Drugs Aging (2014) 31:215–224 DOI 10.1007/s40266-014-0154-8

ORIGINAL RESEARCH ARTICLE

Anti-Dementia Drugs and Co-Medication Among Patients with Alzheimer’s Disease Investigating Real-World Drug Use in Clinical Practice Using the Swedish Dementia Quality Registry (SveDem) Seyed-Mohammad Fereshtehnejad • Kristina Johnell • Maria Eriksdotter

Published online: 5 February 2014 Ó Springer International Publishing Switzerland 2014

Abstract Background There is a substantial risk of drug-interactions, adverse events, and inappropriate drug use (IDU) among frail Alzheimer’s disease (AD) patients; however, there are few studies about co-medication and IDU in clinical settings. Objectives To investigate anti-dementia drugs, associated characteristics of cholinesterase inhibitors (ChEIs) and NMDA antagonists, co-medication, and IDU in a large population of outpatients with mild AD. Methods In this cross-sectional analysis of medication characteristics, we analyzed data from the Swedish Dementia Quality Registry (SveDem) on 5,907 newly diagnosed AD patients who were registered in memory clinics. SveDem is a national quality registry in Sweden, which was established in 2007 to evaluate and improve dementia healthcare. Comparisons were performed concerning co-medications, use of C3 psychotropic drugs (IDU) and polypharmacy (C5 drugs) based on antidementia treatment (ChEIs or NMDA antagonists). Information on baseline characteristics such as age, sex, living conditions, cognitive evaluation based on the Mini-Mental

S.-M. Fereshtehnejad (&)
M. Eriksdotter Division of Clinical Geriatrics, Department of Neurobiology, Care Sciences, and Society (NVS), Karolinska Institutet, Novum 5th floor, 14186 Stockholm, Sweden e-mail: [email protected] K. Johnell Aging Research Center, Karolinska Institutet and Stockholm University, Stockholm, Sweden M. Eriksdotter Department of Geriatric Medicine, Karolinska University Hospital, Stockholm, Sweden

State Examination (MMSE) score, and diagnostic work-up was also evaluated. Results The majority of the AD patients were in the mild stage of the disease. Overall, 4,342 (75.4 %) patients received any ChEI, 438 (7.6 %) used an NMDA antagonist and 74 (1.3 %) patients were treated with both. However, 907 (15.7 %) patients were not treated with any anti-dementia drug. While polypharmacy was seen in 33.5 % of patients, only 2.6 % concurrently used C3 psychotropic medications. Patients on ChEIs were significantly younger, had a higher MMSE score and were treated with a smaller number of medications (a proxy for overall co-morbidity). Co-medication with antipsychotics [3.3 vs. 7.6 %; adjusted odds ratio (OR) 0.55 (95 % CI 0.38–0.79)] and anxiolytics [5.8 vs. 10.9 %; adjusted OR 0.62 (95 % CI 0.46–0.84)] was significantly lower in the ChEI? group than in those with no anti-dementia treatment. Conclusion Patients taking ChEIs were treated with less antipsychotics and anxiolytics than those not taking ChEIs. More research is warranted to elucidate whether use of ChEIs in clinical practice can reduce the need for psychotropic drugs in AD patients.

216

Key Points Alzheimer’s disease (AD) patients diagnosed in memory clinics in Sweden are mostly in the early stage of the disease. AD patients who are treated with cholinesterase inhibitors (ChEIs) are younger, with a higher MiniMental State Examination score and treated with fewer medications than AD patients who are not treated with ChEIs. The frequency of inappropriate drug use is quite low (less than 3 %) in the treatment of AD patients in Sweden, showing acceptable adherence to the national guidelines. Almost 34 % of the AD patients were under treatment with at least five drugs. More attention should be paid on co-medication with psychotropics and cardiovascular drugs in AD patients.

1 Introduction Alzheimer’s disease (AD) is the most common neurodegenerative disorder and dementia diagnosis in the elderly [1]. Despite large research efforts, no cure is available to date. However, symptomatic treatment with cholinesterase inhibitors (ChEIs) to increase the level of acetylcholine in synaptic clefts [2] is recommended by national guidelines for mild to moderate AD [3, 4]. This treatment is based on the knowledge that loss of cholinergic neurons is an early feature in the pathophysiology of AD [5]. ChEIs comprises donepezil, rivastigmine, and galantamine. Another pharmacological intervention in AD is the N-methyl-D-aspartate (NMDA) receptor antagonist memantine, which is the treatment of choice in moderate to severe AD [6]. Memantine inhibits neuronal death by blocking the activity of the toxic neurotransmitter glutamate [7]. Tolerability, safety, and drug–drug interactions of ChEIs and NMDA antagonists are important clinical issues. These concerns are even more important when considering that AD patients often have comorbidities that require multiple medications [8]. Therefore, the risk of potential drug–drug interactions, adverse events, and inappropriate drug use (IDU) increases [9]. Although many clinical trials have evaluated the pharmaceutical properties of ChEIs and NMDA antagonists in AD patients [2], there are few studies in clinical settings about co-medication and IDU [10–12]. We therefore, aimed to investigate the use of anti-

S.-M. Fereshtehnejad et al.

dementia drugs, associated characteristics of ChEIs and NMDA antagonists and co-medication in a large population of outpatients with mild to moderate AD by using data from the national Swedish Dementia Quality Registry (SveDem).

