Effects of Physical Exercise on the Incidence of Delirium and Cognitive Function in Acutely Hospitalized Older Adults: A Systematic Review with Meta-Analysis

Abstract

Background:

Acute care hospitalization increases the likelihood of developing cognitive impairment and delirium in older adults.

Objective:

To summarize evidence about the effectiveness of exercise and physical rehabilitation interventions on the incidence of delirium and cognitive impairment in acutely hospitalized older patients.

Methods:

Relevant articles were systematically searched (PubMed, Web of Science, and CINHAL databases) until 26 August 2021. Randomized and nonrandomized controlled trials of in-hospital physical exercise interventions and rehabilitation programs compared to usual care performed for older patients (> 65 years) hospitalized for an acute medical condition were selected. The primary endpoints were changes in the incidence of delirium and cognition during acute hospitalization. The secondary endpoints included functional independence, psychological measures, well-being status, length of hospital stay, transfer after discharge, fall occurrence, hospital readmissions, and mortality rate. The endpoints were evaluated at different time points (at admission, at discharge, and after discharge).

Results:

Eleven studies from 8 trials (n = 3,646) were included. The methodological quality of the studies was mostly high. None of the studies reported any adverse events related to the intervention. Early rehabilitation improved cognitive function at 3 months postdischarge (Hedge’s g = 0.33, 95% confidence interval [CI] 0.19 to 0.46, p < 0.001). No between-group differences were found for incident delirium and cognitive impairment during hospitalization (all p > 0.05).

Conclusion:

In-hospital physical exercise and early rehabilitation programs seem to be safe and effective interventions for enhancing cognitive function after discharge in older patients hospitalized for an acute medical condition. However, no potential benefits were obtained over usual hospital care for the incidence of delirium.

Keywords

Physical exercise rehabilitation cognitive impairment delirium hospitalized older adults

INTRODUCTION

Acute medical illness and subsequent hospitalization are major events that usually dramatically change older patients’ life trajectories. Hospitalization for medical illness has become a sentinel and highly morbid event leading to long-term disability in older individuals [1], even when the acute disease that causes hospital admission is successfully treated. In addition to the functional decline that has been widely investigated [2], acute care hospitalization increases the likelihood of developing cognitive impairment and delirium (i.e., acute confusional state) in older adults [3, 4]. However, the pathophysiological mechanisms of hospital-acquired cognitive decline and delirium remain poorly understood in the literature [4]. Previous evidence has suggested that neurotransmitter disturbances, especially acetylcholine deficiency and dopamine excess, could be one main reason for the development of delirium [5, 6]. Additionally, hospitalization per se seems to contribute to neurocognitive brain changes [7] by promoting systemic inflammation that has been related to smaller brain volume [8], blood–brain barrier dysfunction [9], and changes in cerebral metabolism [10].

The development of delirium and both functional and cognitive decline are highly prevalent during hospitalization in older populations and are independently associated with multiple adverse outcomes, including higher health care utilization, increased length of hospital stay, readmissions, institutionalization, and death [11 –13]. Thus, the focus of health care professionals has recently been placed on the development of preventive strategies for avoiding or reducing the adverse effects associated with hospitalization, including functional and cognitive impairments, in acutely hospitalized older adults [14].

Low-mobility and prolonged bed rest episodes are common occurrences during hospitalization [15]. Acute hospitalized older patients, including those who can walk independently, spend most of their hospital stay lying in bed (i.e., an average of 20 out of every 24 hours spent in bed) [16]. Reduced in-hospital physical activity levels seems to play a key role in the disabling process of patients, and it is an independent predictor of poor hospital outcomes at discharge, specifically declines in activities of daily living (ADLs), new institutionalization, and mortality [15]. Physical exercise has been proposed as a safe and effective therapy for enhancing functional capacity in nonacute settings, including at the community level, in institutionalized older adults, or in patients hospitalized for rehabilitation purposes. Moreover, recent evidence has highlighted physical exercise as a cornerstone for counteracting the functional decline usually associated with hospitalization in acutely hospitalized older adults [17]. In addition to functional benefits, exercise interventions have a positive effect on brain health in older adults [18], and seem to also provide potential benefits on different cognitive domains (i.e., visuospatial skills, orientation, working memory, memory recall, executive function, and verbal fluency ability) in older patients admitted to an Acute Care for Elderly (ACE) unit [19]. However, the influence of exercise on the occurrence of incident delirium remains unclear in this population. Although exercise benefits on functional status are well established [2], the evidence is less consistent regarding the effects of exercise interventions on the incidence of delirium and cognitive impairment in hospitalized older medical patients. To the best of our knowledge, there is no meta-analytic evidence supporting the effectiveness of in-hospital exercise programs for the enhancement of cognitive function and the preventive effect of exercise on the incidence of delirium in older patients admitted for an acute medical condition.

