Hintergrund: Eine deutsche Demenzstudie (Ageing, Cognition and Dementia in Primary Care Patients, AgeCoDe) weist darauf hin, dass eine leichte Demenz bei Patienten/innen ohne Partner/in seltener von Hausärzten/innen erkannt wird. Woran das liegt, ist nicht bekannt. Im vorliegenden Artikel werden zwei Fragen bearbeitet: 1) Ist es wirklich speziell der Partnerstatus oder ist es eher die gesamte soziale Einbettung eines/r Patienten/in (Partner/in + Familie/Freunde + soziale Aktivitäten), die mit der hausärztlichen Demenzerkennung zusammenhängt? 2) Zeigen Patienten/innen je nach Partnerstatus und sozialer Einbettung Unterschiede in ihrer kognitiven Leistung (als möglicher Grund für die unterschiedliche Demenzerkennung)?
Methoden: Die Daten einer Teilstichprobe der prospektiven Kohortenstudie AgeCoDe wurden querschnittlich exploriert. Hausarztpatienten/innen zwischen 75 und 89 Jahren wurden mittels strukturierten Interviews inkl. kognitiver Testung in häuslicher Umgebung von Wissenschaftlern/innen untersucht und mit dem Social Integration Index (SII) zu Partnerstatus, Kontakt zu Freunden/Angehörigen und sozialen Aktivitäten befragt (soziale Einbettung). Diagnostische Referenz bildete die klinische Demenzdiagnose auf Basis der Interviewergebnisse und der Konsensussitzungen mit Interviewern/innen und Experten/innen. Unabhängig von den Interviews füllten die Hausärzte/innen der teilnehmenden Patienten/innen je einen Fragebogen pro Patient/in aus, in dem u.a. nach der Einschätzung des kognitiven Status und einer Demenz gefragt wurde (Global Deterioration Scale). Die Übereinstimmung zwischen Demenzdiagnose und hausärztlicher Einschätzung ergab die Werte für Sensitivität und Spezifität des Hausarzturteils hinsichtlich der Demenzerkennung. Zu Frage 1) wurden zwei Mehrebenenanalysen mit dem Kriterium übersehener bzw. überschätzter Demenz und den Prädiktoren sozialer Einbettung/Partnerstatus, Geschlecht, Bildung, globale kognitive Leistung durchgeführt. Zu Frage 2) wurden Patientengruppen in unterschiedlicher sozialer Einbettung mit bivariaten Gruppenvergleichen hinsichtlich mehrerer kognitiver Teilleistungen miteinander verglichen.
Ergebnisse: Eine Stichprobe von 91 Patienten/innen mit inzidenter Demenz und eine Stichprobe von 743 Patienten/innen ohne Demenz wurden analysiert. 1) Der Partnerstatus, nicht aber die generelle soziale Einbettung, ist unabhängiger Prädiktor der hausärztlichen Demenzerkennung. 2) Patienten/innen mit Demenz in hoher sozialer Einbettung, aber ohne Partner/in zeigen schlechtere Leistungen in sprachlich-intellektuellen Fähigkeiten. In der Stichprobe ohne Demenz zeigen diejenigen in guter sozialer Einbettung und mit Partner/in die geringste Depressivität, Patienten/innen mit Partner/in geringere Alltagsfunktionalität.
Schlussfolgerungen: Es ist speziell der Partnerstatus, nicht die gesamte soziale Einbettung, der die Demenzerkennung beeinflusst. Des Weiteren sprechen die Ergebnisse gegen ein kognitiv „intakteres“ Bild von Patienten/innen ohne Partner/in als Erklärung für die geringere hausärztliche Demenzerkennung. Aus diesen beiden Ergebnissen lässt sich die Hypothese ableiten, dass dem Hausarzt/der Hausärztin bei Patienten/innen ohne Partner/in eventuell diagnostisch relevante fremdanamnestische Angaben vom Partner/von der Partnerin fehlen, wodurch die Demenzerkennung erschwert wird. Die Berücksichtigung des Partnerstatus als Einflussfaktor hausärztlicher Demenzerkennung sowie sprachlich-intellektueller Fähigkeiten als relevanter kognitiver Domäne könnten helfen, in zukünftiger Forschung sowohl das hausärztliche Urteil als auch diagnostische Instrumente zu verbessern.
