Relevance of Coded Prodromal Mild Cognitive Impairment in the Routine Treatment of Patients with Dementia in Germany

Abstract

Little is known about the impact of prior mild cognitive impairment (MCI, ICD-10: F06.7) diagnosis on the time to dementia diagnosis, anti-dementia drug therapy, and treatment persistence in patients with dementia (PWD). Patients with dementia diagnoses who started anti-dementia therapy between January 2010 and December 2016 were selected from 203 neurological/psychiatric practices in the Disease Analyzer databank (IQVIA). Patients with a history of MCI were compared to non-MCI controls in terms of demographic characteristics, anti-dementia therapy, and the rate of persistence with anti-dementia drugs. For persistence analyses, a 1:1 matching procedure was used based on age, gender, type of residence, and depression and dementia diagnosis. Persistence was represented using Kaplan-Meier curves. A Cox regression analysis was used to determine the influence of MCI diagnosis on persistence with anti-dementia drugs. 339 PWD with MCI diagnoses and 339 controls were available for analysis. PWD with MCI were younger (78.9 versus 80.4 years), less likely to live in a nursing home (8.5% versus 22.5%), more frequently received donepezil (40.1% versus 33.7%), and more likely to exhibit comorbid depression (29.6% versus 16.9%). There was no association between the risk of treatment discontinuation and prior MCI diagnosis. After 24 months, 40% versus 41.1% of patients had discontinued treatment. The prior MCI diagnosis presumably led to an earlier diagnosis of dementia and earlier anti-dementia treatment. Treatment continuity did not differ, which would suggest that it does not depend on prior MCI diagnosis but on the behavior of patients and their caregiving relatives.

Keywords

Alzheimer’s disease anti-dementia drugs Germany mild cognitive impairment preclinical dementia real-world-data

INTRODUCTION

Worldwide, dementia prevalence is expected to increase from 47 million dementia patients (PWD) in 2015 to almost 150 million in 2050 [1]. In Germany, more than 1.5 million patients suffered from the condition in 2014 [2]. By 2050, this number is expected to grow to over 2.7 million [3]. The medical treatment of these patients already poses major challenges for national health care systems, and such an increase in numbers will inevitably lead to changes in the structure of care. For example, between 2005 and 2015, symptomatic anti-dementia therapy was improved in England and Germany [4, 5]. New, disease-modifying treatment strategies that address the biological causes of dementia and that can be used even in patients with mild cognitive impairment (MCI), i.e., before the onset of dementia symptoms, will more than double the number of patients to be treated in the future, as patients with MCI will increasingly be included in treatment approaches [6, 7].

Even now, these prodromal disease stages can be coded using the ICD-10 code for MCI (ICD-10: F 06.7) [8]. In addition to the recommendation to diagnose patients as early as possible, the German dementia guideline was supplemented with references to the prodromal stages of dementia in 2016, without, however, providing differentiated rules for MCI diagnosis and treatment [9]. Nevertheless, there is a trend in university memory assessment clinics: between 1985 and 2009, patients with cognitive disorders were increasingly referred even when presenting with only MCI or exhibiting early symptoms of dementia [10]. Similar developments can be observed in other European countries [11]. Routine care by general practitioners in private practice has shown that prior diagnosis of MCI represents a significant risk (OR = 2.12) for a subsequent dementia diagnosis [12]. In specialist care practices, it was shown that 34.5% of 4,633 MCI patients treated between 2005 and 2015 transitioned to dementia within 5 years [8].

However, little is known about whether dementia is diagnosed earlier (for example at a younger age) in patients with a prior MCI diagnosis compared to PWD without a prior MCI diagnosis. It is also unknown whether these patients could benefit from better drug treatment continuity, for example when receiving anti-dementia treatment. Therefore, we need to examine whether PWD with a prior MCI diagnosis are younger when beginning therapy, whether there are indicators of them being less severely ill, and whether they may benefit from anti-dementia treatment for a longer period than those without a prior MCI diagnosis.