Thus, the main aim of the present systematic review with meta-analysis was to summarize the evidence on the effectiveness of exercise intervention or early rehabilitation programs for acutely hospitalized older medical patients on the incidence of delirium and cognitive impairment. The secondary aims were to describe these exercise training programs and summarize clinical outcomes, including functional performance, mental health, well-being status, length of hospital stay, falls occurrence, transfer to long-term care institutions, hospital readmissions, and mortality, to assess the effectiveness of the interventions.

METHODS

Search strategy

This systematic review was undertaken in accordance with the guidelines described by Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [20] and the method used was based on the minimum criteria established by the Cochrane Back Review Group (CBRG) [21]. It was registered in the PROSPERO database under registration number CRD42020191545.

Queries of the literature were performed based on the electronic databases PubMed, Web of Science and CINAHL (until 26th August 2021), using a combination of key words and controlled vocabulary (Supplementary Material). The search strategy was designed for PubMed and adapted to the other databases. We supplemented the search using complementary databases, such as Google Scholar. Studies reported in languages other than English were not explored.

Selection criteria

Trials that investigated the effects of exercise interventions or early rehabilitation programs on the incidence of delirium and cognitive function in acutely hospitalized older patients (> 65 years) were selected. Randomized with parallel and cross-over design and non-randomized controlled trials were included. Case-reports, case-series, single-case studies, and dissertations were excluded.

Those articles in which the effects of physical exercise or physical rehabilitation as part of a multidomain intervention program were assessed at least one outcome related to cognition or the incidence of delirium were included in the review. Intervention groups were focused on physical exercise or rehabilitation (i.e., physical therapy) programs and had a control group of patients who mainly received usual-care. At least 10 patients per intervention group were required for inclusion. We excluded studies if participants had been admitted in the hospital due to orthopedic surgery, knee/hip replacement, falls, stroke, psychiatric, or neurological disorders as well as those conducted in a post-acute setting (e.g., rehabilitation units). Studies specifically focused on cognitive training were also excluded from this meta-analytic study. In order to isolate the effects of in-hospital physical exercise or rehabilitation program, in those studies that a follow-up period was included (i.e., post-discharge) the exercise intervention should be finished at discharge.

Data extraction

Peer-review process was conducted using the Sysrev tool to organize and facilitate the data extraction. Two authors (MLSA and CC-L) independently screened the titles and abstracts of potentially eligible studies identified by the search strategy. If necessary, a third researcher (AG-H) was consulted. If one abstract did not provide enough information for evaluation based on the inclusion and exclusion criteria, full articles were retrieved for a full text assessment.

The following relevant data was extracted from the included studies: country and year, study design, number of participants within each group, participant's characteristics, methods, intervention protocol description, main outcomes, measurement methods, and main results. When the aforementioned data was missing, we contacted the corresponding author to clarify any uncertainty or to request additional data.

Data collection process

For each study, the following data were extracted 1) first author’s last name and year of publication; 2) sample size, participant’s characteristics and mean age; 3) exercise or rehabilitation protocol characteristics (exercise modality, intensity, frequency and duration) and control intervention; 4) endpoints; and 5) main results.

Endpoints

Primary endpoints were cognitive function and/or incidence of delirium. Secondary endpoints were the following: functional independence, psychological measures, length of hospital stay, transfer after discharge, falls occurrence, hospital readmissions and mortality rate. Authors were contacted to provide missing data related to primary endpoints or to clarify if data were duplicated in multiple publications.