Despite the benefits for physicians, patients and relatives  of early recognizing dementia, mild dementia is frequently overlooked in family medicine . Current research priorities propose a more effective role played by family practitioners (FPs) in the diagnostic process . Still, dementia recognition can be difficult due to disease characteristics (e.g. insidious onset), patients’ intentions (e.g. denial, masking symptoms), or FPs’ attitudinal barriers (e.g. nihilism regarding therapy and support) . Furthermore, certain patient characteristics have been associated with a lower level of dementia recognition by FPs: in a previous analysis of the German Study on Ageing, Cognition and Dementia in Primary Care Patients (AgeCoDe), we found that having no partner was the only independent risk factor for missing incident dementia . The reasons remain unclear, but primarily two assumptions arise: is it the lack of diagnostic information from a partner that makes it difficult for FPs to recognize dementia in patients without partner? Or do these dementia patients have a unique clinical presentation which could explain why FPs have difficulties identifying their dementia? The first assumption is supported by qualitative findings from FP interview studies . The second assumption relates to reported associations between cognitive performance and social integration . Considering both, we address the following research questions in this article: 1) Is it the patient’s partner status in particular that impedes dementia recognition, or does partner status merely act as pars pro toto of the overall social integration construct ? 2) Does the cognitive performance of elderly FP patients vary in relation to their level of social integration?
This paper presents a cross-sectional explorative analysis of elderly German family medicine patients, performed in a subsample of the prospective AgeCoDe cohort study (which has been described in various reports, e.g. ).
Participants were recruited from FP practices in two German study centers. Inclusion criteria were age 75 to 89 and at least one contact with an FP within the past 12 months. Individuals were excluded in case of dementia at baseline, home visit-only consultations, residence in a nursing home, any potentially fatal illness, insufficient German language proficiency, deafness or blindness, inability to provide consent, and not being a regular patient of the participating practice.
Structured interviews were conducted by psychologists and trained physicians during visits to participants’ homes at all assessments. Follow-up assessments (FU) were scheduled at 18-month intervals; data from four assessments were included in the presented analyses (baseline, FU1, FU2, FU3). Interviews included psychometric testing, clinical anamnesis, and sociodemographic data. All assessments and their function in analyses are detailed in table 1. The social integration index SII  was assessed at baseline in all patients of the Leipzig and Düsseldorf study centers. Clinical dementia diagnosis was based on the SIDAM (Structured Interview for Diagnosis of Dementia of Alzheimer type) : international criteria for dementia are comprised in a diagnostic algorithm by assessing cognitive performance and activities of daily living, including an informant interview in suspected dementia cases. All diagnoses were discussed in consensus conferences with the interviewers and experienced geriatric psychiatrists or geriatricians. Independent from patient interviews by the researchers, FPs filled in questionnaires about each of their participating patients regarding a variety of clinical judgements and information.
An ‘incident dementia sample’ (dementia first diagnosed in FU1 or FU2 or FU3) and a ‘non-dementia sample’ (results from FU2 of patients with no dementia up to FU3) were analyzed separately.
- 1. Dementia recognition: Partner status was analyzed as a separate predictor and as part of social integration (interaction term SII x partner status). Independent predictors of FPs’ sensitivity and specificity (see table 1) were analyzed by multivariate generalized estimating equation models (binary logistic, logit link), adjusting for a cluster effect of FP practices. The number of predictors was restricted to six .
- 2. Cognitive performance: In the dementia sample, cognitive performance at the onset of incident dementia and cognitive decline were analyzed. The latter was operationalized as difference between test scores at dementia onset and test scores at the assessment prior to dementia (18 months before dementia diagnosis). In the non-dementia sample, cognitive performance in FU2 was analyzed. Using non-parametric bivariate Kruskal-Wallis tests, we compared four groups of ‘social integration x partner’ combinations with regard to the independent variables listed in table 1. The groups were: A) ‘low social integration, no partner’; B) ‘low social integration, with partner’; C) ‘high social integration, no partner’; D) ‘high social integration, with partner’. The sample-size independent effect size measure Cohen’s d was derived from Kruskal-Wallis H (for overall effects) and Mann-Whitney U (for pairwise effects) according to published procedures . We focused on robust effect sizes with d ? 0.50 .
