Yokukansan in the Treatment of Behavioral and Psychological Symptoms of Dementia: An Updated Meta-Analysis of Randomized Controlled Trials

Abstract

Background: Previous clinical studies found that yokukansan has a therapeutic effect on behavioral and psychological symptoms of dementia (BPSD) in dementia patients.

Objective: To perform an updated meta-analysis of randomized controlled trials (RCTs) testing yokukansan for patients with BPSD.

Methods: Primary efficacy and safety endpoints were BPSD total scores and all-cause discontinuation, respectively. Secondary outcomes were BPSD subscales, cognitive function scores [Mini-mental state examination (MMSE)], activities of daily living (ADL) scores, discontinuation due to adverse events (AEs), and incidences of AEs.

Results: Five RCTs with 381 patients with BPSD were included. Compared with controls [placebo+usual care (UC)], yokukansan significantly decreased BPSD total scores [standardized mean difference (SMD) = –0.32, 95% confidence interval (CI) = –0.53 to –0.11, p = 0.003, I² = 0%, N = 5 studies, n = 361]. Yokukansan was more efficacious in reducing BPSD subscale scores (delusions: SMD = –0.51, 95% CI = –0.98 to –0.04, hallucinations: SMD = –0.54, 95% CI = –0.96 to –0.12, agitation/aggression: SMD = –0.37, 95% CI = –0.60 to –0.15) than placebo+UC. However, yokukansan was not superior to placebo+UC for BPSD total as well as any subscales scores only in Alzheimer’s disease patients. Compared with UC, yokukansan treatment improved ADL scores (SMD = –0.32, 95% CI = –0.62 to –0.01). MMSE scores did not differ between the yokukansan and placebo+UC treatment groups. No significant differences were found in all-cause discontinuation, discontinuation due to AEs, and incidences of AEs between yokukansan and placebo+UC treatments.

Conclusions: Our results suggest that yokukansan is beneficial for the treatment of patients with BPSD and is well-tolerated; it was not beneficial for BPSD total and any subscale scores only in Alzheimer’s disease patients.

Keywords

Alzheimer’s disease behavioral and psychological symptoms of dementia dementia meta-analysis yokukansan

INTRODUCTION

Behavioral and psychological symptoms of dementia (BPSD), characterized by agitation, aberrant motor behavior, anxiety, elation, irritability, depression, apathy, disinhibition, delusions, hallucinations, sleep disturbances, and appetite changes, are some of the most important problems in the treatment of dementia [1, 2]. Approximately 80–90% of patients with dementia have at least one event of BPSD over the course of their illness [2, 3]. BPSD have been associated with significant distress and poor quality of life of not only the patients with dementia but also their caregivers [2 , 5].

