Trends in Dementia and Mild Cognitive Impairment Prevalence and Incidence in German General and Specialist Practices Between 2015 and 2019

INTRODUCTION

Worldwide, dementia prevalence is expected to increase from 47 million patients with dementia (PWD) in 2015 to almost 131 million in 2050 [1–4]. In Germany, more than 1.7 million patients suffered from dementia in 2016 [5]. Projections indicate that this number will increase to more than 2.8 million by 2050 [5, 6].

In the last 10 years, there has been increasing evidence that the incidence and prevalence of dementia are stagnating or even decreasing in some countries [7–9]. In a meta-analysis based on seven population-based studies with more than 42,000 participants, Röhr et al. [9] confirmed that there is indeed evid-ence of decreasing incidence rates in industrialized nations. Another recent study by Wolters et al. [10] summarized the European and U.S. cohorts from the Alzheimer Cohorts Consortium and found that the incidence rate for dementia had decreased by 13%over the 10-year period.

The extent to which these developments found in cohort studies are also reflected in the actual care situation remains unclear. While the prevalence and incidence of dementia are determined in epidemiolo-gical field studies using uniform criteria and standardized methods, care studies are influenced by the diagnostic behaviors of physicians and the framework conditions of the statutory health insurance co-mpanies. In a study using a representative sample of general and specialist practices, Bohlken et al. [11] and Michalowsky et al. [12] found a significant increase in the prevalence of dementia treatment be-tween 2009 and 2015. However, the authors suggested that this might be due to changes introduced in the billing of early screening procedures. These results were confirmed in another recent study in which almost 90%of the population were examined. In this study, prevalence rates increased from 2.5%to 3.6%between 2009 and 2016. This increase was also attributed to improved diagnostics and specific financial incentive systems [12, 13]. By contrast, Doblhammer et al. [6] showed converging results of decreasing incidence rates between 2009 and 2011. The aim of this study is therefore to show the treatment prevalence and incidence of dementia and mild cognitive impairment (MCI) in German outpatient care settings between 2015 and 2019.

METHODS

RESULTS

DISCUSSION

Both the prevalence and incidence of dementia in general and specialist practices decreased continuously over the 5-year period from 2015 to 2019, which has also led to a decrease (-1%) in the number of people with confirmed dementia diagnoses. This tendency was particularly evident in tho-se over 80 years old compared to younger patients. In the case of MCI, on the other hand, the opposite was the case: A drastic increase of 17%was recorded in the number of patients with MCI, where the prevalence and incidence remain constant.

Studies have already reported that dementia is rarely recognized in those treated in outpatient general care settings and that only 40%of patients receive a formal dementia diagnosis [15, 16]. Other studies have shown that systematical screening supports the detection of cognitive abnormalities and can increase the diagnosis rate dramatically [17]. Among a sample of 598 patients in the United Kingdom, a total of 34%and 44%received a formal diagnosis of dementia between 2008 and 2011 and 2011 to 2013 respectively [16]. In particular, younger (<70 years) and very old (>90 years) patients were diagnosed less often than the age groups in between [16]. This low rate of detection and diagnosis contradicts established findings confirming the advantages of a formal dementia diagnosis. Various previous studies have shown that formally diagnosed patients are prescribed antidementia drugs more frequently, take potentially inadequate medication less often, and have a lower hospitalization rate [18–20].

In contrast to the decrease in the incidence and prevalence of dementia between 2016 and 2019, a drastic increase in the number of dementia diagno-ses in primary care was recorded between 2013 and 2015, which was due to changes in the billing process [11, 12]. Surprisingly, since 2015, the prevalence estimates based on field studies with standardized diagnostic criteria have not differed to any significant extent from the dementia diagnoses documented in routine care provided by GP and NP [5]. This result is astonishing in view of the low rate of psychological testing and imaging diagnostics, which, although it has progressed, still requires further improvement [13]. A large percentage of the patients with dementia were probably diagnosed based only on clinical symptoms. The lack of increase in prevalence found here may be due to the limitations of a purely clinical diagnosis. Patients with early-stage dementia in particular are underdiagnosed due to the lack of relevant psychological tests. However, it is conceivable that patients in the early stages of dementia or with MCI in particular may visit memory clinics more frequen-tly and are thus not included in the sample examined here. Nevertheless, the comparatively low number of memory clinics appear to contrast this possible explanation [21]. In addition, Schulz et al. [22] were able to prove that the regionally different distribution of memory clinics did not have any significant effect on the dementia prevalence in the different regions in 2009 [22]. In view of this, it seems unlikely that there has been a significant migration of possible dementia patients from primary care to the clinical diagnosis process in memory clinics.

It is more likely that changes in medical diagnosis behavior or changes in patient demands have led to the downward trend shown. Although dementia guidelines clearly recommend a dementia-specific treatment with antidementia drugs on the basis of numerous studies, many physicians are rather skeptical of the ascribed effects, mainly due to contradictory findings from other studies showing that anti-dem-entia treatment cannot reduce adverse effects or delay admission to an inpatient care facility [23]. For this reason and because of potential side effects, some experts have therefore been advising against the use of antidementia medication for years [24], causing some health authorities to stop covering the costs of such treatments and possibly also impacting the willingness of GP and NP to make a diagnosis. However, next to the diagnosis rate, the prescription rate of anti-dementia drugs is currently increasing in German. It is therefore questionable whether or not German practitioners have let themselves be influenced by these debates and whether or not the treatment prevalence or dementia prevalence in German GP and NP practices presented in this study corresponds to the actual prevalence among the population. Additional studies are needed to shed light on this issue.