2 Methods 2.1 Swedish Dementia Quality Registry (SveDem) SveDem (http://www.svedem.se) is a national quality registry that was established in 2007 to evaluate and improve dementia healthcare in Sweden. For this purpose, newly diagnosed dementia patients are mostly registered from memory clinics (94 %), as well as an increasing number of primary healthcare units. However, in order to increase the homogeneity of quality assessment, the small proportion of data (6 %) from primary care units or nursing homes were deleted in this study. The exhaustiveness of the SveDem registry has gradually increased to cover data from 12 regions in 2007 up to 21 areas in 2011. Patient work-up is based on quality indicators of the Swedish National Board of Health and Welfare [13] and has shown to be a valuable tool to evaluate dementia workup in specialized and primary care units [14]. SveDem includes information on baseline characteristics such as age, sex, type of dementia disorder, living condition, cognitive evaluation based on the Mini-Mental State Examination (MMSE) score [15], investigations used in diagnostic work-up, and drug treatment. For the current investigation, all patients with newly diagnosed AD registered in SveDem from memory clinics between 1 May 2007 and 31 December 2011 were included. We excluded patients with mixed AD and those who were registered from either primary care units (n = 355) or nursing homes (n = 30). Finally, we recruited 5,907 subjects who were registered in memory clinics from 21 different regions, each of which included the median number of 110 patients with the interquartile range (IQR) of 33–274 subjects. 2.2 Definitions All variables were collected when dementia diagnosis was established. In Sweden, dementia disorders are clinically diagnosed according to the International Classification of Diseases, Tenth Revision (ICD-10) [16]. The onset-age was defined as the age when the AD diagnosis was established. In this investigation, we only recruited data from the memory clinics in Sweden, which are specialized care units where dementia patients are usually referred for more advanced work-ups carried out by geriatricians and/or psychiatrists.

Co-Medication in Alzheimer’s Disease

The total number of medications at the start of the dementia work-up was considered as a proxy for overall co-morbidity [17]. The patients’ medications at the time of diagnosis were categorized into the following groups: ChEIs, NMDA antagonists, cardiovascular drugs, antidepressants, antipsychotics, anxiolytics and hypnotics. Cardiovascular drugs included pharmacological treatment for hypertension, hyperlipidemia, atrial fibrillation, heart failure, ischemic heart disease, and diabetes. Psychotropic medication included antidepressants, antipsychotics, anxiolytics, and hypnotics [18]. Polypharmacy was defined as use of five or more drugs [19]. We used the Swedish National Board of Health and Welfare’s indicator of concurrent use of three or more psychotropic drugs as potential IDU [9]. Of note, it must be taken into account that in the SveDem registry, all medications are recorded at the drug class level regardless of the number of drugs that are possibly prescribed from a single class, if there is any (i.e. two or more antidepressant drugs in one subject). However, this issue is accounted for in the variable ’total number of medications’. As a quality registry, we have no access to the specific drugs that are prescribed for each patient in SveDem, and medication work-up is only recorded based on drug families, as previously mentioned. We assigned all AD patients into two subgroups based on use of anti-dementia medications, i.e. ChEIs and NMDA antagonists. Between-group comparisons were carried out, firstly concerning the ChEI user group (ChEI?) and the ChEI non-user group (ChEI-), and secondly the NMDA antagonist user group (NMDA antagonist?) and the NMDA antagonist non-user group (NMDA antagonist-).

217

age, sex, number of medications, MMSE score, and the referral center in the main comparisons between medication groups. The variable ‘referral center’ was entered into multivariate models to assess whether recruiting subjects from various centers with probable differences in quality of care could have induced any cluster effect. Kernel density plots were used as a non-parametric way to estimate the probability density function of a random variable, which here refers to MMSE score and age at diagnosis. A twotailed p-value of less than 0.05 indicated statistical significant difference in all analytical procedures.

3 Results 3.1 Baseline Characteristics A total number of 17,057 patients with dementia were registered in SveDem during 2007–2011. Of those, 6,292 individuals had AD, where 5,907 of them were registered in memory clinics and entered in the main analysis. Study population consisted of 3,866 (65.4 %) females and 2,041 (34.6 %) males, with a mean age of 76.8 (SD 8.4) years at the time of diagnosis. In general, 4,342 (75.4 %) AD patients received any ChEI, 438 (7.6 %) used an NMDA antagonist, and 74 (1.3 %) patients were treated with both, while 907 (15.7 %) patients were not treated with any antidementia medication. The mean MMSE score was 21.6 (SD 5.1). Other baseline and work-up characteristics of all AD patients are summarized in Table 1. Concurrent use of C3 psychotropic medications was found in 142 (2.6 %) AD patients, and polypharmacy in 1,891 (33.5 %) AD patients in the SveDem registry.

2.3 Ethical Considerations The regional Human Ethics Committee in Stockholm approved this study. All patients in SveDem were informed about their participation in the registry and had the right to decline participation. Moreover, data were anonymized before statistical analysis. 2.4 Statistical Analysis Data were analyzed using the SPSS software version 20 (IBM Corporation, Armonk, NY, USA). To describe quantitative and categorical variables, mean [standard deviation (SD)] and frequency percentage (%) were used. We performed the Chi-square test to compare relative frequency of nominal variables between two study groups. An independent samples t-test was used in order to compare the mean value of continuous variables between medication groups. Multivariate analysis was performed using binary logistic regression models adjusted for onset

3.2 Characteristics of Alzheimer’s Disease (AD) Patients on Cholinesterase Inhibitor (ChEI) Medication Demographic, diagnostic, and co-medication characteristics of AD patients are summarized in Table 2 and compared between the two study subgroups based on ChEI treatment. Considering AD patients with no anti-dementia medication as the reference group, those who received only ChEIs were significantly younger [75.8 (SD 8.3) vs. 80.0 (SD 8.4) years; p \ 0.001; Fig. 1a], with an almost similar feature in the prevalence of females compared with the reference group (66.9 vs. 65.9 %; p [ 0.05). The mean MMSE score was significantly higher in the ChEI? group [22.2 (SD 4.7) vs. 20.3 (SD 5.5); p \ 0.001; Fig. 1b]. The mean number of diagnostic tests used to reach a diagnosis was significantly higher in the ChEI? group, both before [4.7 (SD 1.4) vs. 4.2 (SD 1.4); p \ 0.001] and after multivariate adjustment [adjusted odds ratio (OR) 1.17 (95 %

218

S.-M. Fereshtehnejad et al.