Yokukansan (Tsumura TJ-54) is a traditional Japanese herbal medicine (kampo) that has been approved in Japan to treat patients with nervousness, insomnia, and night terrors and temper tantrums in children [6]. Yokukansan is composed of the following seven medicinal herbs: Atractylodis lanceae rhizoma, Poria sclerotium, Cnidii rhizoma, Japanese angelicae root, Bupleuri root, Glycyrrhiza, and Uncaria hook [6]. The recent studies reported that Glycyrrhiza and Uncaria hook might possess an ameliorative effect against amyloid-β oligomer-induced neuronal apoptosis through the suppression of caspase-3 activation [7, 8]. Another study also reported that Japanese angelicae extract possessed the activity to prevent the neurotoxicity induced by amyloid-β-associated oxidative stress, implying that Japanese angelicae might have a potential role in the prevention of Alzheimer’s disease [9]. Patients with dementia presenting BPSD have been reported to present alterations in the transmission of glutamate, gamma-aminobutyric acid (GABA) and/or serotonin [10, 11]. Yokukansan has been found to have a neuroprotective effect: it improves glutamate uptake and inhibits glutamate-induced neuronal death [12]. Yokukansan has also been reported to have a partial agonistic effect on serotonin (5-HT) 1A receptors, which regulate 5-HT neuronal firing [13]. Therefore, this herbal medicine may impact other 5-HT autoreceptors and heteroreceptors that mediate serotonin in several brain regions and are related to several neural transmission systems, including glutamate, GABA, and dopamine [13]. Additionally, yokukansan has been shown to influence the expression of GABA receptors in mice [14]. Recently, several clinical studies found that yokukansan has a therapeutic role in treating patients with BPSD [15 –18]. Moreover, Matsuda et al. [19] reported a meta-analysis of yokukansan for BPSD in 2013 and suggested that yokukansan has a beneficial effect on the neuropsychiatric inventory (NPI) and on activities of daily living (ADL) scores. Moreover, yokukansan seems to be a well-tolerated treatment [NPI: weighted mean difference (WMD) = –7.20, 95% confidence interval (CI) = –11.4 to –2.96, p = 0.0009, I² = 0%, N = 4 studies, n = 225 patients, ADL: standardized mean difference (SMD) = –0.32, 95% CI = –0.62 to –0.01, p = 0.04, I² = 0%, N = 4 studies, n = 170 patients] [19]. However, the authors pointed out a series of limitations (small sample size, none of the studies included a placebo-controlled group, and heterogeneity of dementia) of the studies included in the meta-analysis [19]. Therefore, we aimed to prepare an updated meta-analysis of yokukansan for the treatment of BPSD. Additionally, we performed a subgroup meta-analysis as per the type of dementia.

MATERIALS AND METHODS

This meta-analysis was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines [20]. We performed a systematic literature review according to the PICO strategy [Patients: patients with BPSD; Intervention: yokukansan; Comparator: placebo or usual care; Outcome: BPSD, ADL, cognitive function, discontinuation rate, and individual adverse events].

Inclusion criteria, search strategy, data extraction, and outcome measures

We only included blinded and non-blinded randomized controlled trials (RCTs) of yokukansan in patients with BPSD. To identify relevant studies, we searched major healthcare databases (PubMed, the Cochrane Library databases, and PsycINFO) and clinical trial registries (ClinicalTrials.gov, ISRCTN, the WHO portal). There were no language restrictions, and we considered all studies published up to April 20, 2016. We used the following keywords for the search: “dementia” OR “Alzheimer’s” OR “Alzheimer” OR “Lewy” AND “Yokukansan” OR “Yigansan”. Additional eligible studies were retrieved from the reference lists of primary articles and relevant reviews.

The first two authors of this meta-analysis (S.M. and T.K.) scrutinized the patient inclusion and exclusion criteria of the identified studies. If data required for this meta-analysis were missing, the first and/or corresponding authors were contacted for additional information, including endpoint scores. Finally, the first two authors (S.M. and T.K.) independently extracted, reviewed, and entered the data into Review Manager (Version 5.3 for Windows, Cochrane Collaboration, http://ims.cochrane.org/revman).

Data synthesis and statistical analysis

Each reported outcome measure was used in at least two out of the five included studies. The primary efficacy measures were BPSD total scores. BPSD were assessed using NPI [21] or NPI questionnaire [22]. The primary safety measure was all-cause discontinuation. Secondary outcomes were BPSD subscales scores (delusions, hallucinations, agitation/aggression, dysphoria, anxiety, euphoria, apathy, disinhibition, irritability/lability, aberrant motor be-havior, night-time disturbance, and eating disturbance), ADL scores (as measured by the Barthel index [23] or the Disability Assessment of Dementia [24]), cognitive function scores [as measured by the Mini-Mental State Examination (MMSE)] [25], and discontinuation rates due to adverse events. Additionally, we pooled the incidences of individual adverse events.

