Associations Between Polypharmacy and Gait Speed According to Cognitive Impairment Status: Cross-Sectional Study in a Japanese Memory Clinic

Abstract

Background:

Polypharmacy, usually defined as the use of 5 or more drugs, is associated with reduced quality of life, adverse events, and frailty. Slow gait speed is a component of physical frailty, and some studies have suggested an association between polypharmacy and slow gait speed.

Objective:

We aimed to determine the effects of polypharmacy on the gait difference according to stages of cognitive decline in a cross-sectional study of memory clinic patients.

Methods:

Participants were 431 outpatients aged 65 year or older who were cognitively normal (CN) or had mild cognitive impairment (MCI) or dementia due to Alzheimer’s disease. Participants were divided into a polypharmacy group and a non-polypharmacy group in each group. Multiple regression analysis and logistic analysis were used for data analysis.

Results:

There were 182 patients in the polypharmacy group and 249 patients in the non-polypharmacy group. Multiple regression analysis revealed that gait speed had significant negative associations with number of medications and polypharmacy status in the CN group (β: –0.026 [–0.041 to –0.0018] and –0.128 [–0.022 to –0.0033], respectively) and MCI group (–0.018 [–0.028 to –0.0009] and –0.100 [–0.166 to –0.0034]). Logistic regression analysis also showed that number of medications was associated with slow gait status (< 1 m/s) in the CN group (OR: 1.336 [1.115 to 1.601]) and MCI group (1.128 [1.022 to 1.244]).

Conclusion:

CN and MCI patients with polypharmacy have slower gait speed. Attention should be paid to decreased gait speed in older adults with polypharmacy even when their cognitive function is relatively preserved.

Keywords

Alzheimer’s disease cross-sectional study gait speed older adults outpatient clinic polypharmacy

INTRODUCTION

Older adults with multimorbidity tend to take many kinds of medications. Polypharmacy is usually defined as the use of 5 or more drugs [1]. Several lines of research have indicated that polypharmacy is a serious problem associated with reduced quality of life, adverse events including falls, and unplanned hospitalization [2, 3]. Some studies have found that polypharmacy is associated with higher incidence of frailty [4].

Gait speed is a marker that can be easily measured in many settings including outpatient clinics. Slow gait speed is one of the most important health issues for older adults. In some studies, it has been reported that a decrease in gait speed is a predictive marker of adverse events such as facility admission, falls, and death. Slow gait speed is also one of the defining characteristics of physical frailty. Furthermore, several lines of evidence have found association between slower gait speed and worse cognitive function [5], and parameters of spatiotemporal gait performance deteriorate in parallel with stage of cognitive decline from mild cognitive impairment (MCI) to moderate dementia [6]. Various pathological changes in the brain have been suggested to explain the parallel declines in cognition and gait [7, 8]. In this way, gait speed may reflect the degree of the pathophysiological burden of dementia. This suggests the possibility that the effects of associated factors on gait speed could differ among cognitive stages.

An association between polypharmacy and slow gait speed has been reported in community-dwelling older adults [9]. Polypharmacy is also associated with slow gait in patients with MCI, who have a high risk of progression to dementia [10]. However, the association between polypharmacy and slow gait has not yet been fully elucidated in dementia. Accordingly, we aimed to determine whether the effects of polypharmacy on gait differed according to stages of cognitive decline in a cross-sectional study of memory clinic patients who were cognitively normal (CN) or had MCI or dementia due to Alzheimer’s disease (AD).

MATERIALS AND METHODS

Patient selection

Participants were patients aged 65 year or older who visited the memory clinic at the National Center for Geriatrics and Gerontology between September 2010 and March 2020. For these patients, we used existing information registered as of March 31, 2020 for the Comprehensive Survey for the Frailty Process Analysis of the Elderly.

This study followed the principles of the Declaration of Helsinki. The protocol for follow-up data collection was approved by the ethics committee of Nagoya University Graduate School of Medicine (approval no. 2020-0285).

Data collection

The data analyzed in this study included age, sex, education, body mass index (BMI), number of medications, number of comorbidities, Barthel index (BI) score [11], Mini-Mental State Examination (MMSE) score [12], Geriatric Depression Scale 15 score (GDS15) [13], grip strength, and gait speed.

