Exercise and Quality of Life in Dementia: A Systematic Review and Meta-Analysis of Randomized Controlled Trials

Abstract

This study appraises evidence for the effectiveness of exercise in improving quality of life (QoL) in persons with dementia (PwDs). The Cochrane, Medline, EMBASE, PsycINFO, and Cumulative Index to Nursing and Allied Health Literature databases were searched for peer-reviewed clinical trials of exercise in PwD. Additional searches of the PubMed for ahead-of-print citations and reference lists of articles were undertaken. Studies not including QoL as an outcome were excluded. Thirteen studies comprising 903 PwDs were identified. Random effects meta-analyses indicate that aerobic exercises may produce larger effect on QoL in PwD. However, overall postintervention results suggest exercise interventions led to small and nonsignificant improvement in QoL. Due to studies showing inconsistent results, the evidence for beneficial effects of exercise on QoL in PwD is inconclusive at this time. Future trials should tailor specific types of exercise programs for well-defined cohorts of PwDs.

Keywords

quality of life exercise psychosocial interventions Alzheimer’s disease

Introduction

Dementia is a chronically disabling disease characterized by progressive deterioration in cognition and functioning which results in the eventual loss of independence and personal identity in persons with the disease (Alzheimer’s Society, 2014). The many losses in dementia have been linked to a reduction in quality of life (QoL) of both the affected individual and their caregivers (Brod, Stewart, Sands, & Walton, 1999; Conde-Sala, Garre-Olmo, Turro-Garriga, Lopez-Pousa, & Vilalta-Franch, 2009).

Even though currently available interventions have shown only modest benefits in modifying the clinical trajectory of dementia, nonpharmacological approaches have been recommended as logical first steps in the management of persons with the disease (National Institute of Health and Clinical Excellence [NICE]/Social Care Institute for Excellence, 2011). This is in light of a shifting paradigm, since the 1990s, from considering dementia within a “standard illness” framework (Kitwood, 1990) to a more social–psychological model with greater emphasis on the person, rather than the disease (Kitwood & Bredin, 1992). Compared with pharmacological treatments (Ballard et al., 2009), nonpharmacological approaches are acknowledged to focus more on the individual with the disease (Edvardsson, Fetherstonhaugh, & Nay, 2010); thus, approaches such as exercise therapy have been used with relative success in the management of cognitive and noncognitive symptoms of dementia (Pitkala, Savikko, Poysti, Strandberg, & Laakkonen, 2013), as well as maintenance of physical functioning overtime (Forbes, Thiessen, Blake, Forbes, & Forbes, 2013).

Following on the notion of benefit of exercises in persons with dementia (PwDs), the World Health Organization (WHO; 2010) currently recommends between 150 and 300 min of moderate intensity exercises per week, including aerobic and strength building components, for both the general population of older adults and those with neurodegerative diseases. Despite these recommendations, it is yet unclear whether exercise-based interventions improve the global construct of QoL (Brod et al., 1999) in PwD. Answer to this question is important because in the absence of definitive treatment for diminishing intellectual and functional capacity of PwD, the ability of available health care interventions to positively affect QoL in PwD, as well as their caregivers, may be a more urgent priority.

QoL in Dementia

QoL in dementia is a dynamic and multidimensional index of health that is central to the overall well-being of persons with the disease. Higher or lower QoL is thought to be the balance between the perception of self-esteem, leisure, happiness and unhappiness, and belongingness in the PwD (Brod et al., 1999). Even though dementia may result in the diminishing of the person’s cognitive and functional ability, many patients with the disease, as well as their caregivers, are able to perceive recognition, respect, trust, as well as embarrassment.

The challenge for both clinicians and researchers is often that QoL in dementia is a difficult phenomenon to measure. It has both objective and subjective components. However, the experience of emotional well-being is mostly a subjective state of awareness, which is likely influenced by the values, current needs, and social circumstances of the PwD. Therefore, even though the objective component of QoL is more commonly assessed, and often by the report of a proxy, it may be equally important to evaluate the subjective component by relying on direct reports from the PwD. However, as cognitive impairment may have a greater influence on self-reports by PwDs (Conde-Sala et al., 2009), there is a valid concern about how much of the questions asked about QoL are well understood.

In line with these limitations, it is often the case that people with dementia rate their QoL better than those rated by their carers, or other observers (Hoe, Hancock, Livingston, & Orrell, 2006; Ready, Ott, & Grace, 2004; Thorgrimsen et al., 2003). Carer’s ratings of patients QoL is more likely influenced by the carer’s own level of burden (Sands, Ferreira, Stewart, Brod, & Yaffe, 2004) and state of emotional well-being (Conde-Sala et al., 2009). As such, it has frequently been observed that better carer rating of patient’s QoL is influenced by lower levels of dependency in the PwD (Hoe et al., 2006).

Exercise and QoL in Dementia

Exercise-based interventions are systematic approaches that use regular movements following a routine that is suggested by a trained therapist. They may include several programs ranging from walking, dance, Yoga, and martial art techniques.

In the population with dementia, exercise techniques are thought to have the potential to improve QoL in two major ways. First, an awareness of the participants own breathing movements occurs as the rate of breathing increases especially during the resting phase after the exercise. This experience of bodily sensation and movement is thought to reduce current tension and anxiety in the participant (Burg & Michalak, 2011). Second, exercises afford PwDs and their caregivers a meaningful connection with other participants. In this way, a sense of enjoyment, trust, and belongingness is created (Wu et al., 2015). These presumed effects of exercises would suggest that it is likely effective in improving five of the 10 conceptually derived domains of QoL in dementia (Brod et al., 1999). These domains include social interaction, interaction capacity, bodily well-being, sense of well-being, and sense of aesthetics.

