Aging Clinical and Experimental Research

, Volume 26, Issue 5, pp 555–559 | Cite as

Prevalence of dementia in nursing home and community-dwelling older adults in Germany

  • Falk Hoffmann
  • Hanna Kaduszkiewicz
  • Gerd Glaeske
  • Hendrik van den Bussche
  • Daniela Koller
Short Communication



We compared the prevalence of dementia in nursing home residents and community-dwelling older adults .


Using health insurance claims data for the year 2009, we estimated the prevalence of at least three of four quarters with a diagnosis of dementia in persons aged ≥65 years.


Of 213,694 persons aged 65+ years, 4,584 (2.2 %) lived in nursing homes. The prevalence of dementia was 51.8 % (95 % CI 50.4–53.3) in nursing home residents and 2.7 % (95 % CI 2.6–2.8) in community-dwelling elderly. Increasing prevalences with age were found in both sexes in community-dwelling elderly. These trends were not seen in nursing home residents where prevalences were already high for the age group 65–69 years (35.7 % in males and 40.9 % in females, respectively).


More than half of nursing home residents suffer from dementia, which is about 19-fold higher than the prevalence in insured living in the community.


Dementia Epidemiology Health services research Care dependency Germany 


Dementia is a major public health problem in the elderly today, and its impact will grow in the future. A recently published review estimated that already 35.6 million people worldwide suffered from dementia in 2010, with numbers expected to almost double every 20 years [1]. The authors concluded that nationally representative studies are scarce but critically important for health politics, long-term care planning and allocation of health resources.

Population ageing also increases the number of dementia patients cared for in institutions. These persons are likely to be underrepresented in most epidemiological studies [2]. In Germany, there are several larger studies on dementia of nursing home residents [3, 4, 5], but they do not enable comparisons with community-dwelling persons. Although there are few studies on the prevalence of dementia that include both persons living in nursing homes and community-dwelling older adults [6, 7, 8, 9], comparisons are not possible due to unavailable data [9] or samples size are rather small [6, 7, 8]. The largest study yet was the LEILA75+ which recruited 1,241 elderly aged 75 years and older living at home and 185 institutionalised persons in 1997/1998 [6]. However, comparisons of the disease prevalence stratified by sex and age are hampered by small sample sizes in nursing home residents.

The aim of this study was to compare the prevalence of dementia in nursing home residents and community-dwelling older adults.


Design and study population

We conducted a cross-sectional study using claims data of the Gmünder ErsatzKasse (GEK) for the year 2009. The GEK is a statutory health insurance company which insured 1.8 million people located in all regions of Germany in the corresponding year. In Germany, health insurance is mandatory for the entire population. About 200 statutory health insurance companies covered a total of 70 million persons (88 % of the German population) in 2009, while the remainder are privately insured. Although there are several differences between the statutory and private system, both provide full-coverage health insurance. Based on the solidarity principle, the statutory health insurance system guarantees identical benefits regardless of income, age or morbidity, and benefits covered are the same for all funds. For historical reasons, single statutory health insurance funds covered specific groups of persons (e.g. blue-collar workers in the GEK). However, since 1996, almost all companies are open to everybody [10, 11].

We included all persons aged 65 years and older who were insured at least 1 day in every quarter of 2009. Due to this criterion, the vast majority of our population at risk is continuously insured throughout this period, but also persons that died in the last quarter of the year could be included [12].

Patients with dementia had to have at least one ICD-10 code for dementia from the following list in at least three of the four quarters in 2009 in ambulatory medical care: F00.x, F01.x, F02.0, F02.3, F03, G30.x, G31.0, G31.1, G31.82, G31.9 and R54. These criteria were also applied in other studies analysing claims data of German statutory health insurance funds [12, 13].

Data of the German long-term care insurance were used to gather information on whether a person lives in nursing homes and on care dependency. Services from the long-term care insurance are provided to those who require support in their activities of daily living including personal hygiene, eating, mobility and—separately from personal care—housekeeping. There are three levels of care dependency corresponding to the estimated time required for assistance indicating moderate (level I), severe (level II) and severest care dependency (level III) [14, 15]. If care dependency changed within the corresponding year, we included the highest level. Persons that received benefits indicating institutional care in 2009 were considered as nursing home residents.

Statistical analysis

The prevalence of dementia was estimated according to place of residents (nursing home vs. community dwelling), sex, age and region of residence (east vs. west). We defined 5-year age groups starting with 65–69, and the highest group was ≥95 years. Descriptive measures and exact 95 % confidence intervals (95 % CI) for proportions were computed.