Table 1 Characteristics of the whole study population (n = 5,907) Characteristic

Value

Sex: female [n (%)]

3,866 (65.4)

Age [years; mean (SD)]

76.8 (8.4)

MMSE score [mean (SD)]

21.6 (5.1)

Place of residence: own home [n (%)]

5,534 (95.6)

Co-resident [n (%)]

2,470 (43.4)

Day care [n (%)]

147 (2.6)

Home care [n (%)]

1,250 (22.0)

Total number of tests needed to establish diagnosis [mean (SD)]

4.6 (1.4)

Total number of medications [at start of the dementia work-up; mean (SD)]

3.6 (2.9)

Medication [at the time of diagnosis; n (%)] Cholinesterase inhibitors

4,416 (76.7)

NMDA antagonist

512 (8.9)

Antidepressants

1,237 (22.9)

Antipsychotics

226 (4.2)

Anxiolytics

370 (6.9)

Hypnotics

720 (13.3)

Cardiovascular drugs

3,082 (56.9)

SD standard deviation, MMSE Mini-Mental State Examination, NMDA antagonist N-methyl-D-aspartate receptor antagonist

CI 1.09–1.26)]. Patients in the ChEI? group were treated with fewer drugs before diagnosis, both before [3.5 (SD 2.9) vs. 4.2 (SD 3.1); p \ 0.001] and after multivariate adjustment [adjusted OR 0.96 (95 % CI 0.94–0.99)]. However, multivariable analysis showed that cardiovascular drugs were significantly more commonly used in the ChEI? group [adjusted OR 1.25 (95 % CI 1.04–1.50)]. Co-medication with antipsychotics [3.3 vs. 7.6 %; adjusted OR 0.55 (95 % CI 0.38–0.79)] and anxiolytic drugs [5.8 vs. 10.9 %; adjusted OR 0.62 (95 % CI 0.46–0.84)] was significantly lower in the ChEI? group than the reference group after adjustment for age, sex, MMSE score, total number of medications, and referral center. Additionally, concurrent use of C3 psychotropic medications was significantly less likely to occur in the ChEI? group (2.1 vs. 4.9 %; p \ 0.001), even after adjustment with the list of confounders [adjusted OR 0.48 (95 % CI 0.31–0.75)]. 3.3 Characteristics of AD Patients on NMDA Antagonist Medication Only 512 (8.9 %) of the study population of 5,907 AD patients received treatment with the NMDA antagonist memantine (either with or without ChEI), of which 438 (7.6 %) subjects were only treated with memantine. Patient characteristics of this subgroup are summarized in Table 3.

AD patients who only received memantine as their antidementia treatment had a similar mean age as the reference group under no anti-dementia treatment [79.9 (SD 6.9) vs. 80.0 (SD 8.4) years; p [ 0.05] at the time of diagnosis, however with a lower frequency of females (58.9 vs. 66.9 %; p \ 0.05]. In addition, the mean MMSE score was significantly lower in the NMDA antagonist? group compared with the reference group [18.8 (SD 6.0) vs. 20.3 (SD 5.5); p \ 0.001; Fig. 1b], which also remained significant after adjustment for the confounders [adjusted OR 0.97 (95 % CI 0.95–0.99)]. The mean number of drugs at the beginning of the diagnostic work-up was almost similar in the NMDA antagonist? group compared with the reference group [4.3 (SD 3.3) vs. 4.2 (SD 3.1); p [ 0.05]. Cardiovascular comedication was used in 251 (61.2 %) patients of this subgroup. Neither the use of cardiovascular drugs nor psychotropic medications were significantly different between the NMDA antagonist? group and the reference group. Even though concurrent use of C3 psychotropic medications was lower in the NMDA antagonist? group (2.9 vs. 4.9 %), the difference was not statistically significant, neither in the univariate comparison with the reference group nor after adjustment for the covariates (both p [ 0.05). 3.4 Comparison between AD Patients Taking ChEIs vs. NMDA Antagonist Medication Further analysis was performed to compare baseline, diagnostic, and medication characteristics between AD patients who received only ChEIs (ChEIs?) as their antidementia medication with those under treatment with only memantine (NMDA antagonist?), in which the ChEI? was considered as the reference group in all comparisons. As described in Table 2, compared in Table 3, and illustrated in Fig. 1, AD patients who received only memantine were significantly older [adjusted OR 1.07 (95 % CI 1.05–1.08); Fig. 1a], with a lower mean MMSE score [adjusted OR 0.89 (95 % CI 0.88–0.91); Fig. 1b] at the time of diagnosis. Even after adjustment for other confounders, including age, MMSE score, and total number of medication as a proxy for co-morbidities, the probability of receiving memantine was significantly lower for females [adjusted OR 0.62 (95 % CI 0.49–0.77)]. The mean number of drugs at the beginning of the diagnostic work-up was significantly higher in the NMDA antagonist? group compared with the ChEIs? group after adjustment for the confounders [adjusted OR 1.07 (95 % CI 1.03–1.11)]. The multivariate logistic regression model showed that even after adjustment for sex, age, MMSE score, and the referral center, the prevalence of polypharmacy was significantly higher among the NMDA

4.2 (3.1)

Total number of medications [at start of the dementia work-up; mean (SD)]

497 (56.9)

Hypnotics

Cardiovascular drugs

2,291 (56.5)

497 (12.3)

233 (5.8)

925 (22.8) 132 (3.3)

3.5 (2.9)

4.7 (1.4)

778 (18.3)