BMI was calculated using the following formula: BMI = Weight (kg)/[Height (m)]². Medications were identified and reviewed using medical records, and medications prescribed at other facilities were confirmed by referring to the patient medication handbooks. Each active ingredient was counted as one drug. Topical, ocular, on demand, and temporal medications were not included. Unfortunately, data on pharmacological aspects such as drug type, dose, duration of use, and adherence were not collected. Polypharmacy was defined as the use of 5 or more drugs. The number of comorbidities was determined by counting the following diseases: hypertension, dyslipidemia, diabetes, liver disease, heart disease, kidney disease, lung disease, stroke, and malignant tumor. The BI (maximum score 100, with a higher score indicating higher dependence in activities) measures performance for the following 10 variables describing basic activities of daily living and mobility: feeding, bathing, grooming, dressing, bowels, bladder, toilet use, transfers, mobility, and stairs [11].

We measured grip strength in the participant’s dominant hand with a portable grip strength dynamometer [14]. To measure gait speed, participants were timed as they walked along a 2.4 m-long straight path with a flat surface. Participants were instructed to walk at their preferred speed. Two markers indicated the start and end points for time measurement, and the timed section was preceded by a 1-m section for initiating gait. Participants who used assistive equipment or had walking disability were excluded. Gait speed (m/s) was calculated by dividing the distance (2.4 m) by the measured time (s) [15], and there was a 1 m section following the timed section to allow for reducing speed prior to stopping. A slow gait speed was defined as < 1 m/s because this criterion has been widely used in the Japanese population as a cut-off value for slow gait speed based on the revised Japanese version of the Cardiovascular Health Study [16] and the Asian Working Group for Sarcopenia [17].

Dementia was defined as a clinical diagnosis by an outpatient physician based on the NIA/AA criteria [18]. MCI was diagnosed in accordance with the Petersen criteria [19].

Statistical analysis

Participants were first divided into cognitive status groups and then subdivided into polypharmacy and non-polypharmacy groups. Background characteristics were compared between the groups using Student’s t-test for continuous variables or the chi-square test for categorical variables Multivariate analysis was adjusted using items showing statistically significant differences between polypharmacy groups. They were performed with both a dichotomous polypharmacy measure and a continuous medication count. All analyses were carried out using the Statistical Package for the Social Sciences, version 26 (SPSS, Chicago, IL, USA). Significance was set at p≤0.05.

To verify a difference of 0.1 m/s in gait speed at the significance level of 5%with statistical power of 80%, a sample size of at least 100 people was required in each of the intervention and control groups, based on the standard deviation of 0.25.

RESULTS

Data of 431 patients in the CN (n = 103), MCI (n = 194), and AD (n = 134) groups were analyzed. Patients’ background characteristics are shown in Tables 1 –3. Mean age was 72.7±6.2, 76.6±5.4, and 79.3±6.3 years in the CN, MCI, and AD groups, respectively. The prevalence of polypharmacy was 44.7%, 45.4%, and 36.6%and the mean number of medications was 4.5±3.2, 4.4±3.5, and 3.9±3.0, respectively. Mean gait speed was 1.1±0.2, 1.0±0.2 and 0.9±0.2 m/s, respectively.

Table 1

Background characteristics in the CN group

	Total	Non-polypharmacy	Polypharmacy	p
Number of patients	103	58 (56.3%)	45 (44.7%)
Age (y)	72.7±6.2	72.0±5.6	70.0±6.9	0.225
Sex (%male)	35 (34.5%)	19 (32.8%)	58 (35.6%)	0.383
Education (y)	12.7±2.5	12.6±2.5	12.6±2.4	0.974
Body mass index(kg/m²)	22.1±3.1	22.3±3.1	22.0±3.1	0.603
Number of medications	4.5±3.2	2.2±1.5	7.5±2.2	< 0.001^**
Barthel index	99.6±1.7	99.8±1.3	99.2±2.1	0.077
MMSE score	28.1±2.2	28.3±2.0	27.8±2.4	0.318
Low grip strength (male; < 28 kg; female; < 18 kg)	18 (17.5%)	7 (13.7%)	11 (24.4%)	0.082
Gait speed (m/s)	1.1±0.2	1.2±0.2	1.1±0.3	< 0.001^**
Slow gait speed (< 1 m/s)	20 (19.4%)	5 (8.6%)	15 (33.3%)	< 0.001^**
GDS score	3.8±3.0	3.6±2.9	4.1±3.2	0.374
Number of comorbid conditions	0.8±0.9	0.6±0.8	1.2±1.0	< 0.001^**
Hypertension (%)	33 (32.0%)	17 (29.3%)	16 (35.6%)	0.248
Dyslipidemia (%)	12 (11.7%)	5 (8.6%)	7 (15.6%)	0.136
Diabetes (%)	19 (18.4%)	11 (19.0%)	8 (17.8%)	0.438
Heart disease (%)	15 (14.6%)	6 (10.3%)	9 (20.0%)	0.083

