Chronic conditions and use of health care service among German centenarians

Abstract

Background

there is limited data comparing conditions and health service use across care settings in centenarians. To improve health service delivery in centenarians, the aim of this study was to compare the proportion of centenarians who have chronic conditions, take medication and use health care services across different care settings.

Methods

this cohort study uses routine data from a major health insurance company serving Berlin, Germany and the surrounding region, containing almost complete information on health care transactions. The sample comprised all insured individuals aged 100 years and older (N = 1,121). Community-dwelling and institutionalised individuals were included. Charlson comorbidity index was based on 5 years of recordings. Hospital stays, medical specialist visits and medication prescribed in the previous year were analysed.

Results

while 6% of the centenarians did not receive any support; 45% received family homecare or homecare by professional care services; 49% were in long-term care. The most frequent conditions were dementia and rheumatic disease/arthritis, with the highest prevalence found among long-term care residents. A total of 97% of the centenarians saw a general practitioner in the previous year. Women were more often in long-term care and less often without any care. Centenarians with long-term care showed higher proportions of comorbidities, greater medication use, and more visits to medical specialists compared with centenarians in other care settings.

Conclusions

the higher prevalence of dementia and rheumatic disease/arthritis in long-term care compared to other care settings emphasises the role of these diseases in relation to the loss of physical and cognitive functioning.

Introduction

Although the number of centenarians has been increasing constantly over the last decades, the specifics of their health conditions, as well as their use of health care, are largely unknown [1]. While health problems increase with age, worldwide trends with respect to the prevalence of chronic conditions and health service use among the oldest old do not follow a clear direction [2, 3]. Although nearly one-third of centenarians have had age-related morbidities for 15 or more years, some centenarians compressed disability toward the very end of their lives [2, 4, 5]. Recent findings from a German cohort study indicate that musculoskeletal (60%) and cardiovascular (57%) diseases are most common, although results may be biased due to the low response rate and the use of self-report measures [1]. Furthermore, very few studies have focused on the characteristics of health care use, doctor or hospital visits and prescription of medication in centenarians (for reviews of centenarian surveys see Refs. [1, 6]).

Data from around the world indicate that around half of centenarians live in the community while the other half live in long-term care settings [7–9]. Dementia was more common in individuals receiving long-term care compared to centenarians in other care settings [9]. There are no data comparing other conditions across care settings in centenarians.

Our aim was to provide detailed information about the health status and health service use across different care settings in German centenarians in order to establish a basis for future health care planning. We analysed health insurance data that contain complete and valid data on all health service transactions and do not suffer from desirability bias, recall bias or selection bias [6].

Methods

Participants and procedure

The sample contains N = 1,121 centenarians living in the city of Berlin, Germany and the surrounding Northeastern region (the federal states of Berlin, Brandenburg and Mecklenburg-Vorpommern). All were policy holders at AOK Nordost (General Local Health Insurance) aged 100 years and older as of December 31, 2013. AOK Nordost insures 52% of the local population above the age of 90 in Berlin, 73% in Brandenburg, and 78% in Mecklenburg-Vorpommern. Insured individuals came from a broad range of socioeconomic backgrounds, with 10.3% of the centenarians receiving welfare benefits as a supplement to their small state pensions. According to the German system of health care service, centenarians were categorised into four groups: living at home (i) without care support, i.e. individuals who did not apply for care support and therefore are most likely to be without care needs; (ii) with family homecare; (iii) with homecare by professional care services; and (iv) in a long-term care residency. The study was approved by the local ethics committee (EA1/194/15).

Measurements

Comorbidities

The assessment of comorbidity was based on the Charlson comorbidity index [10]. Seventeen health conditions were defined and weighted, using established algorithms (see Appendix; [9, 11–13]). Both inpatient and outpatient ICD-10 codes from the past 5 years (2009–13) were used to identify each condition; diagnoses labelled ‘suspected’ were excluded. To be included in the index, each chronic comorbidity had to be documented at least twice in individual medical records, which has been shown be more valid compared to single records [14]. Individuals who left the investigated area within this period were not included in the dataset.

Health service use

Inpatient hospitalisations with a duration of one night and longer in the previous year were recorded. Outpatient health service use was assessed by the proportion of centenarians who saw (either in the practice, at home or at nursing home) general practitioners or medical specialists at least once during the previous year.

Medication

Medications were identified using the WHO Collaborating Centre for Drug Statistics Methodology classification on ATC-Level 05, which distinguishes drugs on the level of the chemical substance [15].

Statistical analyses

Chi-square analyses were conducted to determine group differences for categorical variables; for continuous variables, t-tests for independent samples and analyses of variance have been used.

Results

The vast majority of the N = 1,121 centenarians were female (91.0%); 40.4% were exactly 100 years of age, 54.5% were between 101 and 104 years of age and 5.1% were 105 and above.

Care settings

Overall, 49% of the centenarians lived in long-term care facilities (Table 1). Overall, 51.0% of the centenarians lived in the community, 26.9% received professional care, 17.8% received family care and 6.2% lived without any official care support. Of all, 51.1% of the women were in long-term care, 5.4% were living without any care, while 27.7% of men were in long-term care and 14.9% had no care.

