Care Dependency Factors
Analyses of Assessment Data from the Medical Service of German Statutory Health Insurance Providers
; ; ; ; ;
In view of the current life expectancy, most of the population of Germany can expect to be dependent on care in old age. According to the BARMER care report, 75% of women and 60% of men who died in 2017 were care dependent before their death (1). Those affected, their immediate living environment, and increasingly also the care-related services structures and professional care providers are confronted with substantial quantitative and qualitative challenges. Efforts therefore need to be strengthened to reduce demand for care services by preventing, alleviating, or delaying the need for care. This requires identifying important factors influencing the development of care dependency and describing these in their interactions (2, 3).
In the context of funding from the Central Federal Organization of Health Insurance Funds, the opportunity has arisen for the first time to analyze extensive assessment data from the Medical Service of German Statutory Health Insurance providers (MDK) to ascertain care dependency. Of 72 680 applications for initial care assessments in 2017 in the federal states of Berlin and Brandenburg for persons aged between 50 and 99, 57 572 persons (80%) were assigned to a care-level category. Of these people, 48 827 were living in their own homes. What follows is the description of exclusively the results of this partial group, as the focus was on living conditions and social constellations in people’s own homes. All analyses are based on assessment data from the MDK. The stratification into care categories is based on a points system defined in the German Social Code (SGB) XI, whose cut-off values are often rated as overly restrictive by critics (4).
Method and results
Characteristics of first-time applicants
As Table 1 shows, 60.1% of applicants assigned to a care-level category were women and 39.9% men. The average age was 78.2 years. About half (50.2%) of this group were living with at least one other person in their own household. About one third (34.4%) were living alone in their own home, but received social support—for example, from, partners, children, or neighbors (information from free text fields in the assessments of assistance received). 15.4% were living alone and without social support in their own household. Most of the care-level recommendations for people living at home fell into category 2 (46%) followed by category 1 (37.5%). 14.0% received a recommendation for care-level category 3 and 2.5% for category 4 or 5.
The influence of medical diagnoses
Among the top 10 diagnoses necessitating care, dementias were the most common, followed by primary generalized polyarthrosis, heart failure, chronic obstructive pulmonary disease (COPD) and cerebral infarction (Table 1).
We used adjusted ordinal regression models to analyze which diagnoses were most likely to be associated with a recommendation for a higher care-level category (Table 2). Patients with dementia (odds ratio [OR]: 10.71; 95% confidence interval [7.79; 14.72]), lung cancer (OR: 15.35; [10.71; 21.99]), and cerebral infarction (OR: 11.14; [7.95; 15.61]) had a higher probability of being assigned to a high care-level category (4/5) than patients with a diagnosis of dorsalgia (reference category). Furthermore, age and living situation were important factors influencing the recommended category. With each age decade, the probability of being assigned to a higher care-level category rose by up to 19%.
Compared with people living on their own without social support (reference) the risk for being assigned to a higher care-level category on first application rose in people living with another person (OR range 5.43–14.38). This hints at possible protective effects of close social relationships (especially between partners), which may initially offset a growing need for support and care (2). Progressive health losses over the further course, however, lead to applications and first assignments to higher care-level categories, which reflect worse health and the limitations of social support potentials.
5% of the variance (Pseudo-R2 = 0.0468) for a category recommendation to a care-level can be explained by means of the predictors included in the ordinal regression model. A more detailed explanation requires in particular longitudinal data, so as to establish causal associations and interactions between losses in physical and mental functioning as well as personal and social constellations.
Altogether the findings support the argument that care dependency can be affected preventively and is not unavoidable in every case (5). This is supported on the one hand by the great variance in age when entering care dependency, together with the question of which resources enable a person to indicate a need for care and support at age 88 and not as early as age 78. This question is currently the subject of further study. At the same time, the data reveals that whenever a person is assigned to a care-level category, they are usually assigned to a moderate category at first; nearly 84% of applicants still living in their own home were assigned to categories 1 or 2, while only 2.5% were assigned to categories 4 or 5. Consistent rehabilitation measures and support for existing resources—especially support of the person in need of care by close social relationships—can lead to a situation where care-level categories—and therefore the need for assistance—remain constant for long periods of time. Last but not least, acknowledging the association between dementia and care dependency provides starting points for counteracting progressive cognitive losses in a targeted manner by adapting exercise training, in order to keep care dependencies at constant levels or even reduce them.
Stefan Blüher, Thomas Stein, Susanne Schnitzer, Ralph Schilling, lrike Grittner, Adelheid Kuhlmey
Institute for Medical Sociology and Rehabilitation Science, Berlin (Blüher, Stein, Schnitzer, Kuhlmey) email@example.com
Institute for Social Medicine, Epidemology and Health Economics, Berlin (Schilling)
Institute for Biometry and Clinical Epidemiology, Berlin (Grittner)
Conflict of interest statement
The authors declare that no conflict of interest exists.
Manuscript received on 23 November 2020, revised version accepted on 7 June 2021.
Translated from the original German by Birte Twisselmann, PhD.
Cite this as
Blüher S, Stein T, Schnitzer S, Schilling R, Grittner U, Kuhlmey A: Care dependency factors—analyses of assessment data from the Medical Service of German Statutory Health Insurance providers. Dtsch Arztebl Int 2021; 118: 563–4. DOI: 10.3238/arztebl.m2021.0263
|1.||Rothgang H, Müller R: BARMER Pflegereport 2019. Ambulantisierung der Pflege. Berlin: Barmer 2019.|
|2.||Schnitzer S, Blüher S, Teti A, et al.: Risk profiles for care dependency: cross-sectional findings of a population based cohort study in Germany. J Aging Health 2020; 32: 352–60 CrossRef|
|3.||Blüher S, Schnitzer S, Kuhlmey A: Der Zustand Pflegebedürftigkeit und seine Einflussfaktoren im hohen Lebensalter. In: Jacobs K, Kuhlmey A, Greß S, Klauber J, Schwinger A (eds.): Pflegereport 2017 – Schwerpunkt: Die Versorgung der Pflegebedürftigen. Stuttgart: Schattauer GmbH 2017; 3–11.|
|4.||Schmid, Raimund: MDK schöpft aus den Taschen der Ärzte. ÄrzteZeitung 11/2017. www.aerztezeitung.de/Wirtschaft/MDK-schoepft-aus-den-Taschen-der-Aerzte-308113.html (last accessed on 2 July 2021).|
|5.||Rott C, Wiloth S, Kramer B et al.: Prävention statt Pflege. G+G digital 06/2020. www.gg-digital.de/2020/06/praevention-statt-pflege/index.html (last accessed on 2 July 2021.)|