CN, cognitively normal; MMSE, Mini-Mental State Examination; GDS, Geriatric Depression Scale 15. ^*p < 0.05. ^**p < 0.01.

Table 2

Background characteristics in the MCI group

	Total	Non-polypharmacy	Polypharmacy	p
Number of patients	194	106 (54.6%)	88 (45.4%)
Age (y)	76.6±5.4	75.7±5.6	77.6±5.0	0.012^*
Sex (%male)	65 (33.5%)	35 (33.0%)	30 (34.1%)	0.436
Education (y)	10.7±2.2	11.2±2.4	11.0±2.6	0.327
Body mass index(kg/m²)	21.8±3.3	22.0±3.4	21.6±3.3	0.446
Number of medications	4.4±3.5	1.8±1.4	7.7±2.4	< 0.001^**
Barthel index	98.9±3.4	99.8±2.8	98.7±4.0	0.395
MMSE score	24.5±3.1	24.4±3.0	24.8±3.3	0.419
Low grip strength (male; < 28 kg; female; < 18 kg)	72 (37.1%)	31 (29.2%)	41 (46.6%)	0.006^**
Gait speed (m/s)	1.0±0.2	1.1±0.2	0.9±0.2	< 0.001^**
Slow gait speed (< 1 m/s)	84 (43.3%)	37 (34.9%)	47 (53.4%)	0.005^**
GDS score	3.3±2.7	3.2±2.8	3.5±2.6	0.401
Number of comorbid conditions	1.0±0.9	0.7±0.8	1.3±0.9	< 0.001^**
Hypertension (%)	61 (31.4%)	36 (34.0%)	25 (29.1%)	0.233
Dyslipidemia (%)	21 (10.8%)	14 (13.2%)	7 (8.0%)	0.131
Diabetes (%)	24 (12.4%)	11 (10.4%)	13 (14.8%)	0.177
Heart disease (%)	18 (9.2%)	13 (12.3%)	5 (5.7%)	0.058

MCI, mild cognitive impairment; MMSE, Mini-Mental State Examination; GDS, Geriatric Depression Scale 15. ^*p < 0.05. ^**p < 0.01.

Table 3

Background characteristics in the AD group

	Total	Non-polypharmacy	Polypharmacy	p
Number of patients	134	85 (63.4%)	49 (36.6%)
Age (y)	79.3±6.3	79.2±6.1	79.6±5.6	0.670
Sex (%male)	40 (29.9%)	28 (30.9%)	12 (24.5%)	0.153
Education (y)	10.6±2.4	10.6±2.4	10.4±2.5	0.569
Body mass index(kg/m²)	21.9±3.6	21.8±3.5	22.1±3.6	0.596
Number of medications	3.9±3.0	2.0±1.3	7.0±2.1	<0.001^**
Barthel index	96.3±8.7	95.6±9.7	98.3±3.8	0.063
MMSE score	18.9±4.0	18.5±4.0	18.6±4.0	0.957
Low grip strength (male; < 28 kg; female; < 18 kg)	71 (53.0%)	45 (52.9%)	26 (53.1%)	0.491
Gait speed (m/s)	0.9±0.2	0.9±0.3	1.0±0.2	0.331
Slow gait speed (< 1 m/s)	76 (56.7%)	48 (56.5%)	28 (57.1%)	0.473
GDS score	3.0±2.9	3.0±2.9	3.0±3.0	0.856
Number of comorbid conditions	0.9±0.8	0.7±0.7	1.3±0.9	< 0.001^**
Hypertension (%)	45 (33.6%)	24 (28.2%)	21 (42.9%)	0.041^*
Dyslipidemia (%)	21 (15.7%)	12 (14.1%)	9 (18.4%)	0.255
Diabetes (%)	18 (13.4%)	9 (10.6%)	9 (18.4%)	0.101
Heart disease (%)	23 (17.2%)	13 (15.3%)	10 (20.4%)	0.225