Quality assessment

The methodological quality of the randomized controlled trials (RCT) was assessed with the PEDro scale. Two authors (MLSA and CC-L) independently scored each study, and disagreements were resolved through discussion with a third author (NM-V). The PEDro scale uses a rating scale from 0 to 10 by counting the number of criteria satisfied [22]. Study quality was rated as poor (PEDro score≤3), fair (4–5) or high (> 5).

Statistical analysis

All analyses were conducted using the Der-Simonian-Laird random-effects inverse-variance model using STATA software (version 17; StataCorp, College Station, TX, USA). Data were pooled only if outcomes were reported in at least three studies. Changes in both parameters in the trials were calculated by subtracting the differences between the intervention and control groups (i.e., usual care) using the pooled standard deviation (SD) of change in both groups (Hedge’s g). If change score SDs were not available, they were calculated from 95% confidence intervals (CI) for either change outcome or exercise training effect differences, as well as pre- and post-SD values [23]. The odds ratio was also converted to Hedge's g according to their corresponding formula [24].

Heterogeneity across trials was calculated using the inconsistency index (I2). Small-study effects and publication biases were examined using the Doi plot and Luis Furuya-Kanamori (LFK) index. Both tests have been shown to be superior to the traditional funnel plot and Egger’s regression intercept test [25]. No asymmetry, minor asymmetry or major asymmetry was considered with values of –1, between –1 and –2, and < –2, respectively [25]. Each study was deleted from the model once to analyze the influence of each study on the overall results. No subgroup analysis was performed due to limited number of studies.

RESULTS

Study selection

A total of 11 studies from 8 trials were included in the meta-regression analysis. The PRISMA flow-diagram is shown in Fig. 1. The characteristics of the included studies are summarized in Tables 1 2. Three studies [19 , 27] were secondary analyses of other RCTs [17], so only the participants of the main analysis (n = 370) were selected and included in the meta-analysis. A total of 3,646 participants were eventually considered in the systematic review.

Fig. 1

PRISMA flow diagram of literature search and study selection.

Table 1

Characteristics of the studies which included a physical exercise program

Study	Study design	Sample demographics (n, sex, average age)	Exercise intervention	Endpoints	Main results
Martinez-Velilla et al. (2019) [17]	RCT	–EXP: n = 185 patients (100 female), 88 years –CT: n = 185 patients (109 female), 87 years	5–7 days/week, twice a day, 20–30 min/session. Morning sessions included low-moderate intensity resistance (squats rising from a chair, leg press, bilateral knee extension, and seated bench-press [2–3 sets, 8–10 repetitions, 30–60% 1-RM], balance, and walking exercises. Supervised. Evening sessions consisted of knee extension and flexion, hip abduction with light loads (0.5–1 kg anklets and handgrip ball) and walking exercises based on the exercise “Vivifrail” guide. Unsupervised.	On admission and discharge: Cognitive assessment: MMSE Delirium assessment: CAM Functional assessment: ADL function (Barthel Index), SPPB scale, handgrip strength. Quality of life: EQ-5D Depression/anxiety: GDS Clinical assessment: length of hospital stay, falls during hospitalization. On admission and 3-months post-discharge: Clinical assessment: Transfer after discharge, readmissions, and mortality rate.	On admission and discharge:
					–↑ MMSE
					–↑ ADL function, SPPB score, and handgrip strength
					–↑ EQ-5D
					–↑ GDS
					–No differences between groups in the remaining outcomes
Sáez de Asteasu et al. (2019) [19]	RCT	–EXP: n = 185 patients (100 female), 88 years –CT: n = 185 patients (109 female), 87 years	Same as above.	On admission and discharge:	On admission and discharge:
				Cognitive assessment: 6-meter verbal dual-task; 6-meter arithmetic dual-task; MMSE; TMT-A; verbal fluency test	–↑ verbal and arithmetic GVT, MMSE, TMT-A, and verbal fluency score.
Sáez de Asteasu et al. (2019) [26]	RCT	–EXP: n = 185 patients (100 female), 88 years –CT: n = 185 patients (109 female), 87 years	Same as above.	On admission and discharge:	On admission and discharge:
				Cognitive assessment: MMSE	–↑ Rs and ↓ NRs and ↓ ARs MMSE
				Functional assessment: SPPB, GVT, handgrip strength	–↑ Rs and ↓ NRs and ↓ ARs SPPB, GVT, handgrip strength
				Follow-up 1 year post-discharge:	–↑ mortality rate ARs in both CT and EXP groups
				Clinical assessment: mortality
Sáez de Asteasu et al. (2021) [27]	RCT	–EXP: n = 185 patients (100 female), 88 years –CT: n = 185 patients (109 female), 87 years	Same as above.	On admission and discharge:	–Between 26–32% of the changes on physical function (Δ SPPB) are based on cognitive changes (Δ MMSE and verbal fluency score, respectively)
				Cognitive assessment: MMSE, verbal fluency test
				Functional assessment: SPPB

