog, I., Turrina, C., Versporten, A., Copeland, J.R.M., 2005. Physical health and depressive symptoms in older Europeans. Results from EURODEP. British Journal of Psychiatry 187, 35–42. Brauns, H., Steinmann, S., 1999. Educational reform in France, WestGermany and the United Kingdom. ZUMA-Nachrichten 44, 7–44. British Medical Association, 1995. Alcohol: guidelines on sensible drinking. British Medical Association, London. Cole, M.G., Dendukuri, N., 2003. Risk factors for depression among elderly community subjects: a systematic review and metaanalysis. American Journal of Psychiatry 160 (6), 1147–1156. D'Ath, P., Katona, P., Mullan, E., Evans, S., Katona, C., 1994. Screening, detection and management of depression in elderly primary care attenders. I: The acceptability and performance of the 15 item Geriatric Depression Scale (GDS 15) and the development of short versions. Family Practitioner 11, 260–266. Ganguli, M., Du, Y., Dodge, H.H., Ratcliff, G.G., Chang, C.C., 2006. Depressive symptoms and cognitive decline in late life. Archives of General Psychiatry 63, 153–160. Gauggel, S., Birkner, B., 1999. Validität und Reliabilität einer deutschen Version der Geriatrischen Depressionsskala (GDS). Zeitschrift für Klinische Psychologie 28, 18–27.
Geda, Y.E., Knopman, D.S., Mrazek, D.A., Jicha, G.A., Smith, G.E., Negash, S., Boeve, B.F., Ivnik, R.J., Petersen, R.C., Pankratz, V.S., Rocca, W.A., 2006. Depression, apolipoprotein E genotype, and the incidence of mild cognitive impairment: a prospective cohort study. Archives of Neurology 63, 435–440. Green, R.C., Cupples, L.A., Kurz, A., Auerbach, S., Go, R., Sadovnick, D., Duara, R., Kukull, W.A., Chui, H., Edeki, T., Griffith, P.A., Friedland, R.P., Bachmann, D., Farrer, L., 2003. Depression as a risk factor for Alzheimer disease. Archives of Neurology 60, 753–759. Jongenelis, K., Pot, A.M., Eisses, A.M.H., Beekman, A.T.F., Kluiter, H., Ribbe, M.W., 2004. Prevalence and risk indicators of depression in elderly nursing home patients: the AGED study. Journal of Affective Disorders 83, 135–142. Jorm, A.F., 1987. Sex and age differences in depression: a quantitative synthesis of published research. Australian and New Zealand Journal of Psychiatry 21, 46–53. Linden, M., Kurtz, G., Baltes, M.M., Geiselmann, B., Reischies, F.M., 1998. Depression bei Hochbetagten. Ergebnisse der Berliner Altersstudie. Nervenarzt 69, 27–37. Luck, T., Riedel Heller, S.G., Kaduszkiewicz, H., Bickel, H., Jessen, F., Pentzek, M., Wiese, B., Koelsch, H., van den Bussche, H., Abholz, H.-H., Moesch, E., Gorfer, S., Angermeyer, M.C., Maier, W., Weyerer, S., 2007a. Mild cognitive impairment in general practice: age-specific prevalence and correlate results from the German Study on Ageing, Cognition and Dementia in primary care patients (AgeCoDe). Dementia and Geriatric Cognitive Disorders 24, 307–316. Luck, T., Zaudig, M., Wiese, B., Riedel-Heller, S.G., 2007b. SIDAM: Alters-und bildungsspezifische Normen des kognitiven Leistungsteils nach der neuen CASMIN-Bildungsklassifikation. Zeitschrift für Gerontopsychologie und-psychiatrie 20 (1), 31–38. Luppa, M., Heinrich, S., Matschinger, H., Sandholzer, H., Angermeyer, M.C., König, H.-H., Riedel-Heller, S.G., 2008. Direct costs associated with depression in old age in Germany. Journal of Affective Disorders 105, 195–204. Lyness, J.M., Caine, E.D., King, D.A., Conwell, Y., Duberstein, P.R., Cox, C., 2002. Depressive disorders and symptoms in older primary care patients: One year outcomes. American Journal of Geriatric Psychiatry 10, 275–282. Margraf, J., 1994. Mini-DIPS. Diagnostisches Kurz-Interview bei psychischen Störungen. Springer, Berlin. McDougall, F.A., Kvaal, K., Matthews, F.E., Paykel, E., Jones, P.B., Dewey, M.E., Braine, C., 2007. Prevalence of depression in older people in England and Wales: the MRC CFA Study. Psychological Medicine 1–9. Modrego, P.J., Ferrández, J., 2004. Depression in patients with mild cognitive impairment increases the risk of developing dementia of Alzheimer Type. Archives of Neurology 61, 1290–1293. Osborn, D.P., Fletcher, A.E., Smeeth, L., Stirling, S., Bulpitt, C.J., Breeze, E., Ng, E.S., Nunes, M., Jones, D., Tulloch, A., 2003. Factors associated with depression in a representative sample of 14217 people aged 75 and over in the United Kingdom: results from the MRC trial of assessment and management of older people in the community. International Journal of Geriatric Psychiatry 18, 623–630. Palsson, S., Skoog, I., 1997. The epidemiology of affective disorders in the elderly: a review. International Clinical Psychopharmacology 12 (suppl. 7), 3–13. Petersen, R.C., Smith, G.E., Waring, S.C., Ivnik, R.J., Tangalos, E.G., Kokmen, E., 1999. Mild cognitive impairment: clinical characterization and outcome. Archives of Neurology 56, 303–308.
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163 Prince, M.J., Harwood, R.H., Blizard, R.A., Thomas, A., Mann, A.H., 1997. Social support deficits, loneliness and life events as risk factors for depression in old age. The Gospel Oak Project VI. Psychological Medicine 27, 323–332. Reischies, F.M., Neu, P., 2000. Comorbidity of mild cognitive disorder and depression — a neuropsychological analysis. European Archives of Psychiatry and Clinical Neuroscience 250, 186–193. Riedel-Heller, S.G., Busse, A., Angermeyer, M.C., 2006. The state of mental health in old-age across the ’old’ European Union — a systematic review. Acta Psychiatrica Scandinavica 113, 388–401. Sheikh, J.I., Yesavage, J.A., 1986. Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. Clinical Gerontologist 5, 165–173. Speer, D.C., Schneider, M.G., 2003. Mental health needs of older adults and primary care: Opportunity for interdisciplinary geriatric team practice. Clinical Psychology: Science and Practice 21, 85–101. Vinkers, D.J., Gussekloo, J., Stek, M.L., Westendorp, R.G.J., van der Mast, R.C., 2004. Temporal relation between depression and cognitive impairment in old age: prospective population based study. British Medical Journal 329, 1–4.