AD, Alzheimer’s dementia; MMSE, Mini-Mental State Examination; GDS, Geriatric Depression Scale 15. ^*p < 0.05. ^**p < 0.01.

Gait speed was significantly slower in the polypharmacy group than in the non-polypharmacy group for the CN and MCI patients after adjustment with significantly different variables in univariate analysis (Tables 1 2), but not for the AD patients (Table 3). Significant differences between the polypharmacy and non-polypharmacy groups were found in number of medications and number of comorbidities in the CN patients; in age, number of medications, low grip strength, and number of comorbidities in the MCI patients; and in number of medications and number of comorbidities in the AD patients.

Multiple regression analysis was carried out to investigate the association between gait speed and number of medications (Table 4, model 1). The association with polypharmacy status (number of medications≥5) was also analyzed (Table 4, model 2). The analysis was adjusted using the factors that were significantly different in univariate analysis (number of comorbidities in CN patients, age, low grip strength, number of medications, and number of comorbidities in MCI patients). This revealed that gait speed had a significant negative association with number of medications and polypharmacy status in the CN and MCI patients after adjustment for potential confounders.

Table 4

Multiple regression analysis of gait speed with number of medications and polypharmacy status in CN and MCI patients

	CN						MCI
	Model 1			Model 2			Model 1			Model 2
	B	95%	p	B	95%	p	B	95%	p	B	95%	p
Age (y)	—	—	—	—	—	—	–0.009	–0.015 to –0.003	0.003^**	–0.009	–0.015 to –0.004	0.003^**
Low Grip strength	—	—	—	—	—	—	–0.101	–0.006 to –0.037	0.002^**	–0.102	–0.167 to –0.034	0.003^**
Number of medications	–0.026	–0.041 to –0.018	<0.001^**	—	—	—	–0.018	–0.028 to –0.009	< 0.001^**	—	—	—
Polypharmacy status	—	—	—	–0.128	–0.022 to –0.033	0.009^*	—	—	—	–0.100	–0.166 to –0.034	0.003^**
Number of comorbidities	–0.038	–0.089 to 0.013	0.145	–0.048	–0.099 to 0.004	0.070	0.011	–0.024 to 0.047	0.533	0.003	–0.032 to 0.039	0.836

CN, cognitively normal; MCI, mild cognitive impairment. CN Model 1: Explanatory variables are number of medications, and number of comorbidities. CN Model 2: Explanatory variables are polypharmacy status, and number of comorbidities. MCI Model 1: Explanatory variables are age, low grip strength, number of medications, and number of comorbidities. MCI Model 2: Explanatory variables are age, low grip strength, polypharmacy status, and number of comorbidities.

Logistic regression analysis to investigate the associations of slow gait speed (< 1 m/s) with polypharmacy status and number of medications (Table 5) revealed that slow gait speed was associated with number of medications (model 1) and polypharmacy status in the CN patients (model 2). In the MCI patients, slow gait speed was associated with number of medications (model 1) but not with polypharmacy status (model 2).

Table 5

Logistic regression analysis of slow gait speed with number of medications and polypharmacy status in CN and MCI patients

	CN						MCI
	Model 1			Model 2			Model 1			Model 2
	Odds ratio	95%	p	Odds ratio	95%	p	Odds ratio	95%	p	Odds ratio	95%	p
Age (y)	—	—	—	—	—	—	1.073	1.009 to 1.140	0.024^**	1.077	1.014 to 1.144	0.015^*
Low Grip strength	—	—	—	—	—	—	2.214	1.167 to 4.200	0.015^*	2.231	1.182 to 4.211	0.013^*
Number of medications	1.336	1.115 to 1.601	0.002^**	—	—	—	1.128	1.022 to 1.244	0.016^*	—	—	—
Polypharmacy status	—	—	—	4.225	1.348 to 13.244	0.013^*	—	—	—	1.676	0.879 to 3.199	0.117
Number of comorbidities	1.627	0.916 to 2.892	0.097	1.674	0.959 to 2.921	0.07	0.956	0.0663 to 1.376	0.807	1.029	0.724 to 1.463	0.873

