Sports Medicine (2020) 50:1095–1106

https://doi.org/10.1007/s40279-020-01259-y

SYSTEMATIC REVIEW

Safety and Effectiveness of Long‑Term Exercise Interventions in Older

Adults: A Systematic Review and Meta‑analysis of Randomized
Controlled Trials
Antonio García‑Hermoso1,2,3 · Robinson Ramirez‑Vélez1,3 · Mikel L. Sáez de Asteasu1,3 · Nicolás Martínez‑Velilla1,3 ·
Fabricio Zambom‑Ferraresi1,3 · Pedro L. Valenzuela4 · Alejandro Lucia3,5,6 · Mikel Izquierdo1,3

Published online: 4 February 2020

© Springer Nature Switzerland AG 2020

Abstract
Background Physical exercise is beneficial to reduce the risk of several conditions associated with advanced age, but to our
knowledge, no previous study has examined the association of long-term exercise interventions (≥ 1 year) with the occur-
rence of dropouts due to health issues and mortality, or the effectiveness of physical exercise versus usual primary care
interventions on health-related outcomes in older adults (≥ 65 years old).
Objective To analyze the safety and effectiveness of long-term exercise interventions in older adults.
Methods We conducted a systematic review with meta-analysis examining the association of long-term exercise interventions
(≥ 1 year) with dropouts from the corresponding study due to health issues and mortality (primary endpoint), and the effects
of these interventions on health-related outcomes (falls and fall-associated injuries, fractures, physical function, quality of
life, and cognition) (secondary endpoints).
Results Ninety-three RCTs and six secondary studies met the inclusion criteria and were included in the analyses (n = 28,523
participants, mean age 74.2 years). No differences were found between the exercise and control groups for the risk of dropouts
due to health issues (RR = 1.05, 95% CI 0.95–1.17) or mortality (RR = 0.93, 95% CI 0.83–1.04), although a lower mortal-
ity risk was observed in the former group when separately analyzing clinical populations (RR = 0.67, 95% CI 0.48–0.95).
Exercise significantly reduced the number of falls and fall-associated injuries, and improved physical function and cognition.
These results seemed independent of participants’ baseline characteristics (age, physical function, and cognitive status) and
exercise frequency.
Conclusions Long-term exercise training does not overall influence the risk of dropouts due to health issues or mortality in
older adults, and results in a reduced mortality risk in clinical populations. Moreover, exercise reduces the number of falls
and fall-associated injuries, and improves physical function and cognition in this population.

1 Introduction physical activity) [3]. In this respect, regular exercise has

multi-system anti-aging effects [4], and can attenuate the
The growing prevalence of physical inactivity among older deleterious effects of inactivity while increasing healthy
adults represents an important public health problem in light lifespan [5].
of global population aging [1, 2]. For instance, only 27–44% Recent systematic reviews and meta-analyses have con-
of older U.S. adults meet international physical activity rec- cluded that physical exercise is beneficial to reduce the risk
ommendations (≥ 150 min per week of moderate-to-vigorous of several conditions associated with an advanced age, such
as frailty [6], cognitive decline [7], low muscle power, or
poor functional capacity [8]. There is also evidence that
Electronic supplementary material The online version of this exercise intervention programs can prevent falls in older
article (https​://doi.org/10.1007/s4027​9-020-01259​-y) contains
people with minor concomitant adverse effects [9–11]. To
supplementary material, which is available to authorized users.
the best of our knowledge, however, no study has specifically
* Mikel Izquierdo analyzed the risk of dropouts from an exercise program due
mikel.izquierdo@gmail.com to health issues (adverse events, medical problems) in older
Extended author information available on the last page of the article persons, or the effect of this type of intervention on risk of

Vol.:(0123456789)
1096 A. García‑Hermoso et al.

2.3 Eligibility Criteria and Study Selection

Key Points
The criteria for a study to be included in the system-
Long-term exercise does not influence the risk of drop- atic review were the following: (i) adults aged ≥ 65 years
outs due to health issues or mortality in older adults and assessed at baseline; (ii) a randomized controlled trial
results in a reduced mortality risk in clinical populations. (RCT) design, where the control group received usual care
Exercise reduces the number of falls and fall-associated combined or not with a nutritional or educational interven-
injuries and improves physical function and cognition in tion, but no structured physical exercise intervention; (iii)
this population. the study assessed the occurrence of dropouts due to health
issues and/or mortality; and (iv) the exercise intervention
Long-term exercise training interventions are an effec- lasted ≥ 1 year. Regarding exercise programs, no restric-
tive and beneficial strategy in older adults for the preven- tions were made in terms of combination or not with a con-
tion falls and fall-associated injuries and improvement of comitant nutritional intervention, supervision or structure
physical function and cognition. (e.g., supervised group based, unsupervised home based or
a combination thereof), setting (e.g., community dwellers or
institutionalized), or exercise modality, frequency or inten-
mortality, with both outcomes being indicators of the safety sity (i.e., whether focusing only on either muscle strength
and practical applicability of this type of lifestyle interven- [‘resistance’] or aerobic exercises, including multicompo-
tion in this population segment. It also remains to be deter- nent exercise training [combined aerobic, muscle strength,
mined whether the type of participant (age, health status) and balance exercises], or using Tai Chi). However, when
or intervention influences the risk of dropouts or mortality. the study included a nutritional intervention (e.g., health
It was, therefore, the aim of our study to systematically education), both the control and exercise groups had to par-
review and meta-analyze (i) the association of long-term ticipate in this intervention.
exercise programs (≥ 1 year) with dropouts due to health We excluded those studies with the following character-
issues and risk of mortality in older adults (primary end- istics: (i) the exercise intervention was not well defined; (ii)
points), and (ii) the effects of this type of intervention on included subjects aged < 65 years; and (iii) did not include
health-related outcomes, including physical function, cogni- a control (non-exercise) group for comparison. Titles,
tion status, health-related quality of life, risk of hospitaliza- abstracts and full texts were assessed for eligibility indepen-
tion, falls and fall-associated injuries, and fractures (second- dently by two authors (AG-H and RR-V) for potential inclu-
ary endpoints). sion. If necessary, a third researcher (MI) was consulted.
Finally, only those papers written in the English language
were included.
2 Methods
2.4 Data Collection Process
2.1 Protocol and Registration
For each study, the following data were extracted (i) first
This study followed the Preferred Reporting Items for Sys- author’s last name; (ii) year of publication; (iii) country;
tematic Reviews and Meta-Analyses (PRISMA) statement (iv) study setting; (v) sample size, participants’ character-
checklist [12] and is registered in the International Prospec- istics and mean age; and (vi) characteristics of the exercise
tive Register of Systematic Reviews (PROSPERO) (Regis- (exercise modality, frequency, and duration) and control
tration number: CRD42020139614) (Electronic Supplemen- intervention.
tary Material Appendix S1).
2.5 Endpoints
2.2 Information Sources and Search
Primary endpoints were dropouts (defined as the number of
Searches in PubMed, Cochrane Controlled Trials Registry, randomized participants having no post-intervention meas-
and SPORTDiscus combined with manual searches of exist- urements available due to health issues [medical problems,
ing literature were done from inception to September 16, adverse events]), and mortality. Secondary endpoints were
2019. The search strategy applied is available in Electronic the following: physical function, cognition status, health-
Supplementary Material Appendix S2. In addition, the refer- related quality of life, risk of hospitalization (inpatient hos-
ence list of each of the included studies was reviewed to find pitalization ≥ 24 h), falls (people who fell at least once) and
potential studies that could be used in this review. fall-associated injuries, and fractures (number of people who
Safety and Effectiveness of Exercise in Older Adults 1097

sustained a fracture). When there was insufficient informa- primary endpoints (dropouts due to health issues, and risk
tion, the corresponding author of the study in question was of mortality).
contacted.

