EXERCISE IS MEDICINE

Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

A Scoping Review of Multiple-modality Exercise

and Cognition in Older Adults: Limitations
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023

and Future Directions

Narlon C. Boa Sorte Silva, BSc, PhD;1,2 Dawn P. Gill, PhD;2,3 and
Robert J. Petrella, MD, PhD, FCFP, FACSM1,2,3,4

literature has focused on aerobic exer-

Abstract cise training (AET) interventions alone
The effects of multiple-modality exercise (MME) on brain health warrants (13), with some evidence suggesting that
further elucidation. Our objectives were to report and discuss the current AET appears to benefit cognition, par-
evidence regarding the influence of MME on cognition and neuroimaging ticularly executive functioning (14), as
outcomes in older adults without dementia. We searched the literature well as neuroplasticity (15), neural effi-
for studies investigating the effects of MME on measures of cognition, ciency (16), and hippocampal size in
brain structure, and function in individuals 55 years or older without healthy older adults (12) and in those with
dementia. We include 33 eligible studies. Our findings suggested that mild cognitive impairment (MCI) (17).
MME improved global cognition, executive functioning, processing speed, Despite promising evidence, the im-
and memory. MME also improved white and gray matter and hippocampal pact of AET on cognitive function in
volumes. These findings were evident largely when compared with the aging population remains equivocal
no-treatment control groups but not when compared with active (e.g., (18). A recent Cochrane review suggests
health education) or competing treatment groups (e.g., cognitive training). that there is insufficient evidence to con-
MME may improve brain health in older adults without dementia; however, clude that cognitive improvements fol-
because of possible confounding factors, more research is warranted. lowing AET are solely due to AET itself,
even when improvement in cardiovascu-
lar fitness is observed (19). The current state of knowledge al-
Introduction lows for exploration of exercise interventions that could have
Changes to lifestyle, such as engaging in regular exercise, additive benefits to cognition beyond AET alone. Encouraging
are postulated as important strategies to prevent or slow the findings have suggested resistance exercise training (RET) as an
progression of dementia in the aging population (1–3), this in- effective exercise modality to impart benefits to cognition, with
cludes those with high genetic risk (4). Exercise has been asso- positive effects of RET on executive functioning (20,21), as well
ciated with preserved age-related cognitive functioning in as brain functional plasticity (22), and white matter structure (23).
observational studies (3,5–8) and has improved cognition Findings from other meta-analytic studies have indicated a
(9), as well as shown positive functional (10,11) and structural lack of consistency across different exercise studies, which
(12) brain changes in longitudinal interventional studies. Most could be due to variability in cognitive tests applied, sensitivity
of cognitive tests in detecting treatment effects, cognitive and
1
School of Kinesiology, Faculty of Health Sciences, Western University, physical health at baseline, as well as characteristics of the ex-
London, ON, CANADA; 2Centre for Studies in Family Medicine, Department ercise programs administered (18,24). Moreover, most studies
of Family Medicine, Schulich School of Medicine and Dentistry, Western
University, London, ON, CANADA; 3School of Health Studies, Faculty of have failed to comply with current guidelines for exercise in
Health Sciences, Western University, London, ON, CANADA; and older adults with regard to exercise type, intensity, frequency,
4
Department of Family Practice, Faculty of Medicine, University of British and duration (25,26). These guidelines highly emphasize the
Columbia, Vancouver, BC, CANADA importance of multiple-modality exercise (MME) programs
Address for correspondence: Robert J. Petrella, MD, PhD, FCFP, FACSM,
to enhance overall health and quality of life (26). A recent
Department of Family Practice, Faculty of Medicine, University of British meta-analysis demonstrated the potential of MME to induce
Columbia, 320 - 5950 University Boulevard, Vancouver, British Columbia V6T clinically relevant fitness improvements in older adults, includ-
1Z3, Canada. E-mail: [email protected]. ing cardiovascular fitness and functional capacity (27). How-
1537-890X/1908/298–325
ever, no previous literature review has focused solely on
Current Sports Medicine Reports investigating effects of MME on cognition and neuroimaging
Copyright © 2020 by the American College of Sports Medicine outcomes (28,29).

298 Volume 19  Number 8  August 2020 Exercise and Cognition in Older Adults

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 Therefore, the objectives of this scoping review were to: original publication; however, we excluded those reporting
1) document the current state of evidence of the impact of sensitivity analyses of primary outcomes already reported in
MME on cognition and neuroimaging in older adults without the original publication.
dementia; 2) discuss the current state of evidence with regard
to exercise prescription and implementation in these studies; Results
and 3) propose future directions for research in the field.
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Selection of Sources of Evidence

Methods Our search results, including identification, screening, eligi-
Full details of our methods, including study protocol, bility, and selected articles, are presented in Supplementary
search strategy, and data charting can be found in the Supple- Figure 1 (http://links.lww.com/CSMR/A65). The original search
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023

mentary Material (http://links.lww.com/CSMR/A65). Briefly, led to 2945 results, of which remaining 33 studies (original
the PICO(T) (population, intervention, comparison, outcome, search, 25; added from other sources, 8) were considered eligi-
and [type]) (30) approach was used to develop our research ble for inclusion in this scoping review.
question, while the Preferred Reporting Items for Systematic
Reviews and Meta-Analyses extension for Scoping Reviews Characteristics of Sources of Evidence
(30) were utilized as a guideline in this review. Our research Study design, sample size, and participant characteristics
question was as follows: “What are effects of MME interven- are reported in Table 1. We included 33 studies (31–63), of
tions aimed at improving cognition and neuroimaging out- which 26 were RCT studies (31–37,39,40,42,44–48,50–52,54,
comes in older adults without dementia?” Between August 55,57,58,60–63) and seven were quasiexperimental studies
and October 2019, we searched the following bibliographical (38,41,43,49,53,56,59). We included data from 30 original re-
databases for potentially relevant documents: Cochrane Cen- search articles (31–37,39,41–51,53–63) and three articles that
tral Register of Controlled Trials, EMBASE, MEDLINE, and were analyses of secondary outcomes (38,39,52). A total of
Scopus. We also contacted authors directly to identify addi- 4458 individuals (excluding counts from secondary outcomes
tional relevant material and to further determine eligibility of articles) were studied. Sample sizes varied between 19 and 1476
articles selected for full-text review (30). (mean [SD] = 148.6 [270.8]), with age range between 62.2 and
We searched the literature for peer-reviewed, published 82.3 years (mean [SD] = 72.5 [4.8]) and the majority of partici-
randomized controlled trials (RCTs), and nonrandomized in- pants being females (mean [SD] = 71.3% [16.8]). Most studies
tervention studies (i.e., quasi-experimental) examining the ef- included healthy older adults (e.g., preserved physical and cogni-
fects of MME interventions on cognition and/or neuroimaging tive function, n = 13), followed by studies including older adults
outcomes (see Supplementary Table 1, http://links.lww.com/ who were sedentary (n = 8), had cognitive impairment but not
CSMR/A65, for a sample of our search strategy). We defined dementia (n = 7), were frail (n = 3), had diabetes (n = 1), or were
MME interventions as those that included a combination of obese (n = 1). Further, considering the study groups (i.e., com-
AET aimed at improving aerobic capacity or cardiovascular fit- parators), the majority of study designs included no-treatment
ness, and RET aimed at improving muscle strength, endurance, control groups (n = 23), followed by at least one competing
or power (26,27). We also included studies that combined AET treatment group (n = 17) and an active control group (n = 9).
and RET with balance or flexibility exercises (26,27). Other ac-
tivities referred to as “warm-up,” “cool-down,” or “recovery” MME Protocols
were not considered. Considering the nature of this scoping Of the MME protocols included in the 33 studies, 18 in-
review, we did not specify minimum or maximum length of volved a combination of AET, RET, plus balance/flexibility
exercise programs, whether components of AET or RET were training, while 15 studies included AET and RET only (see
administered in the same session or different sessions, and Table 2). The MME protocols administered varied from
whether interventions were supervised, home-based, or both 1 d·wk−1 to 7 d·wk−1 (mean [SD] = 3.1 d·wk−1 [1.5 d·wk−1]),
(see Supplementary Material, http://links.lww.com/CSMR/ and between 30 min·d−1 and 90 min·d−1 (mean[SD] = 62.7 min·d−1
A65, for details). [15.5 min·d−1]) from 1.5 to 24 months (mean [SD] = 6.8 months
In summary, we included studies that met the following in- [6.3 months]). Because of the limited reporting and high incon-
clusion criteria: 1) MME studies combining both AET and sistency of measures of exercise intensity, it was not feasible to
RET with or without additional balance/flexibility training, summarize exercise intensity for all studies; however, intensities
as defined above; 2) included older adults 55 years or older; varied between low and high for the exercise components. Only
3) included individuals with or without cognitive impairment, 24 studies reported adherence data (i.e., attendance to sessions),
but not dementia (i.e., cognitively healthy, self-reported cogni- ranging from 34.7% to 100% (mean [SD] = 73.9% [15.9%]).
tive or memory complaints, subjective cognitive/memory de- Below, we further describe each component individually.
cline or impairment [SMI, MCI]); included at least one
measure of cognition (e.g., global or domain-specific cognitive Aerobic exercise training
function), and/or neuroimaging outcomes relevant to cogni- Across all studies, the AET component was prescribed on
tive function (e.g., functional network connectivity, gray mat- average 3.1 d·wk−1 (SD = 1.6 d·wk−1, n = 33 reported), for
ter volume); 5) included a comparator group (i.e., competing an average of 32.6 min·d−1 (SD = 13 min·d−1, n = 27), with
treatment group, active control group, or no-treatment con- studies using low (n = 2), moderate (n = 12), and moderate
trol group); 6) published in English between January 1990 to high (n = 7) intensity. As mentioned earlier, 21 studies re-
and October 2019; and 7) published in a peer-reviewed jour- ported measures of AET intensity with high variability in
nal. We also included other articles from the same parent tracking methods, which consisted of rating of perceived exer-
study that reported different relevant outcomes from the tion (RPE) (n = 7) percentage of maximum heart rate (HR)

www.acsm-csmr.org Current Sports Medicine Reports 299

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Table 1.

300
Study and participant baseline characteristics.

First Author (Year) Study Design Population Sample, n Age, mean (SD) Females, n (%) MMSEa or equivalent
Ansai (2015) (31) RCT, 16-wk intervention with Sedentary older adults 69 82.4 (2.4) 47 (68.1) 24.9 (3.3)
6-wk follow-up
Berryman (2014) (32) RCT, 8-wk intervention Healthy older adults 51 70.6 (5.6) 29 (56.9) 23.8 (1.1)
Boa Sorte Silva RCT, 24-wk intervention, single-blind Older adults with SCC 127 67.5 (7.3) 90 (70.9) 29.1 (1.1)
(2018) (33) with 28-wk follow-up
Callisaya (2017) (34) RCT, 6-mo intervention, single blind Older adults with diabetes 50 66.2 (4.9) 24 (48) ≥28b
Carral (2008) (35) RCT, 5-mo intervention Community-dwelling women 62 68.4 (3.4) 62 (100) 23.2 (3.9)
Damirchi (2018) (36) RCT, 8-wk intervention with Sedentary older women with MCI 54 68.4 (4.3) 54 (100) 23.4 (2)
6-mo follow-up

Volume 19  Number 8  August 2020

Eggenberger (2015) (37) RCT, 6-mo intervention with Healthy older adults 71 78.9 (5.4) 46 (64.8) 28.2 (1.4)
1-yr follow-up
Fissler (2017) (38) QE, 10-wk intervention with Older adults with SMI 39 72 (5.3) 23 (59) 28 (1.9)
3-mo follow-up
Gajewski (2012, 2018) (39,40) RCT, 4-mo intervention Healthy older adults 141 70.9 (5.2) 84 (59.6) 28.5 (1.7)
Ji (2017) (41) QE, 6-wk intervention Healthy older adults 24 70 (7.2) 12 (50) ≥24
Klusmann (2010) (42) RCT, 6-mo intervention, single-blind Healthy older women 259 73.6 (4.2) 259 (100) 28.8 (1)
Küster (2016) (43) QE, 6-mo intervention with Older adults with SMI 54 71.4 (5.8) 30 (55.6) 27.9 (2.2)
3-mo follow-up
Langlois (2013) (44) RCT, 3-mo intervention Nonfrail and frail older adults 72 72.4 (5.7) 56 (77.8) ≥25
Leon (2015) (45) RCT, 12-wk intervention Healthy older adults 138 71.4 (5.6) 106 (76.8) Not reported
Linde (2014) (46) RCT, 4-mo intervention, single-blind Healthy older adults 55 67 (3.34) 41 (74.5) Not reported
with 12-wk follow-up
Lord (2003) (47) Cluster RCT, 12-mo intervention Frail older adults 551 79.5 (6.4) 474 (86) ≥20
Napoli (2014) (48) RCT, 12-mo intervention, single-blind Sedentary obese older adults 107 69.9 (4) 67 (62.6) 95.7 (0.8)c
Nascimento (2014) (49) QE, 16-wk intervention, single-blind Older adults with and without MCI 67 67.6 (6.2) 44 (65.7) 23.8 (4.3)d
Okumiya (1996) (50) RCT, 6-mo intervention, single-blind Healthy older adults 42 78.8 (4.7) 24 (57.1) 27.9 (2.6)
Rehfeld (2018) (51) RCT, 6-mo intervention Healthy older adults 38 68.4 (3.5) 20 (52.6) 28.6 (0.9)
Rosano (2017) (52) RCT, 24-mo intervention, single-blind Sedentary older adults 26 75.1 (7.7) 21 (80.8) 92.3 (8.5)e
Shah (2014) (53) QE, 16-wk intervention Healthy older adults 222 67.6 (5.2) 153 (68.9) 28.6 (1.4)
Silva (2019) (54) RCT, 3-mo intervention, single-blind Older adults with MCI 19 75 (5.5) 11 (57.9) 29 (26 to 30)f
Sink (2015) (55) RCT, 24-mo intervention, single-blind Sedentary older adults 1476 78.9 (5.2) 999 (67.7) 91.7 (5.4)g

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Styliadis (2015) (56) QE, 8-wk intervention Older adults with MCI 70 70.6 (5.2) 45 (64.3) 25.7 (2.3)
Tarazona-Santabalbina RCT, 24-wk intervention, single-blind Frail sedentary older adults 100 80 (3.65) 54 (54) 26.9 (5.6)
(2016) (57)

Exercise and Cognition in Older Adults

 (n = 6), percentage of HR reserve (n = 3), percentage of HR
peak (n = 1), and other methods (n = 4). AET types included
continuous endurance activities such as walking, cycling,
Not reported

not reported and dancing (Table 2).