We performed all statistical analyses with SAS for Windows version 9.2 (SAS Institute Inc., Cary, NC).


A total of 213,694 persons aged 65 years and older were insured at least 1 day in all four quarters of 2009. Of them, 4,584 (2.2 %) lived in nursing homes. Baseline characteristics according to place of residence are shown in Table 1. Nursing home residents were on average older, more often female and had a higher degree of care dependency than community-dwelling elderly.
Table 1

Baseline characteristics of our study cohort of persons aged 65+ according to place of residence

Baseline characteristics

Nursing home residents

(n = 4,584)

Community dwelling

(n = 209,110)

Mean age, in years (SD)

82.0 (7.8)

72.4 (5.9)

Age groups (in years)


7.7 %

38.6 %


11.7 %

31.4 %


15.2 %

16.3 %


25.2 %

9.4 %


24.2 %

3.7 %


11.0 %

0.6 %


5.0 %

0.1 %



37.9 %

57.0 %


62.1 %

43.0 %

Level of long-term care


0.0 %a

94.8 %


34.0 %

3.2 %


42.9 %

1.6 %


23.0 %

0.5 %

Region of residence


7.4 %

13.6 %


92.6 %

86.4 %

a All residents of nursing homes are assigned to a level of care

Overall, the prevalence of dementia was about 19-fold higher in nursing home residents compared with community-dwelling elderly (51.8 vs. 2.7 %). These differences were found in both sexes, but males have a slightly lower prevalence (Table 2).
Table 2

Prevalence of dementia stratified by place of residence, age and sex


Nursing home residents (n = 4,584)

[95 % CI]

Community dwelling (n = 209,110)

[95 % CI]








Age groups (in years)


35.7 % [29.4–42.4]

40.9 % [32.3–50.0]

37.6 % [32.5–42.9]

0.8 % [0.8–0.9]

0.7 % [0.6–0.8]

0.8 % [0.7–0.8]


46.4 % [40.7–52.2]

44.0 % [37.5–50.6]

45.4 % [41.1–49.7]

1.8 % [1.7–2.0]

1.7 % [1.6–1.9]

1.8 % [1.7–1.9]


51.8 % [46.0–57.6]

49.2 % [44.2–54.3]

50.4 % [46.6–54.1]

4.0 % [3.7–4.3]

3.8 % [3.4–4.1]

3.9 % [3.7–4.1]


48.2 % [43.5–53.0]

57.0 % [53.3-60.7]

53.6 % [50.7–56.5]

7.1 % [6.6–7.6]

7.2 % [6.7–7.8]

7.2 % [6.8–7.6]


57.0 % [51.2–62.6]

57.1 % [53.6–60.5]

57.1 % [54.1–60.0]

10.6 % [9.6–11.7]

11.4 % [10.5–12.4]

11.1 % [10.4–11.8]


54.8 % [45.2–64.1]

53.7 % [48.6–58.7]

54.0 % [49.5–58.4]

14.6 % [11.6–18.1]

19.2 % [16.3–22.2]

17.3 % [15.2–19.6]


43.6 % [27.8–60.4]

56.6 % [49.2–63.8]

54.4 % [47.7–61.0]

17.4 % [10.1–27.1]

27.9 % [22.0–34.4]

24.9 % [20.1–30.2]

Region of residence



55.5 % [47.3–63.5]

61.8 % [54.4–68.8]

58.9 % [53.5–64.2]

2.2 % [2.0–2.5]

2.2 % [2.0–2.5]

2.2 % [2.0–2.4]


48.1 % [45.6–50.6]

53.1 % [51.2–55.0]

51.3 % [49.7–52.8]

2.6 % [2.5–2.7]

3.0 % [2.9–3.2]

2.8 % [2.7–2.9]


48.8 % [46.4–51.1]

53.7 % [51.9–55.6]

51.8 % [50.4–53.3]

2.6 % [2.5–2.6]

2.9 % [2.8–3.0]

2.7 % [2.6–2.8]

Increasing prevalences with age were found in both sexes in community-dwelling elderly. In the age groups up to 85–89 years the prevalence doubles about every 5 years, this strong effect lessens in older persons. These trends were not seen in nursing home residents where prevalence was already 35.7 % in males and 40.9 % in females aged 65–69 years. With 57 %, the prevalence was highest in both sexes in the age group 85–89 years and no further increase is found thereafter. Differences between men and women are visible, especially after the age of 70 and in persons living in the community. However, as the wide confidence intervals indicate, the prevalence for higher age groups is less precise because of the smaller number of people in these groups.