Chronic conditions

The most frequent chronic conditions were dementia and rheumatic disease/arthritis, at 65.9% and 63.9%, respectively (Table 2). The Charlson comorbidity index was 0 in 3.1% of the centenarians, low (1–2) in 22.7%, moderate (3–4) in 37.6% and high (≥5) in 36.6%.

The long-term care facilities housed the largest proportion of centenarians with high comorbidity (41.2%), whereas centenarians who did not need care, or who lived at home with private care, were more likely to show low comorbidity values (24.3 and 27.0%; Table 2). Of those patients in long-term care 84.9% were diagnosed with dementia and 27.1% of those without any care. For rheumatic diseases, 67.4% were affected in long-term-care and 51.4% without care.

Medication

On average, 6.2 different drugs were prescribed per person during the previous year (Table 3). The most frequent prescriptions were diuretics (prescribed for 54.4% of individuals) and agents acting on the renin angiotensin system (49.5%). The most frequently prescribed mental health drugs were psycholeptics (mostly antipsychotics) and psychoanaleptics (mostly antidepressants), 26.2 (19.5%) and 16.8% (12.8%), respectively.

Centenarians with long-term care were prescribed 6.6 different drugs on average (SD = 3.6); centenarians with no care (M = 5.0, SD = 3.4) and private home care (M = 5.1, SD = 2.9) received significantly less drugs (Table 3).

Hospital stays

A total of 29% of the centenarians had hospital treatment during the previous year (Table 3). A significantly higher proportion of centenarians in professional home care than of those with no care were hospitalised, 39.7 and 18.6%, respectively. Those in professional home care had a longer average duration of 15.7 (SD = 19.8) days compared with those in long-term care (M = 10.4 days; SD = 9.5). Hospitalisation among those living with private home care was 22.0%, among those in long-term care at a slightly higher 26.8%.

Physician visits

In the previous year, 97.0% of the sample saw a general practitioner (Table 3). Otorhinolaryngologists (31.3%), ophthalmologists (29.5%), neurologists/psychiatrists (23.8%) and dermatologists (22.0%) were also commonly visited. Almost all individuals with private homecare, professional home care, or in long-term care (97.7, 99.3 and 96.2%) and only slightly fewer of those with no care (92.9%) saw a general practitioner. Across all specialisations, centenarians in a long-term care setting showed a higher proportion of specialist visits.

Discussion

The aim of the study was to identify the proportions of chronic conditions, medication prescriptions, and use of health service among centenarians differently for care settings. Extending findings from the German Heidelberg Centenarian Studies, this is, to our knowledge, the first centenarian study using routine data in Germany [1, 16, 17].

Care settings

Half of the centenarians were in long-term care. Similar proportions have been reported by several international studies [7, 9]. Of all, 6% of the centenarians did not receive any official care support and there is a chance that some of them may be considered to be independent, although the amount of informal care support is unknown. Japanese studies confirm our results regarding number of individuals without need of care, which are lower than in Ontario and Denmark [8, 9]. Of all, 90% of the centenarians were women, which is a common finding in centenarian studies [8, 9, 18]. Our findings support evidence from previous centenarian studies which describe women as having lower prevalence of chronic diseases and [13, 19] as having worse cognitive and physical functioning compared with men [2, 19–21].

Chronic conditions

Only 3% of the individuals had no chronic conditions. Similar results have been found in Japan. In Canada, this proportion was higher (28.0%), in Denmark lower (<1.0%) [7, 9, 19]. However, comparability is limited due to different methods (for a review see Refs. [1, 6]). One-third had medium and one-third had high comorbidity values. Similar proportions of medium and high comorbidity are reported in Canada. In Italy, using a classification scheme which takes the absence or presence of different comorbidities as well as the activities of daily living into account, 22% of centenarians had good health status, 30% had medium, and 42% poor health [9, 21]. In accordance with previous findings, the most frequent chronic conditions in our sample were dementia, rheumatic disease/arthritis, and congestive heart failure [7, 9, 17]. Regarding the prevalence of dementia (65%), similar results have been found in France, Denmark, and Canada [3, 7, 9, 22]. The Heidelberg Centenarian Study suggested prevalence estimates between 52 and 59% [23]. The prevalence of rheumatic diseases/arthritis (63.9%) in our study is slightly higher compared to findings from Canada, Denmark, and Germany [1, 7, 9].

The proportion of individuals with a high comorbidity burden was greater in long-term care compared to other care settings. The higher number of comorbidities in long-term care is also comparable with previous findings [7, 13]. In line with the assumption about the compression of morbidity and findings from the New England Centenarian Study, significant proportions of centenarians living in no-care settings had low comorbidity values, indicating that the compression of morbidity is prevalent for a minority of centenarians, at the very least [5, 18]. For dementia, rheumatic diseases/arthritis, and chronic pulmonary diseases, this difference was most apparent, whereas proportions of individuals affected by other chronic conditions did not vary across different care settings. In this sample, a larger proportion of centenarians in long-term care were diagnosed with dementia (85%) compared to Canadian findings (79%) [9]. Only 3% of centenarians live without any chronic conditions, nevertheless 23% have a low comorbidity burden. This finding is still encouraging; healthy aging is possible even to such an advanced age. Isolating the characteristics of these so-called ‘escapers’ might be a route for future studies [18].