102 (2.4)

4,202 (96.9) 1,755 (41.0)

22.2 (4.7)

75.8 (8.3)

2,862 (65.9)

0.98 (0.85–1.14)

0.67 (0.55–0.81)**

0.50 (0.39–0.65)**

1.03 (0.86–1.22) 0.41 (0.30–0.56)**

0.92 (0.90–0.95)**

1.28 (1.21–1.35)**

0.41 (0.34–0.47)**

0.77 (0.50–1.18)

3.20 (2.42–4.24)** 0.53 (0.46–0.62)**

1.08 (1.06–1.09)**

0.93 (0.92–0.94)**

0.96 (0.82–1.11)

1.25 (1.04–1.50)*

0.90 (0.72–1.14)

0.62 (0.46–0.84)*

1.01 (0.83–1.24) 0.55 (0.38–0.79)*

0.96 (0.94–0.99)*

1.17 (1.09–1.26)**

0.69 (0.43–1.11)

1.21 (0.74–1.99)

1.77 (1.26–2.48)* 0.67 (0.57–0.80)**

1.07 (1.06–1.09)**

0.94 (0.93–0.96)**

1.11 (0.94–1.31)

Adjusted OR (95 % CI)

251 (61.2)

56 (13.6)

36 (8.8)

86 (21.0) 25 (6.1)

4.3 (3.3)

4.3 (1.3)

133 (31.1)

18 (4.3)

409 (93.6) 181 (42.5)

18.8 (6.0)

79.9 (6.9)

258 (58.9)

NMDA antagonist? (n = 438)

1.20 (0.94–1.02)

0.75 (0.54–1.05)

0.79 (0.53–1.18)

0.92 (0.69–1.23) 0.79 (0.49–1.28)

1.01 (0.97–1.05)

1.05 (0.97–1.14)

0.81 (0.64–1.04)

1.4 (0.76–2.55)

1.51 (0.97–2.36) 0.56 (0.45–0.71)**

0.96 (0.94–0.98)**

1.00 (0.98–1.01)

0.71 (0.56–0.90)*

Unadjusted OR (95 % CI)

Only NMDA antagonist

1.18 (0.88–1.59)

0.81 (0.55–1.19)

0.69 (0.43–1.09)

0.95 (0.68–1.33) 0.61 (0.34–1.08)

1.02 (0.98–1.06)

1.02 (0.91–1.14)

0.75 (0.56–1.02)

1.2 (0.58–2.38)

2.07 (1.21–3.55)* 0.60 (0.45–0.80)**

0.97 (0.95–0.99)*

1.00 (0.98–1.02)

0.69 (0.53–1.02)*

Adjusted OR (95 % CI)

* p \ 0.05, ** p \ 0.001

AD Alzheimer’s disease, SD standard deviation, OR odds ratio, MMSE Mini-Mental State Examination, ChEIs cholinesterase inhibitors, NMDA antagonist N-methyl-D-aspartate receptor antagonist

Adjustment is performed considering age, sex, MMSE score, number of medications (as a proxy for co-morbidities), and the referral center (to assess cluster effect) as the covariate list in the binary logistic regression model. The group who received neither ChEIs nor NMDA antagonist (none) is considered as the reference one for all OR calculations

95 (10.9) 152 (17.4)

Anxiolytics

Antidepressants Antipsychotics

196 (22.4) 66 (7.6)

4.2 (1.4)

Total number of tests needed to establish diagnosis [mean (SD)]

Medication [at the time of diagnosis; n (%)]

27 (3.1) 314 (35.7)

821 (90.6) 498 (56.7)

Place of residence: own home [n (%)] Co-resident [n (%)]

Home care [n (%)]

20.3 (5.5)

MMSE score [mean (SD)]

Day care [n (%)]

607 (66.9) 80.0 (8.4)

Age [years; mean (SD)]

ChEI? (n = 4,342)

None (n = 907)

Unadjusted OR (95 % CI)

Only cholinesterase inhibitor (ChEI)

Reference group

Sex: female [n (%)]

Characteristic

Table 2 Comparison of characteristics of AD patients who received neither ChEIs nor NMDA (n = 907, none) as the reference group, with those under treatment with either ChEI (n = 4,342; ChEI?) or NMDA antagonist (n = 438; NMDA antagonist?)

Co-Medication in Alzheimer’s Disease 219

220

Fig. 1 Kernel density plot of age at diagnosis (a) and MMSE score (b) in AD patients who use ChEIs (black curves) or NMDA antagonist (dotted curves) or none of these (grey) in the SveDem registry, which shows higher distribution of older AD patients with lower MMSE score on NMDA antagonists compared with ChEI treatment. AD Alzheimer’s disease, MMSE Mini-Mental State Examination, ChEIs cholinesterase inhibitors, NMDA antagonist Nmethyl-D-aspartate receptor antagonist

antagonist? group compared with the ChEIs? group [adjusted OR 1.36 (95 % CI 1.09–1.70)]. However, no significant difference was observed, neither in co-medication with antipsychotics and cardiovascular drugs, nor in the prevalence of IDU between the NMDA antagonist? and ChEIs? groups after statistical adjustment for the confounders.