37.6 (3.6)h
24.4 (3.3)
25.9 (2.2)

≥21
Resistance exercise training
RET was prescribed on average 3.2 d·wk−1 (SD = 1.5 d·wk−1,
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

n = 33), lasting on average 23.6 min·d−1 (SD = 11.3 min·d−1,

n = 25), with studies using low to moderate (n = 2), moderate
(n = 8), moderate to high (n = 6), and high (n = 2) intensity.
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023

92 (69.7)

72 (70.6)
29 (100)
49 (100)

Only 18 studies measured RET intensity, using a diversity of

187 (100)
24 (60)

methods, which included RPE (n = 8), maximum repetitions

(n = 6), and others (n = 4). RET type included bodyweight,
machine-based, and free weights, with 1 to 4 sets and 4 to
MMSE score (or equivalent test) to indicate cognitive status of participants at baseline. Data reported as mean (standard deviation) or otherwise indicated.

30 repetitions per muscle group.

80.2 (8.2)
68.9 (3.3)
69.1 (5.7)
69.2 (5.4)

77.4 (4.3)
71.7 (5.4)

Balance and flexibility exercise training

For the 18 studies that included balance and flexibility
training components, these were administered on average
3.4 d·wk−1 (SD = 1.8, n = 18), 16.9 min·d−1 (SD = 10.3,
n = 13) at low (n = 1) to moderate (n = 2) intensity. Only three
studies reported measures of intensity, and these were either
29
40
49

102
187
132

verbally described as “moderate intensity” (n = 2) or reported

as RPE (n = 1). Balance and flexibility training type involved ac-
tivities such a static and dynamic balance, postural sway,
double- and single-leg stance variations, range of motion exer-
cises, stretching, and mobility of main muscle joints.
RCT, 16-wk intervention, single-blind Sedentary older women
Sedentary older adults

RCT, 12-mo intervention, single-blind Sedentary older adults

RCT, 12-mo intervention, single-blind Sedentary older adults

Healthy older women

Healthy older women

Overall Effects of MME on Cognition

MMSE, Mini-Mental State Examination; QE, quasiexperimental; SCC, subjective cognitive complaint.

Details on study intervention, comparator, cognitive do-

mains and tests, as well as main findings are summarized in
Table 3. Also, in Table 4 we report a summary of the tests used
to assess cognitive function. Across all studies and compara-
tors, MME showed superior improvements in three measures
of global cognitive functioning, seven measures of executive
Telephone Interview for Cognitive Status, reported as mean (standard deviation).

functioning, seven measures of memory, and four measures

of processing speed (see Table 4 for more details). When com-
QE, 26-wk intervention, single-blind

piling evidence from the 33 included studies, the effects of

Montreal Cognitive Assessment, reported as median (interquartile range).

MME on cognition were considered mixed and heavily depen-

dent on study designs, comparators, and outcomes. Aiming to
facilitate coherence and to contextualize the evidence, we
RCT, 12-mo intervention
RCT, 3-mo intervention

stratified our findings based on the differences between MME

Telephone Interview for Cognitive Status—Modified score.

Modified MMSE, reported as median (interquartile range).

and comparators (i.e., competing treatment, active control,

Modified MMSE, reported as mean (standard deviation).

and no-treatment control groups) on the outcomes of interest

Modified MMSE, reported as mean (standard error).

MMSE, reported as median (minimum − maximum).

(i.e., cognitive domains and tests used in the studies). Evidence

from studies that included two or more comparators was con-
sidered separately for each applicable comparison, where mul-
tiple comparisons were reported by authors (e.g., MME vs
active control, or competing treatment). Results are reported
in the following subsections.

MME compared with competing treatment

A total of 17 studies included one or more competing treat-
Taylor-Piliae (2010) (58)

Williamson (2009) (63)

ment groups. These included varied forms of cognitive train-

Vaughan (2014) (60)
Vedovelli (2017) (61)
Williams (1997) (62)
Teixeira (2018) (59)

ing (n = 8) (36,38–40,42,43,46,53,56), physical training

(n = 6) (31,32,35,37,51,58), combined physical and cognitive
training (n = 7) (33,36,37,45,46,53,56), or combined physical
training and diet (n = 1) (48). In only two studies, MME
imparted similar improvements to competing treatment com-
pared with no-treatment control group (42,48). For example,
b

h
a

g
c

one study reported that MME was of similar effectiveness

www.acsm-csmr.org Current Sports Medicine Reports 301

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Table 2.

302
MME intervention details, including frequency, intensity, time and type, and comparator(s).
Multimodality
First Author Exercise Intervention Aerobic Training Resistance Balance and/or Flexibility
(Year) Summary Details Training Details Training Details Comparator(s)
Ansai (2015) (31) Aerobic, resistance, and Frequency: 3 d·wk−1 Frequency: 3 d·wk−1 Frequency: 3 d·wk−1 Comparator 1: resistance
balance training for 16 wk for 16 wk for 16 wk training, machine-based
(60 min·d−1, 3 d·wk−1) Intensity: 60% to 85% HRR Intensity: 14 to 17 Intensity: Not (60 min·d−1, 3 d·wk−1)
for 16 wk (n = 23) Time: 13 min·d−1 RPE (20-pt Borg) reported for 16 wk (n = 23)
Adherence = 34.7% Type: Cycling Time: 15 to 20 min·d−1, Time: 10 min·d−1 Adherence = 56.5%
≤15 reps, ≤3 sets Type: Static balance, Comparator 2:
Type: Free-weights, dynamic and static No-treatment control
bodyweight weight transfer group (n = 23)
Adherence = Not

Volume 19  Number 8  August 2020

applicable
Berryman Aerobic, resistance (lower High-intensity interval Frequency: 3 d·wk−1 Not applicable Comparator 1:
(2014) (32) body), and balance training for 2 mo Aerobic, resistance (upper
training (60 min·d−1, Frequency: 2 d·wk−1 Intensity: RM body), and balance
3 d·wk−1) for 2 mo for 2 mo Intensity: Time: 4 sets, 4 to 6 training (60 min·d−1,
(n = 16) Maximal aerobic or 12 to 20 reps Type: 3 d·wk−1) for 2 mo
Attendance = 96.9% power Time: 4 to Upper or lower-body, (n = 15)
7 min (2 sets, 15 s machine-based Attendance = 96.9%
on 15 s off ) Type: Cycling Comparator 2:
Continuous training Stretching, relaxation
Frequency: 1 d·wk−1 and ball manipulation
for 2 mo Intensity: 60% exercises (60 min·d−1,
maximal aerobic power 3 d·wk−1) for
Time: 20 min·d−1 2 mo (n = 16)
Type: Cycling Attendance = 96.9%
Boa Sorte Silva Aerobic, resistance, Frequency: 3 Frequency: 3 d·wk−1 Frequency: 3 d·wk−1 Aerobic and resistance
(2018) (33) and balance training d·wk−1 for 24 wk for 24 wk for 24 wk training combined with
(60 min·d−1, 3 d·wk−1) Intensity: 65% to Intensity: Not Intensity: Not mind-motor training
for 24 wk (n = 64) 85% HRmax reported reported (60 min·d−1, 3 d·wk−1)
Adherence = 68% Time: 20 min·d−1 Time: 10 min·d−1 Time: 15 min·d−1 for 24 wk (n = 63)
Type: Walking, marching Type: Resistance Type: Balance, range Adherence = 72%
and sequenced aerobics bands, wall or of motion, and
chair exercises breathing exercises
Callisaya (2017) (34) Aerobic and resistance Frequency: 2 d·wk−1 Frequency: 3 d·wk−1 Not applicable Upper and lower limb
training (60 min·d−1, (supervised) plus 1 d·wk−1 for 6 mo stretching and gentle
3 d·wk−1) for 6 mo (home-based) for 6 mo Intensity: 14 to 17 RPE movement program
(n = 26) Intensity: 12 to 16 RPE (20-pt Borg) (60 min·d−1, 3 d·wk−1)
Attendance = 79% (20-pt Borg) Time: Time: 30 min·d−1, 3 sets, for 6 mo (n = 24)
30 min·d−1 (supervised) 8 to 12 reps Type: Upper Attendance = 76%

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
plus 60 min·d−1 and lower extremity,
(home-based) Type: bodyweight, machine
Cycling, cross-trainer, and free weights
rower or treadmill

Exercise and Cognition in Older Adults

 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Carral (2008) (35) Water-based aerobic Frequency: 2 d·wk−1 Frequency: 3 d·wk−1 for 5 mo Not applicable Water-based aerobic
and resistance training for 5 mo Intensity: Intensity: 75% 1RM training and calisthenic
(90 min·d−1, 3 d·wk−1) Not reported Time: 45 min·d−1, 3 sets, training (90 min·d−1,
for 5 mo (n = 27) Time: 45 min·d−1 Type: 10 reps Type: Machine-based 3 d·wk−1) for
Adherence = Unclear Water-based movements 5 mo (n = 29)
and continuous swimming Adherence = Unclear

www.acsm-csmr.org
Damirchi (2018) (36) Aerobic and resistance Frequency: 3 d·wk−1 Frequency: 3 d·wk−1 for 8 wk Not applicable Comparator 1:
training (60 min·d−1, for 8 wk Intensity: 13 to 15 RPE Cognitive training (30 to
3 d·wk−1) for 8 wk Intensity: 55% to 75% (20-pt Borg) Time: 60 min·d−1, 3 d·wk−1)
(n = 11) HRR Time: 6 to 30 min·d−1 Type: Muscular for 8 wk (n = 11)
Adherence = 73.3% 25 min·d−1 Type: Walking strength and range of Adherence = 73.3%
motion exercises Comparator 2: aerobic
and resistance training
plus cognitive training
(90 to 120 min·d−1,
3 d·wk−1) for 8 wk
(n = 13)
Adherence = 86.6%
Comparator 3:
No-treatment control
group (n = 9)
Adherence =
Not applicable
Eggenberger Aerobic, resistance, Frequency: 2 d·wk−1 for Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 for 6 mo Comparator 1:
(2015) (37) and balance training 6 mo Intensity: 5 to for 6 mo Intensity: Not reported Time: Video game dancing,
(60 min·d−1, 1 d·wk−1) 7 RPE (10-pt Borg) Intensity: 5 to 7 RPE 20 min·d−1 Type: Double- and strength and balance
for 6 mo (n = 25) Time: 20 min·d−1 Type: (10-pt Borg) Time: single-leg stance variations, on training (60 min·d−1,
Adherence = 85.3% Treadmill walking 20 min·d−1, 1 to 3 sets, the floor or unstable surfaces 1 d·wk−1) for 6 mo
8 to 12 reps Type: (n = 24)
Bodyweight, rubber bands, Adherence = 82.1%
weight vests Comparator 2: treadmill
walking with verbal
memory exercise,
strength, and balance
training (60 min·d−1,
1 d·wk−1) for 6 mo
(n = 22)
Adherence = 88.1%
Fissler (2017) (38)a Aerobic, resistance, Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Not applicable Comparator 1: Cognitive
coordination, balance, (supervised) plus 3 d·wk−1 (supervised) plus 3 d·wk−1 training (60 min·d−1,
and flexibility training (home-based) for 10 wk (home-based) for 10 wk 5 d·wk−1) for 10 wk (n = 11)
(60 min·d−1, 2 d·wk−1) Intensity: Not reported Intensity: Not reported Time: Adherence = Not reported
for 10 wk (n = 12) Time: Not reported Type: Not reported Type: Not Comparator 2: no-treatment
Adherence = Not reported Not reported reported control group (n = 16)

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Adherence = Not applicable
Continued next page

Current Sports Medicine Reports

303
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Table 2.