163
Wancata, J., Alexandrowicz, R., Marquart, B., Weiss, M., Friedrich, F., 2006. The criterion validity of the Geriatric Depression Scale: a systematic review. Acta Psychiatrica Scandinavica 114, 398–410. Wilson, R.S., Barnes, L.L., Mendes de Leon, C.F., Aggarwal, N.T., Schneider, J.S., Bach, J., Pilat, J., Beckett, L.A., Arnold, S.E., Evans, D.A., Bennett, D.A., 2002. Depressive symptoms, cognitive decline, and risk of AD in older persons. Neurology 59, 364–370. Winblad, B., Palmer, K., Kivipelto, M., Jelic, V., Fratiglioni, L., Wahlund, L.-O., Nordberg, A., Bäckman, L., Albert, M., Almkvist, O., Arai, H., Basun, H., Blennow, K., De Leon, M., Decarli, C., Erkinjuntti, T., Giacobini, E., Graff, C., Hardy, J., Jack, C., Jorm, A., Ritchie, K., Van Duijn, C., Visser, P., Petersen, R.C., 2004. Mild cognitive impairment — beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. Journal of Internal Medicine 256, 240–246. Zaudig, M., Hiller, W., 1996. SIDAM-Handbuch. Strukturiertes Interview für die Diagnose einer Demenz vom Alzheimer Typ, der MultiInfarkt-(oder vaskulären) Demenzen und Demenzen anderer Ätiologien nach DSM-III-R, DSM-IV und ICD-10. Huber, Bern.
Research report
Prevalence and risk factors for depression in non-demented primary care attenders aged 75 years and older Siegfried Weyerer a,⁎, Sandra Eifflaender-Gorfer a , Leonore Köhler a , Frank Jessen b , Wolfgang Maier b , Angela Fuchs c , Michael Pentzek c , Hanna Kaduszkiewicz d , Cadja Bachmann d , Matthias C. Angermeyer e , Melanie Luppa e , Birgitt Wiese f , Edelgard Mösch g , Horst Bickel g for the German AgeCoDe Study group (German Study on Ageing, Cognition and Dementia in Primary Care Patients) 1 a
Central Institute of Mental Health, Mannheim, Germany Department of Psychiatry, University of Bonn, Germany c Department of General Practice, University Medical Center Düsseldorf, Germany Institute of Primary Medical Care, University Medical Center Hamburg-Eppendorf, Germany e Department of Psychiatry, University of Leipzig, Germany f Institute for Biometrics, Hannover Medical School, Germany g Department of Psychiatry, Technical University of Munich, Germany b
d
Received 12 December 2007; received in revised form 12 February 2008; accepted 12 February 2008 Available online 26 March 2008
Abstract Background: Depression among the elderly is an important public health issue. The aims of this study were to report the prevalence of depression and to determine the impact of socio-demographic variables, functional impairment and medical diagnoses, lifestyle factors, and mild cognitive impairment on depression as part of the German Study on Ageing, Cognition and Dementia in Primary Care Patients (AgeCoDe Study). Methods: Included in the cross-sectional survey were 3327 non-demented subjects aged 75 and over attending general practitioners (GPs) (n = 138) in an urban area of Germany. The GDS-15 Geriatric Depression Scale was used to measure depression with a threshold of b6/6+. Associations with social and clinical risk factors were assessed by means of multiple logistic regression models. Results: The prevalence of depression was 9.7% (95% confidence interval 8.7–10.7). In a univariate analysis, the following variables were significantly associated with depression: female gender, increasing age, living alone, divorce, lower educational status, functional impairment, comorbid somatic disorder, mild cognitive impairment, smoking, and abstinence from alcohol. After full adjustment for confounding variables, odds ratios for depression were significantly higher only for functional impairment, smoking, and multi-domain mild cognitive impairment.
⁎ Corresponding author. Tel.: +49 621 1703 6351; fax: +49 621 1703 1305. E-mail address: [email protected] (S. Weyerer). 1 Further members of the German AgeCoDe Study group: Steffi G. Riedel-Heller, Tobias Luck, Heike Kölsch, Hendrik van den Bussche, HeinzHarald Abholz, Gabriela Cvetanovska, Franziska Haller, Michael Wagner, Thomas Zimmermann, Teresa Kaufeler, Jochen Werle, Manfred Mayer, Heinz-Peter Romberg, Hagen Sandholzer, Anja Frenzen, Anja Wollny. 0165-0327/$ - see front matter © 2008 Elsevier B.V. All rights reserved. doi:10.1016/j.jad.2008.02.008
154
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
Limitations: Recruitment procedures might have led to an underestimation of current prevalence. The cross-sectional data did not allow us to analyze the temporal relationship between risk factors and depression. Conclusions: The prevalence of depression in the elderly is high and remains high into old age. In designing prevention programs, it is important to call more attention to the impact of functional and cognitive impairment on depression. © 2008 Elsevier B.V. All rights reserved. Keywords: Depression; Epidemiology; Risk factors; Old age; General practice
1. Introduction Depression in old age is common. Prevalence rates among elderly people seen in primary care range from 6.5 to 9% for major depression (Lyness et al., 2002) and from 10 to 25% for clinically relevant depressive symptoms (Speer and Schneider, 2003). In meta-analyses (e.g., Cole and Dendukuri, 2003), a number of risk factors for the occurrence of depression in old age have been isolated: Socio-demographic factors such as gender and marital status are associated with depression, showing higher rates for women and lower rates for the married. Depression in old age is also associated with more physical disorders and increased functional impairment (e.g., Braam et al., 2005). In addition to poor functioning, depression increases the utilization of medical services and health care costs (Luppa et al., 2008). Although the association between age and depression received considerable attention, very little is known about the prevalence and course of depression among those 75 years of age and older. Studies that treat the group 65+ as one entity are often heavily weighted towards the age group 65–75 (Riedel-Heller et al., 2006). Therefore, the prevalence of depression in the very old is uncertain since many community-based studies lack adequate samples over the age of 75. The findings from the few studies which report prevalence rates for the older old are inconsistent, some showing a rise over the age 75, some a fall (Palsson and Skoog, 1997; McDougall et al., 2007). To study the older old is also important since some crucial risk factors such as bereavement, social isolation, somatic diseases, and functional impairment become more common with increasing age. These factors may exert different effects in the younger old compared to the older old. Knowledge of risk factors is a prerequisite to designing tailored interventions, either to tackle the factors itself or to define high risk groups, since depression is treatable in many cases. Based on a large sample of non-demented primary care attenders (age 75+), the objectives of this study are: • to report age and gender-specific prevalence rates of depression and