We based our analyses on intent-to-treat (ITT) or modified ITT data (i.e., at least 1 dose or at least 1 follow-up assessment). To combine studies, the random effects model by DerSimonian and Laird was used [26]. The random effects model is more conservative than the fixed-effects model and produces wider CIs. For continuous data, we used the WMD and 95% CIs when data, measured using the same scale, were being combined (i.e., MMSE). To combine the data measured using different scales (i.e., BPSD), we used the SMD, combining effect size (Hedges’ g) data and 95% CIs. For dichotomous data, the risk ratio (RR) was estimated with 95% CIs. We explored study heterogeneity using the I² statistic, where values of 50% or higher were considered the evidence of considerable heterogeneity [27]. In cases with I²≥50% for primary outcome measures, we planned to conduct sensitivity analyses to determine the reasons for heterogeneity. However, these analyses were not conducted because no significant study heterogeneity was found for the chosen primary outcomes. However, because we performed a meta-analysis including various types of dementia in the meta-analysis, we performed a subgroup analysis for efficacy outcomes divided by the type of dementia [“studies that included only subgroups of patients with Alzheimer’s disease” versus “studies that included several subgroups of patients with various types of dementia (Alzheimer’s disease, vascular dementia, and dementia with Lewy bodies)”]. We also assessed the methodological quality of the articles included based on Cochrane risk of bias criteria (Cochrane Collaboration; http://www.cochrane.org/). We did not assess the presence of publication bias because only five studies were included in the meta-analysis.

RESULTS

Study characteristics

The primary search yielded a total of 244 studies, of which 149 were duplicates (Fig. 1). We excluded 80 studies after reviewing the title and abstract and other 10 studies were excluded after full-text reviews because these were single-arm studies (N = 9) and a review article (N = 1). Five RCTs testing yokukansan for BPSD treatment were finally included [15–18 , 28].

The characteristics of the trials included in our study are shown in Table 1. The five RCTs included 381 patients with BPSD [15–18 , 28]. The studies included three on yokukansan for Alzheimer’s disease [17 , 28], one on yokukansan for Alzheimer’s disease, vascular dementia, or dementia with Lewy bodies [15], and one on Alzheimer’s disease or dementia with Lewy bodies [16]. Four trials lasted for 4 weeks [15–17 , 28], and 1 lasted for 12 weeks [18]. The study sample size ranged from 15 to 145 patients. The mean age of the pooled study population was 79 y. Three studies were sponsored by the pharmaceutical industry [17 , 28]. All studies were conducted in Japan. The Iwasaki study [15] used tiapiride for some patients during the study (yokukansan group = 0%, usual care group = 44%). The Monji study [18] used sulpiride for all patients during the study. The Okahara study [17] used donepezil for all patients and risperidone for some patients during the study (yokukansan group = 3.45%, usual care group = 3.13%). The Furukawa study [28] used anti-dementia drugs for some patients during the study (yokukansan group = 71%, usual care group = 74%).

We evaluated the methodological quality of all studies using the Cochrane risk of bias criteria (Supplementary Figure 1). One study did not mention detailed information regarding sequence generation [17]. Two studies did not mention detailed information regarding allocation concealment [15, 18]. One study was double-blind, parallel, and placebo-controlled [28], one was open-label, observer-blind, parallel, and non-placebo- (i.e., usual care) controlled [15], two were open-label, parallel, and non-placebo-controlled [17, 18], and one was open-label, non-placebo-controlled, crossover [16]. All studies assessed ITT or modified ITT populations [15–18 , 28]. One study did not report the results of all treatment arms as primary outcome [18].

BPSD total scores (primary outcome for efficacy)

Compared with control treatments, yokukansan significantly decreased the BPSD total scores (SMD = –0.32, 95% CI = –0.53 to –0.11, p = 0.003, I² = 0%, N = 5 studies, n = 361 patients; Fig. 2). There was no significant heterogeneity in BPSD total scores among the studies (I² = 0%).