304
Continued.
Multimodality
First Author Exercise Intervention Aerobic Training Resistance Balance and/or Flexibility
(Year) Summary Details Training Details Training Details Comparator(s)
−1 −1
Gajewski (2012, Aerobic and resistance Frequency: 2 d·wk Frequency: 2 d·wk Not applicable Comparator 1:
2018) (39,40)b training (90 min·d−1, for 4 mo Intensity: for 4 mo Intensity: Cognitive training
2 d·wk−1) for 4 mo (n = 35) Not reported Not reported Time: (90 min·d−1, 2 d·wk−1)
Adherence = Not reported Time: 30 min·d−1 30 min·d−1, 3 sets, for 4 mo (n = 32)
(cardiovascular) plus 15 reps Type: Adherence = Not reported
30 min·d−1 (aerobic) Machine-based Comparator 2: relaxation
Type: Treadmills, bicycles, training (90 min·d−1,
cross trainers, easy step 3 d·wk−1) for 4 mo (n = 34)
and floor movement Adherence = Not reported

Volume 19  Number 8  August 2020

sequences Comparator 3: no-treatment
control group (n = 40)
Adherence = Not applicable
Ji (2017) (41) Aerobic, resistance, and Frequency: 7 d·wk−1 Frequency: 7 d·wk−1 Not applicable no-treatment control
balance training via for 6 wk Intensity: Not for 6 wk Intensity: group (n = 12)
Nintendo Wii Fit reported Time: Not reported Not reported Time: Not Adherence = not applicable
(30 min·d−1, 7 d·wk−1) Type: Nintendo Wii reported Type: Nintendo
for 6 wk (n = 12) fit exercises Wii fit exercises
Adherence = Not reported
Klusmann (2010) Aerobic, resistance, Frequency: 3 d·wk−1 Frequency: 3 d·wk−1 Frequency: 3 d·wk−1 Comparator 1:
(42) balance, and flexibility for 6 mo Intensity: for 6 mo Intensity: for 6 mo Intensity: cognitive training
training (90 min·d−1, Not reported Time: Not reported Not reported Time: (90 min·d−1, 3 d·wk−1)
3 d·wk−1) for 6 mo (n = 80) 30 min·d−1 Type: Cycling Time: Not reported Type: Not reported Type: for 6 mo (n = 81)
Adherence = Not Reported Not reported Not reported Adherence = not reported
Comparator 2:
no-treatment control
group (n = 69)
Adherence = not applicable
Küster (2016) (43) Aerobic, resistance, Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Not applicable Comparator 1:
coordination, balance, (supervised) plus (supervised) plus 3 d·wk−1 cognitive training
and flexibility training 3 d·wk−1 (home-based) (home-based) for 10 wk (60 min·d−1, 5 d·wk−1)
(60 min·d−1, 2 d·wk−1) for 10 wk Intensity: Intensity: Not reported for 10 wk (n = 16)
for 10 wk (n = 18) Not reported Time: Time: Not reported Type: Adherence = 99.8%
Adherence = 77.1% Not reported Type: Not reported Comparator 2:
Not reported no-treatment control
group (n = 20)
Adherence = not applicable
Langlois (2013) (44) Aerobic and resistance Frequency: 3 d·wk−1 Frequency: 3 d·wk−1 Not applicable No-treatment control
training (60 min·d−1, for 12 wk Intensity: for 12 wk Intensity: group (n = 36)

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
3 d·wk−1) for 12 wk Moderate to hard intensity Not reported Time: Adherence = not applicable
(n = 36) RPE (10-pt Borg) Time: 10 min·d−1 Type:
Adherence = 84% 10 to 30 min·d−1 Type: Not reported
Treadmills, recumbent
bikes, and elliptical

Exercise and Cognition in Older Adults

 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Leon (2015) (45) Aerobic and resistance Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Not applicable Comparator 1: aerobic
training (60 min·d−1, for 12 wk Intensity: for 12 wk Intensity: and resistance training
2 d·wk−1) for 12 wk Not reported Time: Not reported Time: combined with cognitive
(n = 46) 30 min·d−1 Type: Dancing 15 min·d−1 Type: training (60 min·d−1,
Adherence = Not Reported and circuit training Bodyweight, 2 d·wk−1) for 12 wk (n = 57)
elastic bands Adherence = not reported

www.acsm-csmr.org
Comparator 2: no-treatment
control group (n = 35)
Adherence = not applicable
Linde (2014) (46) Aerobic and resistance Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Not applicable Comparator 1: cognitive
training (60 min·d−1, for 16 wk Intensity: for 16 wk Intensity: training (30 min·d−1,
2 d·wk−1) for 16 wk 40% to 70% HRR Time: Moderate intensity 1 d·wk−1) for 16 wk (n = 11)
(n = 15) 40 min·d−1 Type: Walking Time: 20 min·d−1, 10 Adherence = 81%
Adherence = 81% or running to 20+ reps Comparator 2: aerobic
Type: Not reported and resistance training
(60 min·d−1, 2 d·wk−1)
combined with cognitive
training (30 min·d−1,
1 d·wk−1) for 16 wk (n = 16)
Adherence = 81%
Comparator 3:
No-treatment control
group (n = 13)
Adherence = Not Applicable
Lord (2003) (47) Aerobic, resistance, and Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Comparator 1: flexibility
balance training for 12 mo for 12 mo Intensity: for 12 mo Intensity: and relaxation exercises
(60 min·d−1, 2 d·wk−1) Intensity: Not reported Not reported Time: 4 to Not reported Time: 60
( min·wk−1, 2 d·wk−1)
for 12 mo (n = 259) Time: Not reported Type: 30 reps Type: Bodyweight Not reported Type: for 12 mo (n = 80)
Adherence = 42.3% Leg, trunk, and Tandem foot standing, Adherence = 45.4%
arm exercises standing on one leg, Comparator 2: no-treatment
altering the base of control group (n = 169)
support, etc. Adherence = not applicable
Napoli (2014) (48) Aerobic, resistance, Frequency: 3 d·wk−1 Frequency: 3 d·wk−1 Frequency: 3 d·wk−1 Comparator 1: diet that
and balance training for 12 mo Intensity: 65 for 12 mo Intensity: for 12 mo Intensity: provided energy deficit
(90 min·d−1, 3 d·wk−1) to 85% HRpeak Time: 65 to 80% 1RM Not reported Time: of 500 to 750 kcal·d−1
for 12 mo (n = 26) 30 min·d−1 Type: Walking, Time: 30 min·d−1, 1 to 15 min·d−1 Type: with goal to achieve
Adherence = 88% cycling and stair climbing 3 sets, 6 to 12 reps Not reported ~10% weight loss (n = 26)
Type: Machine-based Adherence = 83%
Comparator 2: aerobic,
resistance, and balance
training (90 min·d−1,
3 d·wk−1) combined
with diet for 12 mo (n = 28)
Adherence = 83%

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Comparator 3:
No-treatment control

Current Sports Medicine Reports

group (n = 27)
Adherence = Not applicable
Continued next page

305
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Table 2.

306
Continued.
Multimodality
First Author Exercise Intervention Aerobic Training Resistance Balance and/or Flexibility
(Year) Summary Details Training Details Training Details Comparator(s)
−1 −1 −1
Nascimento Aerobic, resistance, Frequency: 1 d·wk Frequency: 1 d·wk Frequency: 1 d·wk No-treatment control
(2014) (49) and balance training for 16 wk Intensity: 60% for 16 wk for 16 wk Intensity: group (n = 32)
(60 min·d−1, 3 d·wk−1) to 80% HRmax Time: Intensity: RM Time: Moderate intensity Adherence = not applicable
for 16 wk (n = 35) 45 min·d−1 45 min·d−1, 3 sets, Time: 45 min·d−1
Adherence ≥75% Type: Walking and marching 15 to 20 reps Type: Type: Recreational
Free weights (e.g., activities stimulating
rubber-bands) visual, vestibular and
somatosensory systems
Okumiya Aerobic, resistance, Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 No-treatment control

Volume 19  Number 8  August 2020

(1996) (50) and balance training for 24 wk Intensity: for 24 wk Intensity: for 24 wk Intensity: group (n = 21)
(60 min·d−1, 2 d·wk−1) Light Time: Not reported Light Time: Not reported Not reported Time: Not Adherence = not applicable
for 24 wk (n = 21) Type: Walking, game playing Type: Bodyweight reported
Adherence = 86% Type: Not reported
Rehfeld (2018) (51) Aerobic, resistance, Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Dance training
and balance training for 6 mo Intensity: for 6 mo Intensity: for 6 mo Intensity: (90 min·d−1, 2 d·wk−1)
(90 min·d−1, 2 d·wk−1) Not reported Time: Not reported Not reported for 6 mo (n = 18)
for 6 mo (n = 18) 20 min·d−1 Type: Time: 20 min·d−1 Time: 20 min·d−1 Adherence = ≥70%
Attendance ≥70% Bicycle ergometers Type: Free weights (e.g., Type: Not reported
barbells, rubber bands, etc.)
Rosano (2017) (52)c Aerobic, resistance, Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Health and education
balance, and flexibility (supervised) plus 3 to (supervised) plus 3 to (supervised) plus 3 to sessions (60 to
training (≤50 min·d−1, 4 d·wk−1 (home-based) 4 d·wk−1 (home-based) for 4 d·wk−1 (home-based) 90 min·d−1, weekly to
2 d·wk−1 [supervised] for 24 mo 24 mo Intensity: 15 to 16 for 24 mo monthly) for 24 mo (n = 16)
and 3 to 4 d·wk−1 Intensity: 13 RPE (20-pt RPE (20-pt Borg) Time: Intensity: Not reported Adherence = 90.6%
[home-based]) for 24 mo Borg) Time: ≤30 min·d−1 10 min·d−1, 2 sets, 10 reps Time: 10 min·d−1
(n = 10) Type: Walking Type: Lower-extremity, Type: Balance and larger
Adherence = 66.7% ankle weights muscle flexibility
Shah (2014) (53) Aerobic (60 min·d−1, Frequency: 3 d·wk−1 Frequency: 2 d·wk−1 Not applicable Comparator 1: Cognitive
3 d·wk−1) and resistance for 16 wk Intensity: Low for 16 wk Intensity: training (60 min·d−1,
training (40 min·d−1, intensity Time: 60 min·d−1 Not reported Time: 40 5 d·wk−1) for 16 wk (n = 51)
2 d·wk−1) for 16 wk Type: Walking min·d−1 Type: Free Adherence = not reported
(n = 42) weights and bodyweight Comparator 2: aerobic
Adherence = Not reported and resistance training
(60 min·d−1, 5 d·wk−1)
combined with cognitive
training (60 min·d−1,
5 d·wk−1) for 16 wk (n = 44)
Adherence = not reported

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Comparator 3:
No-treatment control
group (n = 35)
Adherence = Not applicable

Exercise and Cognition in Older Adults

 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Silva (2019) (54) Aerobic, resistance, Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 No-treatment control
and balance training for 12 wk Intensity: for 12 wk for 12 wk Intensity: group (n = 12)
(60 min·d−1, 2 d·wk−1) 80% HRmax Time: Intensity: Not reported Not reported Time: Adherence = not applicable
for 12 wk (n = 7) 30 min·d−1 Type: Time: 20 min, 3 sets, 10 min·d−1 Type:
Adherence = 90% Treadmill training 8 to 12 reps Static balance exercises
Type: Machine-based and stretching

www.acsm-csmr.org
prioritizing mobility
of the main joints
Sink (2015) (55) Aerobic, resistance, Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Health and education
balance, and flexibility (supervised) plus 3 to (supervised) plus 3 to (supervised) plus 3 to sessions (60 to
training (≤50 min·d−1, 4 d·wk−1 (home-based) 4 d·wk−1 (home-based) 4 d·wk−1 (home-based) 90 min·d−1, weekly to
2 d·wk−1 [supervised] for 24 mo Intensity: for 24 mo Intensity: 15 to for 24 mo Intensity: monthly) for 24 mo (n = 741)
and 3 to 4 d·wk−1 13 RPE (20-pt Borg) Time: 16 RPE (20-pt Borg) Time: Not reported Time: Adherence = Not reported
[home-based]) for 24 mo ≤30 min·d−1 Type: Walking 10 min·d−1, 2 sets, 10 reps 10 min·d−1 Type:
(n = 735) Type: Lower-extremity, Balance and larger
Adherence = 71% ankle weights muscle flexibility
Styliadis (2015) (56) Aerobic, resistance, Frequency: 5 d·wk−1 Frequency: 5 d·wk−1 Frequency: 5 d·wk−1 Comparator 1:
and balance training for 8 wk Intensity: for 8 wk Intensity: for 8 wk Intensity: Cognitive training
via Nintendo Wii Not reported Time: Not reported Not reported (60 min·d−1, 3 to
(60 min·d−1, 5 d·wk−1) 20 min·d−1 Type: Time: 8 to 10 min·d−1, Time: 10 min·d−1 Type: 5 d·wk−1) for 8 wk (n = 14)
for 8 wk (n = 14) Exergaming via 8 to 10 exercises Type: Exergaming via Adherence = 60.9%
Adherence = 65.2% Nintendo Wii Exergaming Nintendo Wii Comparator 2:
via Nintendo Wii Aerobic, resistance, and
balance training via
Nintendo Wii
(60 min·d−1, ≤5 d·wk−1)
plus Cognitive
(60 min·d−1, ≤5 d·wk−1)
for 8 wk (n = 14)
Adherence = 65.5%
Comparator 3:
Active control group (e.g.,
watching documentaries
60 min·d−1, 5 d·wk−1)
for 8 wk (n = 14)
Adherence = 67.1%
Comparator 4:
No-treatment control group
(n = 14)
Adherence = Not applicable
Tarazona-Santabalbina Aerobic, resistance, Frequency: 3 d·wk−1 Frequency: 2 d·wk−1 Frequency: 5 d·wk−1 No-treatment control
(2016) (57) balance, and flexibility for 24 wk Intensity: 40% for 24 wk Intensity: for 24 wk Intensity: group (n = 49)
training (65 to to 65% HRmax Time: 25% to 75% 1RM Time: Not reported Time: 5 to
70 min·d−1, 5 d·wk−1) 40 min·d−1 Type: Walking 40 min·d−1 Type: 15 min·d−1 Type: Postural

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
for 24 wk (n = 51) and climbing stairs Resistance bands, sway and dynamic balance,
Adherence = 47.3% isometric, concentric coordination, and flexibility

Current Sports Medicine Reports

and eccentric exercises
Continued next page

307
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Table 2.