One previous Cochrane review investigating the effect of exercise-based interventions on several dementia outcomes (Forbes et al., 2013) could not examine QoL because the authors found only one study (Conradsson, Littbrand, Lindelof, Gustafson, & Rosendahl, 2010) meeting the criteria for quantitative synthesis. Most of the studies available at the time did not examine QoL or did not publish the required data for meta-analyses. Yet there are currently no systematic reviews investigating the effectiveness of exercise-based intervention for QoL in PwD. In the 4 years since the Forbes review (Forbes et al., 2013), the literature on exercise for QoL in PwD has been boosted by the addition of five published randomized controlled trials (RCTs). Despite this progress, the question on whether exercise therapy improves QoL in PwD remains unanswered.

The present systematic review aims to synthesize the results of available clinical trials of exercise-based interventions for QoL in PwD. This is in order to provide a measure of their effectiveness for this dimension of well-being in persons with the disease.

Method

Search Strategy

This review followed conventional recommendations for the methodology and reporting of systematic reviews (Moher, Liberati, Tetzlaff, Altman, & Group, 2009; National Institute for Health and Care Excellence, 2012). First, a scoping search of the Cochrane database was conducted for existing records for systematic reviews or protocols dealing with trials of exercise interventions for QoL in dementia. Next, a more comprehensive literature search was conducted using Medline (Ovid SP 1946, December 23, 2015), EMBASE (Ovid SP 1974, December 24, 2015), PsycINFO (Ovid SP 1806, December 24, 2015), and the Cumulative Index to Nursing and Allied health Literature (CINAHL-EBSCO host, November 6, 2015) databases. The following keywords were used for the searches with the “explode” operator: dementia OR “Alzheimer’s disease,” AND exercise OR “exercise therapy” OR “physical therapy” OR “physical activity” OR physiotherapy OR interventions, AND “quality of life” OR “emotional well-being.” To retrieve “in-process” and “ahead of print” citations, four repeated searches of the PubMed database were conducted between November 25, 2015, and February 21, 2016. For this, the following keywords were combined: (dementia OR “Alzheimer’s disease”) AND (exercise OR “exercise therapy” OR “physical therapy” OR interventions OR “physical activity” OR physiotherapy) AND “quality of life” OR “emotional well-being”). Additional hand searching of the reference list of relevant articles retrieved from the databases was also implemented.

The searches were limited to English language and human literature. Limits on publication dates were not imposed. Eligible studies had to be peer reviewed, reporting on participants with any type of dementia, and having physical exercise therapy as the experimental intervention. We defined physical exercise therapy as exercise techniques or combination of techniques aimed at improving strength, endurance, flexibility, balance, and mobility. Studies not including outcome measures of QoL were excluded.

In subsequent quantitative analyses, studies where participants were not selected based on a randomization scheme and those not including an independent control population were additionally excluded. Also excluded were published trials with no data for participants who did not receive exercise interventions.

Study assessments for inclusion and exclusion criteria as well as subsequent data extraction and group categorization as aerobic, nonaerobic, or combined exercises was done by two independent assessors based on the description of the intervention in the original article. One of the two assessors is an exercise physiologist. It was agreed a priori that in cases of disagreement, a consensus will be reached based on the decision of an experienced colleague who is not directly involved in the present systematic review. The study exclusion criteria as applied by two independent assessors are presented in table 1.

Table 1.

Study Exclusion Criteria as Applied by Two Independent Assessors.

Excluded from qualitative syntheses

1. Trials not reported in the English Language

2. Studies not involving human subjects

3. Trial reports with no evidence of peer review

4. Trials not including experimental interventions meeting a priori definition of physical exercise

Excluded from qualitative syntheses

1. Trials with no evidence of randomization

2. Trials not including an independent control group

3. Trials not including data on participants receiving control interventions

Assessment of Risk of Bias

A standard framework (National Institute for Health and Care Excellence, 2012) was used for judgments about the risk of bias in the selected trials. Risk of bias was classified as low, unclear/unknown, and high. In all, we considered all four major types of biases listed in the NICE criteria to determine the overall risk of bias: selection (randomization, allocation concealment, groups comparability), performance (personnel/patient blinding, whether cointerventions were introduced), attrition (whether there were substantial losses to follow-up, and whether follow-up was comparable in the intervention versus control groups), and detection biases (whether there was adequate length of follow-up, use of reliable outcome measure, and blinding of outcome assessors). We allocated points to individual studies according to the assessed risk contributed by each component of the major types of biases in the NICE criteria. Two points were allocated when the risk of bias was determined to be low, 1 point when this was unclear/unknown, and no points when the risk of bias was clearly high. We then calculated the average risk of bias for a particular study by summing up the total points accrued by that study and dividing this sum by the total number of components assessed. Finally, we classified the overall risk of bias for a particular study as high (when the average risk of bias scores for that study is less than 1), moderate (when this is between 1 and 1.5), and low (when the score is greater than 1.5). Similar to the study selection methods, judgments about risk of bias in the included studies were made independently by two independent assessors.

Meta-Analysis

To explore the overall quantitative effects, postintervention QoL data for both the intervention and control groups from individual RCT providing such data were combined using the Cochrane review manager (Revman) Version 5.3 software (The Cochrane Collaboration, 2014). The study effect measure was the standardized mean difference in postintervention QoL scores. This was computed from the mean and standard deviations of outcome scores of the respective QoL instruments. For studies where both self and caregivers rating of QoL were reported for the same measure, the better rating of the two was considered. The inverse of variance method was used for weighting.

As heterogeneity was expected due to differences in the type, frequency, and intensity of exercise interventions, as well as postintervention QoL measurements across studies, a random effect meta-analysis model was chosen. To reduce the extent of methodological heterogeneity, we limited meta-analyses to RCTs. To determine the extent of statistical heterogeneity, we estimated the percentage of total variation in effect sizes reported across studies that is due to heterogeneity, rather than chance. The extent of this variation is reported using the I² statistics. Because statistical heterogeneity is a function of the variance in effect sizes across studies included in the meta-analysis model, we conducted an adjusted analysis by controlling for study effect outliers in cases of significant statistical heterogeneity. In considering clinical heterogeneity, additional subgroup meta-analyses based on subtype of dementia (i.e., Alzheimer’s disease) and exercise therapy (whether aerobic, nonaerobic, or combined) were conducted.