When divided into “studies that included only subgroups of patients with Alzheimer’s disease” and “studies that included several subgroups of patients with various types of dementia”, yokukansan significantly decreased BPSD total scores than that by control treatments in “studies that included several subgroups of patients with various types of dementia” (SMD = –0.53, 95% CI = –0.86 to –0.21, p = 0.001, I² = 0%, N = 2 studies, n = 150 patients). However, no significant differences in BPSD total scores were found in the “studies that included only subgroups of patients with Alzheimer’s disease”.

All-cause discontinuation (primary outcome for safety)

The all-cause discontinuation was similar between the yokukansan and control treatment groups (RR = 1.2, 95% CI = 0.51 to 2.83, p = 0.67, I² = 0%, N = 5, n = 381; Fig. 3). There was no significant heterogeneity in the discontinuation rate among the studies (I² = 0%).

BPSD subscale scores

In the subscale of BPSD subscale scores, yokukansan treatment was superior to control treatments with respect to the improvement of delusions (SMD = –0.51, 95% CI = –0.98 to –0.04, p = 0.03, I² = 67%, N = 5 studies, n = 284 patients; Supplementary Figure 2), hallucinations (SMD = –0.54, 95% CI = –0.96 to –0.12, p = 0.01, I² = 48%, N = 5 studies, n = 248 patients; Supplementary Figure 2), agitation/aggression subscale scores (SMD = –0.37, 95% CI = –0.60 to –0.15, p = 0.001, I² = 0%, N = 5 studies, n = 311 patients; Supplementary Figure 2).

When divided into “studies that included only subgroups of patients with Alzheimer’s disease” and “studies that included several subgroups of patients with various types of dementia”, yokukansan significantly decreased delusions (SMD = –0.66, 95% CI = –1.32 to –0.00, p = 0.05, I² = 63%, N = 2 studies, n = 102 patients; Supplementary Figure 2), hallucinations (SMD = –0.84, 95% CI = –1.29 to –0.38, p = 0.0003, I² = 0%, N = 2 studies, n = 81 patients; Supplementary Figure 2), agitation/aggression (SMD = –0.53, 95% CI = –0.90 to –0.16, p = 0.005, I² = 0%, N = 2 studies, n = 118 patients; Supplementary Figure 2), and night-time disturbance (SMD = –0.63, 95% CI = –1.18 to –0.07, p = 0.03, I² = not applicable, N = 1 study, n = 52 patients; Supplementary Figure 2) subscale scores compared to control treatments in the “studies that included several subgroups of patients with various types of dementia”. However, compared with control treatments, yokukansan marginally decreased agitation/aggression subscale scores in the “studies that included only subgroups of patients with Alzheimer’s disease” (SMD = –0.28, 95% CI = –0.56 to 0.01, p = 0.06, I² = 0%, N = 3 studies, n = 193 patients; Supplementary Figure 2) and marginally increased night-time disturbance subscale scores (SMD = 0.28, 95% CI = –0.05 to 0.60, p = 0.09, I² = 0%, N = 2 studies, n = 150 patients; Supplementary Figure 2). No significant differences in other BPSD subscales were found in the “studies that included only subgroups of patients with Alzheimer’s disease” (Supplementary Figure 2).

ADL scores

Additionally, compared with control treatments, yokukansan significantly decreased ADL scores (SMD = –0.32, 95% CI = –0.62 to –0.01, p = 0.04, I² = 0%, N = 4 studies, n = 170 patients; Supplementary Figure 2).

MMSE scores

No significant differences were found between groups in the MMSE scores (Supplementary Figure 2).

When divided into “studies that included several subgroups of patients with various types of dementia”, no significant differences in the MMSE scores were found in both subgroups (Supplementary Figure 2).

Safety outcomes

The all-cause discontinuation rates were similar between yokukansan and control treatment groups (Supplementary Figure 2). No significant differences were found between groups in the incidence of serious adverse events, hypokalemia, leg edema, or loss of appetite (Supplementary Figure 2).