• to determine the impact of putative risk factors: socio-demographic variables (marital status, education), functional impairment and medical diagnoses, lifestyle factors (consumption of alcohol and nicotine), and mild cognitive impairment. 2. Methods 2.1. Study design and sample The sample consists of all subjects participating in the baseline assessment of a prospective longitudinal study on the early detection of mild cognitive impairment and dementia in primary care. The study was conducted in six centers (Bonn, Düsseldorf, Hamburg, Leipzig, Mannheim, and Munich) representing an urban area of cities with a total population ranging between about 300,000 (Mannheim) and almost 1,8 million (Hamburg). The subjects were recruited between January 1, 2003 and November 30, 2004. In each center, 19 to 29 GPs participated in the recruitment process: all in all, 138 GPs. Inclusion criteria for GP patients were an age of 75 years and over, the absence of dementia in the GP's view, and at least one contact with the GP within the last 12 months. Exclusion criteria were consultations only by home visits, residence in a nursing home, a severe illness which the GP would deem fatal within three months, an insufficient knowledge of the German language, deafness or blindness, inability to consent, and not being a regular patient of the participating GP. On average, each practice comprised 24 patients. Information on sampling frame, eligible subjects, and respondents is given in Fig. 1. Finally, 3327 selected GP patients were assessed by means of structured clinical interviews. Eighty-five (2.6%) of the 3327 interviewed subjects were excluded from the following analyses: 41 (48.2%) were classified as demented, 39 (45.9%) fell short of the age limit of 75 years, and 5 (5.9%) had incomplete neuropsychological test data. The calculation of the prevalence of depression is based on the remaining 3242 subjects. In order to analyze possible non-response bias, data on age and gender was collected for 1770 (99.7%) of the 1775 subjects refusing participation. The
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
155
psychodiagnostic interview (Mini-DIPS; Margraf, 1994), a cut-off score of 6 yielded the best sensitivity (84.0%) and specificity (88.9%) (Gauggel and Birkner, 1999). Therefore, this cut-off was used in the present study. Neuropsychological assessment was based on the Structured Interview for Diagnosis of Dementia of Alzheimer type, Multi-infarct Dementia and Dementia of other Etiology according to DSM-III-R, DSM-IV and ICD-10 (SIDAM; Zaudig and Hiller, 1996). In order to evaluate impairment of cognitive functioning, age- and education-specific norms for the cognitive domains were applied (Luck et al., 2007b). Dementia according to DSM-IV was excluded by means of the SIDAM. Mild Cognitive Impairment (MCI) was diagnosed according to new consensus criteria proposed by the International Working Group on Mild Cognitive Impairment. According to Winblad et al. (2004), four subtypes for MCI were examined based solely on differences in the criterion of objective cognitive decline (Luck et al., 2007a):
Fig. 1. Sampling frame and sample.
mean age of the included subjects was 80.2 years (SD = 3.6) vs. 80.8 years (SD = 3.8) for those refusing participation (t = - 6.104, p = 0.000). 2127 (65.6%) of all participants were female and 1115 (34.4%) were male, among non-participants 1219 (68.9%) were female and 551 (31.1%) were male (χ2 = 5.594, df = 1, p = 0.018). Thus, group of participants was significantly younger and included more males than did the group of nonparticipants. 2.2. Instruments Structured clinical interviews were conducted by trained physicians and psychologists during visits to the participants' homes. Depressive symptoms were assessed by means of the 15-item short version of the Geriatric Depression Scale (GDS; Sheikh and Yesavage, 1986; D'Ath et al., 1994). Due to the simplified yes/no response format and the exclusion of questions on somatic symptoms, it is especially suitable for elderly people. This instrument has good psychometric properties also for Germanspeaking populations (Wancata et al., 2006). Comparing the German version of the GDS-15 with a
• Single-domain amnestic MCI was used for subjects having an objective deficit in memory but not in any other area of cognitive functioning. • Single nonmemory MCI was diagnosed only if a single domain other than memory was impaired. • Multi-domain MCI amnestic was diagnosed if memory and at least one other cognitive domain were impaired. • Multi-domain MCI nonamnestic was diagnosed if at least two cognitive domains other than memory showed an objective impairment. In this study modified versions of the MCI (see Luck et al., 2007a) were used. With the exception of subjective cognitive complaints, each of these modifications was defined by the same subtypes according to Winblad et al. (2004). Lifestyle factors assessed include self-reported alcohol consumption and smoking, and it was differentiated between current consumers and nonconsumers. With regard to alcohol consumption, current consumers were distinguished according to their level of consumption: “normal” or “at risk”. Following widely accepted international guidelines (e.g., British Medical Association, 1995), at-risk drinking was defined as the daily consumption of N 20 g alcohol for women and N 30 g alcohol for men. The measurement of functional impairment (mobility, vision, and hearing) was based on the SIDAMADL-Scale (Zaudig and Hiller, 1996).
156
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
Table 1 Univariate association with depression N (% of total 3242)
Cases on GDS 6+ (%)
Univariate odds ratio (95%CI)
Gender Female 2127 (65.6) 237 (11.1) 1.7 (1.28–2.20) Male 1115 (34.4) 76 (6.8) 1 Age 75–79 1725 (53.2) 149 (8.6) 1 80–84 1210 (37.3) 122 (10.1) 1.2 (0.92–1.53) 85+ 307 (9.5) 42 (13.7) 1.7 (1.16–2.42) Living alone Yes 1661 (51.2) 186 (11.2) 1.4 (1.14–1.83) No 1581 (48.8) 127 (8.0) 1 Marital status Single 203 (6.3) 23 (11.3) 1 Married 1377 (42.5) 99 (7.2) 0.6 (0.38–0.98) Divorced 193 (6.0) 36 (18.7) 1.7 (1.02–3.16) Widowed 1469 (45.3) 155 (10.6) 0.9 (0.58–1.47) Level of education (Based on the revised version of the international CASMIN educational classification; Brauns and Steinmann, 1999) Low 2013 (62.1) 210 (10.4) 1 Middle 886 (27.3) 80 (9.0) 0.9 (0.64–1.10) High 343 (10.6) 23 (6.7) 0.6 (0.36–0.90) Mobility impairment Yes 1146 (35.3) 199 (17.4) 3.7 (2.87–4.66) No 2096 (64.7) 114 (5.4) 1 Vision impairment Yes 469 (14.5) 81 (17.3) 2.3 (1.74–3.01) No 2773 (85.5) 232 (8.4) 1 Hearing impairment Yes 993 (30.6) 126 (12.7) 1.6 (1.26–2.04) No 2249 (69.4) 187 (8.3) 1 Somatic comorbidity No somatic comorbidity 143 (4.3) 9 (6.5) 1 1–4 somatic diagnoses 2233 (68.5) 192 (8.6) 1.4 (0.68–2.73) 5 and more somatic diagnoses 888 (27.2) 112 (12.7) 2.1 (1.04–4.25) Smoking (at present) Yes 244 (7.5) 34 (13.9) 1.6 (1.08–2.31) No 2994 (92.5) 279 (9.3) 1 Alcohol consumption (at present) Completely abstinent 1613 (50.1) 192 (11.9) 1.7 (1.33–2.17) Moderate 1478 (45.9) 109 (7.4) 1 At risk (men: 30 g+ per day; 128 (4.0) 9 (7.0) 1.0 (0.47–1.92) women: 20 g+ per day) Mild Cognitive Impairment (MCI)-modified a Unimpaired 2422 (74.8) 209 (8.6) 1 Single-domain amnestic MCI 97 (3.0) 15 (15.5) 1.9 (1.10–3.42) Single nonmemory MCI 486 (15.0) 44 (9.1) 1.1 (0.75–1.48) Multi-domain MCI amnestic 134 (4.1) 26 (19.4) 2.5 (1.62–4.00) Multi-domain MCI nonamnestic 100 (3.1) 18 (18.0) 2.3 (1.37–3.95) a
p-value b0.001 – – 0.184 0.006 0.002 – – 0.041 0.043 0.736 – 0.247 0.034 b0.001 – b0.001 – b0.001 – – 0.377 0.038 0.020 – b0.001 – 0.886
– 0.023 0.762 b0.001 0.002
Modified subtypes (exclusion of the criterion of a subjective cognitive complaint) of MCI according to Winblad et al. (2004).