Results

Search Results

The combined database and hand searches identified a total of 556 records. Duplicates in the records of either database (54 articles) were excluded. After screening of the titles and abstracts, a further 486 materials were excluded as they did not contain information relevant to the review question. The remaining 16 full texts articles of clinical trials of exercise interventions for PwD were retrieved. After reading through the full text of these papers, a further three articles (Litchke, Hodges, Jan, & Reardon, 2012; Pitkala et al., 2013; Santana-Sosa, Barriopedro, Lopez-Mojares, Perez, & Lucia, 2008) were excluded. Details of included and excluded studies are shown on the PRISMA flow chart (Moher et al., 2009) in Figure 1.

Figure 1.

PRISMA flow chart showing details of included and excluded studies.

Appraisal of the Selected Studies

Key information synthesized from the 13 trials of exercise interventions for QoL in PwD is summarized in Table 2.

Table 2.

Study Characteristics.

Reference	Type of study	Exercise intervention	Frequency	Duration	Control intervention	Outcome measure	Outcome effect estimate	Effect (yes/no)
Roger et al. (2005)	nRCT	Nonaerobic: Progressive strength training exercise	30 min per week	12 weeks	Usual care	Self-rated QOL-AD	0.9^a	No
Williams and Tappen (2007)	Pre-/posttest	Nonaerobic: Strength, balance and walking activity	50 min per week	16 weeks	Nontherapeutic Social conversation	Observed Affect Scale	3.0^a	Yes
Steinberg, Leoutsakos, Podewils, and Lyketsos (2009)	RCT	Aerobic and nonaerobic: Fitness, strength, balance, and flexibility training	Daily completion of exercise tasks	12 weeks	OT-Led safety assessments	Carer-rated ADQRL	−7.8^b	No
Conradsson, Littbrand, Lindelof, Gustafson, and Rosendahl (2010)	Cluster RCT	Nonaerobic: Functional weight bearing activity	120 min per week	12 weeks	OT-Led Sedentary activities	Self-rated well-being (PGCMS)	0.6^c	No
Gallego, Alexey, Ma Clara, Lina, and Reyes (2011)	Pre-/posttest	Aerobic and Nonaerobic: Flexibility, strength, and respiration activity	120 min per week	14 weeks	Preintervention status	Carer-rated ADRQL	−2.38^c	No
Suttanon et al. (2013)	RCT	Nonaerobic: Standing, strengthening, and graduated walking	Daily completion of exercise tasks	48 weeks	General health education	Self-rated AQoL	−1.0^b	No
Lowery et al. (2014)	RCT	Aerobic: Tailored walking regimen	140 min per week	12 weeks	Usual treatment	Carer-rated DEMQoL	2.6^b	No
Yu et al. (2013)	Pre-/posttest	Aerobic: Cycling exercises	30-135 min per week	24 weeks	Preintervention status	Self-rated QoL-AD	0.3^d	Yes
Frederiksen, Sobol, Beyer, Hasselbalch, and Waldemar (2014)	Pre-/posttest	Aerobic: Balance, warm-ups and group aerobic exercises	120 min per week	14 weeks	Preintervention status	Self- and Carer-rated QoL-AD	0.4^c	No
Aguiar, Monteiro, Feres, Gomes, and Melo (2014)	RCT	Aerobic and nonaerobic: Flexibility, strength, and balance	80 min per week	24 weeks	Usual care	Self-rated QoL-AD	0.2^c	Yes
Barnes et al. (2015)	Cross-over nRCT	Nonaerobic: Standardized chair-based exercises preceded by body awareness training	135 min per week	18 weeks	Usual care	Self- and Carer-rated QoL-AD	0.8^c	Yes
Hoffmann et al. (2015)	RCT	Aerobic: Exercises on a treadmill, ergometer bicycle, and cross trainer	180 min per week	16 weeks	Usual care	Self- and Carer-rated EuroQoL index	0.01^c	No
Telenius, Engedal, and Bergland (2015)	RCT	Aerobic and nonaerobic: Warm-ups, strength, and balance	100-120 min per week	12 weeks	OT-led fun activities: reading, listening to music, and social conversation	Carer-rated QUALID	0.0^d	No

Note. RCT = randomized controlled trial; nRCT =non Randomized Controlled Trial; OT= Occupational Therapist; QoL-AD = Quality of Life-Alzheimer’s Disease; ADRQL = Alzheimer’s Disease Related Quality of Life; PGCMS = Philadelphia Geriatric Center Morale Scale; AQoL=Assessment of Quality of Life; ADQRL = Alzheimer’s Disease Related Quality ofLife Scale; EuroQoL = The European Health Related Quality of Life Scale; QUALID = Quality of Life in Late Stage Alzheimer’s Disease Scale; DEMQOL = Dementia Quality of Life measure.

Independent samples ANOVA.

Coefficient of treatment effect overtime.

Between-group mean difference.

Cohen’s standardized.