The dementia sample consisted of 91 patients with incident dementia (29 FU1, 27 FU2, 35 FU3), the non-dementia sample consisted of 743 patients from follow-up 2. Participating physicians were 65 FPs from Leipzig and Düsseldorf, working in separate practices (38 male, 27 female; mean age 48.75 (SD 6.90); mean years in practice 12.52 (SD 6.36)).
Table 2 shows that patient groups with partner were characterized by younger age, more men and higher education level.
Multivariate analyses in the two samples were performed with 58 and 709 patients with complete data (see eTable 1). The required number of clusters  was achieved in both samples.
Figure 1 shows FPs’ sensitivity for incident dementia, with an obvious main effect of partner status.
Analysis 1 (table 3) reveals partner status (as a separate variable, not as part of social integration) as the only significant independent predictor of FP dementia recognition (worse sensitivity in patients without partner). Global cognition was the only significant independent predictor of FPs correctly ruling out dementia (better specificity with higher test scores).
Including additional predictors (not shown) did not modify these patterns of results.
Cognitive performance patterns
Within the dementia sample, the only significant and most robust overall effect was a clear difference in the SIDAM subscale ‘verbal intellectual abilities’ at the time of dementia onset (Cohen’s d = 0.61; p = .041): dementia patients with high social integration but without partner showed considerably worse verbal intellectual performance than the other three groups (strong post-hoc effects with d ? 0.80). This and some other moderate effects (incl. worse orientation and steeper pre-dementia cognitive decline) identified this patient group as being unique in its kind of cognitive presentation. No relevant overall effect sizes (d 0.50) existed for global cognition, memory, depressiveness, and subjective memory impairment.
The profiles in the non-dementia sample showed two clear differences regarding depressiveness and IADL (Instrumental Activities of Daily Living) (d = 0.53 and d = 1.10, respectively, both p .001): least depressiveness in patients with high social integration with partner, and less IADL functioning in both patient groups with partner compared to both patient groups without partner. None of the cognitive parameters differed substantially between groups (d 0.50).
Details on group comparisons are shown in the eTables 2 and 3 (see www.online-zfa.de/).
Taken together, our results support the hypothesis that a partner living with the patient is a major case-finding resource for FPs and that living alone is a risk factor for missed dementia. This corresponds to the relevance FPs assign to relatives in the diagnostic process .
Incident dementia was less likely to be recognized by FPs in patients without a partner, irrespective of their overall social integration. FPs’ ability to rule out dementia was not related to a patient’s level of social integration and partner status.
We did not find that cognitive performance was better in dementia patients without partner. Therefore, a well-preserved appearance of cognitive competence becomes less plausible as a reason for poor sensitivity. Quite the contrary, dementia patients who are socially integrated but live without a partner showed a unique cognitive profile with poorer verbal intellectual abilities and a steeper pre-dementia cognitive decline.
We found no differences in the awareness of deficits (subjective memory impairment, SMI) between patients with a partner and those without. It is possible, however, that the deficits in dementia patients who lived without a partner were less likely to be reported to the FP, thus making recognition more difficult. We did not collect data on this. Levels of depressiveness did not vary between groups.
In accordance with previous research, patients who had a high level of social integration in our non-dementia sample showed the lowest level of depressiveness . Superior everyday functioning (IADL) in patients without partner stood in contrast to earlier studies (e.g. ). This might be due to the higher percentage of women among patients without partner as well as the exclusively urban patient sample of the AgeCoDe study (meaning better access to social contacts and activities than in rural communities, constituting and preserving IADL in people living alone).
There are some limitations to our study: the social integration index was only assessed at baseline, and changes could not be taken into account. On the other hand, social networks in old age have been reported to be stable, though not in all aspects . The social integration index is a rather rough measure of a structural social support component. More differentiated aspects of social support may also be of interest in this context . Confounders of social support and its effects on cognition have not been assessed .
Among the strengths of the study are the focus on incident dementia cases and a valid and reliable reference of dementia diagnosis. We applied robust statistical methods that consider the cluster effects of FP practice and allow for a careful interpretation of explorative analyses. We did not reduce cognition to global mental status, but analyzed various cognitive domains.
Dementia recognition should be improved by focusing on patients without partner, due to their special needs and increased risks (e.g. malnutrition, social isolation) . Further investigations are required to understand why poor verbal intellectual abilities and a steeper cognitive decline constitute the cognitive profile of dementia patients with high social integration without partner.