2.6 Risk of Bias of Individual Studies 3 Results

The risk of bias was evaluated using the Physiotherapy Evi- 3.1 Study Selection
dence Database (PEDro) criteria [13], using an 11-item scale
designed for measuring the methodological quality of RCT. From the retrieved articles, 93 RCTs [1, 20–111] and 6
[112–117] secondary studies met the inclusion criteria and
2.7 Statistical Analyses were included in the analyses (with secondary studies being
only included to pool the secondary outcomes) (Fig. 1).
All analyses were carried out using Comprehensive Meta- However, due to missing data, we have included only 90
analysis Software (2nd version, Biostat; Englewood, NJ, RCTs in the quantity analysis (meta-analysis).
USA) to calculate (i) the risk ratio (RR) for dichotomous
outcomes, that is, dropouts due to health issues, mortality, 3.2 Study Characteristics
hospitalization, falls, and fractures; or (ii) the standardized
mean difference (SMD) for continuous data (i.e., physical The characteristics of the included studies are summarized
function parameters, cognition and health-related quality of in Electronic Supplementary Material Table S1. The final
life parameters) [14]. The RR or SMD of each parameter analysis included 28,523 participants (mean age 74.2 years).
(from baseline to follow-up) between groups [15] was cal- Most studies included apparently healthy older adults, but 25
culated and pooled using a random-effects model (DerSimo- RCTs were conducted in clinical populations (i.e., individu-
nian–Laird approach). For continuous outcomes, the pooled als with cancer [n = 3 studies], mild cognitive impairment,
effect size for SMD was classified as small (0 ≤ SMD ≤ 0.50), Parkinson, Alzheimer or dementia [n = 11], or cardiorespira-
moderate (0.50 < SMD ≤ 0.80) or large (SMD > 0.80) [16]. tory [n = 4], renal [n = 1], musculoskeletal [n = 3], or meta-
The percentage of total variation across studies due to bolic [n = 3] disease). Participants enrolled in the different
heterogeneity (Cochran’s Q test) was used to calculate studies were predominantly from the U.S. (n = 23 articles),
the I2 statistic [17], considering I2 values < 25%, 25–75%, with other studies from Australia (n = 15), Belgium (n = 2),
and > 75% as indicators of small, moderate, and high het- Brazil (n = 2), Canada (n = 5), Chile (n = 2), Denmark (n = 2),
erogeneity, respectively [18]. Each study was removed once Finland (n = 8), France (n = 3), Germany (n = 5), Hong Kong
from the model to analyze its individual influence on the (n = 4), Hungary (n = 2), Japan (n = 2), The Netherlands
overall results. Egger’s regression tests and funnel plots (n = 1), New Zealand (n = 4), South Korea (n = 1), Spain
were performed to detect small study effects and possible (n = 2), Sweden (n = 2), Taiwan (n = 1), Thailand (n = 2), or
publication bias [19]. Data were pooled if outcomes were the UK (n = 5). Most studies included men and women except
reported in at least 10 studies, because the power of the tests four [78, 81, 90, 118] that included only men, and 19 studies
for funnel plot asymmetry is too low to distinguish chance with women only [30, 33, 34, 42, 44, 49, 61, 62, 68, 69, 72,
from real asymmetry when fewer studies are evaluated [14]. 83, 88, 89, 93, 95, 99, 103, 108]. Sample sizes across studies
When possible, a subgroup analysis for the primary ranged from 20 [23, 56] to 6420 [54] participants.
endpoint was conducted taking into account the follow- The primary mode of exercise intervention was multi-
ing potential moderators: type of exercise intervention, component exercise training (n = 47), followed by mus-
participants’ health status (apparently healthy vs clinical cle strength (n = 24) and aerobic training (e.g., walking,
population), intervention structure (supervised group based, dancing) (n = 19), and Tai Chi (n = 4). Most studies used
unsupervised home based, or a combination thereof), pri- group-based supervised exercise training alone (n = 56) or
mary mode of exercise intervention (muscle strength [or combined with home-based unsupervised training (n = 21).
‘resistance’]-based, aerobic-based, multicomponent exercise The duration of the interventions ranged from 52 (in most
training, or Tai Chi), study setting (community dwellers vs studies) to 208 [85] weeks, and training frequency ranged
institutionalized), and risk of bias (PEDro Scale score < 7 from 1 to 7 sessions per week, with a duration of 10–90 min
or ≥ 7) [10]. Additionally, random-effects meta-regression per session. Most control groups were instructed to main-
analyses were used to evaluate the influence of age, physical tain usual activity levels with or without an additional non-
function and cognition characteristics at baseline and the exercise intervention (e.g., health education, social visits,
frequency of exercise (number of sessions per week) in the or telephone calls).
1098 A. García‑Hermoso et al.

Fig. 1  PRISMA flow diagram

3.3 Risk of Bias Within Studies control group, respectively (Electronic Supplementary Mate-
rial Table S3). Exercise and control groups did not differ in
The average total score was 6.1 with a range of 1–8. Thirty- terms of dropouts due to medical problems (RR = 1.05, 95%
six studies scored 7 or higher. Low scores corresponded confidence interval [CI] 0.95–1.17), with low heterogeneity
to studies that failed to conceal allocation (40.2%), or to for this result (I2 = 14.07%).
blind subjects (0%), or had researchers in charge of endpoint All subgroup analyses provided similar results independ-
assessment and/or exercise supervision (1.0%) (Electronic ent of the study setting, type of exercise training, or risk of
Supplementary Material Table S2). bias (Table 1). The meta-regression analyses showed that
participants’ age, physical function or cognition at base-
3.4 Synthesis of Results line, or exercise frequency (number of exercise sessions per
week) was not related to risk of mortality (ß = − 0.13–0.03)
3.4.1 Primary Endpoints (Table 2 and Electronic Supplementary Material Figure
S1–S5).
Dropouts due to health issues Forty-nine studies provided There was no evidence of publication bias in the analysis
information on dropouts due to medical problems for 21,292 of dropouts due to medical problems (p = 0.618), and the
participants. Overall, 16.9% (1638 older adults) and 19.0% results remained the same in the sensitivity analysis. The
(2207 older adults) of participants dropped out of the cor- funnel plot is presented in Electronic Supplementary Mate-
responding study due to health issues in the exercise and rial Figure S6.
Safety and Effectiveness of Exercise in Older Adults 1099

Mortality Fifty-six RCTs provided information on death participants died in the exercise and control groups, respec-
for 26,017 participants. Of these, 19 reported no deaths tively (Electronic Supplementary Material Table S3). All
during the trial [26, 28, 36, 38, 44, 48, 49, 55, 56, 78, 79, subgroup analyses provided similar results (Table 1), except
81, 86, 92, 93, 103, 107, 108, 119]. Exercise intervention for the analysis restricted to clinical populations, which
had no effects on mortality (RR = 0.93, 95% CI 0.83–1.04), showed that exercise significantly (p = 0.024) reduced the
with low heterogeneity between studies (I2= 0%). Specifi- risk of mortality (RR = 0.67, 95% CI 0.48–0.95; I2= 0%)
cally, 5.5% (664 older adults) and 5.8% (820 older adults) of (Table 1).

