Care for patients with dementia before and after initial diagnosis

The diagnosis of dementia is an enormous emotional and financial burden for many of those affected and their relatives. At the same time, it represents a major cultural and political challenge for society. There are no treatment options. It is therefore important for everyone involved to learn to live with dementia and to find ways that enable them to live a good life despite the illness (cf. Klie 2017). Dementia is the most common “mental” illness in older people and one of the most common reasons for the need for care. The most common forms of dementia are Alzheimer's disease (ICD codes F00. & G30.) and vascular dementia (ICD F01.), which is caused by damage to the blood vessels in the brain. There are also secondary forms of dementia (F02.), e.g. B. in connection with Parkinson's or HIV, and unspecified forms of dementia (F04.). These five ICD codes were used in the analysis to identify patients with dementia and a level of care.

>> For the analysis, conducted as part of the 2017 DAK Care Report, routine statutory health insurance data and data from the DAK-Gesundheit social long-term care insurance on dementia patients in need of care were available for the years 2014 to 2016. Data from approximately 100,000 dementia patients were analyzed for each year. In addition to pseudonymized master data, the dataset included information on care provided by outpatient physicians or home nursing care, in hospitals, with medications, medical devices, and assistive devices, as well as travel costs and care services such as inpatient care in nursing homes.
The entire dataset was used to determine the prevalence and incidence of dementia and to investigate regional differences at the district level. In addition, a subset of 17,450 dementia patients who received a dementia diagnosis for the first time in 2015 was evaluated. For the analysis of the period before and after the first dementia diagnosis, relative periods were created, with the day of the first dementia diagnosis or the first day of the corresponding quarter in the outpatient sector representing the individual time point t0. Subsequently, the 364 days before and after the initial diagnosis were examined as a relative year with four relative quarters of 91 days each. The methodological recommendations of the "Good Practice Secondary Data Analysis" were observed (Swart et al. 2015). In order to be able to make statements about the strength of the effect of a dementia diagnosis on service utilization behavior, the relative risk (RR) was calculated.
Prevalence and incidence of dementia
The average administrative prevalence, determined from the documented diagnostic information of the outpatient and inpatient care sectors from 2014 to 2016 based on routine billing data from DAK-Gesundheit, is approximately 1.67%. If this administrative prevalence of the DAK population is extrapolated to the total German population of approximately 82.2 million, an average of approximately 1.4 million people in Germany suffer from dementia (average: 16.69 dementia patients per 1,000 inhabitants). This result is largely consistent with comparable analyses (Ziegler/Doblhammer 2009). Only 2.80% of patients were documented to have "presenile dementia" in those under 65 years of age. Furthermore, dementia primarily affects female patients. Approximately 71.7% of all dementia patients requiring care were female. Women tend to require a slightly higher level of care than men, with approximately three-quarters of all women in the highest level of care, compared to only two-thirds of men. The average number of new cases (incidence) in 2015 according to the billing data was approximately 0.34%, or approximately 3.43 patients per 1,000 inhabitants developing dementia each year. Extrapolated to the total German population, this equates to approximately 282,000 new cases of dementia per year. A new case was assumed if no dementia diagnosis was documented in the statutory health insurance billing data for the entire previous year, 2014. Since dementia is a chronic illness and patients generally have regular contact with their doctor (approximately 96% of dementia patients have contact with outpatient doctors at least once per quarter), the actual incidence of dementia is likely to be only slightly overestimated using this method.
Furthermore, it is striking that more than a quarter of the dementia patients examined were diagnosed with type 2 diabetes, and a similar number suffer from depression. This result is consistent with the literature, where diabetes and depression have been identified as risk factors for dementia (e.g., Katon et al. 2015). Another risk factor for dementia is hypertension (e.g., Lüders/Schrader 2009), which affects approximately three-quarters of the dementia patients examined. Compared to female dementia patients, men have a significantly higher prevalence of cardiovascular diseases such as chronic ischemic heart disease, heart failure, or atrial fibrillation and atrial flutter. Women, on the other hand, showed significantly higher prevalences for the comorbidities of depression,
Osteoporosis, back pain and unspecified pain.
Regional evaluations
DAK Gesundheit's insured persons in need of care in the eastern German states are, on average, up to four years younger than in the rest of Germany. Accordingly, a regional prevalence analysis reveals unusually low rates compared to other studies, particularly in Mecklenburg-Western Pomerania, Brandenburg, Saxony, and Saxony-Anhalt (see Figure 1, left). The highest dementia prevalence rates at the district level were found primarily in densely populated, independent cities.
To compensate for age effects, all further analyses were age-adjusted. Firstly, the proportion of dementia patients cared for in nursing homes was examined (Fig. 1 middle), and secondly, the number of hospital admissions per 1,000 dementia patients (Fig. 1 right). Equally above-average values were found mainly in Bavaria. Otherwise, the number of hospital admissions was above average primarily in states where the proportion of dementia patients cared for as inpatients was low. This is particularly evident in the eastern federal states, where the proportion of dementia patients cared for in nursing homes is low, but the number of hospital admissions per 1,000 dementia patients is above average. Whether there is a connection that, for example, a lack of nursing home structures leads to a greater use of hospital services cannot be directly deduced from the data.
Evaluations relative to
Initial diagnosis
The average age of the 17,450 insured persons requiring care was approximately 83 years at the time of their first dementia diagnosis (women: 84.04; men: 80.73). The proportion of female insured persons, at 69%, is significantly higher than that of male insured persons. The majority of first dementia diagnoses – just under 74% – were made in the outpatient sector. The remaining 26% of dementia cases were diagnosed in hospitals. Fewer than 0.02% of initial diagnoses originate from other sectors, such as home nursing or specialized outpatient palliative care. Due to quarterly billing and data transmission in the outpatient sector, it is possible that multiple diagnoses could be considered as the initial diagnosis, as the date of diagnosis is not recorded in the billing data. Therefore, multiple counts were unavoidable when analyzing the respective ICD codes. The most frequently documented dementia was unspecified dementia (ICD: F03.), slightly more common among female care recipients than among male care recipients (women: 651,000; men: 611,000). Approximately one in four to five insured persons was diagnosed with vascular dementia (F01.) (women: 221,000; men: 251,000), and approximately one in five care recipients was diagnosed with dementia due to Alzheimer's disease (women: 221,000; men: 201,000).
It is striking that women were more likely to develop Alzheimer's dementia and more men to develop vascular dementia. This result is consistent with other scientific studies, for example, by Gao et al. (1998). Dementia due to diseases classified elsewhere occurred in only 31% (women) and 51% (men) of the insured persons examined. Of the nursing care recipients examined, a total of 81% received care level 1 benefits, 22% received care level 2 benefits, and 70% received care level 3 benefits, including hardship cases. Thus, more than two-thirds of the insured persons requiring care with dementia received the highest level of care. A further breakdown by women and men shows that significantly more women required the full range of care required at care level 3.