308
Continued.
Multimodality
First Author Exercise Intervention Aerobic Training Resistance Balance and/or Flexibility
(Year) Summary Details Training Details Training Details Comparator(s)
−1 −1 −1
Taylor-Piliae (2010) Aerobic, resistance, Frequency: 1 to 2 d·wk Frequency: 1 to 2 d·wk Frequency: 1 to 2 d·wk Comparator 1:
(58) balance, and flexibility (supervised) 3 d·wk−1 (supervised) 3 d·wk−1 (supervised) 3 d·wk−1 Tai Chi (60 min·d−1, 1 to
training (60 min·d−1, 1 (home-based) for 12 mo (home-based) for 12 mo (home-based) for 12 mo 2 d·wk−1 [supervised]
to 2 d·wk−1 [supervised] Intensity: Moderate Intensity: Moderate Time: Intensity: moderate time: plus 3 d·wk−1
plus 3 d·wk−1 Time: 15 to 25 min·d−1 15 to 20 min·d−1 not reported [home-based]) for
[home-based]) for 12 mo (supervised) plus (supervised) plus 10 to Type: not reported 12 mo (n = 37)
(n = 39) ≥30 min·d−1 (home-based) 25 min·d−1 (home-based) Adherence = 77%
Adherence = 68% Type: Walking and calisthenics Type: Calisthenics, Comparator 2:
performed to music free-weight and Healthy aging classes

Volume 19  Number 8  August 2020

rubber bands (90 min·d−1, 1 d·wk−1)
for 6 mo (n = 56)
Adherence = 67%
Teixeira (2018) (59) Aerobic and resistance Frequency: 1 d·wk−1 Frequency: 2 d·wk−1 Not applicable No-treatment control
training (20 to 30 min·d−1, for 6 mo Intensity: 70% for 6 mo Intensity: 70% to group (n = 20)
3 d·wk−1) for 6 mo to 90% HRmax Time: 90% HRmax Time: Adherence = Not applicable
(n = 20) 20 to 30 min·d−1 Type: 20 to 30 min·d−1 Type:
Adherence = 66.8% Outdoor walking and Circuit training including
jogging, circuit training resistance exercises using
rubber bands
Vaughan (2014) (60) Aerobic, resistance, Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 No-treatment control
balance, and flexibility for 16 wk Intensity: 3 to for 16 wk Intensity: 4 to for 16 wk Intensity: 3 to group (n = 23)
training (60 min·d−1, 6 RPE (10-pt Borg) Time: 6 RPE (10-pt Borg) Time: 4 RPE (10-pt Borg) Time: Adherence = Not applicable
2 d·wk−1) for 16 wk 10 to 15 min·d−1 Type: 10 to 15 min·d−1 Type: 7 to 30 min·d−1 Type:
(n = 25) Freestyle aerobics and Free-weights, bodyweight Static and dynamic balance,
Adherence = 85.7% circuit training coordination and agility,
and reaction time
Vedovelli (2017) (61) Aerobic and resistance Frequency: 3 d·wk−1 Frequency: 3 d·wk−1 Not applicable No-treatment control
training (60 min·d−1, for 3 mo Intensity: 75% to for 3 mo Intensity: group (n = 9)
3 d·wk−1) for 3 mo 85% HRmax Time: 50% to 75% 1RM Time: Adherence = Not applicable
(n = 22) ≤30 min·d−1 ype: walking 30 min·d−1, 3 sets,
Adherence = 100% 10 reps plus 10 s isometric
holds Type: Resistance
bands, bodyweight
Williams (1997) (62) Aerobic, resistance, Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 Frequency: 2 d·wk−1 No-treatment control
and balance training for 12 mo for 12 mo Intensity: for 12 mo Intensity: group (n = 93)
(60 min·d−1, 2 d·wk−1) Intensity: Not reported not reported Time: not Not reported Time: Adherence = Not applicable
for 12 mo (n = 94) Time: Not reported Type: reported Type: body weight Not reported Type:
Adherence = 72% continuous movement Standing on one leg,

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
of the legs and trunk and ball games, hand-eye and
intermittent arm movement foot-eye coordination

Exercise and Cognition in Older Adults

 compared with cognitive training in improving executive func-
sessions (weekly to monthly)

Adherence = Not reported tioning and memory (42). Another study showed that MME
was equally effective compared with combined physical exer-
cise and diet intervention in improving global cognitive func-
Health and education

for 24 mo (n = 52)

tioning and executive functioning (48).

No other studies reported superiority of MME in improving
cognition when compared with competing treatment groups.
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Furthermore, the overall effects of the competing treatment

groups were seen to be either superior to (n = 7) (36,39,40,45,
46,53,58), or equivalent (n = 8) (31–33,35,37,43,51,56) to
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023

MME in the remaining studies (as reported in Table 3). For

studies showing superiority of competing treatment groups,
the findings showed that cognitive training alone was superior
Frequency: 1 to 3 d·wk−1

to MME in improving measures of processing speed (39) and

5 d·wk−1 (home-based)

Time: 10 min·d−1 Type:

(supervised) plus 1 to

executive functioning (40), and one study showed that physi-

for 24 mo Intensity:

cal training (i.e., Tai Chi) was superior to MME in improving

measures of executive functioning (58). Furthermore, combin-
ing physical (i.e., MME) and cognitive training seemed to
Not reported

not reported

yield the greatest benefits in measures of executive functioning

(36), processing speed (36,46), and memory (53).

MME compared with active control

Nine of 33 studies included an active control group and re-
lower extremity, free weights

ported cognition outcomes (32,34,39,40,47,55,56,58,63). Of

2 sets, 10 repetitions Type:
Frequency: 2 to 3 d·wk−1

these, one study indicated that MME was effective in improv-

Borg) Time: 10 min·d−1,
3 d·wk−1 (home-based)
(supervised) plus 1 to

ing measures of global cognitive function, executive functioning

15 to 16 RPE (20-pt
for 24 mo Intensity:

and memory (34), and another study reported improvements in

measures of processing speed (47). All other studies did not re-
port results supporting MME imparting superior effects in
cognition when compared with active control groups (n = 7)
(32,39,40,55,56,58,63).

MME compared with no-treatment control

Twenty-three studies included a no-treatment control group.
13 RPE (20-pt Borg) Time:
Frequency: 2 to 3 d·wk−1

40 min·d−1 Type: walking

The overall evidence suggested that MME was effective in im-

3 d·wk−1 (home-based)

proving many aspects of cognitive function. For instance, a

(supervised) plus 1 to

number of studies reported improvements in measures of mem-

for 24 mo Intensity:

ory (n = 4) (41,42,59,61), processing speed (n = 5) (44,47,60–62),

executive functioning (n = 7) (42,44,48,54,60–62) and global
cognitive functioning (n = 3) (48,49,57). The remaining stud-
ies did not report significant results (n = 11) (31,36,38–40,43,45,
46,50,53,56).
HRmax, maximum heart rate; RM, maximum repetition.

MME and Neuroimaging Outcomes

training (40 to 60 min·d−1,
1 to 3 d·wk−1 [supervised]

Adherence = Not reported

Nine studies included neuroimaging outcomes (34,38–41,51,

[home-based]) for 24 mo

Secondary outcomes from Gajewski et al 2012.

52,56,59). These involved structural (n = 6) (34,38,41,51,52,59)

balance, and flexibility

Secondary outcomes from Küster et al 2016.

and functional (n = 1) (41) magnetic resonance imaging (MRI)

Aerobic, resistance,

Secondary outcomes from Sink et al 2015.

and 1 to 5 d·wk−1

data, as well as electroencephalogram (EEG, n = 3) (39,40,56)

data (see Table 3 for details).
For MRI outcomes, one study reported no significant dif-
ferences between MME compared with cognitive training
(n = 50)

and no-treatment control (38) in white matter integrity (i.e.,

fractional anisotropy). Another study, however, reported
MME was associated with improvements in gray matter (oc-
cipital and cerebellar regions) and white matter (right tempo-
Williamson (2009)

ral and right occipital regions) volumes, compared with

dance training (51). Compared with active control groups,
two studies reported that MME was associated with greater
improvements in hippocampal volume (34,52), and one study
(63)

reported increased white matter integrity (i.e., fractional anisot-

b
a

ropy) and total brain volume (34). Furthermore, compared with

www.acsm-csmr.org Current Sports Medicine Reports 309

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

310
Table 3.
Summary of study interventions, outcomes, and main findings.
First Author Cognitive Domain(s) Cognitive Test(s)
(Year) Treatment Group(s) Comparator(s) Assessed (Outcomes) (Measures Used) Main Findings
Ansai (2015) (31) Aerobic, resistance, and Comparator 1: 1,2) Global cognitive 1) Montreal Cognitive No within- or
balance training Resistance training, function Assessment between-group
(60 min·d−1, 3 d·wk−1) machine-based 3) Executive function 2) Clock Drawing Test differences at follow-up.
for 16 wk (n = 23) (60 min·d−1, 3 d·wk−1) 3) Verbal Fluency
Adherence = 34.7% for 16 wk (n = 23)

Volume 19  Number 8  August 2020

Adherence = 56.5%
Comparator 2:
No-treatment control
group (n = 23)
Adherence = Not applicable
Berryman (2014) (32) Aerobic, resistance (lower Comparator 1: 1) Inhibition under 1, 2) Random Number Improvements in inhibition
body), and balance Aerobic, resistance single-task condition Generation Task (single-task), inhibition
training (60 min·d−1, (upper body), and balance 2) Working memory 3, 4) Random Number and working memory
3 d·wk−1) for 2 mo training (60 min·d−1, under single-task Generation Task while (dual-task) in all groups.
(n = 16) 3 d·wk−1) for 2 mo (n = 15) condition walking on treadmill No between-group
Attendance = 96.9% Attendance = 96.9% 3) Inhibition under differences at follow-up.
Comparator 2: dual-task condition
Stretching, relaxation and 4) Working memory
ball manipulation exercises under dual-task
(60 min·d−1, 3 d·wk−1) condition
for 2 mo (n = 16)
Attendance = 96.9%
Boa Sorte Silva Aerobic, resistance, and Aerobic and resistance 1) Global cognitive 1 to 5) Cambridge Brain Improvements in global
(2018) (33) balance training training combined with function Sciences Cognitive cognitive function,
(60 min·d−1, mind-motor training 2) Concentration Battery concentration and
3 d·wk−1) for 24 wk (60 min·d−1, 3 d·wk−1) 3) Reasoning reasoning in both
(n = 64) for 24 wk (n = 63) 4) Planning groups. Improvements
Adherence = 68% Adherence = 72% 5) Memory in planning and
memory in combined
multiple-modality and
mind-motor training group.
No between-group
differences at follow-up.

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Exercise and Cognition in Older Adults
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Callisaya (2017) (34) Aerobic and resistance Upper and lower limb Cognition: Cognition: Greater improvements in
training (60 min·d−1, stretching and gentle (1 to 7) Global 1) Victoria Stroop cognition (global
3 d·wk−1) for 6 mo movement program cognitive function Test (Part C-D) cognitive function,
(n = 26) (60 min·d−1, 3 d·wk−1) 1 to 5) Executive 2) Trail Making Test executive function,
Attendance = 79% for 6 mo (n = 24) function (Part B-A) and memory) and brain
Attendance = 76% 6, 7) Memory 3) Digit Symbol structure (fractional

www.acsm-csmr.org
Neuroimaging: Coding Test anisotropy, total and
8) Brain total volume 4) Digit span Test hippocampal brain
9) White matter volume (WAIS-III) volume), compared
10) Hippocampal volume 5) Controlled Oral with control group.
(11) Cortical thickness Word Association Test
(12) Fractional anisotropy 6) Hopkins Verbal
(13) Mean diffusivity Learning Test — Revised
(7) Rey Complex
Figure Test
Neuroimaging:
(8–13) MRI
Carral (2008) (35) Water-based aerobic and Water-based aerobic training Global cognitive function Mini-Mental No between-group
resistance training and calisthenic training State Examination differences at follow-up.
(90 min·d−1, 3 d·wk−1) (90 min·d−1, 3 d·wk−1)
for 5 mo (n = 27) for 5 mo (n = 29)
Adherence = Unclear Adherence = Unclear
Damirchi (2018) (36) Aerobic and resistance Comparator 1: 1) Working memory 1) Digit Span Test (WAIS-III) Greater improvements in
training (60 min·d−1, Cognitive training (30 to 2) Processing speed 2) Digit Symbol Coding working memory and
3 d·wk−1) for 8 wk 60 min·d−1, 3 d·wk−1) for 3) Reaction time Test (WAIS-III) processing speed in
(n = 11) 8 wk (n = 11) 4) Inhibition (error number) 3, 4) Stroop Test combined MME and
Adherence = 73.3% Adherence = 73.3% cognitive training
Comparator 2: compared with
Aerobic and resistance MME group.
training plus cognitive
training (90 to 120 min·d−1,
3 d·wk−1) for 8 wk
(n = 13)
Adherence = 86.6%
Comparator 3:
No-treatment control
group (n = 9)
Adherence = Not applicable
Continued next page

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Current Sports Medicine Reports
311
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Table 3.

312
Continued.
First Author Cognitive Domain(s) Cognitive Test(s)
(Year) Treatment Group(s) Comparator(s) Assessed (Outcomes) (Measures Used) Main Findings
Eggenberger Aerobic, resistance, and Comparator 1: 1) Executive function 1) Trail-Making Test (Part B) Improvements in
(2015) (37) balance training Video game dancing, 2) Working memory 2) Executive Control Task executive function,
(60 min·d−1, 1 d·wk−1) strength and balance 3) Long-term visual memory 3) Paired-Associates working memory,
for 6 mo (n = 25) training (60 min·d−1, 4) Long-term verbal memory Learning Task long-term visual and
Adherence = 85.3% 1 d·wk−1) for 6 mo 5) Short-term verbal memory 4) Logical Memory (Story verbal memory, attention
(n = 24) 6) Attention and recall, WMS-R) and processing speed
Adherence = 82.1% concentration 5) Digit Span Test (WMS-R) in all groups.
Comparator 2: 7,8) Processing speed 6) Age Concentration Tests No between-group
Treadmill walking with A and B differences at follow-up.
verbal memory exercise, 7) Trail-Making Test (Part A)

Volume 19  Number 8  August 2020

strength and balance 8) Digit Symbol Substitution
training (60 min·d−1, Task (WAIS-R)
1 d·wk−1) for 6 mo (n = 22)
Adherence = 88.1%
Fissler Aerobic, resistance, Comparator 1: Fractional anisotropy Diffusion Tensor No within- or
(2017) (38)a coordination, balance, Cognitive training Imaging via MRI between-group
and flexibility training (60 min·d−1, 5 d·wk−1) differences at follow-up.
(60 min·d−1, 2 d·wk−1) for 10 wk (n = 11)
for 10 wk (n = 12) Adherence = Not reported
Adherence = Not reported Comparator 2:
No-treatment control
group (n = 16)
Adherence = Not applicable
Gajewski Aerobic and resistance Comparator 1: Cognition: Cognition: Greater improvements in
(2012) (39) training (90 min·d−1, Cognitive training 1) Reaction times 1,2) Task Switching reaction time variability
2 d·wk−1) for 4 mo (90 min·d−1, 2 d·wk−1) 2) Executive function Test in cognitive training
(n = 35) for 4 mo (n = 32) Neuroimaging: Neuroimaging: group compared with
Adherence = Adherence = Not reported 3) Peak and amplitude 3) EEG MME training and
Not reported Comparator 2: of electrophysiological no-treatment
Relaxation training brain activity control group.
(90 min·d−1, 3 d·wk−1) EEG results suggested
for 4 mo (n = 34) higher improvements in
Adherence = Not reported event-related brain
Comparator 3: action potentials
No-treatment control associated with response
group (n = 40) selection, allocation
Adherence = Not applicable of cognitive resources
and error detection in
cognitive training group.