METHODS

Database

The present study was based on the Disease Analyzer database (IQVIA), which grants anonymous access to a selected sample (of approximately 3%) of medical practices and patients. The data are transmitted by the practice computers via standardized interfaces and provide daily routine information about the illnesses and therapies of the patients. In addition to the data collected from general practices and specialist practices for internal medicine, the database also includes information gathered from various specialist groups in Germany. The database contains only anonymous data in accordance with applicable data protection laws [13]. The selection process for the Disease Analyzer database is based on summary statistics of all physicians in Germany, issued annually by the German Medical Association. The database is considered to be representative [14] and has been used in several published dementia studies [5, 12].

Study population

In the period from January 2010 to December 2016, dementia diagnoses (ICD-10: F01, F03, G30) were first documented in 78,470 patients. Of these patients, 13,054 (16.6%) were prescribed an anti-dementia drug for the first time. In total, 339 patients treated with anti-dementia drugs had received a prior MCI diagnosis (ICD-10: F 06.7), while 12,715 patients treated with anti-dementia drugs had no documented history of MCI. In order to compare these two groups, patients with a history of MCI were matched to controls without MCI using a 1:1 matching procedure based on age, gender, type of residence (nursing home or home care), dementia diagnosis (Alzheimer’s disease (AD), vascular dementia, unspecified dementia), and depression diagnosis (cf. Fig. 1).

Fig. 1.

Flow chart for the selection of study participants for the analysis of the influence of MCI diagnosis on persistence with anti-dementia drugs.

Outcome measure and variables

Patients with a history of MCI were compared to non-MCI controls in terms of demographic characteristics, anti-dementia therapy, and the rate of persistence with anti-dementia drugs. Persistence was estimated as the time of treatment without discontinuation (defined as at least 90 days without the treatment in question). All patients were followed for up to 24 months from the start of therapy. Demographic data included age, gender, and type of residence (home care, nursing home). Clinical data included the type of dementia diagnosis (AD, vascular dementia, and unspecified dementia), initial anti-dementia drug (donepezil, memantine, galantamine, rivastigmine), and depression diagnosis (ICD-10: F32, F33).

Statistical methods

Descriptive statistics were generated and differences in patient characteristics (MCI versus no MCI diagnosis) were determined using the Wilcoxon test for paired samples or McNemar’s test. Persistence was represented using Kaplan-Meier curves. A Cox regression analysis was used to determine the influence of MCI diagnosis on persistence with anti-dementia drugs. A p-value of <0.05 was considered statistically significant. All analyses were performed using SAS 9.4 (SAS Institute, Cary, USA).

RESULTS

Overall, 16.6% (13,075) of all incident PWD (N = 78,470) in the 203 included specialist practices included were treated with anti-dementia drugs between 2010 and 2016. The PWD treated with anti-dementia drugs who had a prior diagnosis of MCI accounted for 2.7% (n = 339) of all PWD treated with an anti-dementia drug for the first time.

PWD with MCI were 78.9 (SD = 7.3) years old on average, those without a prior MCI diagnosis were 80.4 (SD = 7.3) years old on average. In the younger age groups of <60, 61–70, and 71–80, the relative shares of patients with a prior MCI diagnosis were higher (3.2%, 7.6%, and 46.9% versus 2.6%, 6.2%, and 39.2%, respectively), while in the age group of over 80-year-olds the share of patients with prior MCI was lower than in the group of PWD without a prior MCI diagnosis (42.3% versus 52.0%). In contrast, PWD with a prior MCI diagnosis were less likely to live in a nursing home (8.5% versus 22.5%) and more likely to live at home (91.5% versus 77.5%). The AD, vascular dementia, and unspecified dementia groups did not differ. While galantamine and rivastigmine were prescribed equally frequently in both groups (see Table 1), PWD with a prior MCI diagnosis were more frequently treated with donepezil (40.1% versus 33.7%) than memantine (24.4% versus 33.7%). PWD with a prior MCI diagnosis were more likely to exhibit comorbid depression (29.6% versus 16.9%).