Of the 13 trials in Table 3, two (Barnes et al., 2015; Rogers, 2005) were nonrandomized but included a control group, whereas a total of four reports of preintervention and postintervention tests (Frederiksen, Sobol, Beyer, Hasselbalch, & Waldemar, 2014; Gallego, Alexey, Ma Clara, Lina, & Reyes, 2011; Williams & Tappen, 2007; Yu et al., 2013) were identified. The remaining seven were RCTs (Aguiar, Monteiro, Feres, Gomes, & Melo, 2014; Conradsson et al., 2010; Hoffmann et al., 2015; Lowery et al., 2014; Steinberg, Leoutsakos, Podewils, & Lyketsos, 2009; Suttanon et al., 2013; Telenius, Engedal, & Bergland, 2015). In four of the 13 trials (Frederiksen et al., 2014; Hoffmann et al., 2015; Steinberg et al., 2009; Williams & Tappen, 2007), specific diagnoses of Alzheimer’s disease were made according to criteria of the National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association (NINCDS-ADRDA; McKhann et al., 1984; Table 3). The studies led by Gallego (Gallego et al., 2011), Suttanon (Suttanon et al., 2013), Yu (Yu et al., 2013), and Aguiar (Aguiar et al., 2014) also used participants with specific diagnosis of Alzheimer’s disease, although this was based on the “best judgments” of dementia made by a clinician. Using the same method of disease ascertainment (clinician judgments), the remaining studies included participants with a broad definition of dementia. Apart from the study led by Williams (Williams & Tappen, 2007), the other studies excluded participants with severe dementia. Four trials were conducted among community dwelling participants, whereas three included PwD in care or nursing homes, while the other two trials used participants attending out-patient clinics.

Table 3.

Characteristics of Participants With Dementia.

Reference	Intervention/control (N)	Country	Setting	Female (%)	Ascertainment/types of dementia	Baseline MMSEM (±SD)	AgeM (±SD)
Roger et al. (2005)	17/8	United States	Care home	72	Clinician judgment/any dementia	15.7 (9.0)	94.0 (6.0)
Williams et al. (2007)	30^a	United States	Care home	85	NINCDS-ADRDA/Alzheimer’s	10.4 (6.3)	88.0 (6.0)
Steinberg, Leoutsakos, Podewils, and Lyketsos (2009)	14/13	United States	Community	70	NINCDS-ADRDA/Alzheimer’s	20.1 (5.1)	76.5 (3.9)
Conradsson, Littbrand, Lindelof, Gustafson, and Rosendahl (2010)	91/100	Sweden	Care home	73	Clinician judgment/any dementia	17.5 (5.0)	85.3 (6.1)
Gallego, Alexey, Ma Clara, Lina, and Reyes (2011)	8^a	Colombia	Community	88	Clinician judgment/Alzheimer’s	15-19^b	74.8 (9.2)
Suttanon et al. (2013)	19/21	Australia	Community	63	Clinician judgment/Alzheimer’s	20.9 (4.7)	83.4 (5.1)
Lowery et al. (2014)	67/64	United Kingdom	Community	57	Clinician judgment/any dementia	16.3 (7.4)	79.0 (6.8)
Yu et al. (2013)	11 /21^a	United States	Community	63	Clinician judgment/Alzheimer’s	21.8 (5.1)	81.4 (3.6)
Frederiksen, Sobol, Beyer, Hasselbalch, and Waldemar (2014)	8/7^a	Denmark	Community	75	NINCDS-ADRDA/Alzheimer’s	24.3 (4.4)	71.9 (5.4)
Aguiar, Monteiro, Feres, Gomes, and Melo (2014)	22/18	Brazil	Out-patient	74	Clinician judgment/Alzheimer’s	20.1 (4.5)	78.6 (8.4)
Barnes et al. (2015)	7/5	United States	Community	82	Clinician judgment/any dementia	61.8 (15.8)^c	85.7 (5.6)
Hoffmann et al. (2015)	107/93	Denmark	Out-patient	44	NINCDS-ADRDA/Alzheimer’s	24.0 (3.6)	70.5 (7.4)
Telenius, Engedal, and Bergland (2015)	87/83	Norway	Nursing homes	74	CDR-Scale/any dementia	15.7 (5.0)	86.7 (7.4)

Note. MMSE = Mini Mental State Examination; NINCDS-ADRDA = National Institute of Neurological and Communicative Disorders and Stroke and the Alzheimer’s Disease and Related Disorders Association; CDR= Clinical Dementia Rating.

Preintervention/postintervention.

Only the range was provided.

3MS (Modified MMSE).

Table 4 shows the risk of bias assessment profile for the controlled trials combined. The strength of the identified studies includes low levels of attrition in general and adequate accounting for loss to follow-up where they occurred. The only exception is in the study led by Suttanon (Suttanon et al., 2013) where significant losses to follow-up were reported (Table 4). Standardized procedures, including assessor blinding, were used for outcome assessments in all studies. Weakness of the studies mostly includes difficulty with implementation of patients and research personnel (double) blinding. Also, there were important differences between the intervention and control groups in factors such as self perception of health status in the study by Conradsson and colleagues (2010) and mini-mental state scores in Steinberg et al. (2009). Considered together, these limitations contribute to a combined moderate risk of bias in the controlled trials providing relevant data (National Institute for Health and Care Excellence, 2012).

Table 4.

Assessments by Two Independent Raters to Evaluate the Risk of Bias in RCTs of Exercise for QoL in Dementia.

Studies	Selection			Ascertainment		Attrition		Detection			Overall risk of bias = moderate
Studies	A	B	C	A	B	A	B	A	B	C	Average points	Verdict
Suttanon et al. (2013)	2	2	0	2	1	0	0	2	2	2	1.3	Moderate
Lowery et al. (2014)	2	0	1	2	1	2	2	1	2	2	1.5	Moderate
Steinberg, Leoutsakos, Podewils, and Lyketsos (2009)	2	2	0	2	0	1	2	1	2	2	1.4	Moderate
Conradsson, Littbrand, Lindelof, Gustafson, and Rosendahl (2010)	2	2	0	2	1	0	2	2	1	2	1.4	Moderate
Aguiar, Monteiro, Feres, Gomes, and Melo (2014)	2	1	2	2	1	2	2	2	2	2	1.8	Low
Barnes et al. (2015)	0	0	2	2	0	2	2	1	2	2	1.3	Moderate
Hoffmann et al. (2015)	2	1	2	2	1	0	2	2	2	2	1.6	Low
Telenius, Engedal, and Bergland (2015)	1	1	2	2	1	2	2	1	2	1	1.5	Moderate

Note. RCT = randomized controlled trial; QoL-AD = Quality of Life-Alzheimer’s Disease; Selection bias: A = Randomization; B = Allocation concealment; C = Group comparability; Ascertainment bias: A = whether cointervention; B = Personnel/patient blinding; Attrition bias: A = whether follow-up was comparable; B = Whether significant loss to follow-up; Detection bias: A = Adequate length of follow-up; B = Reliable outcome assessment measure; C = Assessor blinding. 2 points = Low risk, 1 point = Unclear/unknown risk, 0 point = Significant risk, Overall: <1 = high risk, 1-1.5 = moderate,>1.5 = low risk.