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Exercise and Cognition in Older Adults
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Gajewski (2018) (40)b Aerobic and resistance Comparator 1: Cognition: Cognition: Improvements in reaction
training (90 min·d−1, Cognitive training 1) Immediate verbal 1) Verbal Learning time in MME group at
2 d·wk−1) for 4 mo (90 min·d−1, 2 d·wk−1) memory and delayed and Memory Test follow-up, no change
(n = 35) for 4 mo (n = 32) word recognition 2) Word Fluency Test in other groups.
Adherence = Adherence = Not reported 2) Long-term 3) Digit Span Test Greater improvements in
Not reported Comparator 2: semantic memory 4) Rey-Osterrieth working memory in

www.acsm-csmr.org
Relaxation training 3) Short-term memory Complex Figure Test cognitive training group
(90 min·d−1, 3 d·wk−1) 4) Visuospatial memory 5) Digit Span Test compared with MME
for 4 mo (n = 34) 5, 6) Working memory 6) N-Back Task and both control groups.
Adherence = Not reported Neuroimaging: Neuroimaging: EEG suggested
Comparator 3: (7) Electrophysiological 7) EEG improvements in
No-treatment control brain activity underlying processing
group (n = 40) associated with
Adherence = working memory.
Not applicable
Ji (2017) (41) Aerobic, resistance, No-treatment control Cognition: Cognition: Greater improvements in
and balance training group (n = 12) 1) Immediate, delayed 1) Hopkins Verbal Learning emotional memory
via Nintendo Wii Fit Adherence = Not applicable and recognition recall Test — Revised recall, gray matter volume,
(30 min·d−1, 7 d·wk−1) 2) Immediate and delayed 2) Rivermead Behavioral and increased functional
for 6 wk (n = 12) story recall Memory Test connectivity in MME
Adherence = Not reported 3, 4) Executive function 3) Trail-Making Test group compared with
5) Working memory (Part B) control group.
6, 7) Processing speed 4) Stroop Test
8) Emotional memory recall 5) Digit Span Test
Neuroimaging: 6) Digit Symbol
(9) Gray matter volumes Substitution Test
(10) Resting state amplitude of 7) Trail-Making Test
low-frequency fluctuations (Part A)
(11) Regional homogeneity (8) Emotional
(12) Functional connectivity Memory task
Neuroimaging:
(9) Structural MRI
(10–12) Resting-state
functional MRI
Klusmann (2010) (42) Aerobic, resistance, Comparator 1: 1) Semantic verbal fluency 1) Verbal Fluency Greater improvements
balance, and flexibility Cognitive training 2, 3) Episodic memory 2) Story Recall in memory (immediate
training (90 min·d−1, (90 min·d−1, 3 d·wk−1) 4, 5) Executive function (RBMT) and delayed story recall,
−1
3 d·wk ) for 6 mo for 6 mo (n = 81) 3) Free and as well as delayed free
(n = 80) Adherence = Cued Selective recall) and executive
Adherence = Not Reported Not Reported Reminding Test function in MME group
Comparator 2: 4) Trail-Making Test as well as cognitive
No-treatment control (Part A/B) training group compared
group (n = 69) 5) Stroop Test with no-treatment
Adherence = control group.
Not Applicable

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Continued next page

Current Sports Medicine Reports

313
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Table 3.

314
Continued.
First Author Cognitive Domain(s) Cognitive Test(s)
(Year) Treatment Group(s) Comparator(s) Assessed (Outcomes) (Measures Used) Main Findings
Küster (2016) (43) Aerobic, resistance, Comparator 1: 1 to 9) Global 1) Phonematic Fluency No between-group
coordination, balance, Cognitive training cognitive function 2) Semantic Fluency differences at follow-up.
and flexibility training (60 min·d−1, 5 d·wk−1) 1 to 6) Executive function 3) Digit Span Test (WAIS-III)
(60 min·d−1, 2 d·wk−1) for 10 wk (n = 16) 6 to 8) Memory 4) Trail-Making Test
for 10 wk (n = 18) Adherence = 99.8% (Part A and B)
Adherence = 77.1% Comparator 2: 5) Digit Symbol Coding
No-treatment control Test (WAIS-III)
group (n = 20) 6) Computation Span (ECB)
Adherence = Not applicable 7) Free Recall (ADAS-cog)
(8) Munich Verbal

Volume 19  Number 8  August 2020

Learning Test
Langlois (2013) (44) Aerobic and resistance No-treatment control 1) Global cognitive function 1) Mini-Mental Greater improvements in
training (60 min·d−1, group (n = 36) 2) Abstract verbal reasoning State Examination processing speed,
3 d·wk−1) for 12 wk Adherence = 3 to 5) Processing speed 2) Similarities Test working memory, and
(n = 36) Not Applicable 6,7) Working memory (WAIS-III) executive function
Adherence = 84% 8) Episodic memory 3) Digit Symbol in MME compared
9,10) Executive function Coding Test with no-treatment
4) Trail-Making Test (Part A) control group.
5) Stroop Test
6) Letter-Number Sequencing
7) Digit Span Test (WAIS-III)
(8) Rey Auditory Verbal
Learning Test
(9) Trail-Making Test
(Part B-A)
(10) Stroop Test
Leon (2015) (45) Aerobic and resistance Comparator 1: 1) Simple reaction time 1 to 4) Vienna Test System Improvements in simple
training (60 min·d−1, Aerobic and resistance 2) Choice reaction time reaction time and choice
2 d·wk−1) for 12 wk training combined with 3) Simple movement time movement time in MME
(n = 46) cognitive training 4) Choice movement time group at follow-up.
Adherence = Not Reported (60 min·d−1, 2 d·wk−1) Greater improvements in
for 12 wk (n = 57) simple movement time,
Adherence = Not Reported choice reaction and
Comparator 2: movement time in
No-treatment control combined MME and
group (n = 35) cognitive training group
Adherence = Not Applicable compared to MME and
no-treatment control.

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Exercise and Cognition in Older Adults
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Linde (2014) (46) Aerobic and resistance Comparator 1: 1) Reasoning 1, 2) Leistungs-Prüf No between-group
training (60 min·d−1, Cognitive training 2) Spatial relations System 50+ differences in
2 d·wk−1) for 16 wk (30 min·d−1, 1 d·wk−1) 3) Concentration 3) d2: Test of Attention treatment groups.
(n = 15) for 16 wk (n = 11) 4) Processing speed 4) Trail-Making Test (Part A) Greater improvements
Adherence = 81% Adherence = 81% 5) Cognitive speed 5) Digit Symbol in cognitive speed in
Comparator 2: 6) Short-term memory Substitution Test (NAI) combined MME and

www.acsm-csmr.org
Aerobic and resistance 6) Word List Test (NAI) cognitive training
training (60 min·d−1, compared with
2 d·wk−1) combined no-treatment control.
with cognitive training Greater improvements in
(30 min·d−1, 1 d·wk−1) concentration in
for 16 wk (n = 16) cognitive training group
Adherence = 81% compared to
Comparator 3: no-treatment control group.
No-treatment control
group (n = 13)
Adherence = Not Applicable
Lord (2003) (47) Aerobic, resistance, and Comparator 1: 1) Choice reaction time 1) Stepping on rectangular Greater improvements
balance training Flexibility and relaxation 2) Simple reaction time panels as quickly in MME group in
(60 min·d−1, 2 d·wk−1) exercises (60 min·wk−1, as possible choice and simple
for 12 mo (n = 259) 2 d·wk−1) for 12 mo 2) Seated, using a light as reaction time
Adherence = 42.3% (n = 80) the stimulus and a hand compared to no-treatment
Adherence = 45.4% press as the response control group, and
Comparator 2: flexibility and relaxation
No-treatment control group, respectively.
group (n = 169)
Adherence =
Not applicable
Napoli (2014) (48) Aerobic, resistance, and Comparator 1: 1) Global cognitive function, 1) Modified Mini-Mental Greater improvements
balance training Diet that provided energy 2) Processing speed State Examination in global cognitive
90
( min·d−1, 3 d·wk−1) deficit of 500 to 750 3, 4) Executive function 2) Trail-Making Test (Part A) function all treatment
for 12 mo (n = 26) kcal·d−1 with goal to 3) Word List Fluency Test groups compared to
Adherence = 88% achieve ~10% weight 4) Trail-Making Test (Part B) no-treatment control group.
loss (n = 26) Greater improvements in
Adherence = 83% executive function in
Comparator 2: MME group and combined
Aerobic, resistance, and group compared
balance training (90 min·d−1, to no-treatment group.
3 d·wk−1) combined with
diet for 12 mo (n = 28)
Adherence = 83%
Comparator 3:
No-treatment control
group (n = 27)
Adherence = Not applicable

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Continued next page

Current Sports Medicine Reports

315
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Table 3.

316
Continued.
First Author Cognitive Domain(s) Cognitive Test(s)
(Year) Treatment Group(s) Comparator(s) Assessed (Outcomes) (Measures Used) Main Findings
Nascimento Aerobic, resistance, and No-treatment control Global cognitive function Montreal Cognitive Greater improvements
(2014) (49) balance training group (n = 32) Assessment in global cognitive
(60 min·d−1, 3 d·wk−1) Adherence = Not applicable function in MCI
for 16 wk (n = 35) participants in MME
Adherence ≥75% group compared to
no-treatment
control group.
Okumiya (1996) (50) Aerobic, resistance, No-treatment control 1, 2) Global cognitive function 1) Mini-Mental No within- or
and balance training group (n = 21) 3) Visual orientation State Examination between-group
60
( min·d−1, 2 d·wk−1) Adherence = Not applicable 2) Hasegawa Dementia differences at follow-up.

Volume 19  Number 8  August 2020

for 24 wk (n = 21) Scale Revised
Adherence = 86% 3) Visuospatial
Performance Test
Rehfeld (2018) (51) Aerobic, resistance, and Dance training (90 min·d−1, Cognition: Cognition: Improvements in attention,
balance training 2 d·wk−1) for 6 mo 1) Attention 1) Alertness, Go/Nogo, immediate and delayed
(90 min·d−1, 2 d·wk−1) (n = 18) 2) Processing speed Divided Attention, and recall in both groups.
for 6 mo (n = 18) Adherence ≥70% 3) Verbal fluency, short-term Flexibility Tasks No between-group
Attendance ≥70% and working memory 2) Trail-Making Test differences.
4) Verbal episodic memory 3) Digit Span Test (WMS) Greater increases in
5) Visuospatial memory 4) Verbal Learning and dance group in gray
Neuroimaging: Memory Task matter (frontal and
6) Gray matter volume 5) Rey-Osterrieth-Complex temporal cortical areas)
(7) White matter volume — Figure Test and white matter
Neuroimaging: (truncus and splenium
(6,7) MRI in corpus callosum)
volumes. Greater
changes in MME in
gray matter (occipital
and cerebella regions)
and white matter (right
temporal and right
occipital) volumes.
Rosano (2017) (52)c Aerobic, resistance, Health and education 1) Hippocampal volume (1 to 3) MRI Greater improvements
balance, and flexibility sessions (60 to 2) Dentate gyrus in left and right
training (≤50 min·d−1, 90 min·d−1, weekly 3) Cornu ammonis hippocampus and left
2 d·wk−1 [supervised] to monthly) for 24 mo ( cornu ammonis in MME
and 3 to 4 d·wk−1 n = 16) group compared with
[home-based]) Adherence = 90.6% active control group.
for 24 mo (n = 10) After adjustments, only
Adherence = 66.7% changes in left

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
hippocampus remained
statistically significant.