Change in the use of services after dementia diagnosis
In all areas, utilization increases in the quarter following the first dementia diagnosis compared to the previous relative quarters and remains at a higher level in subsequent quarters than before the dementia diagnosis. This demonstrates that, as expected, dementia patients generate increased demand for services in all areas. Also interesting is the gradual increase in utilization before the dementia diagnosis in almost all areas and the final sharp increase once the first diagnosis is made (see Table 1).
Regarding hospital stays, it can be observed that in the relative year following the dementia diagnosis, three-quarters (76% of the insured persons in need of care) were treated in hospital at least once, whereas before the diagnosis, this was the case for less than half of the insured persons (46% of the insured persons). Of those treated in hospital after the dementia diagnosis, half had already had at least one hospital stay in the relative year before the dementia diagnosis. Only 8% of the insured persons included in the analysis were treated in hospital before the diagnosis, but not in the relative year following the diagnosis.
The relative risk (RR) of an insured person requiring inpatient care increases by more than two and a half times with a dementia diagnosis (overall: RR = 2.49; men: RR = 2.65; women: RR = 2.44). While 24% of women utilized this service in the relative quarter before the dementia diagnosis and 46% afterward, the figures for men were only 18% and 36%, respectively. In contrast, a larger proportion of men receive home care (before men: 17%, women: 15%; after men: 29%, women: 27%). Overall, the relative risk of an insured person requiring home care increases by more than one and a half times after a dementia diagnosis (RR = 1.75).
In the relative year before the dementia diagnosis, a patient in need of care incurred average costs of €12,768 (of which €8,770 was in statutory health insurance (SHI) and €3,998 in social long-term care insurance (SPV)) and €24,128 in the relative year after the dementia diagnosis (of which €14,178 was in statutory health insurance and €9,950 in social long-term care insurance). The total costs for male patients were slightly higher at €14,341 before and €25,819 after the dementia diagnosis than for female patients at €12,073 before and €23,403 after the dementia diagnosis. However, these higher total costs are due to the costs from the statutory health insurance benefits (men before diagnosis: 13,205 euros; women before diagnosis: 9,211 euros; men after diagnosis: 23,042 euros; women after diagnosis: 16,230 euros). The care costs are significantly higher for female care recipients, especially after a dementia diagnosis (men before diagnosis: 1,136 euros; women before diagnosis: 2,862 euros; men after diagnosis: 2,777 euros; women after diagnosis: 7,173 euros).
Changes in comorbidities and hospital admissions after dementia diagnosis
Comparing documented illnesses before and after dementia diagnosis, the proportion of care recipients with urinary incontinence (unspecified +151 Tp3T points, specified +51 Tp3T points) and gait and mobility disorders (+101 Tp3T points) increased particularly sharply in the outpatient setting. Significant growth was also observed in heart failure, depression, unclassified pain (mainly in women), and other symptoms of the nervous and musculoskeletal systems (mainly R29.6 – tendency to fall), as well as several other illnesses presented in Table 2. However, there are also diagnoses that significantly decline after a dementia diagnosis, such as accommodation disorders and refractive errors (-81% of the total), back pain (-2.51% of the total), metabolic disorders (-1.51% of the total), and, particularly in men, vaccinations (-2.51% of the total) or the documentation of functional implants (-31% of the total) in women. Whether these declines are due to limitations in patients' ability to provide information that can accompany dementia, or to a change of primary care physician or limited access to certain specialist groups in the setting of a residential care facility, can only be speculated at this stage.
Table 3, in contrast, shows the most common reasons for inpatient treatment and their trends relative to the dementia diagnosis. Only significant growth trends can be observed for the top 10 diagnoses by frequency. The proportion of patients requiring hospitalization due to volume depletion, which may be caused by insufficient fluid intake, increased particularly sharply after a dementia diagnosis (+51% T-points). Inpatient discharges following treatment for delirium (+41% T-points), unspecified pneumonia (+31% T-points, a greater increase in men), and heart failure (+41% T-points) increased almost as sharply. In addition, the number of diagnoses of fractures of the femur (+4% points), intracranial injuries (+2% points) and fractures of the lumbar spine and pelvis (+1.5% points) increased, each of which could be caused by the significantly increased tendency to fall (diagnosis R26 in Table 2).