Table 1

Baseline characteristics of dementia patients with a prior MCI diagnosis and dementia controls without a prior MCI diagnosis in neurological/psychiatric practices (Disease Analyzer database)

	Prior to 1:1 matching			After 1:1 matching
Variable	MCI	no MCI	p	MCI	no MCI	p
	N = 339	N = 13,075		N = 339	N = 339
N
Age (mean, SD)	78.9 (7.3)	80.4 (7.3)	<0.001	78.9 (7.3)	78.9 (7.3)	1.000
Males (%)	40.5	35.8	0.077	40.5	40.5	1.000
Females (%)	59.5	64.2	59.5	59.5
≤60 years (%)	3.2	2.6	0.005	3.2	3.2	1.000
61–70 years (%)	7.6	6.2	7.6	7.6
71–80 years (%)	46.9	39.2	46.9	46.9
>80 years (%)	42.3	52.0	42.3	42.3
Institutionalization (%)	8.5	22.5	<0.001	8.5	8.5	1.000
Home care (%)	91.5	77.5	91.5	91.5
Dementia diagnosis (%)
Alzheimer’s disease	70.7	68.3	0.352	70.7	70.7	1.000
Vascular dementia	7.0	8.0	0.518	7.0	7.0	1.000
Unspecified dementia	22.3	23.7	0.545	22.3	22.3	1.000
Initial anti-dementia drug (%)
Donepezil	40.1	33.7	<0.001	40.1	40.1	1.000
Memantine	22.4	35.3	22.4	22.4
Galantamine	19.2	19.4	19.2	19.2
Rivastigmine	18.3	18.6	18.3	18.3
Depression (ICD-10: F32, F33)	29.6	16.9	<0.001	29.6	29.6	1.000

These differences leveled out in the context of the 1:1 matching by age, gender, type of dementia diagnosis, and comorbid depression (see the last three columns of Table 1).

The logistic regression model showed no association between the risk of treatment discontinuation before the end of the second year of treatment and prior MCI diagnosis (r = 1.24 (0.92–1.67), p = 0.167). Kaplan-Meier curves for treatment continuity with anti-dementia prescriptions in PWD with or without a prior MCI diagnosis show no difference. During the two-year observation period following the initiation of anti-dementia treatment, there were no differences in anti-dementia treatment continuity and discontinuation rates between the 1:1 matched groups. After 24 months, 60% of PWD without MCI and 58.9% of PWD with prior MCI still received anti-dementia drugs. 40% and 41.1% of patients had discontinued treatment, respectively (cf. Fig. 2).

Fig. 2.

Kaplan–Meier curves for persistence with anti-dementia drugs in dementia patients with or without prior MCI diagnosis in neurological/psychiatric practices.

DISCUSSION

Between 2010–2016, only approximately one in five (16.6%) of the 78,470 incident PWD received an anti-dementia drug prescription in the 203 specialist practices surveyed (16.6%). In addition, PWD with documented MCI and receiving anti-dementia drugs were rare (2.6%). In view of this low diagnosis frequency, it can be assumed that the MCI concept as a prodromal stage of dementia is of no significant quantitative relevance in the everyday clinical routine in the specialist practices surveyed. Epidemiological estimates suggest that MCI prevalence is more than twice as high as dementia prevalence [6]. In contrast, MCI was rarely documented in routine care. In the present specialist sample, MCI diagnoses were also rare compared to the epidemiologically expected prevalence [6 , 16]. The diagnosis of MCI may be a qualitative indicator of the particular focus of a physician’s practice on the prodromal stages of dementia, but it is of subordinate relevance in terms of the quantitative scope of care [8].