Exercise and Cognition in Older Adults

 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Shah (2014) (53) Aerobic (60 min·d−1, Comparator 1: 1) Premorbid IQ 1) Cambridge Contextual Greater improvements
3 d·wk−1) and resistance Cognitive training 2) Verbal episodic memory Reading Test in verbal episodic
training (40 min·d−1, (60 min·d−1, 5 d·wk−1) 3) Verbal fluency 2) Rey Auditory Verbal memory in combined
2 d·wk−1) for 16 wk for 16 wk (n = 51) 4) Processing speed Learning Test MME and cognitive
(n = 42) Adherence = Not reported 5) Attention 3) Control Word training group compared
Adherence = Not reported Comparator 2: 6) Executive function Association Test to no-treatment

www.acsm-csmr.org
Aerobic and resistance 7) Visual memory 4) Detection control group.
training (60 min·d−1, (CogState Battery)
5 d·wk−1) combined 5) One Back Memory
with cognitive training (CogState Battery)
(60 min·d−1, 5 d·wk−1) 6) Groton Maze Learning
for 16 wk (n = 44) (CogState Battery)
Adherence = Not reported 7) Visual Memory Index
Comparator 3: Score (CogState Battery)
No-treatment control
group (n = 35)
Adherence = Not applicable
Silva (2019) (54) Aerobic, resistance, and No-treatment control 1, 2) Global cognitive function 1) Clinical Dementia Rating Greater improvements in
balance training group (n = 12) 3, 4) Executive function 2) Mini-Mental State executive function in
(60 min·d−1, 2 d·wk−1) Adherence = Not applicable 5) Inhibition Examination MME group
for 12 wk (n = 7) 3) Clock Drawing Test compared to no-treatment
Adherence = 90% 4) Verbal Fluency control group.
5) Stroop Test
Sink (2015) (55) Aerobic, resistance, Health and education 1) Psychomotor speed, 1) Digit Symbol Coding Task No main effects of
balance, and flexibility sessions (60 to attention and 2) Hopkins Verbal Learning MME on any of the
training (≤50 min·d−1, 90 min·d−1, weekly working memory Test — Revised cognition outcomes.
2 d·wk−1 [supervised] to monthly) for 24 mo 2) Word list learning and recall 3) Rey-Osterrieth Complex Subgroup analysis
and 3 to 4 d·wk−1 (n = 741) 3) Visuospatial function and Figure Test revealed greater
[home-based]) for Adherence = Not reported figural memory 4) Boston Naming Test improvements in
24 mo (n = 735) 4) Language 5) Trail-Making Test (Part A) executive function
Adherence = 71% 5) Concentration, attention 6) Trail-Making Test (Part B) in participants with
and psychomotor speed 7) Category Fluency Test poorer physical function
(6–10) Executive function 8) n-Back Task at baseline or aged
(1,2,8–10) Global 9) Eriksen Flanker Task 80+, in the MME group.
cognitive function 10) Task Switching Exercise
Continued next page

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Current Sports Medicine Reports
317
 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Table 3.

318
Continued.
First Author Cognitive Domain(s) Cognitive Test(s)
(Year) Treatment Group(s) Comparator(s) Assessed (Outcomes) (Measures Used) Main Findings
Styliadis (2015) (56) Aerobic, resistance, and Comparator 1: Cognition: Cognition: No within- or between
balance training via Cognitive training 1) Global cognitive function 1) Mini-Mental State group differences in
Nintendo Wii (60 min·d−1, (60 min·d−1, Neuroimaging: Examination global cognitive function.
5 d·wk−1) for 8 wk 3 to 5 d·wk−1) for 2) Electrophysiological brain Neuroimaging: Greater improvements in
(n = 14) 8 wk (n = 14) activity 2) EEG, resting state resting-state
Adherence = 65.2% Adherence = 60.9% electrophysiological
Comparator 2: brain activity in the
Aerobic, resistance, and precuneus/posterior
balance training via Nintendo cingulate cortex in
Wii (60 min·d−1, combined MME and

Volume 19  Number 8  August 2020

≤5 d·wk−1) plus Cognitive cognitive training group
(60 min·d−1, ≤5 d·wk−1) compared to MME group.
for 8 wk (n = 14)
Adherence = 65.5%
Comparator 3:
Active control group (e.g.,
watching documentaries
60 min·d−1, 5 d·wk−1) for
8 wk (n = 14)
Adherence = 67.1%
Comparator 4:
No-treatment control
group (n = 14)
Adherence = Not applicable
Tarazona-Santabalbina Aerobic, resistance, No-treatment control Global cognitive function Mini-Mental Greater improvements
(2016) (57) balance, and flexibility group (n = 49) State Examination in global cognitive
training (65 to 70 min·d−1, function in MME
5 d·wk−1) for 24 wk compared to no-treatment
(n = 51) control group.
Adherence = 47.3%
Taylor-Piliae Aerobic, resistance, Comparator 1: 1) Semantic fluency 1) 60-s Animal Greater improvements in
(2010) (58) balance, and flexibility Tai Chi (60 min·d−1, 2) Attention, concentration, Naming Test attention, concentration
training (60 min·d−1, 1 to 2 d·wk−1 [supervised] and mental tracking 2) Digit Span Test and mental tracking in
1 to 2 d·wk−1 plus 3 d·wk−1 [home-based]) the Tai Chi compared to
[supervised] plus 3 d·wk−1 for 12 mo (n = 37) MME and no-treatment
[home-based]) for Adherence = 77% control groups (6 and
12 mo (n = 39) Comparator 2: 12 months). Improvements
Adherence = 68% Healthy aging classes in both MME and Tai
(90 min·d−1, 1 d·wk−1) Chi groups in semantic
for 6 mo (n = 56) fluency at 12 months

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Adherence = 67% compared to baseline.

Exercise and Cognition in Older Adults

 CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Teixeira (2018) (59) Aerobic and resistance No-treatment control Cognition: Cognition: Greater improvements
training (20 to 30 min·d−1, group (n = 20) 1) Global cognitive function 1) Mini-Mental in memory delayed
3 d·wk−1) for Adherence = Not applicable 2) Memory encoding State Examination recall and increase in
6 mo (n = 20) 3) Memory delayed recall (2 to 4) Rey Auditory Verbal hippocampal volume
Adherence = 66.8% 4) Memory recognition Learning Test in MME group compared
Neuroimaging: Neuroimaging: with no-treatment

www.acsm-csmr.org
5) Cortical and 5) Structural MRI control group.
hippocampal volume
Vaughan (2014) (60) Aerobic, resistance, No-treatment control 1) Inhibition 1) California Older Adults Greater improvements
balance, and flexibility group (n = 23) 2) Verbal fluency Stroop Test in inhibition, verbal
training (60 min·d−1, Adherence = Not applicable 3) Working memory 2) Controlled Oral Word fluency, processing
2 d·wk−1) for 16 wk 4) Reaction time Association Test speed and executive
(n = 25) 5) Processing speed 3) Letter-Number Sequencing function in the MME
Adherence = 85.7% 6) Executive function 4) Deary-Liewald Reaction group compared with
Time Task no-treatment control group
5) Trail-Making Test (Part A)
6) Trail-Making Test (Part B)
Vedovelli (2017) (61) Aerobic and resistance No-treatment control 1) Attention and 1) Digit Span Test (WAIS) Greater improvements
training (60 min·d−1, group (n = 9) working memory 2) Trail-Making Test (Part A) in all cognition
3 d·wk−1) for 3 mo Adherence = Not applicable 2) Processing speed 3) Trail-Making Test (Part B) outcomes in MME
(n = 22) 3) Executive function 4) Logical Memory Test I and II group compared
Adherence = 100% 4) Immediate and 5) Stroop Test with no-treatment
delayed recall control group.
5) Inhibition
Williams (1997) (62) Aerobic, resistance, and No-treatment control 1) Short-term acquisition 1) Digit Span Test (WAIS-R) Greater improvements in
balance training group (n = 93) and retrieval 2) Picture Arrangement reaction time and
(60 min·d−1, 2 d·wk−1) Adherence = Not applicable 2) Nonverbal reasoning ability (WAIS-R) short-term acquisition
for 12 mo (n = 94) 3) Nonverbal reasoning ability 3) Cattell's Matrices and retrieval in MME
Adherence = 72% and problem solving 4) Reaction Time Task group compared with
4) Simple reaction time no-treatment
control group.
Williamson (2009) (63) Aerobic, resistance, balance, Health and education sessions 1) Psychomotor speed and 1) Digit Symbol Test No between-group
and flexibility training (weekly to monthly) for working memory Substitution differences in any of the
40
( to 60 min·d−1, 1 to 24 mo (n = 52) 2) Inhibition 2) Modified Stroop Test cognition outcomes.
3 d·wk−1 [supervised] Adherence = Not reported 3) Global cognitive function 3) Modified Mini-Mental
and 1 to 5 d·wk−1 4) Short- and long-term State Examination
[home-based]) verbal memory 4) Rey Auditory Verbal
for 24 mo (n = 50) Learning Test
Adherence = Not reported
a
Secondary outcomes from Küster et al 2016.
b
Secondary outcomes from Gajewski et al 2012.
c
Secondary outcomes from Sink et al 2015.
EEG, electroencephalography; MRI, magnetic resonance imaging; NAI, Neuropsychological Aging Inventory; WMS, Wechsler Memory Scale; WAIS-R, Wechsler Adult Intelligence Scale.

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
Current Sports Medicine Reports
319
 Table 4.
Overall effects of MME on cognitive tests compared with competing treatment, active control, and no-treatment control groups.a
MME vs
MME vs Competing MME vs Active No-treatment
treatment control group control Total
No No No No
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Cognition measures Studies Sample effect Improvement effect Improvement effect Improvement effect Improvement
Global cognitive function
Mini-Mental State 7 405 2 — 1 — 5 1 8 1
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023

Examination
Modified Mini-Mental 2 209 1 — 1 — — 1 2 1
State Examination
Montreal Cognitive 2 136 1 — — — 1 1 2 1
Assessment
Clock drawing test 2 88 1 — — — 2 — 3 —
Cambridge Contextual 1 224 1 — — — 1 — 2 —
Reading Test
Clinical dementia rating 1 19 — — — — 1 — 1 —
Global Cognitive 1 127 1 — — — — — 1 —
Function (CBS)
Hasegawa Dementia 1 42 — — — — 1 — 1 —
Scale Revised
Visuospatial 1 42 — — — — 1 — 1 —
performance test
Executive function
Digit span test 12 993 7 — 4 — 6 2 17 2
Stroop test 10 730 2 — 2 — 3 5 7 5
Trail-Making Test (Part B) 10 2191 4 — 1 — 2 5 7 5
Semantic fluency 6 2169 5 — 3 — 3 1 11 1
Digit symbol coding test 5 1706 2 — 2 — 2 2 6 2
Digit symbol substitution test 4 252 2 — 1 — 1 — 4 —
Controlled Oral Word 3 323 1 — — — 1 2 2 2
Association Test
Verbal fluency 3 142 2 — — — 2 1 4 1
Letter-number sequencing 2 121 — — — — 2 — 2 —
N-Back Task 2 1617 1 — 2 — 1 — 4 —
Task switching test 2 1617 1 — 2 — 1 — 4 —
Age Concentration 1 71 1 — — — — — 1 —
Tests A and B
Alertness 1 38 1 — — — — — 1 —
Boston Naming Test 1 1476 — — 1 — — — 1 —
Cattell's Matrices 1 187 — — — — 1 — 1 —
Computation Span (ECB) 1 54 1 — — — 1 — 2 —
Concentration (CBS) 1 127 1 — — — — — 1 —
d2: Test of Attention 1 55 1 — — — 1 — 2 —
Divided Attention 1 38 1 — — — — — 1 —
Eriksen Flanker Task 1 1476 — — 1 — — — 1 —
Executive Control Task 1 71 1 — — — — — 1 —
Flexibility 1 38 1 — — — — — 1 —
Go/Nogo 1 38 1 — — — — — 1 —

Continued next page

320 Volume 19  Number 8  August 2020 Exercise and Cognition in Older Adults

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 Table 4.
Continued.
MME vs
MME vs Competing MME vs Active No-treatment
treatment control group control Total
No No No No
Cognition measures Studies Sample effect Improvement effect Improvement effect Improvement effect Improvement
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

Groton Maze Learning 1 224 1 — — — 1 — 2 —

(CogState Battery)
Leistungs-Prüf System 50+ 1 55 1 — — — 1 — 2 —
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023

One Back Memory 1 224 1 — — — 1 — 2 —

(CogState Battery)
Picture arrangement 1 187 — — — — 1 — 1 —
(WAIS-R)
Planning (CBS) 1 127 1 — — — — — 1 —
Random number 1 51 1 — 1 — — — 2 —
generation task
Reasoning (CBS) 1 127 1 — — — — — 1 —
Similarities test (WAIS-III) 1 72 — — — — 1 — 1 —
Memory
Rey Auditory Verbal 4 438 1 — 1 — 2 1 4 1
Learning Test
Rey-Osterrieth Complex 4 1705 2 — 2 — 1 1 5 1
Figure Test
Hopkins Verbal Learning 3 1550 — — 1 — — 1 1 1
Test — Revised
Logical memory 2 100 1 — — — — 1 1 1
Verbal learning and 2 179 2 — 1 — 1 — 4 —
memory test
Emotional memory task 1 24 — — — — — 1 — 1
Free and cued selective 1 259 1 — — — — 1 1 1
reminding test
Free recall (ADAS-cog) 1 54 1 — — — 1 — 2 —
Memory (CBS) 1 127 1 — — — — — 1 —
Munich verbal memory test 1 54 1 — — — 1 — 2 —
Paired-associates 1 71 1 — — — — — 1 —
learning task
Rivermead Behavioral 2 24 1 — — — — 1 1 1
Memory Test
Visual Memory Index 1 224 1 — — — 1 — 2 —
Score (CogState Battery)
Word List Test (NAI) 1 55 1 — — — 1 — 2 —
Processing speed
Trail-Making Test (Part A) 10 1975 5 — 1 — 3 3 9 3
Simple reaction time 4 925 1 — — 1 2 2 3 3
Choice reaction time 3 738 1 — 1 — 2 1 4 1
Choice movement time 1 138 1 — — — — 1 1 1
Detection (CogState Battery) 1 224 1 — — — 1 — 2 —
Simple movement time 1 138 1 — — — 1 — 2 —
a
Data reported in columns “No Effect” and “Improvement” represent number of comparisons made between MME and comparator groups across stud-
ies, considering that a given study may report one or more comparisons depending on how many groups were included in the study design. Empty cells (—)
indicate either that the comparison is not applicable, or no significant result were reported.
CBS, Cambridge Brain Sciences cognitive battery; ECB, Everyday Cognition Battery; ADAS-cog, Alzheimer's disease Assessment Scale-cognitive subscale;
WAIS-III, Wechsler Adult Intelligence Scale.