Change in drug therapy after dementia diagnosis
The largest increase in patients prescribed medication after a dementia diagnosis was observed for the main therapeutic group of psycholeptics. Here, the proportion of insured persons prescribed a medication from this category nearly doubled, from approximately 30% before the dementia diagnosis to approximately 54% after the dementia diagnosis. Antipsychotic and sedative medications, which are often used as adjunctive therapy for dementia, accounted for the majority of the increase. Psycholeptics are associated with an increased tendency to fall and a higher fracture rate. Overall, approximately 10% of dementia patients prescribed psycholeptics had at least one hospital admission with a fracture. The proportion of patients prescribed antipsychotics more than doubled, increasing by 26%, while the proportion of patients prescribed hypnotics and sedatives increased by 3%.
A significant increase of 16% was also observed in the main therapeutic group of psychoanaleptics. This can be attributed primarily to antidementia drugs, which are often used shortly after diagnosis of the disease to reduce the breakdown of acetylcholine. The proportion of patients prescribed this drug group increased significantly from approximately 3% before the initial diagnosis to over 18% after the dementia diagnosis. The fact that some dementia patients requiring care were already taking such medications suggests a lack of documentation of dementia immediately after its diagnosis in a small proportion of those requiring care. In addition to the drugs already listed, there were also significant increases in the proportion of people receiving at least one drug from the respective main therapeutic group for analgesics (+15% points), with the proportion of dementia patients prescribed metamizole rising sharply from 36% before diagnosis to 50% after diagnosis.
outlook
Patients with dementia are among the most expensive and care-intensive patients in the German healthcare system. The above analysis revealed several areas where significant action is needed to improve the care of dementia patients:
• Avoid unnecessary hospital admissions, which are known to significantly increase the risk of mortality.
• A drug therapy that conforms to S3 guidelines, which is clearly not followed in practice and suggests significant technical and legal deficits.
• More integrated care for dementia patients by nurses, general practitioners, specialists, pharmacologists and relatives, which could optimise care and reduce the aforementioned inadequate care.

All of this would contribute to more adequate care and create better conditions for living with dementia. Whether a good life with dementia is possible depends crucially on regional and local conditions. This is also clearly demonstrated by the special analysis of the DAK routine data.