When MCI was documented in a patient who was later diagnosed with dementia, it can be assumed that this patient received the dementia diagnosis earlier than would have been the case without the prior MCI diagnosis: Patients with prior MCI diagnoses were almost 2 years younger, they usually lived at home, and they more frequently received donepezil than memantine, with the latter only being approved for later stages. The likelihood of dementia patients being diagnosed and treated at an earlier age appears to be higher when MCI was previously diagnosed.

In addition, it can be assumed that patients with prior documented MCI diagnoses in specialist practices received special attention. However, it is unclear whether MCI diagnoses have undergone a similar increase as dementia diagnoses. Bohlken et al. were able to show that the number of diagnoses increased by more than 50% and that the initiation of anti-dementia therapy more than doubled in the period from 2005 to 2015 [5]. Similar developments can be seen in the Netherlands [11] and the United Kingdom [4].

Interestingly, PWD with a prior MCI diagnosis exhibited a depression rate of almost 30%, while PWD without a prior MCI diagnosis had a depression rate of only 16%. This corresponds to the internationally established prevalence rates. The review by Ismail et al. shows a comparable prevalence rate of comorbid depression in MCI patients [18]. Pooled data from 57 studies with a total of 20,392 patients with MCI showed a depression prevalence of 32%. These findings are clinically significant because the frequency and severity of comorbid depression are good predictors of progression to dementia [19].

There was no correlation between the anti-dementia drug discontinuation rate or treatment continuity and prior MCI diagnosis. Treatment discontinuation is usually initiated by the patient [20]. MCI patients are often patients who are particularly worried about their memory problems [21]. It is not known whether this attitude fosters a willingness to discontinue therapy more quickly in the case of adverse events or if, on the contrary, the treatment is continued for a long time precisely because of these concerns about memory. Moreover, the reasons for the lack of treatment continuity may well be distributed differently in the 1:1 matched groups, i.e., the reasons for therapy discontinuation in the control group may differ from those in the study group. However, age differences were excluded in the matching procedure. It is known that younger age cohorts exhibit lower discontinuation rates than older age cohorts [20, 22]. The discontinuation rates in older-age patients were more often associated with lower Mini-Mental State Examination scores [23], a higher polypharmacy rate [24], and a higher side-effect rate [25].

Little is known about the extent to which prescribing behavior influences treatment continuity What is known is that guideline compliance, and thus treatment continuity, is particularly high in Scandinavian countries [26]. Furthermore, it has been shown that treatment continuity was positively impacted in practices with a particularly high number of PWD [17]. Finally, it should be noted that there are medical reasons for discontinuing anti-dementia therapy [27].

The concept of MCI as per the ICD-10 (ICD-10: F 06.7) is unspecific in routine care and can cover a broad range of disorders. The guidelines used in the German health system [9] are currently not extensive enough to articulate the etiologic background of the MCI diagnosis in as clear a manner as is recommended for general practitioners in the United States, for example [6]. No information was available in this study about whether MCI diagnoses were made without a test-psychological evaluation. The available dataset contained no test-psychological or biomarker-relevant results.

The distinction between early and late stages of dementia could also only be made indirectly with respect to age, the proportion of PWD living in nursing homes, and the type of anti-dementia drugs administered. However, it can be assumed that later— and therefore more severe— dementia stages were more common in older, institutionalized, and memantine-treated PWD. In contrast, earlier and milder stages of dementia can be more commonly expected in younger and home-based patients, as well as in PWD who are being treated with acetylcholinesterase inhibitors, such as donepezil, that have been approved for the treatment of dementia symptoms in early stages.

Finally, the group of patients with initial dementia diagnoses and treatment also includes patients who actually have MCI but have been diagnostically upgraded for indication reasons, to justify the prescribed anti-dementia drug.

One strength of the study is the large number of patients available for analysis. Another strength is the use of routine data (real-world data) from specialized medical practices at which the diagnoses and prescriptions have been continuously documented. This allows for a study without recall bias. Finally, this was the first study to research the association between prior MCI diagnosis and continuity of anti-dementia treatment in PWD in routine care.