DISCUSSION

In this study, gait speed was negatively associated with number of medications and the polypharmacy group had slower gait speed than the non-polypharmacy group in the CN and MCI patients even after adjustment for potential confounders. However, these associations were not found in the AD patients. To our knowledge, this is the first study to investigate the association between polypharmacy and gait speed in older outpatients with CN, MCI, and AD.

Previous studies have reported an association between polypharmacy and slow gait speed in CN community-dwelling older adults [9] and in MCI patients [10]. A study in community-dwelling older adults also showed that polypharmacy was cross-sectionally associated with poor gait performance and longitudinally associated with gait decline and fall incidence [20]. Our results here are consistent with these studies. However, previous studies examined the relationship between gait speed and polypharmacy in particular populations, such as community-dwelling older adults and MCI patients. In the present study, on the other hand, we simultaneously analyzed the association between gait speed and polypharmacy in three different groups with differing cognitive status, and we found that polypharmacy was related to gait speed in CN and MCI patients but not in AD patients. Thus, we should pay attention to decreased gait speed and risk of falls in older adults with polypharmacy even when their cognitive function is relatively preserved.

Slow gait speed is a predictive marker for several adverse events such as facility admission, falls, and death in older adults [15, 21]. Slow gait speed is also known to be a strong risk factor for the development of dementia. It has been reported that community-dwelling older adults with lower baseline gait speed were at higher risk of developing dementia and that those with a greater decline in gait speed were at greater risk [22]. Slow gait is also a strong risk factor for conversion from normal cognitive function to MCI [23]. Moreover, co-occurrence of MCI and slow gait is a risk factor for progression to dementia [24]. Our results demonstrate that CN and MCI patients with polypharmacy have slower gait speed and polypharmacy could therefore be a risk factor for progression to dementia in these populations. Longitudinal studies to explore the effects of deprescription on gait speed and cognition are warranted because polypharmacy is a potentially modifiable status [25].

In this study, polypharmacy and slow gait were associated in the CN and MCI patients. There has been some discussion about a possible mechanism. Previous studies have reported that comorbidities affected gait speed [26, 27]. In our study, however, polypharmacy was still significantly associated with slow gait speed even after adjustment for comorbidities. Unfortunately, we did not include musculoskeletal disorders and neurological disorders such as arthritis or Parkinson’s disease which could directly affect gait speed. Moreover, the severity of each comorbidity was not considered in the present analysis. Therefore, the possibility that comorbidity affected gait speed in the polypharmacy groups may not be completely excluded. In addition, we should consider other mechanisms. Gait is a complex process requiring various functions of multiple organ systems including the central and peripheral nervous systems, the respiratory and cardiovascular systems, and the musculoskeletal system [28, 29]. Various effects of medication could affect these systems. For example, antihistaminergic and anticholinergic medications can negatively affect domains of cognition that are critical for controlling gait and balance including attention, processing speed, and executive function [30, 31]. Antihypertensive medications can induce dizziness and orthostatic hypotension, thus affecting dynamic gait balance [30]. Anticholinergics, proton pump inhibitors, antidepressants, benzodiazepines, and other drugs have been suggested to have an effect on cognitive decline [32], which is associated with slow gait speed [33]. In the CN and MCI patients in this study, gait speed might have been decreased by use of these drugs through drug-induced cognitive impairment. The type of drug would also be important, as well as the number of drugs, when considering the underlying mechanism of the present findings. Unfortunately, however, our data included only the number of medications taken and not precise pharmacological information such as the type of medication or dosage. Further research including the type of medication is needed.

Slow gait speed is one of the defining characteristics of physical frailty. A previous systematic review determined a significant association between an increased number of medications and frailty [34]. Palmer et al. reported that polypharmacy is more common in frail persons than in robust persons and that there was higher risk of either incident prefrailty or frailty in persons with polypharmacy [35, 36]. Therefore, the association between polypharmacy and slow gait could be bidirectional.