Table 1  Association of long- n RR 95%CI p value I2 Egger’s p

term exercise interventions
with primary endpoints (i.e., Dropouts due to health issues
dropouts due to health issues,
Overall 49 1.05 0.95–1.17 0.338 14.07 0.618
and mortality). Sensitivity
analyses as a function of Characteristics of older adults
participant’s characteristics, Clinical population 16 0.82 0.61–1.11 0.212 0 0.902
exercise type, study setting and Apparently healthy 37 1.07 0.94–1.22 0.308 30.63 0.357
risk of bias are also shown
Type of exercise training
Multicomponent 26 1.03 0.92–1.17 0.580 20.25 0.676
Strength 13 0.85 0.62–1.15 0.297 0 0.943
Aerobic 8 –
Tai Chi 2 –
Intervention structure
Home based 11 0.97 0.59–1.61 0.918 60.59 0.434
Group based 27 1.05 0.99–1.12 0.102 0 0.559
Home and group based 11 1.07 0.97–1.19 0.181 0 0.564
Study setting
Community dwellers 42 1.05 0.89–1.23 0.564 25.34 0.981
Institutionalized 6 –
Risk of bias
PEDro score < 7 24 1.06 0.98–1.14 0.101 3.27 0.455
PEDro score ≥ 7 24 1.01 0.78–1.30 0.941 28.30 0.550
Mortality
Overall 56 0.93 0.83–1.04 0.182 0 0.097
Characteristics of older adults
Clinical population 16 0.67 0.48–0.95 0.024 0 0.512
Apparently healthy 39 0.96 0.87–1.06 0.441 0 0.514
Type of exercise training
Multicomponent 33 0.95 0.85–1.06 0.376 0 0.002
Strength 12 1.02 0.75–1.40 0.933 6.56 0.075
Aerobic 7 –
Tai Chi 3 –
Intervention structure
Home based 13 1.07 0.83–1.39 0.604 2.89 0.869
Group based 33 0.95 0.84–1.07 0.400 0 0.030
Home and group based 9 –
Study setting
Community dwellers 48 0.91 0.81–1.01 0.082 0 0.075
Institutionalized 7 –
Risk of bias
PEDro score < 7 24 0.95 0.85–1.06 0.347 2.91 0.410
PEDro score ≥ 7 31 0.86 0.67–1.10 0.240 0 0.072

CI confidence interval, PEDro Physiotherapy Evidence Database, RR risk ratio

Significant p values are in bold
1100 A. García‑Hermoso et al.

The meta-regression analyses showed that neither age nor 4 Discussion

physical function nor cognition at baseline was associated
with risk of mortality (ß = − 0.20– 0.09) (Table 2 and Elec- The present meta-analysis found that long-term (≥ 1 year)
tronic Supplementary Material Figure S7–S11). exercise training interventions in older adults do not cause
There was no evidence of publication bias in the analysis more dropouts due to health issues and do not affect mor-
of mortality (p = 0.097), and the results remained the same tality compared with a control (usual care) group. Similar
in the sensitivity analysis. The funnel plot is presented in results were found for risk of hospitalization and fractures.
Electronic Supplementary Material Figure S12. Importantly, however, these types of interventions are asso-
ciated with a lower risk of falls and fall-induced injuries,
3.4.2 Secondary Endpoints and with improvements in muscle strength, balance, physical
function, and cognition. Moreover, a lower risk of mortal-
The secondary endpoint measures (which were pooled when ity was found in a separate analysis for patient populations
there were ≥ 10 studies per endpoint) were: physical func- only. Therefore, long-term physical exercise is overall safe
tion (assessed with balance tests, gait speed, knee-extension and effective in older adults, and its benefits appear to be
strength, timed get-up-and-go test, sit-to-stand test, and the independent of participants’ age, physical function or cogni-
short physical fitness battery [SPPB]), cognition status tion status at baseline.
(assessed with the mini-mental state examination [MMSE]), To determine the safety of exercise training, we regis-
health-related quality of life (assessed with the Short Form tered the number of dropouts per group due to health issues,
Health Survey [SF]-36 or SF-12), and risk of hospitalization, and the pooled analysis suggests that physical exercise does
falls, fall-associated injuries, and fractures. not increase the risk of dropouts compared with the con-
Compared with the control group, exercise interventions trol group, and thus, compliance to this type of intervention
were associated with a significantly lower number of falls does not seem to be compromised by health issues related
and fall-associated injuries, as well as with improvements to the intervention itself. In the same line, a recent system-
in all analyzed physical function parameters (balance, gait atic review reported only minor harms of exercise inter-
speed knee-extension strength, SPPB, sit-to-stand and timed ventions aiming at preventing falls in community-dwelling
get-up-and-go tests) and cognition (MMSE) (Table 3). The older adults [11]. Therefore, our results, together with the
forest plots are shown in Electronic Supplementary Material aforementioned study [11], suggest that, beyond its practical
Figure S13–S25. benefits, exercise does not cause more harm among older
Finally, Fig. 2 illustrates how long-term physical exercise adults than maintaining usual activity levels or usual care.
programs are safe and effective among older adults. Similar to the conclusions of our review, recent meta-anal-
yses have reported that long-term exercise training does not
increase the overall risk of mortality in older adults [9, 11].
The magnitude of the association for the risk of mortality in
our review in patients and healthy individuals combined (56
Table 2  Meta-regression analysis of the association between partici- RCTs, RR = 0.93, 95% CI 0.83–1.04) is in general agreement
pants’ age, physical function and cognition at baseline with the pri- with the results reported by Barreto et al. [9] (RR = 0.96,
mary endpoints 95% CI 0.85–1.09) and Guirguis-Blakeet al. [11] (RR = 0.96,
n β 95% CI p 95% CI 0.79–1.17). In line with recent research [9], the
risk of mortality in the general population of older adults
Dropouts due to health issues
is unlikely to be affected by exercise interventions alone.
Age 49 − 0.01 − 0.00 to 0.01 0.758
However, an important finding of our review is that long-
SPPB 16 0.03 − 0.03 to 0.10 0.308
term exercise decreases mortality risk specifically in clinical
Gait speed 22 − 0.13 − 1.51 to 1.25 0.858
populations of older adults (16 RCTs; RR = 0.67, 95% CI
Cognition (MMSE) 15 − 0.01 − 0.05 to 0.05 0.915
0.48–0.95). In this respect, a recent population-based cohort
Frequency per week 43 − 0.01 − 0.08 to 0.04 0.508
study reported significantly lower risk of mortality if becom-
Mortality
ing physically active among middle-aged and older patients
Age 56 0.01 − 0.01 to 0.02 0.112
with cardiovascular disease and cancer, which reinforces the
SPPB 15 0.09 − 0.21 to 0.39 0.553
role of physical exercise as a core therapeutic element for
Gait speed 22 − 0.20 − 1.43 to 1.03 0.749
treating prevalent diseases [120].
Cognition (MMSE) 23 − 0.01 − 0.05 to 0.03 0.683
Subgroup analyses revealed that type of physical exercise
Frequency per week 49 0.03 − 0.05 to 0.12 0.450
training, setting of the study, and risk of bias do not appear to
MMSE mini-mental state examination, SPPB short physical fitness influence the effect of exercise on the primary endpoint; that
battery
Safety and Effectiveness of Exercise in Older Adults 1101