4 Discussion The SveDem registry covered up to 93 % of Swedish patients with dementia in memory clinics during 2011, with the most common diagnosis being AD [13]. However, the rate of coverage is almost 14 % for the entire incident population of AD throughout Sweden each year. The incidence rate of dementia in Sweden is estimated to be about 24,000 new cases per year [20], of which 40 % is AD when mixed cases are excluded, resulting in 9,600 incident cases each year and almost 43,200 patients during our study period (4.5 years). Of these, we have studied 5,900 newly diagnosed AD cases, which led to a coverage rate of almost 14 %; however, this figure does not take into account where the patients were diagnosed. In SveDem, the majority of the newly diagnosed dementia patients,

S.-M. Fereshtehnejad et al.

including those with AD, are registered. Therefore, the coverage figure is difficult to discuss in the context of the whole country. Nevertheless, it seems clear that data from our study is only relevant for those AD patients diagnosed in specialized units/memory clinics in Sweden. Up to 2011, most registered cases were from specialized units (memory clinics) rather than other sources such as primary care units and nursing homes where most patients are visited by general physicians. These points must be taken into account while generalizing the findings of our investigation. This study on medications used in AD is an observational investigation reflecting clinical practice and quality of care in a large sample of AD patients with mild disease. The percentage of AD patients using ChEIs in this large AD cohort is similar to a previous report on ChEI use in a smaller sample [17]. We showed that ChEIs were administered to younger AD patients with less severe cognitive impairment. In contrast, the NMDA antagonist memantine is prescribed to older AD patients with significantly lower MMSE score. These results are in line with the recommendations to use ChEIs in mild to moderate AD, and NMDA antagonists in moderate to severe AD [21]. Recently, a very large Canadian registry including more than 104,000 home-care clients with dementia assessed the factors associated with dementia pharmacotherapy [22]. Compared to no active medication, they showed that age greater than 64 years and moderate to severe cognitive impairment were among the factors most strongly associated with the use of ChEI monotherapy [22]; however, it must be noted that their studied population included different types of dementia (not only AD) and the comparison group that was used as the reference arm of the study was also different from our report. We also evaluated use of other medications among AD patients in the SveDem database. The frequency of concurrent use of C3 psychotropic medications was quite low (less than 3 %), which probably reflects that the studied AD patients were in the early stages of disease where neuro-psychiatric symptoms are less common. Furthermore, it may confirm the adherence to national guidelines, which advocates the avoidance of IDU and non-pharmacological treatment as first choice [23]. Similarly, data from the Swedish Prescribed Drug Register has found that, with the exception of antidepressant drugs, the use of psychotropic medications has recently decreased [24]. Another explanation might be that medication data are recorded with regard to drug class and not individual drugs in SveDem. This may lead to an underestimation of the concurrent use of C3 psychotropic medications. Based on real data from the Swedish Prescribed Drug Register with more than 1,300,000 records, institutionalized elderly are more likely to use psychotropic

Co-Medication in Alzheimer’s Disease

221

Table 3 Comparison of characteristics of AD patients who received only ChEIs (n = 4,342, ChEI?) as the reference group, with those under treatment with only memantine as their anti-dementia medication (n = 438, NMDA antagonist?) Characteristic

Unadjusted OR (95 % CI)

Sexa

0.74 (0.61–0.91)

Age

1.08 (1.06–1.09)

MMSE score Place of residenceb

0.89 (0.87–0.91) 0.47 (0.31–0.72)

Co-resident Day care

Unadjusted p-Value

Adjusted OR (95 % CI)

Adjusted p-Value

0.003*

0.62 (0.49–0.77)

\0.001**

\0.001**

1.07 (1.05–1.08)

\0.001**

\0.001** \0.001**

0.89 (0.88–0.91) 1.04 (0.63–1.74)

\0.001** 0.871

1.06 (0.87–1.30)

0.558

0.89 (0.70–1.13)

0.348

1.81 (1.08–3.01)

0.022*

1.07 (0.60–1.89)

0.822

Home care

2.02 (1.62–2.51)

\0.001**

1.13 (0.87–1.48)

0.346

Total number of tests needed to establish diagnosis

0.82 (0.76–0.88)

\0.001**

0.84 (0.76–0.93)

0.001*

Total number of medications (at start of the dementia work-up)

1.09 (1.06–1.13)

\0.001**

1.07 (1.03–1.11)

\0.001**

Antidepressants

0.90 (0.70–1.15)

0.392

1.03 (0.78–1.35)

0.852

Antipsychotics

1.92 (1.24–2.98)

0.003*

1.21 (0.73–2.00)

0.468

Anxiolytics

1.57 (1.09–2.26)

0.015*

1.29 (0.86–1.94)

0.217

Hypnotics

1.12 (0.84–1.51)

0.440

1.03 (0.74–1.42)

0.875

Cardiovascular drugs

0.196

Medication (at the time of diagnosis)

1.22 (0.99–1.50)

0.066

1.16 (0.93–1.45)

IDU (C3 psychotropic medications)

1.40 (0.76–2.59)

0.277

1.13 (0.54–2.35)

0.751

Polypharmacy (C5 drugs)

1.56 (1.28–1.92)

\0.001**

1.36 (1.09–1.70)

0.008*

Adjustment is performed considering age, sex, MMSE score, total number of medications (as a proxy for co-morbidities), and the referral center (to assess cluster effect) as the covariate list in the binary logistic regression model. The group who received only ChEIs (ChEIs?) is considered as the reference one for all OR calculations AD Alzheimer’s disease, SD standard deviation, OR odds ratio, MMSE Mini-Mental State Examination, ChEIs cholinesterase inhibitors, NMDA antagonist N-methyl-D-aspartate receptor antagonist, IDU inappropriate drug use a

The reference condition is male sex

b

The reference condition is living in places other than their own home

* p \ 0.05, ** p \ 0.001

medications compared with community-dwelling subjects [25]. Our patients were mostly residing in their own home, a fact that can explain the low rate of IDU in the SveDem data. The Swedish Prescribed Drug Register has also shown that people aged 75 years and older use, on average, 5.4 prescription drugs per person [9] and those aged 80 years and over receive 5.7 [24] prescription drugs per person. This polypharmacy may be even more problematic among frail AD patients where cognitive impairment increases the risk of misadministration of drugs. By definition (use of at least five drugs), polypharmacy was reported in almost 34 % of the AD patients in our study, which may be worrying given that increased medication burden can lead to adverse outcomes among these frail patients [26]. On the other hand, the polypharmacy may also be an effect of increasing age, where it is well know that comorbidity increases and, subsequently, the number of drugs needed to treat these disorders [27, 28]. In another large study on the Swedish Prescribed Drug Register, it has recently been shown that 39 % of community-dwelling elderly are at risk of polypharmacy; this feature is even as high as 76 % among institutionalized older subjects in