Despite their obvious diagnostic relevance, informant reports and scales  should be evaluated not only for validity, but also in terms of patient-centered handling. The development of valid questions concerning subjective cognitive complaints, specifically for patients living alone, may enhance the diagnostic relevance of patient reports in dementia recognition. Furthermore, current dementia tests have a strong focus on orientation and types of memory . It would be interesting to see if items on verbal intellectual abilities can improve sensitivity in patients without partner.
This study is part of the German Research Network on Dementia (Kompetenznetz Demenzen, KND) and the German Research Network on Degenerative Dementia (Kompetenznetz Degenerative Demenzen, KNDD) and was funded by the German Federal Ministry of Education and Research (grants KND: 01GI0102, 01GI0422, 01GI0423, 01GI0429, 01GI0431, 01GI0434; grants KNDD: 01GI0710, 01GI0711, 01GI0713, 01GI0714, 01GI0716). This study is moreover published in affiliation with the Health Service Research Initiative (study on needs, health service use, costs and health-related quality of life in a large sample of oldest-old primary care patients (85+; AgeQualiDe)) that was also funded by the German Federal Ministry of Education and Research (grants: 01GY1322A, 01GY1322B, 01GY1322C, 01GY1322F, 01GY1322G).
Ethical approval was obtained from the following study centers: Ethics Commission of the Medical Faculty of the Heinrich Heine University Düsseldorf 2079/2002, 2999/2008; Ethics Commission of the Medical Association Hamburg OB/08/02, 2817/2007; Ethics Commission of the University of Bonn 050/02, 258/07; Ethics Commission at the Medical Center of the University of Leipzig 143/2002, 309/2007. All participants gave written informed consent before taking part.
Acknowledgements: We would like to thank all participating patients as well as their family practitioners for their good collaboration.
Conflicts of interest: None of the authors report relevant conflicts of interest.
Supplementary material available at: www.online-zfa.de
Dr. rer. nat. Michael Pentzek (Dipl.-Psych.)
Institute of General Practice (ifam)/<br/>Centre for Health and Society (chs)
Medical Faculty<br/>Heinrich Heine University Düsseldorf
Werdener Straße 4
40227 Düsseldorf, Germany
Phone: +49 0211 81 16 818
1. National Health Service (NHS). Benefits of early dementia diagnosis. NHS Dementia Guide. 2015. www.nhs.uk/Conditions/dementia-guide/Pages/dementia-early-diagnosis-benefits.aspx (retrieved February 16, 2017)
2. van den Dungen P, van Marwijk HW, van der Horst HE, et al. The accuracy of family physicians’ dementia diagnoses at different stages of dementia: a systematic review. Int J Geriatr Psychiatry 2012; 27: 3423–354
3. Kelly S, Lafortune L, Hart N, Cowan K, Fenton M, Brayne C. Dementia priority setting partnership with the James Lind Alliance. Age Ageing 2015; 44: 985–993
4. Koch T, Iliffe S. Rapid appraisal of barriers to the diagnosis and management of patients with dementia in primary care: a systematic review. BMC Fam Pract 2010; 11: 1
5. Pentzek M, Wollny A, Wiese B, et al. Apart from nihilism and stigma: what influences general practitioners’ accuracy in identifying incident dementia? Am J Geriatr Psychiatry 2009; 17: 965–975
6. van Hout HPJ, Vernooij-Dassen MJ, Stalman WA. Diagnosing dementia with confidence by GPs. Fam Pract 2007; 24: 616–621
7. Litwin H, Stoeckel KJ. Social network, activity participation, and cognition – a complex relationship. Res Aging 2016; 38: 76–97
8. Berkman LF, Glass T. Social integration, social networks, social support, and health. In: Berkman LF, Kawachi I (eds.). Social epidemiology. Oxford: Oxford University Press, 2000: 137–173
9. Luck T, Riedel-Heller SG, Luppa M, et al. Risk factors for incident mild cognitive impairment. Acta Psychiatr Scand 2010; 121: 260–272
10. Berkman LF, Melchior M, Chastang JF, Niedhammer I, Leclerc A, Goldberg M. Social integration and mortality: a prospective study of French employees of Electricity of France. Am J Epidemiol 2004; 159: 167–174
11. Zaudig M, Hiller W. SIDAM-Handbuch. Bern: Huber, 2000