Quantitative Effect of Exercise Intervention on QoL

Figure 2 is a forest plot showing the total and subscores for participants in the exercise and control groups in six RCTs meeting criteria for quantitative synthesis (i.e., provided postintervention QoL scores for both participants and control groups). The study by Steinberg et al. (2009) did not provide appropriate data for participants not receiving exercise interventions in reporting coefficients of treatment effects overtime. The 95% confidence intervals for the postintervention QoL scores together with their quantitative summary are also presented.

Figure 2.

Forest plot showing RCTs of exercise interventions for QoL in dementia.

There was no significant difference in the QoL of participants receiving exercise interventions compared with controls in five out of the six studies. Overall, the effect of exercise interventions on QoL in PwD was small and nonsignificant (Z = 1.2, p = .23), with results comprising all six RCTs indicating high statistical heterogeneity (Tau² = 0.4, p < .01, I² = 92%; Figure 2). Statistical heterogeneity in the RCTs included in the model was found to be due to the effect size outlier produced by the study led by Conradsson (Conradsson et al., 2010). In adjusted analysis excluding the outlier, the overall effect size of exercise in improving QoL in PwD was further reduced (Z = 0.7, p = .47) with results indicating absence of heterogeneity (Tau² = 0.0, p = .73, I² = 0%).

The results of subgroup analyses are presented in Figure 3. Although not statistically significant, evidence synthesized from two studies indicate that aerobic exercises may produce larger effect on QoL in PwD compared with nonaerobic or combined exercises (Figure 3). In the same figure, caregivers had better overall outcome in terms of QoL after exercise interventions for the PwD (Z = 1.7, p = .10; heterogeneity p = .18, I² = 45%).

Figure 3.

Forest plot showing subgroup analyses of the effect of exercise interventions on QoL in dementia.

Discussion

The current systematic review found very little evidence supporting exercise interventions for improving QoL in PwD at this time. This limits the ability to draw firm conclusions as to the direction of the overall effect of this category of psychosocial-based interventions for QoL in the population with the disease. Although not statistically significant, evidence synthesized from meta-analysis of two RCTs comprising 331 participants indicate that aerobic exercises may potentially produce larger effect on QoL in PwD compared with nonaerobic or combined exercises. In all, it would appear that caregivers benefit more, in terms of QoL, after exercise interventions for the PwD.

There are no previous meta-analyses of exercise interventions for QoL in PwD on which the results of the present systematic review may be compared. A previous Cochrane review investigating the effect of exercise interventions on several outcomes in PwD (Forbes et al., 2013) could not examine QoL because the authors did not identify studies with appropriate data at the time of investigation. According to the results of our systematic literature search, Conradsson et al. (2010) and Steinberg et al. (2009) would have been the only two published RCTs investigating the question as to whether exercise interventions improved QoL in PwD at the time of conducting the Cochrane review led by Forbes (Forbes et al., 2013). The study by Steinberg et al. (2009) did not publish relevant data for meaningful quantitative synthesis while Conradsson et al. (2010) used a proxy measure, the Philadelphia Geriatric Center Morale Scale, thought to measure a construct qualitatively similar to QoL in the general population. Such generic measures do not adequately measure conceptually derived domains of QoL relevant to the population with dementia (Joseph, Becker, Elwick, & Silburn, 2012); thus, for example, the inclusion of the study by Conradsson et al. (2010) in our unadjusted meta-analysis produced results with high statistical heterogeneity.

We note that only a handful of good quality trials examining exercise techniques for improving QoL in dementia were found in the present study. For example, about 50% of trials identified were nonrandomized, and most of these nonrandomized trials were preintervention postintervention tests. Only two (Aguiar et al., 2014; Hoffmann et al., 2015) out of seven RCTs could be assessed as having a low risk of bias. The RCTs were mostly pilot studies with small sample sizes, and in none was QoL investigated as a primary outcome. Rather, the construct was investigated as part of a group of several secondary outcomes. In this way, none of the studies identified was specifically powered to detect statistically significant difference in QoL between those receiving exercise interventions and the controls. Therefore, well designed RCTs are still required to clarify whether exercise interventions result in meaningful overall improvements in QoL in PwD. In reviewing the existing evidence, a number of theoretical and methodological issues have emerged that are of potential relevance for the design of such future trials.

Ascertainment of QoL in dementia: Exercise procedures are typically designed to improve strength, flexibility and physical functioning (Pahor et al., 2014). However, many of the techniques have also been successfully used to improve cognitive functioning (Chang, Nien, Tsai, & Etnier, 2010) as well as overall health and well-being (Irwin, Olmstead, & Motivala, 2008; Pahor et al., 2014) in the general older population. There is an assumption that the success of exercise interventions in improving QoL in the general older population may be transferable to populations with neurodegenerative diseases, including the dementias (WHO, 2010). However, QoL in PwD is a difficult to measure phenomenon. It is yet unclear whether it is best to measure QoL in this population according to the judgment of a proxy, or that of the person whose QoL is being ascertained. In the present review, some of the identified studies have relied on proxy rating of QoL (Gallego et al., 2011; Lowery et al., 2014; Steinberg et al., 2009; Telenius et al., 2015; Williams & Tappen, 2007) while the others have used direct rating made by the PwD. As discussed in preceding sections of this review, proxy ratings of patients’ QoL are known to be influenced by, for example, the rater’s own level of burden (Sands et al., 2004) or state of emotional well-being (Conde-Sala et al., 2009). As such, it is often observed that better caregivers’ ratings of patients’ QoL are influenced by lower levels of dependency in PwD (Hoe et al., 2006). On the contrary, cognitive impairment may affect the interpretation of self report of QoL in persons with more severe dementia. There is thus the possibility that these factors may limit the measured effect of interventions, including exercise programs, on QoL in PwD. Due to these challenges, an approach that includes the combination of proxy and self-reports, as well as a physical functioning examination reports to generate a summary score may provide the most comprehensive information about QoL in PwD in future trials.