Sweden [25]. In the SveDem registry, a small proportion (\5 %) of AD patients were living in places other than their own homes, which is one reason to explain similar polypharmacy features as that of the community-dwelling elderly, from the census data of the Swedish Prescribed Drug Register. Interestingly, we found that co-medication with antipsychotic drugs was almost 45 % less common in AD patients taking ChEIs than among those without any antidementia drug. Similarly, the rate of anxiolytic medication was 38 % lower in the ChEI? subgroup of patients with AD. The total number of drugs was also lower in AD patients who received ChEIs. One can speculate that AD patients on ChEIs are less frail, with higher MMSE score and younger age. However, the use of less antipsychotic and anxiolytic drugs in the ChEI? group also remained significant after adjustment for age, total number of medications, MMSE score, and gender. In the large populationbased study of community-dwelling elderly receiving home-care services in Canada, Maxwell et al. [22] also showed a lower rate for the administration of anxiolytic medication in dementia patients undergoing treatment with

222

ChEIs. Conversely, their findings showed that dementia patients who received monotherapy with ChEIs were more likely to use an antipsychotic compared with those under no pharmacotherapy [22]. Some reports claim that concomitant use of ChEIs and antipsychotics may increase the risk of extrapyramidal symptoms by deregulating the acetylcholine/dopamine balance in the striatum [29, 30]. However, the co-administration of rivastigmine and risperidone in an open-labeled clinical trial showed no clinically relevant adverse interactions [31]. On the other hand, in a cohort study by Rochon et al. [32], a significantly higher number of serious adverse events were developed in elderly dementia patients receiving either typical or atypical antipsychotics. Another meta-analysis on 15 clinical trials showed a slightly higher mortality rate in dementia patients receiving atypical antipsychotics [33]. Potentially, many of these patients were also undergoin treatment with dementia medications such as ChEIs. However, several studies have also shown positive effects of ChEIs on neuropsychiatric symptoms, although the primary outcome measure has been cognition [34–36]. We have recently shown that the ChEI galantamine and the atypical antipsychotic risperidone are equally effective in treating neuropsychiatric (unless psychotic) symptoms, where the ChEI with its more benign safety profile would be preferred [37]. These data highlight the need for caution in the concomitant use of antipsychotics with dementia medication in AD patients, which is in line with our findings. In addition, more than one-fifth of the AD patients used antidepressants, reflecting how common depression is in the mild stage of AD. However, this high rate of antidepressant prescription could also stem from other psychiatric features such as anxiety and aggression in people with AD. Similar findings have also been found in other studies where almost 23 % of AD patients had minor depression [38]. Recently, Tifratene et al. [39] also showed that, in a large national databank of French patients, antidepressants were the most commonly prescribed psychotropic agent (26 %) among AD patients receiving antiAlzheimer’s medication Furthermore, our data from the SveDem registry also showed that 57 % of the AD patients used cardiovascular drugs. Cardiovascular disease is common in AD and cardiovascular drugs can have positive effects on functional decline among these patients [40]. Recent data from our group showed that the risk for myocardial infarction is reduced in AD patients treated with ChEIs [41]. The current study has some limitations, including the cross-sectional design that makes it impossible to establish causal relationships. As SveDem is a naturalistic registry, no in-depth information is available on either depressive or psychotic symptoms among recruited patients. Detailed information on different types of medications, such as first-

S.-M. Fereshtehnejad et al.

or second-generation neuroleptics, benzodiazepines or nonbenzodiazepine anxiolytics and hypnotics, selective serotonin reuptake inhibitors (SSRIs) or tricyclic antidepressants, are not available in the SveDem registry. Moreover, as all patients in the study were receiving care from memory clinics, there is a possible risk of selection bias of AD patients. In specialized units in Sweden, AD patients are usually visited by either geriatricians or psychiatrists with higher updated knowledge on dementia care, which increases the likelihood of adherence to national guidelines. Having recruited patients from several centers (memory clinics), there could be a chance for the risk of cluster bias due to the different quality of care in each center. Therefore, we entered the variable ‘referral center’ as a potential confounder in all multivariate analysis and no cluster effect was observed. The majority of patients in SveDem were in an early stage of AD at the time of diagnosis. The mean MMSE score of 21.6 (IQR 19–25) reflected an overall mild to moderate stage of AD in our sample. Hence, our findings are not applicable for moderate to severe AD. This is probably one reason for the low proportion (4.2 %) who received antipsychotic agents. Data from the ICTUS cohort has shown that more than 30 % of the AD patients were receiving treatment with psychotropic medication [42]. The Swedish national guidelines on the treatment of neuropsychiatric symptoms advocate drug treatment as the last resort [4], which may reflect the low use in mild AD in our study. Even though registry-based studies such as SveDem lack the benefits of randomized clinical trials to show causal inference, the real-world nature of these registries is valuable for policy making, clinical assessment of medications, and the real benefits of these. In addition, the large sample size of AD patients in the SveDem registry makes these findings unique and highly generalizable to earlystage AD patients in memory clinics.