This study found that the association between gait speed and polypharmacy was attenuated in the AD group. Mean gait speed was slowest in the AD group (0.9 m/s in AD, but 1.1 m/s in CN and 1.0 m/s in MCI), and we did not find a significant difference in gait speed between the polypharmacy and non-polypharmacy groups in the AD patients. A previous study has shown that poorer baseline cognition is associated with slow gait speed [37]. Also, attention deficits are recognized as early and consistent manifestations of AD [38], and the inability to divide or select attention has negative effects on motor function [39]. Gait speed was already very slow in the AD patients, even those with non-polypharmacy. Thus, the effects of polypharmacy on gait speed could be masked by a stronger influence of cognitive decline in AD compared with MCI or CN. Moreover, in patients with AD, drug management becomes difficult from a relatively early stage due to a decline in self-management ability [40]. In the present study, the number of medications tended to decrease as the cognitive function declined. Physicians may try to simplify prescriptions and reduce the number of medications to maintain good adherence in this population. The reduced number of medications in the AD group might have influenced the present results.

The strengths of our study include a large and well-characterized cohort of older adults who attended an outpatient clinic. Additionally, quantitative gait analysis was performed with adjustments for known strong confounders such as age, sex, BMI, MMSE score, and GDS-15 score. In addition, analyses of both a dichotomous polypharmacy measure and a continuous medication count were performed in the multivariate analysis. However, our study is not free of limitations. First, this study had a cross-sectional design, so the causal relationship between polypharmacy and gait speed cannot be discussed. Moreover, this study could not determine if polypharmacy is associated with progression from CN/MCI status to dementia due to the cross-sectional study design. A longitudinal study is needed to investigate these issues. Second, we excluded those who used a walking aid, had walking disability, or were unable to follow the gait instructions. Such patients would probably have slow gait and polypharmacy, which would have also affected the results. Indeed, the excluded subjects had more medications compared with those included in the AD group. Third, because information on pharmacological aspects including dose and duration of use were not included and adherence to the prescription was not considered, their direct causal relationships on gait performance are not known. Finally, not all comorbidities, such as musculoskeletal diseases, neurological disorders and otolaryngology diseases were examined, and their severities were not recorded. Therefore, we might not have fully adjusted for interactions with comorbidity.

In conclusion, this study found that gait speed was associated with polypharmacy in outpatients with CN and MCI but not AD.

Footnotes

ACKNOWLEDGMENTS

The study was supported by Research Funding for Longevity Sciences from the National Center for Geriatrics and Gerontology, Japan.

Authors’ disclosures available online ().

References

Gnjidic

, Hilmer

, Blyth

, Naganathan

, Waite

, Seibel

, McLachlan

, Cumming

, Handelsman

, Le Couteur

(2012) Polypharmacy cutoff and outcomes: 5 or more medicines were used to identify community dwelling older men at risk of different adverse outcomes. J Clin Epidemiol 65, 989–995.

Lai

, Liao

(2013) Optimal cutoff of polypharmacy and outcomes. J Clin Epidemiol 66, 465.

Maher

, Hanlon

, Hajjar

(2014) Clinical consequences of polypharmacy in elderly. Expert Opin Drug Saf 13, 57–65.

Gutiérrez-Valencia

, Izquierdo

, Cesari

, Casas-Herrero

, Inzitari

, Martínez-Velilla

(2018) The relationship between frailty and polypharmacy in older people: A systematic review. Br J Clin Pharmacol 84, 1432–1444.

Peel

, Alapatt

, Jones

, Hubbard

. (2019) The association between gait speed and cognitive status in community-dwelling older people: A systematic review and meta-analysis. J Gerontol A Biol Sci Med Sci 74, 943–948.

Allali

, Annweiler

, Blumen

, Callisaya

, De Cock

, Kressi

, Srikanth

, Steinmetz

, Verghese

, Beauchet

(2016) Gait phenotype from mild cognitive impairment to moderate dementia: Results from the GOOD initiative. Eur J Neurol 23, 527–541.

Murray

, Senjem

, Petersen

, Hollman

, Preboske

, Weigand

, Knopman

, Ferman

, Dickson Jack

DW CR

(2010) Functional impact of white matter hyperintensities in cognitively normal elderly subjects. Arch Neurol 67, 1379–1385.