For each study participant, his or her GP filled in a questionnaire to assess comorbidity. The following clinical diagnoses were predetermined by the questionnaire (answer: yes/no/I don't know): diabetes mellitus, high blood pressure, cardiac arrhythmia, coronary heart disease, myocardial infarction, hyperlipidemia, hyper-
cholesterolemia, peripheral arterial obstructive disease, Parkinson's disease, epilepsy, carotic artery stenosis (N 80%), transient ischemic attack (TIA), stroke, hypothyroidism, hyperthyroidism, renal insufficiency, and liver disease. Based on this information, somatic comorbidity was defined: no comorbidity/1–4 diagnoses/5+ diagnoses.
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
2.3. Statistical analyses The data were collected in the centers via an internetbased Remote-Data-Entry-System into a central ORACLE version 9 database. The statistical analyses were performed with SPSS for Windows, version 12.01. According to the defined cut-off (6+), prevalence rates of depression were estimated as the percentage of the completely assessed non-demented subjects aged 75 years and over. In addition to prevalence, 95% confidence intervals (95% CI) were calculated. Associations between risk factors and depression (scoring 6+ on the GDS-15) were analyzed using multiple logistic regression. Univariate Odds Ratios (OR) were generated for each variable and the independence of any association was initially controlled for sex and age, then additionally by entering all variables into the model. A p-value less than 0.05 was considered statistically significant. 2.4. Ethical approval The ethics committees of the participating centers approved the study. Written informed consent was obtained from all participants. 3. Results 3.1. Demographic characteristics Of the 3242 patients interviewed, 2127 were women (65.6%) and 1115 men (34.4%). Their average age was 80.2 years (SD = 3.6; range: 75–99). Based on the revised CASMIN classification according to Brauns and Steinmann (1999), 2013 (62.1%) of the subjects exhibited a low level of education, 886 (27.3%) a middle level of education, and 343 (10.6%) a high level of education. The majority of the patients were widowed (45.3%) or married (42.5%); only 203 (6.3%) of the subjects were never married, while a further 193 subjects (6.0%) were divorced.
157
the rate among female subjects was 11.1% (95% CI 9.76–12.44). The prevalence of depression increased with age: In the group of 75–79 year-olds, 8.6% (95% CI 7.28– 9.92), among the 80–84 year-olds already 10.1% (95% CI 8.40–11.80), and in the group of those over 85 years of age and older, 13.7% (95% CI 9.85–17.55) were affected by a depressive disorder. Above and beyond this, Table 2 contains the Odds Ratios for each potential risk factor, adjusted first of all for age and gender, and in a further step, for all the variables studied. In order to better visualize the adjustment effects, the univariate correlations were entered into a second column. 3.3. Socio-demographic factors Independent of age, women suffer from a depressive disorder significantly more often than men (OR = 1.7, 95% CI 1.27–2.18). Independent of their gender, the subjects 85 years of age and older (OR = 1.7, 95% CI 1.16–2.42) ran a 70% higher risk of depressive disorder compared to those in the group aged 75–79 years. With regard to marital status, a greater number of depressive subjects is found in the group of divorced persons (18.7%), as well as in the unmarried group (11.3%), while in the widowed group (10.6%), and especially in the married group (7.2%), the respective number of depressive subjects is much lower (OR = 0.6, 95% CI 0.38–0.98). The prevalence of depression was significantly higher among those living alone (11.2%) compared to those not living alone (8.0%). Adjusted for all variables, age, gender, marital status, and living alone are not significantly associated with depression. Univariate analysis revealed a significant correlation between depression and the level of education, with a low prevalence (6.7%) in subjects with a high level of education (OR = 0.6, 95% CI 0.36–0.90) compared to the group with a middle level of education (9.0%), respectively a low level of education (10.4%). After controlling for age and gender, this correlation was no longer significant (OR = 0.7, 95% CI 0.46–1.15).
3.2. Age- and gender-specific prevalence rates of depressive disorders
3.4. Functional impairment and comorbidity
The numbers of depressive subjects, including their proportional distribution within the categories of the various independent variables, are depicted in Table 1. According to the case definition, at baseline 313/3242 subjects (9.7%; 95% CI 8.68–10.72) could be identified as “cases”. While only 6.8% (95% CI 5.32–8.28) of the male subjects suffered from depressive symptoms,
Substantial differences in prevalence rates of depression are found in the area of functional limitations. The depression rate is 17.4% among those subjects impaired in their mobility, compared to a rate of 5.4% among those who are mobile. The association was also highly significant in the fully adjusted model (OR = 2.9, 95% CI 2.26–3.78).