5 Conclusion The results of our observational investigation show that AD patients diagnosed in memory clinics are mostly in the early stage of the disease. Those who were treated with ChEIs are younger, with a higher MMSE score, and treated with fewer medications than those who do not receive ChEIs. Those taking ChEIs also use about 45 % fewer antipsychotic drugs. This difference remains significant even after adjustment for age, gender, and cognitive level. Hence, our findings further support the benefits of ChEIs among AD patients in clinical practice, and strengthen the recommendation for use of these drugs in national AD guidelines. Although previous reports on quality of treatment have shown that use of anti-dementia drugs generally

Co-Medication in Alzheimer’s Disease

agree with Swedish guidelines [17, 43], more attention should be paid to co-medication with psychotropics and cardiovascular drugs in AD patients. Noteworthy, this study was based upon data collected from specialized units where the chance of adherence to national guidelines could be higher compared with primary care units. This makes it eligible to generalize these findings mostly to the memory clinics rather than the other sources. Our findings provide some preliminary and fundamental data to build upon for future studies regarding outcomes from anti-dementia drugs and co-medication in AD patients in clinical practice.

223

13.

14.

15.

16. Acknowledgments Seyed-Mohammad Fereshtehnejad, Kristina Johnell and Maria Eriksdotter have no conflicts of interest to declare in relation to the content of this paper. The authors are grateful to SveDem (http://www.svedem.se) for providing data for this study, as well as many thanks to all participants in SveDem (patients, caregivers, and staff). This study was supported financially by the Swedish Brain Power network, the Swedish Association of Local Authorities and Regions, and the Swedish Research Council.

17.

18.

19.

References 1. Henderson AS, Jorm A. Definition and epidemiology of dementia: a review. In: Maj M, Sartorius N, editors. Dementia. Chichester: Wiley; 2000. 2. Muayqil T, Camicioli R. Systematic review and meta-analysis of combination therapy with cholinesterase inhibitors and memantine in Alzheimer’s disease and other dementias. Dement Geriatr Cogn Dis Extra. 2012;2(1):546–72. 3. National Institute for Health and Clinical Excellence (Great Britain); Social Care Institute for Excellence (Great Britain); National Collaborating Centre for Mental Health (Great Britain). Dementia: supporting people with dementia and their carers in health and social care. NICE clinical guideline 42. London: National Institute for Health and Clinical Excellence; 2006. 4. The Swedish National Board of Health and Welfare. National guidelines for treatment and care in dementia, 2010. 5. Birks J. Cholinesterase inhibitors for Alzheimer’s disease. Cochrane Database Syst Rev 2006;(1):CD005593. 6. Rainer M, Wuschitz A, Jagsch C, Erb C, Chirikdjian JJ, Mucke HA. Memantine in moderate to severe Alzheimer’s disease: an observational post-marketing study. J Neural Transm. 2011;118(8):1255–9. 7. Jain KK. Evaluation of memantine for neuroprotection in dementia. Expert Opin Investig Drugs. 2000;9(6):1397–406. 8. Defilippi JL, Crismon ML. Drug interactions with cholinesterase inhibitors. Drugs Aging. 2003;20(6):437–44. 9. Johnell K, Fastbom J, Rosen M, Leimanis A. Inappropriate drug use in the elderly: a nationwide register-based study. Ann Pharmacother. 2007;41(7):1243–8. 10. Montastruc F, Gardette V, Cantet C, Piau A, Lapeyre-Mestre M, Vellas B, et al. Potentially inappropriate medication use among patients with Alzheimer disease in the REAL.FR cohort: be aware of atropinic and benzodiazepine drugs! Eur J Clin Pharmacol. 2013;69(8):1589–97. 11. Epstein NU, Saykin AJ, Risacher SL, Gao S, Farlow MR. Differences in medication use in the Alzheimer’s disease neuroimaging initiative: analysis of baseline characteristics. Drugs Aging. 2010;27(8):677–86. 12. Haasum Y, Fastbom J, Johnell K. Institutionalization as a risk factor for inappropriate drug use in the elderly: a Swedish

20.

21.

22.

23.

24. 25.

26.

27.

28.

29.

30.