Table 3  Association of long- n RR/SMD 95% CI p I2 Egger’s p

term exercise interventions
with secondary endpoints (i.e., Adverse events and hospitalization
adverse events, hospitalization,
Number of falls 44 0.89 0.83–0.96 0.002 69.57 0.674
cognition, health-related quality
of life and physical function Fall-associated injuries 16 0.78 0.67–0.92 0.003 46.67 0.309
parameters) Fractures 24 0.94 0.80–1.11 0.490 0 0.376
Hospitalization 12 1.04 0.97–1.11 0.217 19.17 0.439
Cognition
MMSE 10 0.24 0.05–0.42 0.011 68.52 0.028
Health-Related Quality of Life
Physical functioning (SF-36 or SF-12) 12 0.03 − 0.03–0.10 0.336 46.00 0.635
Mental health (SF-36 or SF-12) 12 0.05 − 0.01–0.11 0.124 27.99 0.615
Physical function parameters
Balance 21 0.31 0.21–0.42 < 0.001 54.95 0.058
Gait speed 19 0.13 0.03–0.23 0.008 46.60 0.932
Knee-extension strength 15 0.28 0.13– 0.44 < 0.001 72.89 0.012
SPPB 10 0.16 0.01–0.30 0.040 70.97 0.587
Sit to stand 15 − 0.27 − 0.38 to − 0.17 < 0.001 44.42 0.133
Timed up and Go 18 − 0.20 − 0.34 to − 0.06 0.005 74.64 0.501

CI confidence interval, MMSE mini-mental state examination, SF-12 12-Item Short Form Survey, SF-36
36-Item Short Form Survey, SPPB short physical fitness battery
Non-italicized entries = Standardized Mean Difference values; italicized entries = Risk Ratio values
Significant p values are in bold

is, dropouts due to medical problems and mortality. We also Some limitations of our review should be acknowledged.
evaluated the influence of age, cognition, physical function First, health issues were recorded through dropouts and
and exercise frequency on these results. With regard to age, we cannot determine whether they were directly associ-
although one could expect a priori that older participants are ated with physical exercise. In this regard, however, most
at greater risk of having adverse events when enrolling in RCTs reported only minor harms associated with exercise
an exercise program, this hypothesis was not corroborated interventions. Second, the included RCTs were heterogene-
by our results. In this respect, Hewitt et al. [41] showed a ous with respect to exercise training protocols and study
lower risk of dropouts due to health issues (RR = 0.35, 95% population (healthy vs clinical population); however, we
CI 0.15–0.91) in the exercise group, in individuals with low performed subgroup analyses to account for heterogeneity
physical function level (SPPB score < 9) [121]. However, among studies. Third, most of the analyzed studies included
the results of the present meta-regression analysis should be community-dwelling older adults, and the small number of
interpreted with caution because the low number of studies studies with institutionalized elderly people (n = 9) pre-
might have affected the results. vented us from performing a separate subgroup analysis,
With regard to secondary endpoints, our findings cor- thereby limiting the generalizability of our results. On the
roborate the results of a recent meta-analysis showing that other hand, a main strength of our meta-analysis is that it is,
exercise decreased the risk of falls and fall-associated inju- to our knowledge, the first to determine both the safety and
ries [9], and similar results were reported by Guirguis-Blake effectiveness of long-term physical exercise interventions
et al. for the US Preventive Services Task Force [11]. Thus, among older adults. Our meta-analysis provides an update
there seems to be compelling evidence to support the posi- and extension of the evidence reported in an earlier reviews
tive effects of exercise against fall-related outcomes in older [9], including more studies and other health outcomes. Fur-
adults. Furthermore, our review extends this literature by thermore, according to the Cochrane Handbook, and to avoid
examining, for the first time, the association of long-term publication bias, we pooled only the results of those out-
exercise training with several health outcomes in an elderly comes that were reported by at least 10 studies [14], thereby
population, showing significant improvements in physical avoiding any loss of statistical power.
function and cognition parameters, and providing further
support to the notion that regular exercise is necessary for
healthy aging and offers multi-systemic health benefits
[122].
1102 A. García‑Hermoso et al.

Fig. 2  This figure illustrates

how long-term physical exercise
programs are safe and effec-
tive among older adults. Green
arrows represent significant
results and blue equal signs
depict non-negative effects.
Numbers in brackets represent
data results for each parameter.
RR risk ratio, SMD standardized
mean difference. *The results
are shown in Table 3

5 Conclusion los III-CP18/0150). NM-V received funding from “la Caixa” Founda-
tion (ID 100010434), under agreement LCF/PR/PR15/51100006.

Compared with usual care, long-term (≥ 1 year) exercise

Conflicts of Interest Antonio García-Hermoso, Robinson Ramirez-
training interventions in older adults do not increase the Vélez, Mikel López Sáez de Asteasu, Nicolás Martínez-Velilla, Fabri-
risk of dropouts due to health issues, mortality, hospitaliza- cio Zambom-Ferraresi, Pedro Valenzuela, Alejandro Lucia and Mikel
tion, or fractures, but do decrease the risk of falls and fall- Izquierdo declare that they have no conflicts of interest relevant to the
associated injuries and mortality (although the latter only content of this review.
in clinical populations), while improving physical function
and cognition. Therefore, health care professionals have an
important role to play in reducing the prevalence of inactiv- References
ity and fostering regular exercise among older adults.
1. Pahor M, Guralnik JM, Ambrosius WT, Blair S, Bonds DE,
Church TS, et al. Effect of structured physical activity on preven-
Data Availability Statement The data that support the findings of this
tion of major mobility disability in older adults: the LIFE study
review are available on request from the corresponding author (Mikel
randomized clinical trial. JAMA [Internet]. 2014;311:2387–96
Izquierdo).
(cited 2019 Aug 4).
2. He W, Goodkind D, Kowal P. International population reports.
Compliance with Ethical Standards P95/16-1, An aging world: 2015. US Census Bureau, editors.
Washington, DC: US Government Publishing Office, 2016.
Funding This work was supported by the Biomedical Research 3. Keadle SK, McKinnon R, Graubard BI, Troiano RP. Prevalence
Networking Center on Frailty and Healthy Aging (CIBERFES) and trends in physical activity among older adults in the United
and FEDER funds from the European Union. It was further funded States: a comparison across three national surveys. Prev Med
by a grant from the Spanish Ministry of Economy, “Ministerio de (Baltim). 2016;89:37–43.
Economía y Competitividad,” Instituto de Salud Carlos III, FEDER, 4. Garatachea N, Pareja-Galeano H, Sanchis-Gomar F, Santos-
(PI17/01814). AGH is a Miguel Servet Fellow (Instituto de Salud Car- Lozano A, Fiuza-Luces C, Morán M, et al. Exercise attenuates
the major hallmarks of aging. Rejuvenation Res. 2015;18:57–89.
Safety and Effectiveness of Exercise in Older Adults 1103