ADL, activities of daily living; ARs, adverse responders; CAM, Confusion Assessment Method; CT, control group; EQ-5D, EuroQol-5 Dimension; EXP, experimental group; GDS, Geriatric Depression Scale; GVT, Gait Velocity Test; MMSE, Mini-Mental State Examination; NRs, non-responders; Rs, responders; RM, repetition maximum; SPPB, Short Physical Performance Battery; TMT-A, Trail Making Test Part A. Significance was established at p < 0.05.

Table 2

Characteristics of the studies that included physical rehabilitation combined with other treatments

Study	Study design	Sample demographics (n, sex, average age)	Intervention	Endpoints	Main results
Landefeld et al. (1995) [28]	RCT	–EXP: n = 327 patients (223 female), 80 years –CT: n = 324 patients (212 female), 80 years	Walking or standing 3 times a day. Daily walking to the exercise rooms and for meals. Weight support, resistance, and aerobic exercises. In addition to: ACE unit with special environmental measures to prevent cognitive impairment Patient-centered care Nursing plans to prevent functional decline Discharge planning Prevention of iatrogenic complications Daily visits.	On admission and discharge: Cognitive assessment: First 21-items of MMSE Functional assessment: ADL function (Katz Index), IADL function (Lawton-Brody), ability to walk Depression/Anxiety: GDS Clinical assessment: Length of hospital stay, mortality On admission and 3 months post-discharge: Functional assessment: ADL function (Katz Index), IADL function (Lawton-Brody) Clinical assessment: Readmissions, transfer to long-term care	On admission and discharge:
					–↑ ADL function
					–↓ GDS
					On admission and 3-months post-discharge:
					- ↓ long-term care
					–No differences between groups in the remaining outcomes
Inouye et al. (1999) [34]	Non-RCT	–EXP: n = 426 patients (259 female), 79 years –CT: n = 426 patients (259 female), 79 years	Early mobilization. In addition to: Orientation protocol Therapeutic-activities protocol Non-pharmacologic sleep protocol Sleep-enhancement protocol Vision protocol Hearing protocol Hydration protocol	On admission:	On admission and discharge:
				Cognitive assessment: MMSE, Digit Span Functional assessment: ADL function (Katz Index)	–↑ MMSE –↓ CAM –No differences between groups in the remaining outcomes
				On admission and discharge:
				Cognitive assessment: MMSE
				Delirium: CAM
				Functional assessment: ADL function (Katz)
				Clinical assessment: length of hospital stay, mortality
Asplund et al. (2000) [30]	RCT	–EXP: n = 190 patients (111 female), 80 years –CT: n = 223 patients (140 female), 81 years	Early rehabilitation. In addition to: ACE unit with one geriatrician, physical therapist and occupational therapist. The interdisciplinary team focused on early and intensive rehabilitation and discharge planning.	On admission and discharge: Clinical assessment: Length of hospital stay On admission and 3 months post-discharge: Cognitive assessment: MMSE Delirium: CAM Functional assessment: ADL function (Barthel Index) Well-being status: PGCM Clinical assessment: Readmissions, transfer after discharge, mortality	On admission and discharge:
					–↓ length of hospital stay
					On admission and 3 months post-discharge:
					–↑ MMSE
					–No differences in the remaining outcomes post-discharge