www.acsm-csmr.org Current Sports Medicine Reports 321

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 no-treatment control groups, MME yielded increases in cortical MME session (i.e., Linde et al (46) and Shah et al (53), see also
gray matter (41) and hippocampal volume (59) in two studies. Table 3 for more details). Therefore, it remains to be investi-
For EEG outcomes, two studies reported that MME was gated whether a combination of cognitive and physical train-
not effective in improving event-related brain action potentials ing can impart improvements to cognition because of intrinsic
(i.e., peak and amplitude of activations) in two executive func- aspects of these interventions only or because of prolonged ex-
tioning tasks (39,40) compared with cognitive training, active posure to treatment stimuli.
control, and no-treatment control groups. Similarly, another Nevertheless, considering that MME was not superior to ac-
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

study reported greater improvements in resting-state EEG tive control groups in seven (32,39,40,55,56,58,63) of nine
brain activity (precuneus/posterior cingulate cortex) following tudies included, we also must consider other factors influencing
a combined cognitive and physical training group compared the effects of MME beyond prolonged exposure to treatment.
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023

with MME alone (56). Confounding effects of socialization, for instance, are present
when these interventions are administered in sessions with mul-
Discussion tiple participants exercising together. In fact, social interaction
In this review, we explored the overall effects of MME may provide significant cognitive stimulation (55) and partially
compared with competing treatment, active control, and account for improvements in cognition (64,65). Furthermore, in
no-treatment control conditions in global and domain cogni- a previous review (28), greater effect sizes were observed follow-
tive function, and neuroimaging outcomes in older adults ing exercise in older adults compared with no-treatment control
without dementia. We also had interest in the characteristics groups, but not in comparison to active control groups (28).
of the MME programs administered in the studies (i.e., fre- Therefore, the lack of superiority of MME when compared with
quency, intensity, time, and type) with hopes that our findings active control groups could be attributed to effects of socialization.
would aid in informing translation of current findings into Our findings suggested that only when compared with
practice and provide direction for future research. Our main no-treatment control groups, MME yielded improvements in
findings and recommendations are discussed below. cognition (i.e., memory (41,42,59,61), processing speed
(44,47,60–62), executive functioning (42,44,48,54,60–62)
MME and Cognitive Function and global cognitive functioning (48,49,57)). These findings
Our findings indicated that when compared with com- suggest the potential of MME to impart improvements in cog-
peting treatment groups, apart from two studies (42,48), nition in individuals with different clinical characteristics,
the majority of studies indicated that MME was inferior given that the studies included healthy or sedentary older
to competing treatments in improving cognition outcomes adults (41,42,59–62), as well as frail (44,47,57), obese (48),
(31–33,35–40,43,45,46,51,53,56,58). Similarly, only two and MCI (49) individuals. Nevertheless, caution must be
studies reported that MME was superior to active control exercised when interpreting these findings, as essential limita-
groups in improving cognition (34,47), while the remaining tions must be considered. For instance, three of the included
studies including active control groups did not find MME to studies were nonrandomized (i.e., quasi-experimental), and
be superior (32,40,55,56,58,63). The only scenario in which therefore, bias is inflicted in study results owing to confound-
MME was primarily effective in improving global and ing factors (e.g., selection bias). Another confounding variable
domain-specific cognitive function was when compared with introduced by including no-treatment control groups is that
no-treatment control groups (41,42,44,47–49,54,57,59–62). participants exposed to MME interventions also are exposed
Moreover, as reported in Table 4, most studies investigated to other factors, such as attention and social interaction (as
changes in measures of executive functioning, followed by mentioned above). This is a crucial aspect of the studies in-
measures of memory, global cognitive function, and process- cluded, since six (42,44,47,49,57,60) of the included studies
ing speed. In all of these measures, apart from one study in explicitly reported that the MME sessions were administered
which processing speed (47) was improved compared with ac- in groups of at least three participants.
tive control groups, MME was only superior in improving Altogether, the literature suggests MME may be an effective
global or domain-specific cognitive function when compared strategy to improve global and domain-specific cognitive func-
with no-treatment control groups. tion; albeit, there is limited evidence from studies including ac-
Important considerations must be made when discussing tive control or competing treatment groups. Considerations
these findings. Many studies included competing treatment and limitations regarding the MME protocols administered
groups that combined both cognitive and physical training in these studies are discussed in the subsequent sections.
(33,36,37,45,46,53,56). Considering the studies showing su-
periority of combining both treatments when compared with Effects of MME in Neuroimaging Outcomes
MME alone, we observed improvements in measures of exec- Evidence from nine studies suggested mixed effects of
utive functioning (36), processing speed (36,46), and memory MME on white matter structure, but more consistent effects
(53). One confounding aspect of these findings is that by re- on cortical and subcortical gray matter. Fissler and colleagues
ceiving both physical and cognitive training, study subjects (38) reported no differences in white matter integrity (i.e., frac-
would receive prolonged exposure to treatment effects during tional anisotropy) in older adults with SMI following cogni-
each session. As identified in the study by Damirchi and col- tive training or MME, compared with a no-treatment
leagues (36), participants in the combined treatment group re- control group. Conversely, Callisaya and colleagues (34) re-
ceived prolonged intervention (90 to 120 min·d−1, 3 d·wk−1) ported improvements in fractional anisotropy in older adults
compared with the MME group (60 min·d−1, 3 d·wk−1). Sim- with diabetes compared with an active control group, and
ilarly, two studies showed superiority of cognitive plus physi- Rehfeld and colleagues (51) noted greater increases in white
cal training sessions lasting longer (i.e., min·d−1) than the matter in temporal and occipital lobes following MME compared

322 Volume 19  Number 8  August 2020 Exercise and Cognition in Older Adults

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 with dance training. Dance training, nonetheless, yielded electrophysiological brain activity in the precuneus/posterior
greater changes in the white matter of other brain regions cingulate cortex compared with MME alone (56). Overall,
(see Table 3) suggesting training-specific adaptations. these findings suggest that MME alone has limited influence
Although the evidence is limited, these findings suggest that in brain activity measured via EEG outcomes when compared
MME may be effective in imparting improvements in white with competing treatment groups. Owing to limited literature
matter; however, the extent to which these improvements included in this review, this topic needs to be further explored.
are superior to other interventions (e.g., dance training or cog-
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

nitive training) warrants further exploration. Some relevant Recommendations and Future Directions
contrasts among these studies also must be considered. Fissler One key aspect to be further investigated is whether compli-
and colleagues (38) included older adults with SMI, a marker ance with international guidelines for exercise in older adults
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023

of increased risk of dementia (66), while Callisaya (34) studied and increasing adherence to exercise will aid in strengthening
older individuals with diabetes—comprising a different risk the effects of MME on cognitive function. For example, for
profile for dementia (67) — and Rehfeld and colleagues included the studies included in this review, the average frequency of
only healthy individuals (51). The most notable difference be- MME sessions was 3.1 (SD = 1.5) d·wk−1, lasting on average
tween studies, however, could be the length of these programs, 62.7 (SD = 15.5) min·d−1. However, the average time spent
with one lasting only 10 wk (38), while the other two studies in each MME component was 32.6 (SD = 13) min·d−1 for
(34,51), which showed in part positive effects of MME on white AET, 23.6 (SD = 11.3) min·d−1 for RET, and 16.9 (SD = 10.3)
matter outcomes, lasted 6 months. As such, longer intervention min·d−1 for the balance/flexibility component. In this context,
periods may result in greater positive changes in white matter. the average time per component is relatively low compared
Regarding changes in the gray matter of cortical and sub- with previous recommendations (26,73). It is important that
cortical structures, compared with no-treatment control future research addresses whether complying with recommen-
groups, Ji and colleagues (41) reported that MME was associ- dations would yield greater benefits above and beyond con-
ated with increases in cortical gray matter, while Teixeira and founding variables influencing cognition (e.g., socialization).
colleagues (59), reported increases in hippocampal volume. In This is pertinent when contemplating that previous studies
comparison to active control groups (34,52), MME was asso- have provided strong evidence for the positive effects of AET
ciated with greater increases in total brain volume (34) and (15) and RET (21) on cognition. These are examples of
hippocampal volume (34,52). Furthermore, among the stud- well-conducted RCTs, with detailed exercise programs, and
ies included, Ji and colleagues (41) were the only ones to inves- measures of cognitive function sensitive to the effects of exer-
tigate functional connectivity changes via fMRI. Using a cise (15,21). Consequently, with a detailed MME program,
resting-state fMRI protocol, the authors reported increased administered with appropriate frequency, duration, and inten-
functional connectivity between the posterior cingulate sity, it is plausible to expect additive effects of combining AET
cortex/precuneus and the right striatum, and other regions and RET, and potentially balance/flexibility training (1,74).
compared with controls — while controls suffered atrophy Finally, because of the heterogeneity across studies, it was
of the striatum region, suggesting protective effects of MME. challenging to gather and harmonize information on the ele-
Altogether, the main findings of MRI and fMRI studies point ments of the exercise programs administered (i.e., frequency,
toward MME imparting positive changes in brain function and intensity, time, and type). Future studies should consider a
structure, particularly marked by multiple studies reporting sig- standardized and detailed method of reporting exercise train-
nificant increases in hippocampal volume (34,52,59). Clini- ing protocols, which will facilitate appreciation and under-
cally, these results could have relevance to prevent and/or standing of the effects of exercise on variables of interest
delay onset of cognitive impairment. Both hippocampi are im- (27). To this end, we suggest reporting on exercise training
plicated in memory function (68–70), and are hallmark regions variables following previous recommendations (26,27), includ-
where pathophysiological changes in MCI and early/prodromal ing the following: a) exercise frequency (e.g., d·wk−1); b) objec-
stages of Alzheimer’s disease occur (e.g., amyloid beta deposi- tive or subjective measures of intensity (e.g., target HR, RPE,
tion) (71), including cortical atrophy preceding Alzheimer’s dis- maximum repetitions, etc.); c) time allocated to each compo-
ease diagnosis (72). Nonetheless, three of the studies reporting nent (e.g., min·d−1) and d) type of exercise administered (e.g.,
positive effects of MME were nonrandomized interventions running, walking, machine-based, bodyweight). If with stronger
(38,41,59) and their findings should be interpreted with caution. study designs, clearer training methodology, and well-defined
Finally, three studies explored EEG outcomes as surrogate study populations, MME is proven to be efficient to improve
measures of brain activity (39,40,56). All three studies in- brain health, it will then be plausible to discuss long-term ef-
cluded competing treatment groups and their results suggested fects and follow-ups, feasibility, and translation of these pro-
that MME was not superior to other treatment conditions in grams in real-world community settings (27).
improving resting-state and task-based brain activity. For in-
stance, in an early study Gajewski and Falkenstein (39) re- Conclusions
ported that cognitive training yielded higher improvements in Our findings indicated that MME has the potential to impart
event-related brain action potentials associated with response positive changes in global and domain-specific cognitive func-
selection, allocation of cognitive resources, and error detection tion, as well as white matter, cortical gray matter, and hippo-
compared with MME. Similarly, in a secondary study, the same campal volume when compared with no-treatment control
authors (40) reported improvements in underlying processing as- groups. The lack of superiority of MME when compared with
sociated with working memory following cognitive training competing treatment or active controls suggests that extrinsic
only. Accordingly, Styliadis and colleagues reported additive ef- factors, such as socialization, could yield improvements inde-
fects of combining cognitive and physical training in resting-state pendent of MME effects. Summary data from the MME

www.acsm-csmr.org Current Sports Medicine Reports 323

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 protocols administered, including frequency, intensity, and 24. Gates N, Singh MA, Sachdev PS, Valenzuela M. The effect of exercise training on
cognitive function in older adults with mild cognitive impairment: a meta-analysis
time of MME programs administered did not seem to be fully of randomized controlled trials. Am. J. Geriatr. Psychiatry. 2013; 21:1086–97.
aligned with current guidelines for exercise for older adults, 25. American College of Sports Medicine. ACSM’s Guidelines for Exercise Testing
which could have hindered MME effects. It is plausible that and Prescription. 9th ed. Baltimore (MD): Wolters Kluwer/Lippincott Williams
combining different treatment conditions may provide addi- & Wilkins Health; 2014. p. 456.

tive effects to cognitive function; however, the feasibility of 26. Chodzko-Zajko WJ, Proctor DN, Fiatarone Singh MA, et al. Exercise and phys-
ical activity for older adults. Med. Sci. Sports Exerc. 2009; 41:1510–30.
such programs to be translated to real world settings remains
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

27. Hurst C, Weston KL, McLaren SJ, Weston M. The effects of same-session com-
to be explored in future research. bined exercise training on cardiorespiratory and functional fitness in older adults:
a systematic review and meta-analysis. Aging Clin. Exp. Res. 2019; 31:1701–17.
28. Northey JM, Cherbuin N, Pumpa KL, et al. Exercise interventions for cognitive
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023