Type, frequency, and duration of exercise interventions and QoL in dementia: Some recommendations from the WHO appear to suggest that between 150 and 300 min of moderate intensity exercises per week, including aerobic and strength building components, may be of greater overall benefit for both the general population of older adults and those with neurodegerative diseases, including the dementias (WHO, 2010). Aspects of this recommendations, such as the inclusion of aerobic components, is supported by evidence from individual trials (Burgener, Yang, Gilbert, & Marsh-Yant, 2008; Wu et al., 2015) and systematic reviews of such trials (Groot et al., 2015) as leading to overall improvements in both cognitive and noncognitive symptoms in PwD. Although contrary to the WHO recommendations, the meta-analytic study led by Groot (Groot et al., 2015) found that lower intensity exercises of between 40 and 45 min per week appear to be associated with greater overall quantitative benefits for cognitive functioning, as an example, in PwD compared with higher intensity regimen. The differences between the WHO guidelines and the systematic review (Groot et al., 2015) may indicate that certain types of exercise strategies benefit some outcomes of dementia more than others. Studies included in the present investigation were heterogeneous in the type, frequency, and intensity of exercise interventions delivered. For instance, only one of the trials included in the present meta-analysis investigated exercise program lasting up to 150 min per week. We thus did not find a consistent pattern of intensity of exercises among trials reporting significant improvement in postintervention QoL scores. Out of four studies using aerobic exercise techniques, only one preintervention and postintervention test (Yu et al., 2013) found significant improvement in postintervention QoL scores. However, our subgroup meta-analysis comprising two RCTs including 331 participants appeared to indicate that aerobic exercises may potentially produce larger effect on QoL in PwD compared with nonaerobic or combined exercises.

Exercise, subtypes of dementia, and QoL: The interpretation of findings from some animal studies (Belarbi et al., 2011; Leem et al., 2009; Yuede et al., 2009) would suggest that exercise potentially reduces aggregation of defective proteins that have been associated with subtypes of dementia. For example, amyloid plaques in Alzheimer’s type dementia (Yuede et al., 2009), and Tau in various subtypes of dementia (Belarbi et al., 2011; Leem et al., 2009). Given these findings, there is a small possibility that exercise may be effective in improving specific outcomes, for example, QoL, in certain types of dementia, but not in others. Our subgroup meta-analysis comprising three studies including 280 persons with clinically diagnosed Azheimer’s-type dementia did not give an indication of better outcome in terms of QoL in these participants (Z = 0.3, p = .79; heterogeneity tests p = .83, I² = 0%). It is yet possible that patients with other types of dementia, for example, those due to accumulation of vascular risk factors may benefit from exercises by impeding additional accumulations of disease causative factors.

Strength and Limitations

It is important to highlight the limitations of the present systematic review. The interpretation of our results is limited by the small number of good quality trials examining exercise techniques for improving QoL in PwD. Many of the studies meeting criteria for syntheses were pilot studies with small sample sizes. Nevertheless, the selected trials have included a total of 903 PwD. One RCT (Steinberg et al., 2009) was excluded from meta-analyses as the required data for participants not receiving interventions were not reported. We made one unsuccessful contact with the authors of that study to provide additional data on their control population.

Despite these limitations, there are several areas of strength in the present study. First, there are no previous systematic reviews and meta-analyses of the effect of exercise-based interventions on QoL in PwD, thus justifying the present synthesis of currently available evidence as to whether exercise-based interventions are effective for improving QoL in PwD. This study is important and timely to guide clinicians and policy advisors in the planning of rehabilitation and long term care in PwD. Second, our search strategy has focused on the largest repositories of biomedical literature, with additional strategy to cover recent citations that might not have been included in the Medline, EMBASE, CINAHL, and PsycINFO databases. Furthermore, manual searches of the references of the appraised articles were also undertaken. On the contrary, gray literature, which may be a valuable source of materials dealing with the specific review question, has not been considered. Third, to reduce heterogeneity, we conducted an adjusted analysis by controlling for study effect outlier. Additional subgroup meta-analyses based on clinical subtype of dementia (i.e., Alzheimer’s disease) and exercise therapy (whether aerobic, nonaerobic, or combined) were also undertaken.

Conclusion

Even though there is strong theoretical support for possible benefit of exercise for improving QoL in PwD, empirical evidence from properly designed trials is yet inconclusive due to small number of studies showing inconsistent results. Future RCTs should consider specific subtypes of dementia. This may help to better tailor specific exercise programs for well-defined clinical cohorts of PwDs.

Footnotes

Acknowledgements

The authors acknowledge support from the Department of Old-Age Psychiatry, Institute of Psychiatry, Psychology and Neuroscience, King’s College London.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Aguiar

Monteiro

Feres

Gomes

Melo

(2014). Rivastigmine transdermal patch and physical exercises for Alzheimer’s disease: A randomized clinical trial. Current Alzheimer Research, 11, 532-537.