nationwide register-based study. Ann Pharmacother. 2012;46(3):339–46. Religa D, Spangberg K, Wimo A, Edlund AK, Winblad B, Eriksdotter-Jonhagen M. Dementia diagnosis differs in men and women and depends on age and dementia severity: data from SveDem, the Swedish Dementia Quality Registry. Dement Geriatr Cogn Disord. 2012;33(2–3):90–5. Wimo A, Religa D, Spa˚ngberg K, Edlund AK, Winblad B, Eriksdotter M. Costs of diagnosing dementia in Sweden: results from SveDem, the Swedish Dementia Registry. Int J Geriatr Psychiatry. 2013;28(10):1039–44. Folstein MF, Folstein SE, McHugh PR. ‘‘Mini-mental state’’. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975;12(3):189–98. World Health Organization. International statistical classification of diseases and related health problems, 10th ed. (ICD-10). Geneva: World Health Organization; 2009. Johnell K, Religa D, Eriksdotter M. Differences in drug therapy between dementia disorders in the Swedish Dementia Registry: a nationwide study of over 7,000 patients. Dement Geriatr Cogn Disord. 2013;35(5–6):239–48. Johnell K, Fastbom J. Antiepileptic drug use in community-dwelling and institutionalized elderly: a nationwide study of over 1,300,000 older people. Eur J Clin Pharmacol. 2011;67(10):1069–75. Gnjidic D, Hilmer SN, Blyth FM, Naganathan V, Waite L, Seibel MJ, McLachlan AJ, Cumming RG, Handelsman DJ, Le Couteur DG. Polypharmacy cutoff and outcomes: five or more medicines were used to identify community-dwelling older men at risk of different adverse outcomes. J Clin Epidemiol. 2012;65(9): 989–95. Skoldunger A, Wimo A, Johnell K. Net costs of dementia in Sweden: an incidence based 10 year simulation study. Int J Geriatr Psychiatry. 2012;27(11):1112–7. Jones RW. A review comparing the safety and tolerability of memantine with the acetylcholinesterase inhibitors. Int J Geriatr Psychiatry. 2010;25(6):547–53. Maxwell CJ, Vu M, Hogan DB, Patten SB, Jantzi M, Kergoat MJ, Jette N, Bronskill SE, Heckman G, Hirdes JP. Patterns and determinants of dementia pharmacotherapy in a population-based cohort of home care clients. Drugs Aging. 2013;30(7):569–85. The Swedish Medical Products Agency. Treatment of behavioral and psychological symptoms of dementia (BPSD). Vol. 5; The Swedish Medical Products Agency; 2008. Holm L. Health care report 2009. Stockholm: National Board of Health and Welfare; 2010. Johnell K, Fastbom J. Comparison of prescription drug use between community-dwelling and institutionalized elderly in Sweden. Drugs Aging. 2012;29(9):751–8. Lau DT, Mercaldo ND, Shega JW, Rademaker A, Weintraub S. Functional decline associated with polypharmacy and potentially inappropriate medications in community-dwelling older adults with dementia. Am J Alzheimers Dis Other Demen. 2011;26(8):606–15. Haider SI, Johnell K, Weitoft GR, Thorslund M, Fastbom J. The influence of educational level on polypharmacy and inappropriate drug use: a register-based study of more than 600,000 older people. J Am Geriatr Soc. 2009;57(1):62–9. Wastesson JW, Parker MG, Fastbom J, Thorslund M, Johnell K. Drug use in centenarians compared with nonagenarians and octogenarians in Sweden: a nationwide register-based study. Age Ageing. 2012;41(2):218–24. Bentue-Ferrer D, Tribut O, Polard E, Allain H. Clinically significant drug interactions with cholinesterase inhibitors: a guide for neurologists. CNS Drugs. 2003;17(13):947–63. Johnell K, Fastbom J. Concurrent use of anticholinergic drugs and cholinesterase inhibitors: register-based study of over 700,000 elderly patients. Drugs Aging. 2008;25(10):871–7.

224 31. Weiser M, Rotmensch HH, Korczyn AD, Hartman R, Cicin-Sain A, Anand R. A pilot, randomized, open-label trial assessing safety and pharmacokinetic parameters of co-administration of rivastigmine with risperidone in dementia patients with behavioral disturbances. Int J Geriatr Psychiatry. 2002;17(4):343–6. 32. Rochon PA, Normand SL, Gomes T, Gill SS, Anderson GM, Melo M, Sykora K, Lipscombe L, Bell CM, Gurwitz JH. Antipsychotic therapy and short-term serious events in older adults with dementia. Arch Intern Med. 2008;168(10):1090–6. 33. Schneider LS, Dagerman KS, Insel P. Risk of death with atypical antipsychotic drug treatment for dementia: meta-analysis of randomized placebo-controlled trials. JAMA. 2005;294(15): 1934–43. 34. Feldman H, Gauthier S, Hecker J, Vellas B, Subbiah P, Whalen E. A 24-week, randomized, double-blind study of donepezil in moderate to severe Alzheimer’s disease. Neurology. 2001;57(4): 613–20. 35. Raskind MA, Peskind ER, Wessel T, Yuan W. Galantamine in AD: a 6-month randomized, placebo-controlled trial with a 6-month extension. The Galantamine USA-1 Study Group. Neurology. 2000;54(12):2261–8. 36. Rodda J, Morgan S, Walker Z. Are cholinesterase inhibitors effective in the management of the behavioral and psychological symptoms of dementia in Alzheimer’s disease? A systematic review of randomized, placebo-controlled trials of donepezil, rivastigmine and galantamine. Int Psychogeriatr. 2009;21(5): 813–24. 37. Freund-Levi Y, Jedenius E, Tysen-Ba¨ckstro¨m A-C, La¨rksa¨ter M, Wahlund LO, Eriksdotter M. Galantamine versus risperidone

S.-M. Fereshtehnejad et al.

38.

39.

40.

41.

42.

43.

treatment of neuropsychiatric symptoms in patients with probable dementia, an open randomized trial. Am J Geriatr Psychiatry. Epub 10 Sep 2013 Vida S, Des Rosiers P, Carrier L, Gauthier S. Prevalence of depression in Alzheimer’s disease and validity of research diagnostic criteria. J Geriatr Psychiatry Neurol. 1994;7(4):238–44. Tifratene K, Duff FL, Pradier C, Quetel J, Lafay P, Schuck S, Benzenine E, Quantin C, Robert P. Use of drug treatments for Alzheimer’s disease in France: a study on a national level based on the National Alzheimer’s Data Bank (Banque Nationale Alzheimer). Pharmacoepidemiol Drug Saf. 2012;21(9):1005–12. Rosenberg PB, Mielke MM, Tschanz J, Cook L, Corcoran C, Hayden KM, Norton M, Rabins PV, Green RC, Welsh-Bohmer KA, et al. Effects of cardiovascular medications on rate of functional decline in Alzheimer disease. Am J Geriatr Psychiatry. 2008;16(11):883–92. Nordstro¨m P, Religa D, Wimo A, Winblad B, Eriksdotter M. The use of cholinesterase inhibitors and the risk of myocardial infarction: a nationwide cohort study in subjects with Alzheimer’s disease. Eur Heart J. 2013;34(33):2585–91. Reynish E, Cortes F, Andrieu S, Cantet C, Olde Rikkert M, Melis R, et al. The ICTUS Study: a prospective longitudinal observational study of 1,380 AD patients in Europe. Study design and baseline characteristics of the cohort. Neuroepidemiology. 2007;29(1–2):29–38. Sonde L, Johnell K. Is drug treatment for dementia followed up in primary care? A Swedish study of dementia clinics and referring primary care centres. PLoS One. 2013; 8(2):e57161.