158
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
Table 2 Univariate and adjusted associations with depression Univariate odds ratio
Odds ratio adjusted for age and sex
(95%CI)
(95%CI)
Gender Female 1.7 (1.28–2.20) Male 1 Age 75–79 1 80–84 1.2 (0.92–1.53) 85+ 1.7 (1.16–2.42) Living alone Yes 1.4 (1.14–1.83) No 1 Marital status Single 1 Married 0.6 (0.38–0.98) Divorced 1.7 (1.02–3.16) Widowed 0.9 (0.58–1.47) Level of education (CASMIN) Low 1 Middle 0.9 (0.64–1.10) High 0.6 (0.36–0.90) Mobility impairment Yes 3.7 (2.87–4.66) No 1 Vision impairment Yes 2.3 (1.74–3.01) No 1 Hearing impairment Yes 1.6 (1.26–2.04) No 1 Somatic comorbidity No somatic comorbidity 1 1–4 somatic diagnoses 1.4 (0.68–2.73) 5 and more somatic diagnoses 2.1 (1.04–4.25) Smoking (at present) Yes 1.6 (1.08–2.31) No 1 Alcohol consumption (at present) Completely abstinent 1.7 (1.33–2.17) Moderate 1 At risk (men: 30 g+ per day; 1.0 (0.47–1.92) women: 20 g+ per day) Mild Cognitive Impairment (MCI)-modified a Unimpaired 1 Single-domain amnestic MCI 1.9 (1.10–3.42) Single nonmemory MCI 1.1 (0.75–1.48) Multi-domain MCI amnestic 2.5 (1.62–4.00) Multi-domain MCI nonamnestic 2.3 (1.37–3.95) a
p-value
Odds ratio fully adjusted for all variables
p-value
(95%CI)
1.7 (1.27–2.18) 1
b0.001 –
1.3 (0.90–1.76) 1
0.178 –
1 1.2 (0.92–1.52) 1.7 (1.16–2.42)
0.196 0.006
1 0.9 (0.70–1.21) 1.2 (0.79–1.75)
– 0.540 0.419
1.2 (0.89–1.52) 1
0.266 –
0.8 (0.57–1.26) 1
0.411 –
1 0.7 (0.44–1.19) 1.7 (0.99–3.10) 0.9 (0.55–1.40)
– 0.199 0.055 0.576
1 0.6 (0.33–1.10) 1.4 (0.79–2.61) 0.8 (0.48–1.26)
– 0.097 0.240 0.300
1 0.8 (0.61–1.05) 0.7 (0.46–1.15)
– 0.109 0.173
1 0.8 (0.61–1.09) 0.7 (0.41–1.10)
– 0.168 0.112
3.5 (2.71–4.45) 1
b0.001 –
2.9 (2.26–3.78) 1
b0.001 –
2.1 (1.61–2.82) 1
b0.001 –
1.7 (1.25–2.26) 1
b0.001 –
1.6 (1.26–2.07) 1
b0.001 –
1.4 (1.10–1.83) 1
0.008 –
1 1.4 (0.68–2.73) 2.2 (1.08–4.46)
– 0.379 0.029
1 1.4 (0.64–3.16) 1.8 (0.80–4.09)
– 0.387 0.154
1.6 (1.10–2.39) 1
0.016 –
1.6 (1.03–2.36) 1
0.038 –
1.5 (1.15–1.93) 1 0.9 (0.46–1.87)
0.003 – 0.822
1.3 (0.95–1.64) 1 0.9 (0.43–1.84)
0.106 – 0.760
1 2.0 1.0 2.3 2.0
– 0.017 0.876 b0.001 0.014
1 1.6 (0.86–2.86) 1.0 (0.73–1.50) 2.1 (1.30–3.43) 1.5 (0.88–2.81)
– 0.140 0.798 0.003 0.125
(1.13–3.56) (0.73–1.45) (1.46–3.68) (1.15–3.41)
Modified subtypes (exclusion of the criterion of a subjective cognitive complaint) of MCI according to Winblad et al. (2004).
The prevalence of depression among subjects with impaired vision and impaired hearing was significantly higher than among subjects without such deficits. The extent of the relationship was influenced by the variables age and gender; however, a significant correlation
with depression remained for subjects with impaired vision (OR = 2.1, 95% CI 1.61–2.82), as well as for those with impaired hearing (OR = 1.6, 95% CI 1.26– 2.07). After controlling for potential risk factors, the relationship decreases further, but there still remains a
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
significantly higher risk of depression among subjects with impaired vision (OR = 1.7, 95% CI 1.25–2.26) and among those with impaired hearing (OR = 1.4, 95% CI 1.10–1.83). Univariate analysis (results not shown in table) revealed that in 5 out of 17 somatic diseases assessed by the primary care physician, depression rates are statistically significant higher: for Parkinson's disease (OR = 3.6, 95% CI 1.99–6.66), transient ischemic attack (OR = 1.7, 95% CI 1.15–2.40), coronary heart disease (OR = 1.5, 95% CI 1.16–1.86), cardiac arrhythmia (OR = 1.4, 1.06–1.73), and diabetes mellitus (OR = 1.3, 95% CI 1.04–1.73). Compared to subjects without any somatic comorbidity, the prevalence of depression among the group of multimorbid subjects (5 or more somatic diagnoses) is almost twice as high (12.7% vs. 6.5%). The relationship between depression and multimorbidity remains significant even when age and gender are taken into account (OR = 2.2, 95% CI 1.08–4.46), but falls below the level of significance after controlling for all potential risk factors (OR = 1.8, 95% CI 0.80–4.09). 3.5. Lifestyle factors While the rate of depression is higher among current smokers than among non-smokers (13.9% vs. 9.3%), the opposite is true in terms of alcohol consumption. Here, the number of depressive subjects is higher in the group of abstinent subjects (11.9%) than in the groups whose alcohol consumption is either “normal” (7.4%) or “at risk” (7.0%). The risk of depression is significantly higher for smokers, both after controlling for age and gender (OR = 1.6, 95% CI 1.10–2.39) and in the fully adjusted model (OR = 1.6, 95% CI 1.03–2.36). With regard to alcohol consumption, the correlation with depression was no longer significant once all variables had been adjusted. 3.6. Mild Cognitive Impairment (MCI) Compared to the cognitively unimpaired subjects, subjects with “single nonmemory MCI” do not exhibit an increased risk of depression (OR = 1.1, 95% CI 0.75– 1.48). In comparison thereto, subjects with “singledomain amnestic MCI” (OR = 1.9, 95% CI 1.10–3.42), “multi-domain MCI nonamnestic” (OR = 2.3, 95% CI 1.37–3.95) and “multi-domain MCI amnestic” (OR = 2.5, 95% CI 1.62–4.00) have a significantly higher risk of depression. While age and gender influence these correlations only slightly, after controlling for all
159
potential determinants only those subjects with “multidomain MCI amnestic” (OR = 2.1, 95% CI 1.30–3.43) exhibited a significantly increased risk of depression ( p = 0.003). 4. Discussion The current study is one of the largest studies worldwide to have been conducted on depression in very old patients in primary care. We found that depression in urban areas of Germany is a common condition: One in ten subjects 75 years of age and older is affected. A direct comparison with other studies is difficult, however, due to the varying operationalization of depressive disorders and the heterogeneous age groups. In a large-scale study of general practices carried out by Osborn et al. (2003) in the United Kingdom, the same age group (75+) was studied, excluding, as in our study, people in long-term care, with terminal diseases, or with moderate/severe cognitive impairment. In both studies, the 15-item Geriatric Depression Scale was administered. At an identical cut-off (6+), the prevalence rate of 9.7% which we determined was slightly higher than that in the English study (7.7%). With regard to the total population of the elderly in both studies, depression rates are probably underestimated, since residents in nursing homes were excluded. It is well known that not only rates of dementia but also rates of depression are particularly high among the institutionalized elderly. For instance, Jongenelis et al. (2004) reported that the prevalence rates of depression found among nursing home residents were three to four times higher than in the communitydwelling elderly. Numerous community-based studies indicate that also in old age depression rates are significantly higher among women compared to men. This was also confirmed in our study. However, similar to the results reported by Osborn et al. (2003), after controlling for potential risk factors there was no significant correlation between female gender and depressive disorder. This finding corroborates the hypothesis of Jorm (1987) that the overly proportional rate of depressive disorders among middle-aged women is less markedly so with increasing age. The age effect reported by Osborn et al. (2003), with significantly higher depression rates in the older age groups (85+), could also be confirmed. However, controlling for all variables this association was no longer significant. Univariate analysis revealed significant associations between depression and marital status and education
160
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