5. Chodzko-Zajko WJ, Proctor DN, Fiatarone Singh MA, Minson 24. Becker C, Kron M, Lindemann U, Sturm E, Eichner B, Walter-
CT, Nigg CR, Salem GJ, et al. Exercise and physical activity for Jung B, et al. Effectiveness of a multifaceted intervention on falls
older adults. Med Sci Sport Exerc. 2009;41:1510–30. in nursing home residents. J Am Geriatr Soc. 2003;51:306–13.
6. Silva RB, Aldoradin-Cabeza H, Eslick GD, Phu S, Duque G. The 25. Blumenthal JA, Emery CF, Madden DJ, Coleman RE, Riddle
effect of physical exercise on frail older oersons: a systematic MW, Schniebolk S, et al. Effects of exercise training on cardi-
review. J Frailty Aging. 2017;6:91–6. orespiratory function in men and women older than 60 years of
7. Zheng G, Xia R, Zhou W, Tao J, Chen L. Aerobic exercise age. Am J Cardiol. 1991;67:633–9.
ameliorates cognitive function in older adults with mild cogni- 26. Bogaerts A, Verschueren S, Delecluse C, Claessens AL, Boonen
tive impairment: a systematic review and meta-analysis of ran- S. Effects of whole body vibration training on postural control
domised controlled trials. Br J Sports Med. 2016;50:1443–50. in older individuals: a 1 year randomized controlled trial. Gait
8. Chase J-AD, Phillips LJ, Brown M. Physical activity intervention Posture. 2007;26:309–16.
effects on physical function among community-dwelling older 27. Brandon LJ, Gaasch DA, Boyette LW, Lloyd AM. Effects of
adults: a systematic review and meta-analysis. J Aging Phys Act. long-term resistive training on mobility and strength in older
2017;25:149–70. adults with diabetes. Journals Gerontol Ser A Biol Sci Med Sci.
9. De Souto Barreto P, Rolland Y, Vellas B, Maltais M. Association 2003;58:M740–5.
of long-term exercise training with risk of falls, fractures, hos- 28. Brandon LJ, Boyette LW, Lloyd A, Gaasch DA. Resistive train-
pitalizations, and mortality in older adults: a systematic review ing and long-term function in older adults. J Aging Phys Act.
and meta-analysis. JAMA Intern Med. 2019;179:394–405. 2004;12:10–28.
10. Sherrington C, Michaleff ZA, Fairhall N, Paul SS, Tiedemann 29. Bunout D, Barrera G, Avendaño M, de la Maza P, Gattas V, Leiva
A, Whitney J, et al. Exercise to prevent falls in older adults: an L, et al. Results of a community-based weight-bearing resistance
updated systematic review and meta-analysis. Br J Sports Med. training programme for healthy Chilean elderly subjects. Age
2017;51:1750–8. Ageing. 2005;34:80–3.
11. Guirguis-Blake JM, Michael YL, Perdue LA, Coppola EL, Beil 30. Campbell AJ, Robertson MC, Gardner MM, Norton RN, Tilyard
TL. Interventions prevent falls in older adults: updated evidence MW, Buchner DM. Randomised controlled trial of a general
report and systematic review for the US preventive services task practice programme of home based exercise to prevent falls in
force. JAMA. 2018;319:1705–16. elderly women. BMJ. 1997;315:1065–9.
12. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, 31. Clemson L, Fiatarone Singh MA, Bundy A, Cumming RG,
Ioannidis JPA, et al. The PRISMA statement for reporting sys- Manollaras K, O’Loughlin P, et al. Integration of balance and
tematic reviews and meta-analyses of studies that evaluate health strength training into daily life activity to reduce rate of falls in
care interventions: explanation and elaboration. PLoS Med. older people (the LiFE study): randomised parallel trial. BMJ.
2009;6:e1000100. 2012;345:e4547.
13. Maher C, Sherrington C, Herbert RD, et al. Reliability of the 32. Dangour AD, Albala C, Allen E, Grundy E, Walker DG, Aedo
PEDro scale for rating quality of randomized controlled trials. C, et al. Effect of a nutrition supplement and physical activ-
academic.oup.com. https​://acade​mic.oup.com/ptj/artic​le-abstr​ ity program on pneumonia and walking capacity in Chilean
act/83/8/713/28052​87. Accessed 29 June 2019. older people: a factorial cluster randomized trial. PLoS Med.
14. Higgins JP, Green S, editors. Cochrane handbook for systematic 2011;8:e1001023.
reviews of interventions. Chichester: Wiley; 2008. 33. Ebrahim S, Thompson PW, Baskaran V, Evans K. Randomized
15. Morris S-18th annual conference of the S, 2003 U. Estimating placebo-controlled trial of brisk walking in the prevention of
effect size from the pretest-posttest-control design. http://facul​ postmenopausal osteoporosis. Age Ageing. 1997;26:253–60.
ty.cas.usf.edu/mbrann​ ick/papers​ /conf/esppcs​ iop03​ .pdf. Accessed 34. El-Khoury F, Cassou B, Latouche A, Aegerter P, Charles M-A,
29 June 2019. Dargent-Molina P. Effectiveness of two year balance training pro-
16. Cohen J. Statistical power analysis for the behavioral sciences. gramme on prevention of fall induced injuries in at risk women
2013. aged 75-85 living in community: ossébo randomised controlled
17. Higgins JPT, Thompson SG. Quantifying heterogeneity in a trial. BMJ. 2015;351:h3830.
meta-analysis. Stat Med. 2002;21:1539–58. 35. Elley CR, Robertson MC, Garrett S, Kerse NM, McKinlay E,
18. Higgins J, Thompson S, Deeks J, Altman DG. Measuring incon- Lawton B, et al. Effectiveness of a falls-and-fracture nurse coor-
sistency in meta-analyses. BMJ. 2003;327(7414):557–60. dinator to reduce falls: a randomized, controlled trial of at-risk
19. Egger M, Smith G, Schneider M, Minder C. Bias in meta-analysis older adults. J Am Geriatr Soc. 2008;56:1383–9.
detected by a simple, graphical test. BMJ. 1997;315:629–34. 36. Erickson KI, Voss MW, Prakash RS, Basak C, Szabo A, Chad-
20. Arrieta H, Astrugue C, Regueme S, Durrieu J, Maillard A, dock L, et al. Exercise training increases size of hippocampus
Rieger A, et al. Effects of a physical activity programme to pre- and improves memory. Proc Natl Acad Sci. 2011;108:3017–22.
vent physical performance decline in onco-geriatric patients: a 37. Fairhall N, Sherrington C, Kurrle SE, Lord SR, Lockwood K,
randomized multicentre trial. J Cachexia Sarcopenia Muscle. Cameron ID. Effect of a multifactorial interdisciplinary inter-
2019;10:287–97. vention on mobility-related disability in frail older people: ran-
21. Barnett A. Community-based group exercise improves balance domised controlled trial. BMC Med. 2012;10:120.
and reduces falls in at-risk older people: a randomised controlled 38. Foster ER, Golden L, Duncan RP, Earhart GM. Community-
trial. Age Ageing. 2003;32:407–14. based Argentine tango dance program is associated with
22. Batchelor FA, Hill KD, Mackintosh SF, Said CM, Whitehead increased activity participation among individuals with Parkin-
CH. Effects of a multifactorial falls prevention program for peo- son’s disease. Arch Phys Med Rehabil. 2013;94:240–9.
ple with stroke returning home after rehabilitation: a randomized 39. Gianoudis J, Bailey CA, Ebeling PR, Nowson CA, Sanders KM,
controlled trial. Arch Phys Med Rehabil. 2012;93:1648–55. Hill K, et al. Effects of a targeted multimodal exercise program
23. Baum EE, Jarjoura D, Polen AE, Faur D, Rutecki G. Effective- incorporating high-speed power training on falls and fracture
ness of a group exercise program in a long-term care facility: a risk factors in older adults: a community-based randomized con-
randomized pilot trial. J Am Med Dir Assoc. 2003;4:74–80. trolled trial. J Bone Miner Res. 2014;29:182–91.
40. Graessel E, Stemmer R, Eichenseer B, Pickel S, Donath C, Korn-
huber J, et al. Non-pharmacological, multicomponent group
1104 A. García‑Hermoso et al.