Higher prevalences in the western part of Germany are found in community-dwelling elderly, while in nursing home residents, the trend is reverse to that.


Findings, comparison with other studies and interpretation

To our knowledge, this is the largest German study comparing the prevalence of dementia according to place of residence. We found that nursing home residents were 19-fold more likely to suffer from dementia than those living in the community. Our estimate of 51.8 % dementia patients in nursing home residents is quite the same as published in the LEILA75+ study (47.6 %) conducted in 1997/1998 [6] and a more recent survey from 2008/2009 (53.0 %) [3]. The finding that prevalences did not increase in higher age groups is not supported by the LEILA75+ study [6]. Other German studies did not report age-specific prevalences [3, 4, 5]. However, one has to keep in mind that the highest age group in LEILA75+ was 85+ years [6]. When reported, most studies in nursing home residents found that males were more often affected than females [5, 6].

In field studies, comparisons of prevalences between persons living in nursing homes and community-dwelling elderly are hampered by small sample sizes and the inclusion of different age groups [6, 7, 8]. The BASE cohort consists of persons aged 70 and older [8], the LEILA75+ included only those aged 75+ years [6], and Fichter et al. [7] drew a sample of persons aged 85 years and older. When reported, nursing home residents are much more affected by dementia than those living in the community [6, 8]. This result was also found in our data.

Four German studies reported the prevalence of dementia in the general population including both persons living in nursing homes and community-dwelling older adults [6, 7, 8, 9]. The published health insurance analysis included data as of 2002 [9]. All field studies used also older data [6, 7, 8], and the most actual one (LEILA75+) was conducted in 1997/1998 [6]. Although using a different inclusion criteria, the analysis of a sample of 2.3 million German insurants found quite comparable prevalences in 2002 to our estimates for 2009 [9]. Trends found in the field studies for the general population [6, 7, 8] were also visible in our community-dwelling elderly, even though the prevalences in older age groups are comparably smaller in our cohort.

Regional variations with higher prevalence in the Eastern part of Germany in males and females above the age 85 were first described by Ziegler and Doblhammer [9]. We also found higher prevalences in the East in nursing home residents and in older age groups (data not shown). Unfortunately, due to small sample sizes with only 341 nursing home residents living in East Germany, no further regional analyses according to sex, age groups or federal states were possible.

Strengths and limitations

Using claims data for research has many advantages but also disadvantages. On the one hand, selection bias and non-response are reduced, as claims data allow us to include persons irrespective of their current health status, cognitive impairments, old age, frailty or institutionalisation. These problems often arise in field studies on the prevalence of dementia [2]. Alternative data sources such as medical records or claims data can overcome these drawbacks. On the other hand, however, field studies can apply psychometric tests, physical and psychological examinations enabling the researchers to distinguish between mild cognitive impairment or mild, moderate and severe dementia or actively screen for this disease, which is not possible when using claims data. Therefore, data validity is the major concern in claims data research. We used a conservative algorithm for validity reasons. Of course, this has an influence on our results. When we changed the definition from three to only one quarter of diagnosis, considerably higher prevalences would be the result (65.1 % in nursing homes and 4.4 % in community-dwelling elderly). Also, since awareness is a prerequisite for a diagnosis, patients with mild dementia are more likely to be unidentified in claims data. This problem of underdiagnosis might be greater in community-dwelling adults because nursing home residents are more regularly seen by health professionals.

Furthermore, data of just one German health insurance fund were used, and differences between these funds exist, for instance regarding age, sex, socio-economic status and morbidity [10, 11]. The GEK population consists of a higher proportion of males, which can also be seen in nursing home residents when compared to samples of field studies [3, 4, 5, 6]. Therefore, extrapolations of analyses of single funds to the whole German population seem not appropriate, and this was one reason not to standardise our results. Another was that the German census conducted in 2011 showed an overestimation of the population. However, no data on persons aged 75 years and older have been published yet. Finally, it would not be appropriate to equalise the age distribution between nursing home residents and those living in the community.


Our findings show that more than half of nursing home residents suffer from dementia, which is about 19-fold higher to those living in the community. Further studies should focus on whether prevalences also increase in higher age groups in nursing homes and on regional differences.



We thank the Gmünder ErsatzKasse (GEK, now BARMER GEK) for providing the data. This study was supported by grants from the Jackstädt-Stiftung. The funder had no role in the study design, or in the collection, analysis, interpretation or presentation of the information.

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.