that at first led one to presume a protective effect for subjects with a high level of education and for married subjects. However, the association was no longer significant after adjustment for sex and age. The results of the Berlin Aging Study (Linden et al., 1998) show a similar trend with regard to marital status, with a lesser number of depressive disorders found among married subjects. In the Gospel Oak Project (Prince et al., 1997) only married men ran less risk of a depressive disorder, while the risk was noticeably higher among married women. In correspondence to the results of the Berlin Aging Study (Linden et al., 1998), we found increased depression rates among subjects with impaired mobility, vision, or hearing. The extent of the relationship was modified by the variables age and gender, and in the further course also by the other potential risk factors, but remained significantly correlated with a depressive disorder for all functionally impaired groups. With regard to the relationship between physical illnesses and depressive disorders, Osborn et al. (2003) discovered that the subjects with only one severe illness showed no greater risk than those without such an illness. However, those subjects with two or more illnesses ran a 60% greater risk of depression (GDS). Different from our study, this particular study did not consider functional impairments as potential risk factors. In our study depression was also associated with somatic comorbidity (1–4 diagnoses: OR = 1.4; 5+ diagnoses: OR = 2.4). However, even for the multimorbid subjects (5 and more illnesses), the association in the fully adjusted model failed to reach the level of significance. McDougall et al. (2007) have arrived at similar results after controlling for potential risk factors, finding no significant correlation between depression and comorbid disorders. In this study as well, the strongest associations resulted between functional impairments and depression. Our results indicate that impairments of mobility, vision and hearing exhibit a stronger association with depression than do individual somatic illnesses, respectively the number of medical diagnoses. Similar results are reported by the EURODEP Study (Braam et al., 2005), a consortium of 14 research groups in 11 European States, according to which the association of depressive symptoms with functional disability was stronger than that with chronic physical conditions. With regard to the relationship between lifestyle factors (consumption of nicotine and alcohol) and depression, our results supported the findings reported by Osborn et al. (2003): No significant correlations between consumption of alcohol and depressive symp-
toms were found in either study; however, the correlations between current smoking and depressive symptoms were highly significant, even after controlling for potential risk factors. In a cross-sectional study of older adults attending a representative sample of general practitioners in Western Australia, Almeida and Pfaff (2005) also found that current smoking is associated with increased frequency and severity of depression. There is general agreement on the cross-sectional correlation between depressive and cognitive symptoms (Vinkers et al., 2004; Ganguli et al., 2006; Geda et al., 2006). Thus, cognitive impairments frequently occur in the course of a depression, which can take the shape of attention and concentration disorders, slower cognitive processing, and weaknesses in the storage of new information (Reischies and Neu, 2000). The significant correlation between depressive disorders and the “multidomain MCI amnestic” confirms in this context that the concurrent manifestation of depressive and cognitive disorders is characterized particularly by the occurrence of mild impairments in various functional areas of cognition. Still unclear, however, is whether or not depressive symptoms lead to cognitive deterioration over the course of time, respectively, whether or not these represent an independent risk factor for a subsequent dementia disorder. Several studies (e.g., Wilson et al., 2002; Green et al., 2003; Modrego and Ferrández, 2004) have shown an association between depressive symptoms and an increased risk of developing Alzheimer's dementia. Modrego and Ferrández (2004) report that particularly those subjects with mild cognitive impairments (MCI according to Petersen et al., 1999) and depression as compared to non-depressives, run a more than twofold greater risk of developing an Alzheimer dementia within the course of three years. The analysis of the longitudinal data of our study will help us to clarify the direction of the relationship between depression and cognitive impairment. 5. Limitations A number of factors may have influenced the findings and should be considered when interpreting the results: The general practices taking part were not randomly selected. Since only about 50% of the randomly selected patients consented to participate in the study, selection bias might also play a role. The survey excluded institutionalized elderly people, subjects who were unable to attend a primary care physician, and those who suffered from dementia. Since depressive disorders occur more frequently among those in
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
161
residential care and those with functional impairment and dementia, prevalence rates in our study are probably underestimated. Cases were defined according to scores on the Geriatric Depression Scale (GDS-15), not according to clinical diagnostic criteria. However, this widely used instrument has shown good psychometric properties and allows numerous comparisons with epidemiological studies in different countries. Finally, the results are based on a cross-sectional analysis and therefore provide no insight into the direction of effects.
Role of the funding source This study is part of the German Competence Network Dementia and was funded by the German Federal Ministry of Education and Research (BMBF, grant O1 GI 0420). The funding source had no involvement.
6. Conclusions
This study was funded by the German Ministry of Education and Research within the framework of the Competence Network Dementia (BMBF, grant O1 GI 0420). Our thanks go to all participating patients and their general practitioners for their good collaboration:
With the demographic change in the forthcoming decades, more emphasis should be put on large-scale epidemiological studies of the older old since in many countries the increase in this age group will be particularly high. Our study revealed that the prevalence of depression among elderly patients in general practices is high and remains high into old age. Short scales such as the GDS-15 may be useful in helping GPs and practice staff to identify elderly patients with depressive symptoms (D'Ath et al., 1994). In agreement with previous studies, we found that depression is more prevalent among women compared to men. However, after controlling for confounding variables, gender-specific differences are no longer significant. We found high levels of functional disability, mild cognitive impairment, and smoking to be the most important risk factors for depression. The strong association between functional impairment and depression might stem from a reciprocally causal relationship between the two conditions. Either condition might precipitate the other or prolong its duration, thus raising the prevalence rate of their concurrence. Primary care practice offers ample opportunity to treat mental health problems such as depression that occur in relation to physical disability. If functional impairment accounts for higher rates of depression, the modification of this risk factor could well lower rates of depression among the elderly. Functional impairment can be modified by a large array of measures: maintaining physical health, change in personal health habits, exercise, correcting or compensating functional deficits by means of medical and surgical treatments, walking aids, as well as environmental modifications. Intervention strategies that focus on maintaining health and preventing functional impairment may reduce the prevalence and impact of depression. There is an obvious need for intervention studies of this type.