therapy in patients with degenerative dementia: a 12-month ran- stiffness and health-related quality of life in patients with chronic
domizied, controlled trial. BMC Med. 2011;9:129. kidney disease: a pilot study. Int Urol Nephrol. 2011;43:1133–41.
41. Hewitt J, Goodall S, Clemson L, Henwood T, Refshauge K. Pro- 57. Nowalk MP, Prendergast JM, Bayles CM, D’Amico FJ, Colvin
gressive resistance and balance training for falls prevention in GC. A randomized trial of exercise programs among older indi-
long-term residential aged care: a cluster randomized trial of the viduals living in two long-term care facilities: the FallsFREE
Sunbeam Program. J Am Med Dir Assoc. 2018;19:361–9. program. J Am Geriatr Soc. 2001;49:859–65.
42. Karinkanta S, Heinonen A, Sievänen H, Uusi-Rasi K, Pasanen 58. Olazarán J, Muñiz R, Reisberg B, Peña-Casanova J, del Ser T,
M, Ojala K, et al. A multi-component exercise regimen to pre- Cruz-Jentoft AJ, et al. Benefits of cognitive-motor intervention
vent functional decline and bone fragility in home-dwelling in MCI and mild to moderate Alzheimer disease. Neurology.
elderly women: randomized, controlled trial. Osteoporos Int. 2004;63:2348–53.
2007;18:453–62. 59. Pahor M, Blair SN, Espeland M, Fielding R, Gill TM, Guralnik
43. King AC, Pruitt LA, Phillips W, Oka R, Rodenburg A, Haskell JM, et al. Effects of a physical activity intervention on measures
WL. Comparative effects of two physical activity programs on of physical performance: results of the Lifestyle Interventions
measured and perceived physical functioning and other health- and Independence for Elders Pilot (LIFE-P) Study. J Gerontol
related quality of life outcomes in older adults. J Gerontol A Biol Ser A Biol Sci Med Sci. 2006;61:1157–65.
Sci Med Sci. 2000;55:M74–83. 60. Pakkala I, Read S, Leinonen R, Hirvensalo M, Lintunen T, Ran-
44. Liu-Ambrose T, Nagamatsu LS, Graf P, Beattie BL, Ashe MC, tanen T. The effects of physical activity counseling on mood
Handy TC. Resistance training and executive functions. Arch among 75- to 81-year-old people: a randomized controlled trial.
Intern Med. 2010;170:170. Prev Med (Baltim). 2008;46:412–8.
45. Liu-Ambrose T, Davis JC, Best JR, Dian L, Madden K, Cook 61. Park H, Kim KJ, Komatsu T, Park SK, Mutoh Y. Effect of com-
W, et al. Effect of a home-based exercise program on subsequent bined exercise training on bone, body balance, and gait ability:
falls among community-dwelling high-risk older adults after a a randomized controlled study in community-dwelling elderly
fall. JAMA. 2019;321:2092. women. J Bone Miner Metab. 2008;26:254–9.
46. Lord SR, Castell S, Corcoran J, Dayhew J, Matters B, Shan A, 62. Patil R, Uusi-Rasi K, Tokola K, Karinkanta S, Kannus P,
et al. The effect of group exercise on physical functioning and Sievänen H. Effects of a multimodal exercise program on physi-
falls in frail older people living in retirement villages: a rand- cal function, falls, and injuries in older women: a 2-year com-
omized, controlled trial. J Am Geriatr Soc. 2003;51:1685–92. munity-based, randomized controlled trial. J Am Geriatr Soc.
47. Luukinen H, Lehtola S, Jokelainen J, Väänänen-Sainio R, 2015;63:1306–13.
Lotvonen S, Koistinen P. Pragmatic exercise-oriented preven- 63. Pérula LA, Varas-Fabra F, Rodríguez V, Ruiz-Moral R, Fernán-
tion of falls among the elderly: a population-based, randomized, dez JA, González J, et al. Effectiveness of a multifactorial inter-
controlled trial. Prev Med (Baltim). 2007;44:265–71. vention program to reduce falls incidence among community-
48. MacRae PG, Feltner ME, Reinsch S. A 1-year exercise program living older adults: a randomized controlled trial. Arch Phys Med
for older women: effects on falls, injuries, and physical perfor- Rehabil. 2012;93:1677–84.
mance. J Aging Phys Act. 1994;2:127–42. 64. Petrella RJ, Koval JJ, Cunningham DA, Paterson DH. Can pri-
49. Madureira MM, Takayama L, Gallinaro AL, Caparbo VF, Costa mary care doctors prescribe exercise to improve fitness? The
RA, Pereira RMR. Balance training program is highly effective Step Test Exercise Prescription (STEP) project. Am J Prev Med.
in improving functional status and reducing the risk of falls in 2003;24:316–22.
elderly women with osteoporosis: a randomized controlled trial. 65. Pitkälä KH, Pöysti MM, Laakkonen M-L, Tilvis RS, Savikko N,
Osteoporos Int. 2007;18:419–25. Kautiainen H, et al. Effects of the Finnish Alzheimer Disease
50. Merom D, Mathieu E, Cerin E, Morton RL, Simpson JM, Exercise Trial (FINALEX). JAMA Intern Med. 2013;173:894.
Rissel C, et al. Social dancing and incidence of falls in older 66. Pothirat C, Chaiwong W, Phetsuk N, Liwsrisakun C, Bumroong-
adults: a cluster randomised controlled trial. PLoS Med. kit C, Deesomchok A, et al. Long-term efficacy of intensive
2016;13:e1002112. cycle ergometer exercise training program for advanced COPD
51. Messier SP, Mihalko SL, Legault C, Miller GD, Nicklas BJ, patients. Int J Chron Obstruct Pulmon Dis. 2015;10:133–44.
DeVita P, et al. Effects of intensive diet and exercise on knee 67. Prescott E, Hjardem-Hansen R, Dela F, Ørkild B, Teisner AS,
joint loads, inflammation, and clinical outcomes among over- Nielsen H. Effects of a 14-month low-cost maintenance training
weight and obese adults with knee osteoarthritis: the IDEA ran- program in patients with chronic systolic heart failure: a rand-
domized clinical trial. JAMA. 2013;310:1263–73. omized study. Eur J Cardiovasc Prev Rehabil. 2009;16:430–7.
52. Mian OS, Thom JM, Ardigò LP, Morse CI, Narici MV, Minetti 68. Pruitt LA, Taaffe DR, Marcus R. Effects of a one-year high-
AE. Effect of a 12-month physical conditioning programme on intensity versus low-intensity resistance training program
the metabolic cost of walking in healthy older adults. Eur J Appl on bone mineral density in older women. J Bone Miner Res.
Physiol. 2007;100:499–505. 1995;10:1788–95.
53. Möller UO, Kristensson J, Midlöv P, Ekdahl C, Jakobsson U. 69. Raso V, Benard G, Silva DA, Duarte AJ, Natale VM. Effect
Effects of a one-year home-based case management intervention of resistance training on immunological parameters of healthy
on falls in older people: a randomized controlled trial. J Aging elderly women. Med Sci Sport Exerc. 2007;39:2152–9.
Phys Act. 2014;22:457–64. 70. Reinsch S, MacRae P, Lachenbruch PA, Tobis JS. Attempts to
54. Munro JF. Cost effectiveness of a community based exercise prevent falls and injury: a prospective community study. Geron-
programme in over 65 year olds: cluster randomised trial. J Epi- tologist. 1992;32:450–6.
demiol Community Heal. 2004;58:1004–10. 71. Rejeski WJ, Ambrosius WT, Burdette JH, Walkup MP, Marsh
55. Muscari A, Giannoni C, Pierpaoli L, Berzigotti A, Maietta P, Fos- AP. Community weight loss to combat obesity and disabil-
chi E, et al. Chronic endurance exercise training prevents aging- ity in at-risk older adults. J Gerontol A Biol Sci Med Sci.
related cognitive decline in healthy older adults: a randomized 2017;72:1547–53.
controlled trial. Int J Geriatr Psychiatry. 2010;25:1055–64. 72. Rhodes EC, Martin AD, Taunton JE, Donnelly M, Warren J,
56. Mustata S, Groeneveld S, Davidson W, Ford G, Kiland K, Manns Elliot J. Effects of one year of resistance training on the relation
B. Effects of exercise training on physical impairment, arterial between muscular strength and bone density in elderly women.
Br J Sports Med. 2000;34:18–22.
Safety and Effectiveness of Exercise in Older Adults 1105