Alzheimer’s Society, Dementia UK. (2014). London: Alzheimer’s Society, 2014. Retrieved from: http://www.alzheimers.org.uk/dementia2014

Ballard

Hanney

M. L.

Theodoulou

Douglas

McShane

Kossakowski

, . . . DART-AD Investigators. (2009). The dementia antipsychotic withdrawal trial (DART-AD): Long-term follow-up of a randomised placebo-controlled trial. The Lancet Neurology, 8, 151-157. doi:10.1016/S1474-4422(08)70295-3

Barnes

D. E.

Mehling

Beristianos

Yaffe

Skultety

Chesney

M. A.

(2015). Preventing loss of independence through exercise (PLIE): A pilot clinical trial in older adults with dementia. PLoS ONE, 10(2), e0113367. doi:10.1371/journal.pone.0113367

Belarbi

Burnouf

Fernandez-Gomez

F. J.

Laurent

Lestavel

Figeac

. . . Blum

(2011). Beneficial effects of exercise in a transgenic mouse model of Alzheimer’s disease-like Tau pathology. Neurobiology of Disease, 43, 486-494. doi:10.1016/j.nbd.2011.04.022

Brod

Stewart

A. L.

Sands

Walton

(1999). Conceptualization and measurement of quality of life in dementia: The Dementia Quality of Life Instrument (DQoL). Gerontologist, 39, 25-35.

Burg

J. M.

Michalak

(2011). The healthy quality of mindful breathing: Associations with rumination and depression. Cognitive Therapy and Research, 35, 179-185.

Burgener

S. C.

Yang

Gilbert

Marsh-Yant

(2008). The effects of a multimodal intervention on outcomes of persons with early-stage dementia. American Journal of Alzheimer’s Disease & Other Dementias, 23, 382-394. doi:10.1177/1533317508317527

Chang

Y. K.

Nien

Y. H.

Tsai

C. L.

Etnier

J. L.

(2010). Physical activity and cognition in older adults: The potential of Tai Chi Chuan. Journal of Aging and Physical Activity, 18, 451-472.

10.

The Cochrane Collaboration. (2014). Review Manager (RevMan) (Version 5.3.) [Computer program]. Copenhagen, Denmark: The Nordic Cochrane Centre, The Cochrane Collaboration.

11.

Conde-Sala

J. L.

Garre-Olmo

Turro-Garriga

Lopez-Pousa

Vilalta-Franch

(2009). Factors related to perceived quality of life in patients with Alzheimer’s disease: The patient’s perception compared with that of caregivers. International Journal of Geriatric Psychiatry, 24, 585-594. doi:10.1002/gps.2161

12.

Conradsson

Littbrand

Lindelof

Gustafson

Rosendahl

(2010). Effects of a high-intensity functional exercise programme on depressive symptoms and psychological well-being among older people living in residential care facilities: A cluster-randomized controlled trial. Aging & Mental Health, 14, 565-576. doi:10.1080/13607860903483078

13.

Edvardsson

Fetherstonhaugh

Nay

(2010). Promoting a continuation of self and normality: Person-centred care as described by people with dementia, their family members and aged care staff. Journal of Clinical Nursing, 19, 2611-2618. doi:10.1111/j.1365-2702.2009.03143.x

14.

Forbes

Thiessen

E. J.

Blake

C. M.

Forbes

S. C.

Forbes

(2013). Exercise programs for people with dementia. Cochrane Database of Systematic Review, 12, CD006489.

15.

Frederiksen

K. S.

Sobol

Beyer

Hasselbalch

Waldemar

(2014). Moderate-to-high intensity aerobic exercise in patients with mild to moderate Alzheimer’s disease: A pilot study. International Journal of Geriatric Psychiatry, 29, 1242-1248. doi:10.1002/gps.4096

16.

Gallego

Alexey

Ma Clara

Lina

Reyes

(2011). Effects of Hatha-Yoga program on a small group with Alzheimer’s disease. Journal of Yoga & Physical Therapy, 1(3), 1000104, doi:10.4172/2157-7595.1000104

17.

Groot

Hooghiemstra

A. M.

Raijmakers

P. G.

van Berckel

B. N.

Scheltens

Scherder

E. J.

. . . Ossenkoppele

(2015). The effect of physical activity on cognitive function in patients with dementia: A meta-analysis of randomized control trials. Ageing Research Reviews, 25, 13-23. doi:10.1016/j.arr.2015.11.005

18.

Hoe

Hancock

Livingston

Orrell

(2006). Quality of life of people with dementia in residential care homes. British Journal of Psychiatry, 188, 460-464. doi:10.1192/bjp.bp.104.007658

19.

Hoffmann

Sobol

N. A.

Frederiksen

K. S.

Beyer

Vogel

Vestergaard

. . . Hasselbalch

S. G.

(2015). Moderate-to-high intensity physical exercise in patients with Alzheimer’s disease: A randomized controlled trial. Journal of Alzheimer’s Disease, 50, 443-453. doi:10.3233/JAD-150817

20.

Irwin

M. R.

Olmstead

Motivala

S. J.

(2008). Improving sleep quality in older adults with moderate sleep complaints: A randomized controlled trial of Tai Chi Chih. Sleep, 31, 1001-1008.

21.

Joseph

Becker

Elwick

Silburn

(2012). Adult Carers Quality of Life Questionnaire (AC-QoL): Development of an evidence-based tool. Mental Health Review Journal, 17, 57-69.

22.

Kitwood

(1990). The dialectics of dementia: With particular reference to Alzheimer’s disease. Ageing and Society, 10, 177-196.

23.

Kitwood

Bredin

(1992). Towards a theory of dementia care: Personhood and well-being. Ageing & Society, 12, 269-287.

24.

Leem

Y. H.

Lim

H. J.

Shim

S. B.

Cho

J. Y.

Kim

B. S.

Han

P. L.

(2009). Repression of tau hyperphosphorylation by chronic endurance exercise in aged transgenic mouse model of tauopathies. Journal of Neuroscience Research, 87, 2561-2570. doi:10.1002/jnr.22075

25.

Litchke

L. G.

Hodges

J. S.

Jan

Reardon

R. F.

(2012). Benefits of chair yoga for persons with mild to Severe Alzheimer’s disease. Activities, Adaptation & Aging, 36, 317-328.