The authors declare no conflict of interest and do not have function in adults older than 50: a systematic review with meta-analysis. Br. J.
any financial disclosures. Sports Med. 2018; 52:154–60.
29. Falck RS, Davis JC, Best JR, et al. Impact of exercise training on physical and
cognitive function among older adults: a systematic review and meta-analysis.
References 2019; 79:119–30.
1. Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention, intervention, 30. Tricco AC, Lillie E, Zarin W, et al. PRISMA extension for scoping reviews
and care. Lancet. 2017; 390:2673–734. (PRISMA-ScR): checklist and explanation. Ann. Intern. Med. 2018; 169:467–73.
2. Alzheimer Society of Canada. Rising tide: The impact of dementia on Canadian 31. Ansai JH, Rebelatto JR. Effect of two physical exercise protocols on cognition
society. Executive summary. Dementia. 2010; 1–24. and depressive symptoms in oldest-old people: a randomized controlled trial.
3. Barnes DE, Yaffe K, Satariano WA, Tager IB. A longitudinal study of cardiore- Geriatr. Gerontol. Int. 2015; 15:1127–34.
spiratory fitness and cognitive function in healthy older adults. J. Am. Geriatr. 32. Berryman N, Bherer L, Nadeau S, et al. Multiple roads lead to Rome: combined
Soc. 2003; 51:459–65. high-intensity aerobic and strength training vs. gross motor activities leads to
4. Lourida I, Hannon E, Littlejohns TJ, et al. Association of lifestyle and genetic risk equivalent improvement in executive functions in a cohort of healthy older
with incidence of dementia. JAMA. 2019; 322:430. adults. Age (Dordr.). 2014; 36:9710.
5. Abbott RD, White LR, Ross GW, et al. Walking and dementia in physically ca- 33. Boa Sorte Silva NC, Gill DP, Owen AM, et al. Cognitive changes following
pable elderly men. JAMA. 2004; 292:1447–53. multiple-modality exercise and mind-motor training in older adults with subjec-
6. Weuve J, Kang JH, Manson JE, et al. Physical activity, including walking, and tive cognitive complaints: the M4 study. PLoS One. 2018; 13:1–17.
cognitive function in older women. JAMA. 2004; 292:1454–61. 34. Callisaya ML, Daly RM, Sharman JE, et al. Feasibility of a multi-modal exercise
7. Bugg JM, Shah K, Villareal DT, Head D. Cognitive and neural correlates of aer- program on cognition in older adults with type 2 diabetes—a pilot randomized
obic fitness in obese older adults. Exp. Aging Res. 2012; 38:131–45. controlled trial. BMC Geriatr. 2017; 17:237.
8. Bugg JM, Head D. Exercise moderates age-related atrophy of the medial tempo- 35. Carral JM, Pérez CA. Effects of high-intensity combined training on women over
ral lobe. Neurobiol. Aging. 2011; 32:506–14. 65. Gerontology. 2008; 53:340–6.
9. Lautenschlager NT, Cox KL, Flicker L, et al. Effect of physical activity on cognitive 36. Damirchi A, Hosseini F, Babaei P. Mental training enhances cognitive function
function in older adults at risk for Alzheimer disease. JAMA. 2008; 300:1027–37. and BDNF more than either physical or combined training in elderly women with
mci: a small-scale study. Am. J. Alzheimers Dis. Other Demen. 2018; 33:20–9.
10. Voss MW, Prakash RS, Erickson KI, et al. Plasticity of brain networks in a ran-
domized intervention trial of exercise training in older adults. Front. Aging 37. Eggenberger P, Schumacher V, Angst M, et al. Does multicomponent physical
Neurosci. 2010; 2:32. exercise with simultaneous cognitive training boost cognitive performance in
older adults? A 6-month randomized controlled trial with a 1-year follow-up.
11. Chirles TJ, Reiter K, Weiss LR, et al. Exercise training and functional connectiv-
Clin. Interv. Aging. 2015; 10:1335–49.
ity changes in mild cognitive impairment and healthy elders. J. Alzheimers Dis.
2017; 57:845–56. 38. Fissler P, Müller HP, Küster OC, et al. No evidence that short-term cognitive or
12. Erickson KI, Voss MW, Prakash RS, et al. Exercise training increases size of hippocam- physical training programs or lifestyles are related to changes in white matter in-
pus and improves memory. Proc. Natl. Acad. Sci. U.S.A. 2011; 108:3017–22. tegrity in older adults at risk of dementia. Front. Hum. Neurosci. 2017; 11:110.

13. Erickson KI, Kramer AF. Aerobic exercise effects on cognitive and neural plastic- 39. Gajewski PD, Falkenstein M. Training-induced improvement of response selec-
ity in older adults. Br. J. Sports Med. 2008; 43:22–4. tion and error detection in aging assessed by task switching: effects of cognitive,
physical, and relaxation training. Front. Hum. Neurosci. 2012; 6:130.
14. Colcombe S, Kramer AF. Fitness effects on the cognitive function of older adults:
A meta-analytic study. Psychol. Sci. 2003; 14:125–30. 40. Gajewski PD, Falkenstein M. ERP and behavioral effects of physical and cogni-
tive training on working memory in aging: A randomized controlled study. Neu-
15. Colcombe SJ, Kramer AF, Erickson KI, et al. Cardiovascular fitness, cortical ral Plast. 2018; 2018:3454835.
plasticity, and aging. Proc. Natl. Acad. Sci. 2004; 101:3316–21.
41. Ji L, Zhang H, Potter GG, et al. Multiple neuroimaging measures for examining
16. Smith JC, Nielson KA, Antuono P, et al. Semantic memory functional MRI and
exercise-induced neuroplasticity in older adults: a quasi-experimental study.
cognitive function after exercise intervention in mild cognitive impairment.
Front. Aging Neurosci. 2017; 9:102.
J. Alzheimers Dis. 2013; 37:197–215.
42. Klusmann V, Evers A, Schwarzer R, et al. Complex mental and physical activity
17. ten Brike LF, Bolandzadeh N, Nagamatsu LS, et al. Aerobic exercise increases
in older women and cognitive performance: a 6-month randomized controlled
hippocampal volume in older women with probable mild cognitive impairment:
trial. J. Gerontol. A Biol. Sci. Med. Sci. 2010; 65:680–8.
a 6-month randomized controlled trial. Br. J. Sports Med. 2014; 49:248–54.
18. Smith PJ, Blumenthal JA, Hoffman BM, et al. Aerobic exercise and neurocogni- 43. Küster OC, Fissler P, Laptinskaya D, et al. Cognitive change is more positively
tive performance: a meta-analytic review of randomized controlled trials. associated with an active lifestyle than with training interventions in older adults at risk
Psychosom. Med. 2010; 72:239–52. of dementia: a controlled interventional clinical trial. BMC Psychiatry. 2016; 16:315.

19. Young J, Angevaren M, Rusted J, Tabet N. Aerobic exercise to improve cogni- 44. Langlois F, Vu TT, Chassé K, et al. Benefits of physical exercise training on cog-
tive function in older people without known cognitive impairment. Cochrane nition and quality of life in frail older adults. J. Gerontol. B Psychol. Sci. Soc. Sci.
Database Syst. Rev. 2015; 4:CD005381. 2013; 68:400–4.
20. Nagamatsu LS, Handy TC, Hsu CL, et al. Resistance training promotes cogni- 45. León J, Ureña A, Bolaños MJ, et al. A combination of physical and cognitive ex-
tive and functional brain plasticity in seniors with probable mild cognitive im- ercise improves reaction time in persons 61–84 years old. J. Aging Phys. Act.
pairment. Arch. Intern. Med. 2012; 172:666–8. 2015; 23:72–7.
21. Liu-Ambrose T, Nagamatsu LS, Graf P, et al. Resistance training and executive func- 46. Linde K, Alfermann D. Single versus combined cognitive and physical activity ef-
tions: a 12-month randomized controlled trial. Arch. Intern. Med. 2010; 170:170–8. fects on fluid cognitive abilities of healthy older adults: a 4-month randomized
controlled trial with follow-up. J. Aging Phys. Act. 2014; 22:302–13.
22. Liu-Ambrose T, Nagamatsu LS, Voss MW, et al. Resistance training and func-
tional plasticity of the aging brain: a 12-month randomized controlled trial. 47. Lord SR, Castell S, Corcoran J, et al. The effect of group exercise on physical
Neurobiol. Aging. 2012; 33:1690–8. functioning and falls in frail older people living in retirement villages: a random-
ized, controlled trial. J. Am. Geriatr. Soc. 2003; 51:1685–92.
23. Bolandzadeh N, Tam R, Handy TC, et al. Resistance training and white matter
lesion progression in older women: exploratory analysis of a 12-month random- 48. Napoli N, Shah K, Waters DL, et al. Effect of weight loss, exercise, or both on cog-
ized controlled trial. J. Am. Geriatr. Soc. 2015; 63:2052–60. nition and quality of life in obese older adults. Am. J. Clin. Nutr. 2014; 100:189–98.

324 Volume 19  Number 8  August 2020 Exercise and Cognition in Older Adults

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 49. Nascimento C, Pereira J, Andrade L, et al. Physical exercise in MCI elderly pro- 61. Vedovelli K, Giacobbo BL, Corrêa MS, et al. Multimodal physical activity in-
motes reduction of pro-inflammatory cytokines and improvements on cognition creases brain-derived neurotrophic factor levels and improves cognition in insti-
and BDNF peripheral levels. Curr. Alzheimer Res. 2014; 11:799–805. tutionalized older women. GeroScience. 2017; 39:407–17.
50. Okumiya K, Matsubayashi K, Wada T, et al. Effects of exercise on neurobehav- 62. Williams P, Lord SR. Effects of group exercise on cognitive functioning and
ioral function in community-dwelling older people more than 75 years of age. J. mood in older women. Aust. N. Z. J. Public Health. 1997; 21:45–52.
Am. Geriatr. Soc. 1996; 44:569–72. 63. Williamson JD, Espeland M, Kritchevsky SB, et al. Changes in cognitive function
51. Rehfeld K, Lüders A, Hökelmann A, et al. Dance training is superior to repetitive in a randomized trial of physical activity: results of the lifestyle interventions and
physical exercise in inducing brain plasticity in the elderly. PLoS One. 2018; 13:1–15. independence for elders pilot study. Journals Gerontol - Ser A Biol Sci Med Sci.
Downloaded from http://journals.lww.com/acsm-csmr by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCyw

52. Rosano C, Guralnik J, Pahor M, et al. Hippocampal response to a 24-month 2009; 64:688–94.
physical activity intervention in sedentary older adults. Am. J. Geriatr. Psychia- 64. Li R, Zhu X, Yin S, et al. Multimodal intervention in older adults improves
try. 2017; 25:209–17. resting-state functional connectivity between the medial prefrontal cortex and
53. Shah T, Verdile G, Sohrabi H, et al. A combination of physical activity and com- medial temporal lobe. Front. Aging Neurosci. 2014; 6:1–13.
CX1AWnYQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 12/14/2023

puterized brain training improves verbal memory and increases cerebral glucose 65. Mortimer JA, Ding D, Borenstein AR, et al. Changes in brain volume and cognition
metabolism in the elderly. Transl. Psychiatry. 2014; 4:e487. in a randomized trial of exercise and social interaction in a community-based
54. de Oliveira Silva F, Ferreira JV, Plácido J, et al. Three months of multimodal sample of non-demented chinese elders. J. Alzheimers Dis. 2012; 30:757–66.
training contributes to mobility and executive function in elderly individuals 66. Buckley RF, Ellis KA, Ames D, et al. Phenomenological characterization of mem-
with mild cognitive impairment, but not in those with Alzheimer’s disease: a ran- ory complaints in preclinical and prodromal Alzheimer's disease. Neuropsychol-
domized controlled trial. Maturitas. 2019; 126:28–33. ogy. 2015; 29:571–81.
55. Sink KM, Espeland MA, Castro CM, et al. Effect of a 24-month physical activity 67. Crane PK, Walker R, Hubbard RA, et al. Glucose levels and risk of dementia. N.
intervention vs health education on cognitive outcomes in sedentary older adults: Engl. J. Med. 2013; 369:540–8.
the LIFE randomized trial. JAMA. 2015; 314:781–90.
68. Alvarez P, Squire LR. Memory consolidation and the medial temporal lobe: a
56. Styliadis C, Kartsidis P, Paraskevopoulos E, et al. Neuroplastic effects of com- simple network model. Proc. Natl. Acad. Sci. 1994; 91:7041–5.
bined computerized physical and cognitive training in elderly individuals at risk
69. Jeneson A, Squire LR. Working memory, long-term memory, and medial tempo-
for dementia: an eLORETA controlled study on resting states. Neural Plast.
ral lobe function. Learn. Mem. 2011; 19:15–25.
2015; 2015:172192.
70. Shirer WR, Ryali S, Rykhlevskaia E, et al. Decoding subject-driven cogni-
57. Tarazona-Santabalbina FJ, Gómez-Cabrera MC, Pérez-Ros P, et al. A multicom-
tive states with whole-brain connectivity patterns. Cereb. Cortex. 2012;
ponent exercise intervention that reverses frailty and improves cognition, emo-
22:158–65.
tion, and social networking in the community-dwelling frail elderly: a
randomized clinical trial. J. Am. Med. Dir. Assoc. 2016; 17:426–33. 71. Taylor KI, Probst A. Anatomic localization of the transentorhinal region of the
58. Taylor-Piliae RE, Newell KA, Cherin R, et al. Effects of Tai Chi and Western ex- perirhinal cortex. Neurobiol. Aging. 2008; 29:1591–6.
ercise on physical and cognitive functioning in healthy community-dwelling 72. Pettigrew C, Soldan A, Sloane K, et al. Progressive medial temporal lobe atrophy
older adults. J. Aging Phys. Act. 2010; 18:261–79. during preclinical Alzheimer's disease. NeuroImage Clin. 2017; 16:439–46.
59. Teixeira CVL, Ribeiro de Rezende TJ, Weiler M, et al. Cognitive and structural 73. Nelson ME, Rejeski WJ, Blair SN, et al. Physical activity and public health in
cerebral changes in amnestic mild cognitive impairment due to Alzheimer's dis- older adults: recommendation from the American College of Sports Medicine
ease after multicomponent training. Alzheimer's Dement (NY). 2018; 4:473–80. and the American Heart Association. Circulation. 2007; 116:1094–105.
60. Vaughan S, Wallis M, Polit D, et al. The effects of multimodal exercise on cog- 74. Cotman CW, Berchtold NC, Christie LA. Exercise builds brain health: key
nitive and physical functioning and brain-derived neurotrophic factor in older roles of growth factor cascades and inflammation. Trends Neurosci. 2007; 30:
women: a randomized controlled trial. Age Aging. 2014; 43:623–9. 464–72.

www.acsm-csmr.org Current Sports Medicine Reports 325

Copyright © 2020 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.