73. Robertson MC, Devlin N, Gardner MM, Campbell AJ. Effective- 90. Winters-Stone KM, Dobek JC, Bennett JA, Dieckmann NF,
ness and economic evaluation of a nurse delivered home exercise Maddalozzo GF, Ryan CW, et al. Resistance training reduces
programme to prevent falls 1: Randomised controlled trial. BMJ. disability in prostate cancer survivors on androgen deprivation
2001;322:697–701. therapy: evidence from a randomized controlled trial. Arch Phys
74. Rolland Y, Pillard F, Klapouszczak A, Reynish E, Thomas D, Med Rehabil. 2015;96:7–14.
Andrieu S, et al. Exercise program for nursing home residents 91. Wolf SL, Sattin RW, Kutner M, O’Grady M, Greenspan AI,
with Alzheimer’s disease: a 1-year randomized, controlled trial. Gregor RJ. Intense tai chi exercise training and fall occurrences
J Am Geriatr Soc. 2007;55:158–65. in older, transitionally frail adults: a randomized, controlled trial.
75. Sherrington C, Lord SR, Vogler CM, Close JCT, Howard K, J Am Geriatr Soc. 2003;51:1693–701.
Dean CM, et al. A post-hospital home exercise program improved 92. Woo J, Hong A, Lau E, Lynn H. A randomised controlled trial of
mobility but increased falls in older people: a randomised con- Tai Chi and resistance exercise on bone health, muscle strength
trolled trial. Quinn TJ, editor. PLoS One. 2014;9:e104412. and balance in community-living elderly people. Age Ageing.
76. Shumway-Cook A, Silver IF, LeMier M, York S, Cummings P, 2007;36:262–8.
Koepsell TD. Effectiveness of a community-based multifactorial 93. Englund U, Littbrand H, Sondell A, Pettersson U, Bucht G. A
intervention on falls and fall risk factors in community-living 1-year combined weight-bearing training program is beneficial
older adults: a randomized, controlled trial. J Gerontol Ser A. for bone mineral density and neuromuscular function in older
2007;62:1420–7. women. Osteoporos Int. 2005;16:1117–23.
77. Sjösten NM, Vahlberg TJ, Kivelä S-L. The effects of mul- 94. Fox KR, Stathi A, McKenna J, Davis MG. Physical activity and
tifactorial fall prevention on depressive symptoms among mental well-being in older people participating in the Better Age-
the aged at increased risk of falling. Int J Geriatr Psychiatry. ing Project. Eur J Appl Physiol. 2007;100:591–602.
2008;23:504–10. 95. Kemmler W, von Stengel S, Engelke K, Häberle L, Kalender
78. Sundstrup E, Jakobsen MD, Andersen LL, Andersen TR, Rand- WA. Exercise effects on bone mineral density, falls, coronary
ers MB, Helge JW, et al. Positive effects of 1-year football and risk factors, and health care costs in older women. Arch Intern
strength training on mechanical muscle function and functional Med. 2010;170:179.
capacity in elderly men. Eur J Appl Physiol. 2016;116:1127–38. 96. Kennis E, Verschueren SM, Bogaerts A, Van Roie E, Boonen S,
79. Suzuki T, Shimada H, Makizako H, Doi T, Yoshida D, Ito K, Delecluse C. Long-term impact of strength training on muscle
et al. A randomized controlled trial of multicomponent exer- strength characteristics in older adults. Arch Phys Med Rehabil.
cise in older adults with mild cognitive impairment. PLoS One. 2013;94:2054–60.
2013;8:e61483. 97. Kerr J, Rosenberg D, Millstein RA, Bolling K, Crist K, Take-
80. Thomas GN, Hong AWL, Tomlinson B, Lau E, Lam CWK, moto M, et al. Cluster randomized controlled trial of a multilevel
Sanderson JE, et al. Effects of Tai Chi and resistance train- physical activity intervention for older adults. Int J Behav Nutr
ing on cardiovascular risk factors in elderly Chinese subjects: Phys Act. 2018;15:32.
a 12-month longitudinal, randomized, controlled intervention 98. Kerse N, Peri K, Robinson E, Wilkinson T, von Randow M, Kiata
study. Clin Endocrinol (Oxf). 2005;63:663–9. L, et al. Does a functional activity programme improve function,
81. Tsai C-L, Wang C-H, Pan C-Y, Chen F-C. The effects of long- quality of life, and falls for residents in long term care? Cluster
term resistance exercise on the relationship between neurocogni- randomised controlled trial. BMJ. 2008;337:a1445.
tive performance and GH, IGF-1, and homocysteine levels in the 99. Lord SR, Ward JA, Williams P, Strudwick M. The effect of a
elderly. Front Behav Neurosci. 2015;9:23. 12-month exercise trial on balance, strength, and falls in older
82. Underwood M, Lamb SE, Eldridge S, Sheehan B, Slowther A-M, women: a randomized controlled trial. J Am Geriatr Soc.
Spencer A, et al. Exercise for depression in elderly residents 1995;43:1198–206.
of care homes: a cluster-randomised controlled trial. Lancet. 100. Lam LCW, Chau RCM, Wong BML, Fung AWT, Lui VWC,
2013;382:41–9. Tam CCW, et al. Interim follow-up of a randomized controlled
83. Uusi-Rasi K, Patil R, Karinkanta S, Kannus P, Tokola K, Lam- trial comparing Chinese style mind body (Tai Chi) and stretching
berg-Allardt C, et al. Exercise and vitamin D in fall prevention exercises on cognitive function in subjects at risk of progressive
among older women. JAMA Intern Med. 2015;175:703. cognitive decline. Int J Geriatr Psychiatry. 2011;26:733–40.
84. van Uffelen JGZ, Chinapaw MJM, van Mechelen W, Hopman- 101. Lam LCW, Chan WC, Leung T, Fung AWT, Leung EMF. Would
Rock M. Walking or vitamin B for cognition in older adults with older adults with mild cognitive impairment adhere to and ben-
mild cognitive impairment? A randomised controlled trial. Br J efit from a structured lifestyle activity intervention to enhance
Sports Med. 2008;42:344–51. cognition: a cluster randomized controlled trial. PLoS One.
85. Vetter NJ, Lewis PA, Ford D. Can health visitors prevent frac- 2015;10:e0118173.
tures in elderly people? BMJ. 1992;304:888–90. 102. Sjösten NM, Salonoja M, Piirtola M, Vahlberg T, Isoaho R,
86. Villareal DT, Chode S, Parimi N, Sinacore DR, Hilton T, Hyttinen H, et al. A multifactorial fall prevention programme
Armamento-Villareal R, et al. Weight loss, exercise, or both in home-dwelling elderly people: a randomized-controlled trial.
and physical function in obese older adults. N Engl J Med. Public Health. 2007;121:308–18.
2011;364:1218–29. 103. Korpelainen R, Keinänen-Kiukaanniemi S, Heikkinen J,
87. Voelcker-Rehage C, Godde B, Staudinger UM. Cardiovascular Väänänen K, Korpelainen J. Effect of impact exercise on bone
and coordination training differentially improve cognitive perfor- mineral density in elderly women with low BMD: a population-
mance and neural processing in older adults. Front Hum Neuro- based randomized controlled 30-month intervention. Osteoporos
sci. 2011;5:26. Int. 2006;17:109–18.
88. von Stengel S, Kemmler W, Engelke K, Kalender WA. Effects of 104. Kruse RL, LeMaster JW, Madsen RW. Fall and balance outcomes
whole body vibration on bone mineral density and falls: results after an intervention to promote leg strength, balance, and walk-
of the randomized controlled ELVIS study with postmenopausal ing in people with diabetic peripheral neuropathy: “Feet First”
women. Osteoporos Int. 2011;22:317–25. randomized controlled trial. Phys Ther. 2010;90:1568–79.
89. Williams P, Lord SR. Effects of group exercise on cognitive func- 105. Dean CM, Rissel C, Sherrington C, Sharkey M, Cumming RG,
tioning and mood in older women. Aust N Z J Public Health. Lord SR, et al. Exercise to enhance mobility and prevent falls
1997;21:45–52.
1106 A. García‑Hermoso et al.