26.

Lowery

Cerga-Pashoja

Iliffe

Thune-Boyle

Griffin

Lee

. . . Warner

(2014). The effect of exercise on behavioural and psychological symptoms of dementia: The EVIDEM-E randomised controlled clinical trial. International Journal of Geriatric Psychiatry, 29, 819-827. doi:10.1002/gps.4062

27.

McKhann

Drachman

Folstein

Katzman

Price

Stadlan

E. M.

(1984). Clinical diagnosis of Alzheimer’s disease: Report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimer’s Disease. Neurology, 34, 939-944.

28.

Moher

Liberati

Tetzlaff

Altman

D. G.

Group

(2009). Preferred reporting items for systematic reviews and meta-analyses: The PRISMA Statement. Open Medicine, 3, e123-e130.

29.

National Institute for Health and Care Excellence. (2012). The guidelines manual: Process and method guide. Available from www.nice.org.uk

30.

National Institute of Health and Clinical Excellence/Social Care Institute for Excellence. (2011). Dementia: Supporting people with dementia and their carers in health and social care. Retrieved from http://guidance.nice.org.uk/CG42

31.

Pahor

Guralnik

J. M.

Ambrosius

W. T.

Blair

Bonds

D. E.

Church

T. S.

. . . Williamson

J. D.

(2014). Effect of structured physical activity on prevention of major mobility disability in older adults: The LIFE study randomized clinical trial. Journal of the American Medical Association, 311, 2387-2396. doi:10.1001/jama.2014.5616

32.

Pitkala

Savikko

Poysti

Strandberg

Laakkonen

M. L.

(2013). Efficacy of physical exercise intervention on mobility and physical functioning in older people with dementia: A systematic review. Experimental Gerontology, 48, 85-93. doi:10.1016/j.exger.2012.08.008

33.

Ready

R. E.

Ott

B. R.

Grace

(2004). Patient versus informant perspectives of Quality of Life in Mild Cognitive Impairment and Alzheimer’s disease. International Journal of Geriatric Psychiatry, 19, 256-265. doi:10.1002/gps.1075

34.

Rogers

S. D.

(2005). Resistance training for adults with Alzheimer’s disease and related dementias: Feasibility of Program implementation, appropriateness of participant engagement, and effect on physical performance and quality of life (Unpublished doctoral dissertation). Virginia Polytechnic Institute and State University, Blacksburg. Retrieved from https://theses.lib.vt.edu/theses/available/etd-04282005015026/unrestricted/SDRsubmission043005.pdf

35.

Sands

L. P.

Ferreira

Stewart

A. L.

Brod

Yaffe

(2004). What explains differences between dementia patients’ and their caregivers’ ratings of patients’ quality of life? American Journal of Geriatric Psychiatry, 12, 272-280.

36.

Santana-Sosa

Barriopedro

M. I.

Lopez-Mojares

L. M.

Perez

Lucia

(2008). Exercise training is beneficial for Alzheimer’s patients. International Journal of Sports Medicine, 29, 845-850. doi:10.1055/s-2008-1038432

37.

Steinberg

Leoutsakos

J. M.

Podewils

L. J.

Lyketsos

C. G.

(2009). Evaluation of a home-based exercise program in the treatment of Alzheimer’s disease: The Maximizing Independence in Dementia (MIND) study. International Journal of Geriatric Psychiatry, 24, 680-685. doi:10.1002/gps.2175

38.

Suttanon

Hill

K. D.

Said

C. M.

Williams

S. B.

Byrne

K. N.

LoGiudice

Lautenschlager

N. T.

Dodd

K. J.

(2013). Feasibility, safety and preliminary evidence of the effectiveness of a home-based exercise programme for older people with Alzheimer’s disease: A pilot randomized controlled trial. Clinical Rehabilitation, 27, 427-438. doi:10.1177/0269215512460877

39.

Telenius

E. W.

Engedal

Bergland

(2015). Effect of a high-intensity exercise program on physical function and mental health in nursing home residents with dementia: An assessor blinded randomized controlled trial. PLoS ONE, 10(5), e0126102. doi:10.1371/journal.pone.0126102

40.

Thorgrimsen

Selwood

Spector

Royan

de Madariaga Lopez

Woods

R. T.

Orrell

(2003). Whose quality of life is it anyway? The validity and reliability of the Quality of Life-Alzheimer’s Disease (QoL-AD) scale. Alzheimer Disease & Associated Disorders, 17, 201-208.

41.

Williams

C. L.

Tappen

R. M.

(2007). Effect of exercise on mood in nursing home residents with Alzheimer’s disease. American Journal of Alzheimer’s Disease & Other Dementias, 22, 389-397. doi:10.1177/1533317507305588

42.

World Health Organization. (2010). Global recommendations on physical activity for health 65 years and above. Retrieved From http://www.who.int/dietphysicalactivity/factsheet_recommendations/en/

43.

Barnes

D. E.

Ackerman

S. L.

Lee

Chesney

Mehling

W. E.

(2015). Preventing Loss of Independence through Exercise (PLIE): Qualitative analysis of a clinical trial in older adults with dementia. Aging & Mental Health, 19, 353-362. doi:10.1080/13607863.2014.935290

44.

Nelson

N. W.

Savik

Wyman

J. F.

Dyske

Bronas

U. G.

(2013). Affecting Cognition and Quality of Life via Aerobic Exercise in Alzheimer’s disease. Western Journal of Nursing Research, 35, 24-38.

45.

Yuede

C. M.

Zimmerman

S. D.

Dong

Kling

M. J.

Bero

A. W.

Holtzman

D. M.

. . . Csernansky

J. G.

(2009). Effects of voluntary and forced exercise on plaque deposition, hippocampal volume, and behavior in the Tg2576 mouse model of Alzheimer’s disease. Neurobiology of Disease, 35, 426-432. doi:10.1016/j.nbd.2009.06.002

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.05 MB