Pitkälä et al. (2006) [31]	RCT	- EXP: n = 87 patients (66 female), 83 years –CT: n = 87 patients (62 female), 83 years	Rehabilitation based on physiotherapy. No more details are reported. In addition to: Accurate recognition of delirium Comprehensive geriatric assessment Avoiding conventional neuroleptics and administering atypical antipsychotics Orientation Geriatric interventions (nutritional supplements, calcium + Vitamin D supplements, hip protectors) Cholinesterase inhibitors Early discharge planning	On admission: Cognitive assessment: MMSE, Digit Span Delirium: CAM, DSM-IV, Clinical Dementia Rating Scale On admission and discharge: Delirium: MDAS On admission and at 3 and 6-months post-discharge: Cognitive assessment: MMSE Functional assessment: ADL function (Barthel Index) Depression/Anxiety: GDS Follow up 1 year post-discharge: Clinical assessment: Length of hospital stay, transfer to long term-care, mortality	On admission and discharge:
					–↑ MDAS
					On admission and 6-months post-discharge:
					–↑ MMSE at 6 months post-discharge
					Follow-up 1 year post-discharge:
					–↑ Length of hospital stay in acute care EXP versus CT
					–No differences in the remaining outcomes
Jeffs et al. (2013) [32]	RCT	–EXP: n = 305 patients (168 female), 79 years –CT: n = 343 patients (172 female), 79 years	Progressive resistance training. 4 different programs based on functional status: bed, seated, standing, and rails. Gravity, body or light weights were used as resistance as appropriate. Resistance was increased whenever a participant could perform 10 repetitions at the previous level. In addition to: Mobilization Orientation program 20–30 min/session. Twice/daily including weekends.	On admission and discharge:	On admission and discharge:
				Delirium: CAM	–No differences were observed between groups during hospitalization
				Cognitive assessment: MMSE, drawing a clock face, Modified Blessed Dementia Rating Scale
				Functional assessment: ADL function (Modified Barthel Index), IADL function (Lawton-Brody)
				Clinical assessment: Length of hospital stay, transfer to long-term care
Ekerstad et al. (2017) [29]	RCT	–EXP: n = 206 patients (122 female), 85 years –CT: n = 202 patients (108 female), 85 years	Early rehabilitation strategy involving physicians, occupational therapists, physiotherapists, and nurses. In addition to: Structured, systematic interdisciplinary CGA and care by validated instruments focusing on somatic and mental health, medication review, social situation, and functional and activity ability. Early discharge planning	On admission and discharge: Clinical assessment: length of hospital stay	On admission and 3-months post-discharge: –↑ HUI-3 cognition –↑ HUI-3 total score –No differences in the remaining outcomes
				On admission and 3-months post-discharge:
				Cognitive assessment: HUI-3 cognition dimension
				Well-being status: HUI-3, EQ-VAS
				Clinical assessment: Readmissions, transfer to long-term care, mortality
Gorski et al. (2017) [35]	Non-RCT	EXP: n = 65 patients (40 female), 84 years CT: n = 65 patients (40 female), 84 years	Early mobilization	On admission and discharge:	On admission and discharge:
			In addition to:	Delirium: Not detail.	–No differences in the remaining outcomes
			Orientation protocol	Clinical assessment: Length of hospital stay, falls, mortality
			Psychological strategy
			Hydration protocol
			Nutrition protocol
			Vision protocol
			Hearing protocol
			Sleep-enhancement protocol

ADL, activities of daily living; CAM, Confusion Assessment Method; CT, control group; CGA, comprehensive geriatric assessment; DSM-IV: Diagnostic and Statistical Manual of Mental Disorders; EQ-VAS, EuroQoL –visual analog scale; EXP, experimental group; GDS, Geriatric Depression Scale; HUI-3, Health Utilities Index-3; IADL, instrumental activities of daily living; MDAS, Memorial Delirium Assessment Scale; MMSE, Mini-Mental State Examination; MNA, Mini Nutritional Assessment; PGCM, Philadelphia Geriatric Center Morale scale. Significance was established at p < 0.05.