Medication

On average, centenarians in our study were prescribed six different drugs in the previous year. A smaller number (3) was reported in Denmark, a higher number (9) was reported in Canada [7, 9, 13]. Confirming previous findings, the most frequent prescriptions were diuretics and agents acting on renin/angiotensin system. The next most common prescriptions were analgesics, which were prescribed to about half of the sample. This differs from previous findings, where laxatives had been found to be the second most common prescriptions [7, 9, 19].

Centenarians in long-term care were prescribed the highest number of drugs compared to centenarians without care or in private home care. This may be partly due to a stronger morbidity burden, but also partly due to better diagnostic opportunities in long-term care facilities.

Hospital stays

Overall, 30% of our sample did stay in a hospital during the previous year. This corresponds to findings from Canada and Australia, where 36 and 33% of centenarians had been hospitalised in the previous year [9, 24].

Centenarians in professional home care showed a higher proportion of hospital stays compared to individuals in other care settings, which might be related to the fact that long-term care institutions provide a broader range of care competences.

Physician visits

A total of 97% of the centenarians saw general practitioners during the previous year. Similar proportions have been reported in Canada (95%), Australia (98%) and Denmark (87%) [7, 9, 24]. Specialist visits were more frequent for all specialties in our sample compared to results from Canada. This might be due to the fact that specialists in Germany can be contacted directly without referral [9]. Compared to a representative sample of 70–79 years old Germans, higher proportions of our sample saw general practitioners and neurologists/psychiatrists during the previous year [25]. Other specialists were seen by markedly smaller proportions. This only partly confirms previous results, which found an increase in health service use with increasing age [25].

The higher frequency of specialist visits across all specialties in long-term care might be partly generated by a higher burden of disease. Another explanation might be that specialists sign care contracts with nursing homes, which leads to higher number of specialist visits and can be, thus, considered a potential confounding factor.

Altogether, our findings support the important role of the general practitioner across all care settings and compared to specialists. This has important health service implications. General practitioners are the contact point for managing acute and chronic health issues as well as care.

Strengths and limitations

A significant strength of our study was the use of the routine data of a major statutory health insurance company, which is comprehensive and covers the vast majority of health service used by the included participants. Additionally, these data are free of any self-reporting bias, selective case drop out or missing data that would provide an invalid picture of the health care use. However, this data is limited to routinely collected information. Care arrangements or health care use outside the insurance system, including over the counter medication, are not recorded and, thus, not included in the present analyses. Coding mistakes were not detected. In order to improve the validity of diagnoses data, we neglected ICD-10 codes that had been documented only once (19.7% of codes occurred only once); therefore, it is also likely that some correct diagnoses were lost. While health insurance data tend to contain valid diagnoses from physicians, we were not able to prove the validity of the diagnoses. We cannot guarantee that our sample was representative of the Berlin population, though we analysed the data of the largest health insurance company of the region with ensured individuals of a broad range of socioeconomic backgrounds.

Conclusions

While most centenarians require care, there is a remarkable variation among centenarians in terms of health and health care use [9]. Within the group of individuals receiving long-term care, the high comorbidity burden and the prominence of dementia and rheumatic diseases highlight the important role of these long-term health care arrangements. Our findings emphasise the role of rheumatic diseases/arthritis in relation to the loss of physical functioning and dementia in relation to losses in cognitive function. This confirms that these two domains are important indicators for distinguishing healthy from unhealthy centenarians, as proposed by Cevenini et al. [26].

Although centenarians are the fastest growing population segment, the German health care system does not yet focus on these oldest old. Specially trained GPs as well as geriatricians are desperately needed. Mandatory annual check-ups concerning age-related conditions and medication, including polypharmacy, would be necessary to provide adequate care for centenarians. Slowly, policy makers and practitioners start recognising centenarians as a unique but vulnerable group. An action plan to foster health service research in geriatrics and gerontology for the oldest old was outlined recently by the German government. However, further steps are needed in order to optimise future services to better match the needs of patients. Moreover, the complex relationship between longevity, healthy ageing, and health service use should be a target for future research and interventions [27].

Supplementary data

Supplementary data mentioned in the text are available to subscribers in Age and Ageing online.

Author's contributions

All authors made substantial contributions to the concept and design of the study. T.Z. and J.N. cleaned and provided the routine data. P.B. and P.G. analysed the data, and all authors were involved in reviewing the data. P.B., P.G., D.D. and A.K. wrote and reviewed the article. All authors revised the current manuscript for submission. All authors read and approved the final article.

Conflict of interest

The authors declare that there are no conflicts of or competing interests.

Funding

This work was supported by the GeWINO—Health Research Institut—AOK Nordost, Berlin, Germany. The funding sources had no role in the design or conduct of the study; collection, management, analysis or interpretation of the data; or in the preparation, review or approval of the article.

References