DISCUSSION

This study provides an updated, comprehensive meta-analysis of RCTs testing the efficacy of yokukansan in patients with BPSD. The main finding of our study is that compared with control treatments, yokukansan significantly improves BPSD total scores in patients with BPSD. Moreover, compared with control treatments, yokukansan significantly improves delusions, hallucinations, agitation/aggression subscales, and ADL scores in patients with BPSD. Regarding the safety outcomes, there were no significant differences in all-cause discontinuation, discontinuation due to adverse events, and incidences of individual adverse events between yokukansan and control groups. Our results suggest that yokukansan is an efficacious and safe treatment for patients with BPSD. A recent meta-analysis [19] showed similar results to ours. We examined whether there was a difference in efficacy of yokukansan for different types of dementia (Alzheimer’s disease, vascular dementia, and dementia with Lewy bodies). Compared with control treatments, yokukansan showed a trend towards the improvement of agitation/aggression only in patients with Alzheimer’s disease. However, yokukansan was not beneficial for the treatment of overall BPSD in only patients with Alzheimer’s disease patients. Moreover, compared with control treatments, yokukansan showed a trend towards worsening night-time disturbance only in patients with Alzheimer’s disease. Conversely, when performing the meta-analysis with patients with various types of dementia, yokukansan was beneficial for the treatment of overall BPSD. When taking the above evidence together, it was possible that there was a difference in the efficacy of yokukansan among the different types of dementia. Thus, a clinical trial of yokukansan should be performed for each type of dementia.

These conclusions, however, must be considered in light of several study limitations. The main limitation is the small number of included studies. Because a funnel plot is used only if 10 or more studies are included in the meta-analysis, we did not utilize this plot for investigating potential publication bias. Moreover, the individual studies included relatively small patient samples. The second limitation is that the included studies had several risks of bias (Supplementary Figure 1). In particular, there was a high blinding bias because four of the five studies included a non-placebo-controlled group. The third limitation is that all trials included in this meta-analysis were short in duration (4–12 weeks). The fourth limitation is that because most of the studies included in the meta-analysis used concomitant drugs, such as anti-dementia drugs and antipsychotics during the study, we cannot deny the possibility that these drugs could influence the results of all endpoints. The fifth limitation is that all studies were conducted in Japan because Yokukansan has been approved in only Japan. Since our results suggest that yokukansan is beneficial for the treatment of patients with BPSD, further studies using patients with BPSD in other populations are needed to definitively assess this potential benefit. Finally, there are differences across the types of dementia, which is why we chose to include several types of dementia in our meta-analysis.

In conclusion, our results suggest that yokukansan is beneficial for the treatment of patients with BPSD. Furthermore, yokukansan seemed to be well tolerated. However, yokukansan was not beneficial for the treatment of overall BPSD in only patients with Alzheimer’s disease. Moreover, compared with control treatments yokukansan showed a trend towards worsening night-time disturbance only in patients with Alzheimer’s disease. Therefore, because it is possible that there may be a difference in the efficacy of yokukansan between the types of dementia, further clinical trials are needed to assess the efficacy of yokukansan for each type of dementia.

Footnotes

ACKNOWLEDGMENTS

The authors thank Drs. Koh Iwasaki, Nobumitsu Usuda, Hiroyuki Arai, Akira Monji, Katsutoshi Furukawa, and Takuhiro Yamaguchi for providing information for the study.

Dr. Matsunaga had access to all study data and takes full responsibility for its integrity and the accuracy of the analysis. Drs. Matsunaga and Kishi participated in study conception and design, data acquisition, and statistical analysis (including data analysis and interpretation). The manuscript was written by Drs. Matsunaga, Kishi, and Iwata. Dr. Iwata supervised the review.

Authors’ disclosures available online ()

Appendix

References

Fitten

(2006) Management of behavioral and psychiatric symptoms in dementia. J Nutr Health Aging 10, 409.