Conflict of interest All authors declare that there are not any actual or potential conflicts of interest.
Acknowledgements
Hamburg: Gundula Bormann, Winfried Bouché, Doris Fischer-Radizi, Michael Funke, Heike Gatermann, Wolfgang Herzog, Petra Hütter, Stefanie Kavka-Ziegenhagen, Günther Klötzl, Bernd-Uwe Krug, Dietrich Lau, Ursula Linn, Andrea Moritz, Karl-Christian Münter, Detlef Niemann, Klaus Richard-Klein, Walter Schreiber, Ursula SchröderHöch, Gerhard Schulze, Klaus Stelter, Carl-Otto Stolzenbach, Ljudmila Titova, Klaus Weidner, OttoPeter Witt, Eckehard Zeigert; Mannheim: Gerhard Arnold, Veit-Harold Bauer, Werner Besnier, Hanna Böttcher-Schmidt, Hartmut Grella, Gernot Kunzendorf, Ingrid Ludwig, Manfred Mayer, Hubert Mühlig, Arnt Müller, Adolf Noky, Helmut Perleberg, Carsten Rieder, Michael Rosen, Georg Scheer, Michael Schilp, Matthias Schneider, Jürgen Wachter, Brigitte Weingärtner, Hans-Georg Willhauck; Bonn: Jörg Eimers-Kleene, Klaus Fischer, Maria Goebel-Schlatholt, Peter Gülle, Wolf-Dietrich Honig, Hans Jürgen Kaschell, Hanna Liese, Manfred Marx, Eberhard Prechtel, Heinz-Peter Romberg, Heribert Schützendorf, Annemarie Straimer, Martin Tschoke, Karl-Michael Werner; Halstenbek: Herrmut Mayen; Königswinter: Theodor Alfen; Bad Honnef: Klaus Weckbecker; Niederkassel: Inge Bürfent; Alfter-Oedekoven: Johann von Aswege; Erfstadt-Liblar: Arndt Uhlenbrock; Windeck-Herchen: Wolf-Rüdiger Weisbach; Leipzig: Martina Amm, Heinz-Michael Assmann, Horst Bauer, Barbara Bräutigam, Jochen Ebert, Angelika Gabriel, Eva Hager, Gunter Kässner, Ina
162
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163
Lipp, Thomas Lipp, Ute Mühlmann, Gabi Müller, Thomas Paschke, Gabriele Rauchmaul, Ina Schmalbruch, Holger Schmidt, Hans-Christian Taut, Ute Voβ, Bettina Winkler, Sabine Ziehbold; München: Eugen Allwein, Guntram Bloβ, Peter Dick, Johann Eiber, Lutz-Ingo Fischer, Peter Friedrich, Helga Herbst, Andreas Hofmann, Günther Holthausen, Karl-Friedrich Holtz, Ulf Kahmann, Elke Kirchner, Hans Georg Kirchner, Luitpold Knauer, Andreas Koeppel, Heinz Koschine, Walter Krebs, Franz Kreuzer, Karl Ludwig Maier, Christoph Mohr, Elmar Schmid, Gabriel Schmidt, Johann Thaller; Haar: Richard Ellersdorfer, Michael Speth;Düsseldorf: Angela Ackermann, Pauline Berger, Florinela Cupsa, Barbara Damanakis, Klaus-Wolfgang Ebeling, Tim Oliver Flettner, Michael Frenkel, Friederike Ganßauge, Kurt Gillhausen, Hans-Christian Heede, Uwe Hellmessen, Benjamin Hodgson, Bernhard Hoff, Helga Hümmerich, Boguslaw-Marian Korman, Dieter Lüttringhaus, Dirk Matzies, Vladimir Miasnikov, Wolfgang Josef Peters, Birgitt Richter-Polynice, Gerhard Erich Richard Schiller, Ulrich Schott, Andre Schumacher, Harald Siegmund, Winfried Thraen, Roland Matthias Unkelbach, Clemens Wirtz.
References Almeida, O.P., Pfaff, J.J., 2005. Depression and smoking amongst older general practice patients. Journal of Affective Disorders 86, 317–321. Braam, A.W., Prince, M.J., Beekman, A.T.F., Delespaul, P., Dewey, M.E., Geerlings, S.W., Kivelä, S.-L., Lawlor, B.A., Magnússon, H., Meller, I., Pérès, K., Reischies, F.M., Roelands, M., Schoevers, R.A., Saz, P., Skoog, I., Turrina, C., Versporten, A., Copeland, J.R.M., 2005. Physical health and depressive symptoms in older Europeans. Results from EURODEP. British Journal of Psychiatry 187, 35–42. Brauns, H., Steinmann, S., 1999. Educational reform in France, WestGermany and the United Kingdom. ZUMA-Nachrichten 44, 7–44. British Medical Association, 1995. Alcohol: guidelines on sensible drinking. British Medical Association, London. Cole, M.G., Dendukuri, N., 2003. Risk factors for depression among elderly community subjects: a systematic review and metaanalysis. American Journal of Psychiatry 160 (6), 1147–1156. D'Ath, P., Katona, P., Mullan, E., Evans, S., Katona, C., 1994. Screening, detection and management of depression in elderly primary care attenders. I: The acceptability and performance of the 15 item Geriatric Depression Scale (GDS 15) and the development of short versions. Family Practitioner 11, 260–266. Ganguli, M., Du, Y., Dodge, H.H., Ratcliff, G.G., Chang, C.C., 2006. Depressive symptoms and cognitive decline in late life. Archives of General Psychiatry 63, 153–160. Gauggel, S., Birkner, B., 1999. Validität und Reliabilität einer deutschen Version der Geriatrischen Depressionsskala (GDS). Zeitschrift für Klinische Psychologie 28, 18–27.