This retrospective German study showed that PWD with a prior MCI diagnosis are younger, more likely to live at home, and more frequently receive donepezil than memantine compared to patients without a prior MCI diagnosis. Prior MCI diagnosis presumably leads to earlier dementia diagnosis and thus earlier anti-dementia treatment in routine care. However, the earlier diagnosis did not improve the continuity of drug therapy. Treatment continuity seemingly does not depend on the diagnostic behavior of physicians but on the behavior of patients and their caregiving relatives. Further research is needed on the role played by pre-dementia conditions, such as MCI, in the routine care offered to patients later diagnosed with dementia.

DISCLOSURE STATEMENT

Authors’ disclosures available online (https://www.j-alz.com/manuscript-disclosures/18-0567r1).

References

Livingston

, Sommerlad

, Orgeta

, Costafreda

, Huntley

, Ames

, Ballard

, Banerjee

, Burns

, Cohen-Mansfield

, Cooper

, Fox

, Gitlin

, Howard

, Kales

, Larson

, Ritchie

, Rockwood

, Sampson

, Samus

, Schneider

, Selbæk

, Teri

, Mukadam

(2017) Dementia prevention, intervention, and care. Lancet 390, 2673–2734.

Deutsche

Alzheimer

Gesellschaft

e.V.

(2016) Die Häufigkeit von Demenzerkrankungen. https://www.deutsche-alzheimer.de (Accessed on December 16, 2016)‥.

Doblhammer

, Fink

, Fritze

, Günster

(2013) The demography and epidemiology of dementia. Geriatric Mental HealthCare 1, 29–33.

Donegan

, Fox

, Black

, Livingston

, Banerjee

, Burns

(2017) Trends in diagnosis and treatment for people with dementia in the UK from 2005 to 2015: A longitudinal retrospective cohort study. Lancet Public Health 2, e149–e156..

Bohlken

, Kostev

(2018) Diagnostic and prescription behavior of general practitioners and specialist physicians in patients with dementia in 2005 and 2015 in Germany. Psychiatr Prax 45, 154–159.

Langa

, Levine

(2014) The diagnosis and management of mild cognitive impairment: A clinical review. JAMA 312, 2551–2561.

Ritchie

, Russ

, Banerjee

, Barber

, Boaden

, Fox

, Holmes

, Isaacs

, Leroi

, Lovestone

, Norton

, O’Brien

, Pearson

, Perry

, Pickett

, Waldman

, Wong

, Rossor

, Burns

(2017) The Consensus: Preparing for the advent of disease-modifying therapies for Alzheimer’s disease. Alzheimers Res Ther 9, 85. Edinburgh.

Bohlken

, Jacob

, Kostev

(2016) Progression of mild cognitive impairment to dementia in German specialist practices. Dementia (London). 10.1177/1471301216673919

Deutsche Gesellschaft für Psychiatrie, Psychotherapie und Nervenheilkunde und Deutsche Gesellschaft für Neurologie (eds) DGPPN u. DGN (2010) S3-Leitlinie “Demenzen”.

10.

Grimmer

, Beringer

, Kehl

, Alexopoulos

, Busche

, Förstl

, Goldhardt

, Natale

, Ortner

, Peters

, Riedl

, Roßmeier

, Valentin

, Diehl-Schmid

, Kurz

(2015) Trends of patient referral to a memory clinic and towards earlier diagnosis from 1985–2009. Int Psychogeriatr 27, 1939–1944.

11.

van

der Flier WM

, Scheltens

(2018) Amsterdam Dementia Cohort: Performing research to optimize care. J Alzheimers Dis 62, 1091–1111.

12.

Booker

, Jacob

, Rapp

, Bohlken

, Kostev

(2016) Risk factors for dementia diagnosis in German primary care practices. Int Psychogeriatr 28, 1059–1065.

13.

Dombrowski

, Kostev

(2017) Use of electronic medical records in the epidemiological research. Cuvillier Verlag, Göttingen.