after stroke: the community stroke club randomized trial. Neu- 114. Lam LCW, Chau RCM, Wong BML, Fung AWT, Tam CWC,
rorehabil Neural Repair. 2012;26:1046–57. Leung GTY, et al. A 1-year randomized controlled trial com-
106. Morey MC, Peterson MJ, Pieper CF, Sloane R, Crowley GM, paring mind body exercise (Tai Chi) with stretching and toning
Cowper PA, et al. The veterans learning to improve fitness and exercise on cognitive function in older chinese adults at risk of
function in elders study: a randomized trial of primary care-based cognitive decline. J Am Med Dir Assoc. 2012;13:568.e15–20.
physical activity counseling for older men. J Am Geriatr Soc. 115. Fairhall N, Sherrington C, Lord SR, Kurrle SE, Langron C, Lock-
2009;57:1166–74. wood K, et al. Effect of a multifactorial, interdisciplinary inter-
107. Kim H, Yoshida H, Suzuki T. Falls and fractures in participants vention on risk factors for falls and fall rate in frail older people:
and excluded non-participants of a fall prevention exercise pro- a randomised controlled trial. Age Ageing. 2014;43:616–22.
gram for elderly women with a history of falls: 1-year follow-up 116. Bouchonville M, Armamento-Villareal R, Shah K, Napoli N,
study. Geriatr Gerontol Int. 2014;14:285–92. Sinacore DR, Qualls C, et al. Weight loss, exercise or both and
108. Mikó I, Szerb I, Szerb A, Poor G. Effectiveness of balance train- cardiometabolic risk factors in obese older adults: results of a
ing programme in reducing the frequency of falling in established randomized controlled trial. Int J Obes. 2014;38:423–31.
osteoporotic women: a randomized controlled trial. Clin Rehabil. 117. Williamson JD, Espeland M, Kritchevsky SB, Newman AB, King
2017;31:217–24. AC, Pahor M, et al. Changes in cognitive function in a rand-
109. Boongird C, Keesukphan P, Phiphadthakusolkul S, Rattanasiri omized trial of physical activity: results of the lifestyle interven-
S, Thakkinstian A. Effects of a simple home-based exercise tions and independence for elders pilot study. J Gerontol A Biol
program on fall prevention in older adults: a 12-month primary Sci Med Sci. 2009;64:688–94.
care setting, randomized controlled trial. Geriatr Gerontol Int. 118. Morse CI, Thom JM, Mian OS, Muirhead A, Birch KM, Narici
2017;17:2157–63. MV. Muscle strength, volume and activation following 12-month
110. Day L, Hill KD, Stathakis VZ, Flicker L, Segal L, Cicuttini F, resistance training in 70-year-old males. Eur J Appl Physiol.
et al. Impact of tai-chi on falls among preclinically disabled older 2005;95:197–204.
people. A randomized controlled trial. J Am Med Dir Assoc. 119. Morey MC, Snyder DC, Sloane R, Cohen HJ, Peterson B, Hart-
2015;16:420–6. man TJ, et al. Effects of home-based diet and exercise on func-
111. Voukelatos A, Merom D, Sherrington C, Rissel C, Cumming RG, tional outcomes among older, overweight long-term cancer sur-
Lord SR. The impact of a home-based walking programme on vivors. JAMA. 2009;301:1883.
falls in older people: the Easy Steps randomised controlled trial. 120. Mok A, Khaw K-T, Luben R, Wareham N, Brage S. Physical
Age Ageing. 2015;44:377–83. activity trajectories and mortality: population based cohort study.
112. Öhman H, Savikko N, Strandberg TE, Kautiainen H, Raivio MM, BMJ. 2019;365:l2323.
Laakkonen M-L, et al. Effects of exercise on cognition: the Finn- 121. Pavasini R, Guralnik J, Brown JC, di Bari M, Cesari M, Landi F,
ish Alzheimer Disease Exercise Trial: a randomized, controlled et al. Short physical performance battery and all-cause mortality:
trial. J Am Geriatr Soc. 2016;64:731–8. systematic review and meta-analysis. BMC Med. 2016;14:215.
113. Wang X, Hsu F-C, Isom S, Walkup MP, Kritchevsky SB, Good- 122. Fiuza-Luces C, Santos-Lozano A, Joyner M, Carrera-Bastos P,
paster BH, et al. Effects of a 12-month physical activity interven- Picazo O, Zugaza JL, et al. Exercise benefits in cardiovascular
tion on prevalence of metabolic syndrome in elderly men and disease: beyond attenuation of traditional risk factors. Nat Rev
women. J Gerontol A Biol Sci Med Sci. 2012;67:417–24. Cardiol. 2018;15:731–43.

Affiliations

Antonio García‑Hermoso1,2,3 · Robinson Ramirez‑Vélez1,3 · Mikel L. Sáez de Asteasu1,3 · Nicolás Martínez‑Velilla1,3 ·

Fabricio Zambom‑Ferraresi1,3 · Pedro L. Valenzuela4 · Alejandro Lucia3,5,6 · Mikel Izquierdo1,3

1 4
Navarrabiomed, Complejo Hospitalario de Navarra Department of Systems Biology, University of Alcalá,
(CHN), Universidad Pública de Navarra (UPNA), IdiSNA, Madrid, Spain
Pamplona, Navarra, Spain 5
Faculty of Sport Sciences, Universidad Europea de Madrid,
2
Laboratorio de Ciencias de la Actividad Física, el Deporte y Madrid, Spain
la Salud, Universidad de Santiago de Chile, Santiago, Chile 6
Research Institute of the Hospital 12 de Octubre (‘i + 12’),
3
CIBER of Frailty and Healthy Aging (CIBERFES), Instituto Madrid, Spain
de Salud Carlos III, Madrid, Spain