Makizako

, Shimada

, Doi

, Yoshida

, Ito

, Kato

, Shimokata

, Washimi

, Endo

, Suzuki

(2011) The association between decline in physical functioning and atrophy of medial temporal areas in community-dwelling older adults with amnestic and nonamnestic mild cognitive impairment. Arch Phys Med Rehabil 92, 1992–1999.

George

, Verghese

(2017) Polypharmacy and gait performance in community-dwelling older adults. J Am Geriatr Soc 65, 2082–2087.

10.

Umegaki

, Yanagawa

, Komiya

, Matsubara

, Fujisawa

, Suzuki

, Kuzuya

(2019) Polypharmacy and gait speed in individuals with mild cognitive impairment. Geriatr Gerontol Int 19, 730–735.

11.

Mahoney

, Barthel

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

12.

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.

13.

Vinkers

, Gussekloo

, Stek,

, Westendorp

RGJ

, van der Mast

(2004) The 15-item Geriatric Depression Scale (GDS-15) detects changes in depressive symptoms after a major negative life event The Leiden 85-plus Study. Int J Geriatr Psychiatry 19, 80–84.

14.

Matsui

, Fujita

, Harada

, Sakurai

, Nemoto

, Noda

, Toba

(2014) Association of grip strength and related indices with independence of activities of daily living in older adults investigated by a newly-developed grip strength measuring device. Geriatr Gerontol Int 14, 77–86.

15.

Van Kan

, Rolland

, Andrieu

, Bauer

, Beauchet

, Bonnefoy

, Cesari

, Donini

, Guyonnet

, Inzitari

, Nourhashemi

, Onder

, Ritz

, Salva

, Visser

, Vellas

(2009) Gait speed at usual pace as a predictor of adverse outcomes in community-dwelling older people an International Academy on Nutrition and Aging (IANA) Task Force. J Nutr Health Aging 13, 881–889.

16.

Satake

, Arai

(2020) Questionnaire for medical checkup of old-old (QMCOO). Geriatr Gerontol Int 20, 991–992.

17.

Chen

, Woo

, Assantachai

, Auyeung

, Chou

, Iijima

, Jang

, Kang

, Kim

, Kojima

, Kuzuya

, Lee

JSW

, Lee

, Liang

, Lim

, Peng

, Sugimoto

, Tanaka

, Won

, Yamada

, Zhang

, Akishita

, Arai

(2020) Asian Working Group for Sarcopenia: 2019 Consensus Update on Spenia Diagnosis and Treatment. J Am Med Dir Assoc 21, 300–307.

18.

McKhann

, Knopman

, Chertkow

, Hyman

, Jack

Jr. , Kawas

, Klunk

, Koroshetz

, Manly

, Mayeux

, Mohs

, Morris

, Rossor

, Scheltens

, Carrillo

, Thies

, Weintraub

, Phelps

(2011) The diagnosis of dementia due to Alzheimer’s disease: Recommendations from the National Institute on Aging-Alzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease. Alzheimers Dement 7, 263–269.

19.

Petersen

, Doody

, Kurz

, Mohs

, Morris

, Rabins

, Ritchie

, Rossor

, Thal

, Winblad

(2001) Current concepts in mild cognitive impairment. Arch Neurol 58, 1985–1992.

20.

Odasso

, Adamson

, Song

, Bray

, Faria

, Speechley

(2019) Polypharmacy, gait performance, and falls in community dwelling older adults. Results from the Gait and Brain Study. J Am Geriatr Soc 67, 1182–1188.

21.

Viccaro

, Perera

, Studenski

(2011) Is timed up and go better than gait speed in predicting health, function, and falls in older adults? J Am Geriatr Soc 59, 887–892.

22.

Hackett

, Davies-Kershaw

, Cadar

, Orrell

, Steptoe

(2018) Walking speed, cognitive function, and dementia risk in the English longitudinal study of ageing. J Am Geriatr Soc 66, 1670–1675.

23.

Buracchio

, Dodge

, Howieson

, Wasserman

, Kaye

(2010) The trajectory of gait speed preceding mild cognitive impairment. Arch Neurol 67, 980–986.