14.

Becher

, Kostev

, Schröder-Bernhardi

(2009) Validity and representativeness of the “Disease Analyzer” patient database for use in pharmacoepidemiological and pharmacoeconomic studies. Int J Clin Pharmacol Ther 47, 617–626.

15.

Luck

, Luppa

, Briel

, Matschinger

, König

, Bleich

, Villringer

, Angermeyer

, Riedel-Heller

(2010) Mild cognitive impairment: Incidence and risk factors: Results of the Leipzig longitudinal study of the aged. J Am Geriatr Soc 58, 1903–1910.

16.

Dluga

, Weimar

, Wege

, Verde

, Gerwig

, Dragano

, Moebus

, Jockel

, Erbel

, Siegrist

(2010) Prevalence of mild cognitive impairment and its subtypes in the Heinz Nixdorf Recall study cohort. Dement Geriatr Cogn Disord 30, 362–373.

17.

Bohlken

, Kostev

(2017) The influence of caseload on continuous treatment with antidementia drugs. Fortschr Neurol Psychiatr 85, 156–162.

18.

Ismail

, Elbayoumi

, Fischer

, Hogan

, Millikin

, Schweizer

, Mortby

, Smith

, Patten

, Fiest

(2017) Prevalence of depression in patients with mild cognitive impairment: A systematic review and meta-analysis. JAMA Psychiatry 74, 58–67.

19.

Van

der Mussele S

, Fransen

, Struyfs

, Luyckx

, Mariën

, Saerens

, Somers

, Goeman

, De

Deyn PP

, Engelborghs

(2014) Depression in mild cognitive impairment is associated with progression to Alzheimer’s disease: A longitudinal study. J Alzheimers Dis 42, 1239–1250.

20.

Maxwell

, Stock

, Seitz

, Herrmann

(2014) Persistence and adherence with dementia pharmacotherapy: Relevance of patient, provider, and system factors. Can J Psychiatry 59, 624–631.

21.

Jessen

, Wolfsgruber

, Wiese

, Bickel

, Mösch

, Kaduszkiewicz

, Pentzek

, Riedel-Heller

, Luck

, Fuchs

, Weyerer

, Werle

, van

den Bussche H

, Scherer

, Maier

, Wagner

(2014) German study on aging, cognition and dementia in primary care patients. AD dementia risk in late MCI, in early MCI, and in subjective memory impairment. Alzheimers Dement 10, 76–83.

22.

Bohlken

, Weber

, Rapp

, Kostev (2015) Continuous treatment with antidementia drugs in Germany 2003–2013: A retrospective database analysis. Int Psychogeriatr 27, 1335–1342.

23.

Amuah

JE.

, Hogan

, Eliasziw

, Supina

, Beck

, Downey

, Maxwell

(2010) Persistence with cholinesterase inhibitortherapy in a population-based cohort of patients with Alzheimer’s disease. Pharmacoepidemiol Drug Saf 19, 670–679.

24.

Bohlken

, Jacob

, van

den Bussche H

, Kostev

(2018) The influence of polypharmacy on the initiation of anti-dementia therapy: A retrospective study of 21,888 dementia patients in 1,217 general practices in Germany. J Alzheimers Dis. 10.3233/JAD-180382.

25.

Gill

, Anderson

, Fischer

, Bell

, Li

, Normand

, Rochon

(2009) Syncope and its consequences in patients with dementia receiving cholinesterase inhibitors: A population-based cohort study. Arch Intern Med 169, 867–873.

26.

Taipale

, Tanskanene

, Koponena

, Tolppanen

, Tiihonend

, Hartikainena

(2014) Antidementia drug use among community-dwelling individuals withAlzheimer’s disease in Finland: A nationwide register-based study. Intl Clinl Psychopharmacol 29, 216–223.

27.

Parsons

(2016) Withdrawal of antidementia drugs in older people: Who, when and how? drugs Aging 33, 545–556.