Geda, Y.E., Knopman, D.S., Mrazek, D.A., Jicha, G.A., Smith, G.E., Negash, S., Boeve, B.F., Ivnik, R.J., Petersen, R.C., Pankratz, V.S., Rocca, W.A., 2006. Depression, apolipoprotein E genotype, and the incidence of mild cognitive impairment: a prospective cohort study. Archives of Neurology 63, 435–440. Green, R.C., Cupples, L.A., Kurz, A., Auerbach, S., Go, R., Sadovnick, D., Duara, R., Kukull, W.A., Chui, H., Edeki, T., Griffith, P.A., Friedland, R.P., Bachmann, D., Farrer, L., 2003. Depression as a risk factor for Alzheimer disease. Archives of Neurology 60, 753–759. Jongenelis, K., Pot, A.M., Eisses, A.M.H., Beekman, A.T.F., Kluiter, H., Ribbe, M.W., 2004. Prevalence and risk indicators of depression in elderly nursing home patients: the AGED study. Journal of Affective Disorders 83, 135–142. Jorm, A.F., 1987. Sex and age differences in depression: a quantitative synthesis of published research. Australian and New Zealand Journal of Psychiatry 21, 46–53. Linden, M., Kurtz, G., Baltes, M.M., Geiselmann, B., Reischies, F.M., 1998. Depression bei Hochbetagten. Ergebnisse der Berliner Altersstudie. Nervenarzt 69, 27–37. Luck, T., Riedel Heller, S.G., Kaduszkiewicz, H., Bickel, H., Jessen, F., Pentzek, M., Wiese, B., Koelsch, H., van den Bussche, H., Abholz, H.-H., Moesch, E., Gorfer, S., Angermeyer, M.C., Maier, W., Weyerer, S., 2007a. Mild cognitive impairment in general practice: age-specific prevalence and correlate results from the German Study on Ageing, Cognition and Dementia in primary care patients (AgeCoDe). Dementia and Geriatric Cognitive Disorders 24, 307–316. Luck, T., Zaudig, M., Wiese, B., Riedel-Heller, S.G., 2007b. SIDAM: Alters-und bildungsspezifische Normen des kognitiven Leistungsteils nach der neuen CASMIN-Bildungsklassifikation. Zeitschrift für Gerontopsychologie und-psychiatrie 20 (1), 31–38. Luppa, M., Heinrich, S., Matschinger, H., Sandholzer, H., Angermeyer, M.C., König, H.-H., Riedel-Heller, S.G., 2008. Direct costs associated with depression in old age in Germany. Journal of Affective Disorders 105, 195–204. Lyness, J.M., Caine, E.D., King, D.A., Conwell, Y., Duberstein, P.R., Cox, C., 2002. Depressive disorders and symptoms in older primary care patients: One year outcomes. American Journal of Geriatric Psychiatry 10, 275–282. Margraf, J., 1994. Mini-DIPS. Diagnostisches Kurz-Interview bei psychischen Störungen. Springer, Berlin. McDougall, F.A., Kvaal, K., Matthews, F.E., Paykel, E., Jones, P.B., Dewey, M.E., Braine, C., 2007. Prevalence of depression in older people in England and Wales: the MRC CFA Study. Psychological Medicine 1–9. Modrego, P.J., Ferrández, J., 2004. Depression in patients with mild cognitive impairment increases the risk of developing dementia of Alzheimer Type. Archives of Neurology 61, 1290–1293. Osborn, D.P., Fletcher, A.E., Smeeth, L., Stirling, S., Bulpitt, C.J., Breeze, E., Ng, E.S., Nunes, M., Jones, D., Tulloch, A., 2003. Factors associated with depression in a representative sample of 14217 people aged 75 and over in the United Kingdom: results from the MRC trial of assessment and management of older people in the community. International Journal of Geriatric Psychiatry 18, 623–630. Palsson, S., Skoog, I., 1997. The epidemiology of affective disorders in the elderly: a review. International Clinical Psychopharmacology 12 (suppl. 7), 3–13. Petersen, R.C., Smith, G.E., Waring, S.C., Ivnik, R.J., Tangalos, E.G., Kokmen, E., 1999. Mild cognitive impairment: clinical characterization and outcome. Archives of Neurology 56, 303–308.
S. Weyerer et al. / Journal of Affective Disorders 111 (2008) 153–163 Prince, M.J., Harwood, R.H., Blizard, R.A., Thomas, A., Mann, A.H., 1997. Social support deficits, loneliness and life events as risk factors for depression in old age. The Gospel Oak Project VI. Psychological Medicine 27, 323–332. Reischies, F.M., Neu, P., 2000. Comorbidity of mild cognitive disorder and depression — a neuropsychological analysis. European Archives of Psychiatry and Clinical Neuroscience 250, 186–193. Riedel-Heller, S.G., Busse, A., Angermeyer, M.C., 2006. The state of mental health in old-age across the ’old’ European Union — a systematic review. Acta Psychiatrica Scandinavica 113, 388–401. Sheikh, J.I., Yesavage, J.A., 1986. Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. Clinical Gerontologist 5, 165–173. Speer, D.C., Schneider, M.G., 2003. Mental health needs of older adults and primary care: Opportunity for interdisciplinary geriatric team practice. Clinical Psychology: Science and Practice 21, 85–101. Vinkers, D.J., Gussekloo, J., Stek, M.L., Westendorp, R.G.J., van der Mast, R.C., 2004. Temporal relation between depression and cognitive impairment in old age: prospective population based study. British Medical Journal 329, 1–4.
163
Wancata, J., Alexandrowicz, R., Marquart, B., Weiss, M., Friedrich, F., 2006. The criterion validity of the Geriatric Depression Scale: a systematic review. Acta Psychiatrica Scandinavica 114, 398–410. Wilson, R.S., Barnes, L.L., Mendes de Leon, C.F., Aggarwal, N.T., Schneider, J.S., Bach, J., Pilat, J., Beckett, L.A., Arnold, S.E., Evans, D.A., Bennett, D.A., 2002. Depressive symptoms, cognitive decline, and risk of AD in older persons. Neurology 59, 364–370. Winblad, B., Palmer, K., Kivipelto, M., Jelic, V., Fratiglioni, L., Wahlund, L.-O., Nordberg, A., Bäckman, L., Albert, M., Almkvist, O., Arai, H., Basun, H., Blennow, K., De Leon, M., Decarli, C., Erkinjuntti, T., Giacobini, E., Graff, C., Hardy, J., Jack, C., Jorm, A., Ritchie, K., Van Duijn, C., Visser, P., Petersen, R.C., 2004. Mild cognitive impairment — beyond controversies, towards a consensus: report of the International Working Group on Mild Cognitive Impairment. Journal of Internal Medicine 256, 240–246. Zaudig, M., Hiller, W., 1996. SIDAM-Handbuch. Strukturiertes Interview für die Diagnose einer Demenz vom Alzheimer Typ, der MultiInfarkt-(oder vaskulären) Demenzen und Demenzen anderer Ätiologien nach DSM-III-R, DSM-IV und ICD-10. Huber, Bern.