Cerejeira

, Lagarto

, Mukaetova-Ladinska

(2012) Behavioral and psychological symptoms of dementia. Front Neurol 3, 73.

Stewart

, Hotopf

, Dewey

, Ballard

, Bisla

, Calem

, Fahmy

, Hockley

, Kinley

, Pearce

, Saraf

, Begum

(2014) Current prevalence of dementia, depression and behavioural problems in the older adult care home sector: The South East London Care Home Survey. Age Ageing 43, 562–567.

Tatsumi

, Nakaaki

, Torii

, Shinagawa

, Watanabe

, Murata

, Sato

, Mimura

, Furukawa

(2009) Neuropsychiatric symptoms predict change in quality of life of Alzheimer disease patients: A two-year follow-up study. Psychiatry Clin Neurosci 63, 374–384.

Miyamoto

, Tachimori

, Ito

(2010) Formal caregiver burden in dementia: Impact of behavioral and psychological symptoms of dementia and activities of daily living. Geriatr Nurs 31, 246–253.

Suzuki

, Mitsuhata

, Yuzurihara

, Kase

(2012) Antiallodynic effect of herbal medicine yokukansan on peripheral neuropathy in rats with chronic constriction injury. Evid Based Complement Alternat Med 2012, 953459.

Kanno

, Kawakami

, Iizuka

, Tabuchi

, Mizoguchi

, Ikarashi

, Kase

(2013) Glycyrrhiza and Uncaria Hook contribute to protective effect of traditional Japanese medicine yokukansan against amyloid beta oligomer-induced neuronal death. J Ethnopharmacol 149, 360–370.

Ahn

, Kim

, Sung

, Ha

(2010) Effect of licorice (Glycyrrhiza uralensis fisch) on amyloid-β-induced neurotoxicity in PC12 cells. Food Sci Biotechnol 19, 1391–1395.

Huang

, Lin

, Chiang

(2008) Protective effects of Angelica sinensis extract on amyloid beta-peptide-induced neurotoxicity. Phytomedicine 15, 710–721.

10.

Lanctot

, Herrmann

, Rothenburg

, Eryavec

(2007) Behavioral correlates of GABAergic disruption in Alzheimer’s disease. Int Psychogeriatr 19, 151–158.

11.

Lanctot

, Herrmann

, Mazzotta

(2001) Role of serotonin in the behavioral and psychological symptoms of dementia. J Neuropsychiatry Clin Neurosci 13, 5–21.

12.

Kawakami

, Kanno

, Ueki

, Terawaki

, Tabuchi

, Ikarashi

, Kase

(2009) Neuroprotective effects of yokukansan, a traditional Japanese medicine, on glutamate-mediated excitotoxicity in cultured cells. Neuroscience 159, 1397–1407.

13.

Meltzer

, Sumiyoshi

(2008) Does stimulation of 5-HT(1A) receptors improve cognition in schizophrenia? Behav Brain Res 195, 98–102.

14.

Egashira

, Nogami

, Iwasaki

, Ishibashi

, Uchida

, Takasaki

, Mishima

, Nishimura

, Oishi

, Fujiwara

(2011) Yokukansan enhances pentobarbital-induced sleep in socially isolated mice: Possible involvement of GABA(A)-benzodiazepine receptor complex. J Pharmacol Sci 116, 316–320.

15.

Iwasaki

, Satoh-Nakagawa

, Maruyama

, Monma

, Nemoto

, Tomita

, Tanji

, Fujiwara

, Seki

, Fujii

, Arai

, Sasaki

(2005) A randomized, observer-blind, controlled trial of the traditional Chinese medicine Yi-Gan San for improvement of behavioral and psychological symptoms and activities of daily living in dementia patients. J Clin Psychiatry 66, 248–252.

16.