12. Brauns H, Steinmann S. Educational reform in France, West Germany and the United Kingdom. ZUMA Nachrichten 1999; 44: 7–44
13. Folstein MF, Folstein SE, McHugh PR. Mini Mental State: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12: 189–198
14. Morris JC, Heyman A, Mohs RC, et al. The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part I. Clinical and neuropsychological assessment of Alzheimer’s disease. Neurology 1989; 39: 1159–1165
15. Sheikh JI, Yesavage JA. Geriatric Depression Scale: recent evidence and development of a shorter version. Clin Gerontol 1986; 5: 165–173
16. Geerlings MI, Jonker C, Bouter LM, Adèr HJ, Schmand B. Association between memory complaints and incident Alzheimer’s disease in elderly people with normal baseline cognition. Am J Psychiatry 1999; 156: 531–537
17. Lawton MP, Brody EM. Assessment of older people: self-maintaining and instrumental activities of daily living. Gerontologist 1969; 9: 179–186
18. Reisberg B, Ferris SH, de Leon MJ, Crook T. The Global Deterioration Scale for assessment of primary degenerative dementia. Am J Psychiatry 1982; 139: 1136–1139
19. Harris RJ. A primer of multivariate statistics. New York: Academic Press, 1985
20. Fritz CO, Morris PE, Richler JJ. Effect size estimates: current use, calculations, and interpretation. J Exp Psychol Gen 2012; 141: 2–18
21. Cohen J. Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum, 1988
22. Ghisletta P, Spini D. An introduction to generalized estimating equations and an application to assess selectivity effects in a longitudinal study on very old individuals. J Educ Behav Stat 2004; 29: 421–437
23. Schwarzbach M, Luppa M, Forstmeier S, König HH, Riedel-Heller SG. Social relations and depression in late life – a systematic review. Int J Geriatr Psychiatry 2014; 29: 1–21
24. Liu H, Zhang Z. Disability trends by marital status among older Americans, 1997–2010. Popul Res Policy Rev 2013; 32: 103–127
25. Wrzus C, Hänel M, Wagner J, Neyer FJ. Social network changes and life events across the life span: A meta-analysis. Psychol Bull 2013; 139: 53
26. Gottlieb BH, Bergen AE. Social support concepts and measures. J Psychosom Res 2010; 69: 511–520
27. Segel-Karpas D, Lachman ME. Social contact and cognitive functioning: the role of personality. J Gerontol B Psychol Sci Soc Sci 2016; gbw079
28. Soto M, Andrieu S, Gares V, et al. Living alone with Alzheimer’s disease and the risk of adverse outcomes. J Am Geriatr Soc 2015; 63: 651–658
29. Ready RE, Ott BR, Grace J. Validity of informant reports about AD and MCI patients’ memory. Alzheimer Dis Assoc Disord 2004; 18: 11–16
30. Tsoi KK, Chan JY, Hirai HW, Wong SY, Kwok TC. Cognitive tests to detect dementia: a systematic review and meta-analysis. JAMA Int Med 2015; 175: 1450–1458
eTable 1 Characteristics of patients in generalized estimating equation analysis samples (with complete data on dependent and independent variables)
eTable 2 Cognitive profile by social integration/partner status in patients with incident dementia (detailed results of bivariate group comparisons)
eTable 3 Cognitive profile by social integration/partner status in patients without dementia (detailed results of bivariate group comparisons)
Table 1 Assessment procedures
Table 2 Characteristics of dementia and non-dementia samples according to social integration (SII) and partner status
Table 3 Factors of FPs’ sensitivity and specificity for incident dementia recognition and rule-out (generalized estimating equation models, adjusted for cluster effect of FP practice)
Figure 1 Percentage of FPs’ correct positive judgements among incident dementia cases
1 Institute of General Practice (ifam), Medical Faculty, Heinrich Heine University Düsseldorf, Germany 2 Institute for General Practice, Hannover Medical School, Germany 3 Department of Psychiatry, University of Bonn, Germany 4 German Center for Neurodegenerative Diseases (DZNE) Bonn, Germany 5 Department of Primary Medical Care, University Medical Center Hamburg-Eppendorf, Germany 6 Institute of Social Medicine and Occupational Health, University of Leipzig, Germany * shared senior authorship Peer reviewed article eingereicht: 07.07.2016, akzeptiert: 26.10.2016 DOI 10.3238/zfa.2017.0305–0311