When comparing the age structure of patients with dementia, as estimated on the basis of field studies, with that of those patients formally diagnosed with dementia found in the present study, it becomes ap-parent that the age group of over 85-year-olds is clearly underrepresented in this analysis. We also fo-und that the decrease in the prevalence and incidence was greatest in this age group. A comparison of the prevalence rates provided by Alzheimer Europe [5] for 2018 and those found in the present study for the same year confirms this finding. The prevalence of dementia in general and specialist care tended to deviate increasingly from the prevalence estimated by Alzheimer Europe with advancing age: 1.3%versus 1.2%(65–69), 3.3%versus 2.9%(70–74), 8.1%versus 6.0%(75–79), 12.1%versus 10.3%(80–84), 21.8%versus 15.8%(85–89), 40.9%versus 19.6%(>89). Similar results were obtained in the study by Aldus et al. [16], which demonstrated that the likelihood of a formal diagnosis in general and specialist practices was lowest for older patients. It was also shown that living alone was associated with a lower likelihood of receiving a diagnosis. The diagnosis itself, however, had no influence on life expectancy or the likelihood of living at home, which is why the authors came to the conclusion that, while the detection rate in primary care was low, the diagnosis itself did not correspond to any improvement in the living or care situation [16]. The frequency of antidementia drug prescription also decreases with increasing age in dementia patients. In the group of 75- to 79-year-olds, more than twice as many patients with dementia received antidementia drugs than in the age group of over 90-year-olds (31.9%versus 13.4%) [25]. It can therefore be assumed that the comparatively low diagnosis and treatment prevalence of dementia in old age is due to the reluctance of physicians to actively treat dementia at this advanced age.

On the other hand, it is also possible that the decrease in treatment prevalence in the practices is due to a reduction in the real actual prevalence and a decrease in the number of people with dementia. One of the few studies that have addressed the decreasing prevalence of dementia in Germany is that of Doblhammer et al. [6]. This study showed that the prevalence decreases by 1%to 2%annually, which corresponds roughly to the findings of our study and those of several previous studies [9, 26–29]. Although the decreases in prevalence presented in these studies are not significant, the authors list a variety of potential reasons for them. One such reason may be the reduction in vascular risks, which are the starting point for cardiovascular diseases and are also associated with dementia. Smoking also plays an important role among lifestyle factors, as there is strong ev-idence that it increases the risk of developing Al-zheimer’s disease. In addition, lack of education and wealth are the two most important demographic risk factors, which have direct effects on the brain or the patient’s cognitive reserves and indirect effects by influencing the health-related behavior of the patients themselves, which are very closely related to social status [22, 30–33]. All of these could explain the decreased incidence and prevalence and the reduction in the number of people with dementia. The latter is noteworthy as the progression of demographic change and the associated aging of society means that every year, the number of people 65 years or older is higher. Thus far, it has been assumed that the increase in older patients will dominate a possible decrease in the incidence. Our results show that this is not the case in routine care, however, and that the number of people diagnosed with dementia in primary care settings is decreasing.

Nevertheless, the results presented with regard to the prevalence and incidence of MCI show the tendencies increasing in general and specialist practices in Germany as expected due to the demographic ch-ange. Overall, the number of patients with MCI in-creased by 17%, despite the fact that incidence and prevalence remained constant. In two studies by Luck et al. [34, 35], the actual prevalence of MCI in patients aged between 60 and 79 years who did not suffer from dementia and in patients aged 75 and older was 20.3%and 15.4%respectively, which is well above the 1%of MCI patients diagnosed in GP and NP practices. Another systematic review by Luck et al. [36] showed that the incidence rate is between 51 and 77 per 1,000 person-years, which is also well above the incident diagnosis rate of approximately 25 per 1,000 person-years in GP and NP practices shown in the present analysis. The study by Alexander et al. [37] systematically extracted age-stratified estimates for this patient population using a methodical review including all studies published since 1995. The majority of studies showed a clear increase in the prevalence of MCI, which was also confirmed in our analysis.

Bohlken et al. [38] demonstrated that the number of cases of MCI diagnosed increased sharply in both general practices and neurological and psychiatric practices between 2007 and 2017. However, the authors point out that this increase corresponds to less than 10%of the actual real prevalence of MCI. It can be assumed that the diagnosis rate will once again increase significantly with the introduction of new Alzheimer’s disease therapies that be used to treat even MCI. Should such medications be approved, however, the associated increase in diagnoses will result in capacity problems (cf. e.g., RAND study [39]. The authors suggest that a shortage of specialists trained in the early detection of Alzheimer’s disease or insufficient capacities to treat those diagnosed could lead to long wait times. The direction in which the diagnosis rate for dementia and MCI will develop in the near future remains to be seen in view of the very complex relationships at play.

CONCLUSION

While the total number of MCI cases in general and specialist practices is increasing as expected despite the stagnating incidence and prevalence and in line with the ongoing demographic change, the opposite is true for dementia, which is characterized by a reduction in prevalence, incidence, and the total number of people with the condition. Reasons for this may lie in the diagnosis and therapy behaviors of general practitioners and specialists, which have changed over the years as a result of increasing doubts among these professionals as to the benefits of dementia diagnosis and therapy. Alternatively, the reason could be the reduction in vascular and other lifestyle factors which significantly reduce the risk of the disease. Additional studies are necessary in order to determine whether the diagnostic process in general and specialist practices is equivalent to the actual dementia prevalence and incidence.

DISCLOSURE STATEMENT

Authors’ disclosures available online https://www.j-alz.com/manuscript-disclosures/20-1385r1.