Mizukami

, Asada

, Kinoshita

, Tanaka

, Sonohara

, Nakai

, Yamaguchi

, Hanyu

, Kanaya

, Takao

, Okada

, Kudo

, Kotoku

, Iwakiri

, Kurita

, Miyamura

, Kawasaki

, Omori

, Shiozaki

, Odawara

, Suzuki

, Yamada

, Nakamura

, Toba

(2009) A randomized cross-over study of a traditional Japanese medicine (kampo), yokukansan, in the treatment of the behavioural and psychological symptoms of dementia. Int J Neuropsychopharmacol 12, 191–199.

17.

Okahara

, Ishida

, Hayashi

, Inoue

, Tsuruta

, Takeuchi

, Yoshimuta

, Kiue

, Ninomiya

, Kawano

, Yoshida

, Noda

, Tomita

, Fujimoto

, Hosomi

, Mitsuyama

(2010) Effects of Yokukansan on behavioral and psychological symptoms of dementia in regular treatment for Alzheimer’s disease. Prog Neuropsychopharmacol Biol Psychiatry 34, 532–536.

18.

Monji

, Takita

, Samejima

, Takaishi

, Hashimoto

, Matsunaga

, Oda

, Sumida

, Mizoguchi

, Kato

, Horikawa

, Kanba

(2009) Effect of yokukansan on the behavioral and psychological symptoms of dementia in elderly patients with Alzheimer’s disease. Prog Neuropsychopharmacol Biol Psychiatry 33, 308–311.

19.

Matsuda

, Kishi

, Shibayama

, Iwata

(2013) Yokukansan in the treatment of behavioral and psychological symptoms of dementia: A systematic review and meta-analysis of randomized controlled trials. Hum Psychopharmacol 28, 80–86.

20.

Moher

, Liberati

, Tetzlaff

, Altman

, PRISMA Group (2009) Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement. BMJ 339, b2535.

21.

Cummings

, Mega

, Gray

, Rosenberg-Thompson

, Carusi

, Gornbein

(1994) The Neuropsychiatric Inventory: Comprehensive assessment of psychopathology in dementia. Neurology 44, 2308–2314.

22.

Kaufer

, Cummings

, Ketchel

, Smith

, MacMillan

, Shelley

, Lopez

, DeKosky

(2000) Validation of the NPI-Q, a brief clinical form of the Neuropsychiatric Inventory. J Neuropsychiatry Clin Neurosci 12, 233–239.

23.

Mahoney

, Barthel

(1965) Functional evaluation: The barthel index. Md State Med J 14, 61–65.

24.

Gelinas

, Gauthier

, McIntyre

, Gauthier

(1999) Development of a functional measure for persons with Alzheimer’s disease: The disability assessment for dementia. Am J Occup Ther 53, 471–481.

25.

Folstein

, Folstein

, McHugh

(1975) “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12, 189–198.

26.

DerSimonian

, Laird

(1986) Meta-analysis in clinical trials. Control Clin Trials 7, 177–188.

27.

Higgins

, Thompson

, Deeks

, Altman

(2003) Measuring inconsistency in meta-analyses. BMJ 327, 557–560.

28.

Furukawa

, Tomita

, Uematsu

, Okahara

, Shimada

, Ikeda

, Matsui

, Kozaki

, Fujii

, Ogawa

, Umegaki

, Urakami

, Nomura

, Kobayashi

, Nakanishi

, Washimi

, Yonezawa

, Takahashi

, Kubota

, Wakutani

, Ito

, Sasaki

, Matsubara

, Une

, Ishiki

, Yahagi

, Shoji

, Sato

, Terayama

, Kuzuya

, Araki

, Kodama

, Yamaguchi

, Arai

(2015) Randomized double-blind placebo-controlled multicenter trial of Yokukansan for neuropsychiatric symptoms in Alzheimer’s disease.. Geriatr Gerontol Int, doi: 10.1111/ggi.12696 [epub ahead of print].