  1. 1.
    Prince M, Bryce R, Albanese E, Wimo A, Ribeiro W, Ferri CP (2013) The global prevalence of dementia: a systematic review and metaanalysis. Alzheimer’s Dement J Alzheimer’s Assoc 9(63–75):e2Google Scholar
  2. 2.
    Riedel-Heller SG, Busse A, Angermeyer MC (2000) Are cognitively impaired individuals adequately represented in community surveys? Recruitment challenges and strategies to facilitate participation in community surveys of older adults. A review. Eur J Epidemiol 16:827–835PubMedCrossRefGoogle Scholar
  3. 3.
    Reuther S, van Nie N, Meijers J, Halfens R, Bartholomeyczik S (2013) Malnutrition and dementia in the elderly in German nursing homes. Results of a prevalence survey from the years 2008 and 2009. Z Gerontol Geriatr 46:260–267PubMedCrossRefGoogle Scholar
  4. 4.
    Weyerer S, Mann AH, Ames D (1995) Prevalence of depression and dementia in residents of old age homes in Mannheim and Camden (London). Z Gerontol Geriatr 28:169–178PubMedGoogle Scholar
  5. 5.
    Schumacher J, Zedlick D, Frenzel G (1997) Depressive mood and cognitive impairment in results of old age nursing homes. Z Gerontol Geriatr 30:46–53PubMedGoogle Scholar
  6. 6.
    Jakob A, Busse A, Riedel-Heller SG, Pavlicek M, Angermeyer MC (2002) Prevalence and incidence of dementia among nursing home residents and residents in homes for the aged in comparison to private homes. Z Gerontol Geriatr 35:474–481PubMedCrossRefGoogle Scholar
  7. 7.
    Fichter MM, Meller I, Schroppel H, Steinkirchner R (1995) Dementia and cognitive impairment in the oldest old in the community. Prevalence and comorbidity. Br J Psychiatr 166:621–629CrossRefGoogle Scholar
  8. 8.
    Reischies FM, Geiselmann B, Gessner R, Kanowski S, Wagner M, Wernicke F, Helmchen H (1997) Dementia in the very elderly. Results of the Berlin Aging Study. Nervenarzt 68:719–729PubMedCrossRefGoogle Scholar
  9. 9.
    Ziegler U, Doblhammer G (2009) Prevalence and incidence of dementia in Germany—a study based on data from the public sick funds in 2002. Gesundheitswesen 71:281–290PubMedCrossRefGoogle Scholar
  10. 10.
    Hoffmann F, Icks A (2012) Structural differences between health insurance funds and their impact on health services research: results from the Bertelsmann Health-Care Monitor. Gesundheitswesen 74:291–297PubMedCrossRefGoogle Scholar
  11. 11.
    Hoffmann F, Icks A (2011) Diabetes prevalence based on health insurance claims: large differences between companies. Diabet Med 28:919–923PubMedCrossRefGoogle Scholar
  12. 12.
    Schulze J, van den Bussche H, Glaeske G, Kaduszkiewicz H, Wiese B, Hoffmann F (2013) Impact of safety warnings on antipsychotic prescriptions in dementia: nothing has changed but the years and the substances. Eur Neuropsychopharmacol 23:1034–1042PubMedCrossRefGoogle Scholar
  13. 13.
    Eisele M, van den Bussche H, Koller D, Wiese B, Kaduszkiewicz H, Mayer W, Glaeske G, Steinmann S, Wegscheider K, Schön G (2010) Utilization patterns of ambulatory medical care before and after the diagnosis of dementia in Germany: results of a case-control study. Dement Geriatr Cogn Disord 29:475–483PubMedCrossRefGoogle Scholar
  14. 14.
    Rothgang H (2010) Social insurance for long-term care: an evaluation of the german model. Soc Policy Adm 44:436–460CrossRefGoogle Scholar
  15. 15.
    Bartholomeyczik S, Hunstein D (2004) Time distribution of selected care activities in home care in Germany. J Clin Nurs 13:97–104PubMedCrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Falk Hoffmann
    • 1
  • Hanna Kaduszkiewicz
    • 2
  • Gerd Glaeske
    • 1
  • Hendrik van den Bussche
    • 2
  • Daniela Koller
    • 1
  1. 1.Division Health Economics, Health Policy and Outcomes Research, Centre for Social Policy ResearchUniversity of BremenBremenGermany
  2. 2.Institute of Primary Medical CareUniversity Medical Center Hamburg-EppendorfHamburgGermany

Personalised recommendations