24.

Doi

, Makizako

, Tsutsumimoto

, Hotta

, Nakakubo

, Makino

, Suzuki

, Shimada

(2018) Combined effects of mild cognitive impairment and slow gait on risk of dementia. Exp Gerontol 110, 146–150.

25.

Rieckert

, Reeves

, Altiner

, Drewelow

, Esmail

, Flamm

, Hann

, Johansson

, Mielke

, Kunnamo

, Löffler

, Piccoliori

, Sommerauer

, Trampisch

, Vögele

, Woodham

, Sönnichsen

(2020) Use of an electronic decision support tool to reduce polypharmacy in elderly people with chronic diseases: Cluster randomised controlled trial. BMJ 369, m1822.

26.

Wei

, Kabeto

, Langa

, Mukamal

(2018) Multimorbidity and physical and cognitive function: Performance of a new multimorbidity-weighted index. J Gerontol A Biol Sci Med Sci 73, 225–232.

27.

Gong

, Wan

, Zhang

, Liu

, Yin

(2018) Correlation between the Charlson comorbidity index and skeletal muscle mass/physical performance in hospitalized older people potentially suffering from sarcopenia. BMC Geriatr 19, 367.

28.

Alexander

(1996) Gait disorders in older adults. J Am Geriatr Soc 44, 434–451.

29.

Montero-Odasso

, Verghese

, Beauchet

, Hausdorff

(2012) Gait and cognition: A complementary approach to understanding brain function and the risk of falling. J Am Geriatr Soc 60, 2127–2136.

30.

Leipzig

, Cumming

, Tinetti

(1999) Drugs and falls in older people: A systematic review and meta-analysis: I. psychotropic drugs. J Am Geriatr Soc 47, 30–39.

31.

Huang

, Mallet

, Rochefort

, Eguale

, Buckeridge

, Tamblyn

(2012) Medication-related falls in the elderly: Causative factors and preventive strategies. Drugs Aging 29, 359–376.

32.

Leelakanok

, Ronilda D’Cunha

(2019) Association between polypharmacy and dementia - A systematic review and metaanalysis. Aging Ment Health 23, 932–941.

33.

Verghese

, Lipton

, Hall

, Kuslansky

, Katz

, Buschke

(2002) Abnormality of gait as a predictor of non-Alzheimer’s dementia. N Eng J Med 347, 1761–1768.

34.

Hamerman

(1999) Toward an understanding of frailty. Ann Intern Med 130, 945–950.

35.

Palmer

, Villani

, Vetrano

, Cherubini

, Cruz-Jentoft

, Curtin

, Denkinger

, Gutiérrez-Valencia

, Gu*omundsson

, Knol

, Mak

, O’Mahony

, Pazan

, Petrovic

, Rajkumar

, Topinkova

, Trevisan

, Tischa van der Cammen

J M

, van Marum

, Wehling

, Ziere

, Bernabei

, Onder

, European Geriatric Medicine Society Pharmacology special interest group (2019) Association of polypharmacy and hyperpolypharmacy with frailty states: A systematic review and meta-analysis. Eur Geriatr Med 10, 9–36.

36.

de Groot

, van Campen

JPCM

, Kosse

, de Vries

, Beijnen

, Lamoth

CJC

(2016) The association of medication-use and frailty-related factors with gait performance in older patients. PLoS One 11, e0149888.

37.

Dyer

, Lawlor

, Kennelly

, and for the NILVAD Study Group (2020) Gait speed, cognition and falls in people living with mild-to-moderate Alzheimer disease: Data from NILVAD. BMC Geriatr 20, 117.

38.

Baddeley

, Baddeley

, Bucks

, Wilcock

(2001) Attentional control in Alzheimer’s disease. Brain 124, 1492–1508.

39.

Sheridan

, Mat

, Kowall

, Hausdorff

(2003) Influence of executive function on locomotor function: Divided attention increases gait variability in Alzheimer’s disease. J Am Geriatr Soc 51, 1633–1637.

40.

Tran

, Baxter

, Hamman

, Grigsby

(2014) Impairment of executive cognitive control in type 2 diabetes, and its effects on health-related behavior and use of health services. J Behav Med 37, 414–422.