Quality assessment and publication bias of the RCTs

The quality of the included RCTs was mostly high (median PEDro score = 7, range 4–8; Table 3). Two studies were categorized as having fair methodological quality [28, 29], and the remaining studies were considered to present high-quality methodology [17 , 30–32].

Table 3

Methodological quality of the included RCT

Methodological quality assessment of included studies—PEDro scale
Study	Year	1	2	3	4	5	6	7	8	9	10	11	Total score (-/10)	Score confirmed
Asplund et al. (2020) [30]	2000	1	1	1	1	0	0	0	1	0	1	1	6	Yes
Landefeld et al. (1995) [28]	1995	1	1	0	1	0	0	0	0	0	1	1	5	Yes
Martinez-Velilla et al. (2019) [17]	2019	1	1	0	1	0	0	1	0	1	1	1	6	Yes
Pitkälä et al. (2006) [31]	2006	1	1	1	1	0	0	0	1	1	1	1	7	Yes
Jeffs et al. (2013) [32]	2003	1	1	1	1	0	0	1	1	1	1	1	8	Yes
Ekerstad et al. (2017) [29]	2016	1	1	0	1	0	0	0	0	1	1	1	5	Yes

Scale of item score 0 = no/ unclear, 1 = yes, The PEDro scale criteria are: (1) specification of eligibility criteria, (2) random allocation, (3) concealed allocation, (4) prognostic similarity at baseline, (5) subject blinding, (6) therapist blinding, (7) assessor blinding, (8) greater than 85% follow up of at least one key outcome, (9) intention to treat analysis, (10) between group statistical comparison for at least one key outcome, (11) point estimates and measures of variability provided for at least one key outcome. Studies included in meta-analysis.

Participants’ characteristics

The included studies involved between 130 and 852 participants (median = 410) whose average age ranged between 79 and 88 years (Tables 1 2). All studies included both male and female older adults who were admitted to ACE units’ due to acute medical conditions (e.g., respiratory, cardiovascular, renal/urological, infectious diseases and digestive conditions). The average length of hospital stay was between 6 and 16 days (median = 7 days).

Intervention characteristics

The exercise multicomponent intervention [17 , 27] consisted of 2 sessions per day, lasted 15–30 min per session, and was performed 5–7 days per week. The morning sessions were supervised and included low-moderate intensity resistance training, and evening sessions consisted of walking and unsupervised resistance exercises with light loads based on the Vivifrail exercise guide [33] (Table 1). Most of the studies included rehabilitation programs, including early mobilization as part of a multidomain intervention that consisted of acute-care integrative care, occupational therapy, cognitive stimulation and pharmacological optimization [28–32 , 35] (Table 2).

Endpoints

Adverse events

Seven of the 11 included studies noted the incidence of adverse events related to the intervention (e.g., disease exacerbation, dizziness, falls, muscu-loskeletal injuries), and none reported any complications [17 , 35].

Cognitive function

Four studies [17 , 27] examined the effects of a multicomponent exercise program on cognition using the Mini-Mental State Examination (MMSE), Trail Making Test Part A, and verbal fluency test. Considering the studies that included physical therapy as part of a multidomain intervention, five studies used the MMSE to measure changes in global cognitive function after the intervention at discharge or postdischarge [28 , 34]. Additionally, Jeffs et al. (2013) [32] included drawing a clock face and the modified Dementia Rating Scale to complete the cognitive assessment, and Ekerstad et al. (2017) [29] used the cognitive dimension of the Health Utilities Index-3 (HUI-3) for monitoring cognitive changes at 3 months postdischarge. The findings indicate that the intervention significantly increased cognition at 3 months after hospital discharge compared with usual care (Hedge’s g = 0.33, 95% CI 0.19 to 0.46, p < 0.001, I2 = 0%). However, no differences were observed between the groups for cognitive function at hospital discharge (Hedge’s g = –0.20, 95% CI –0.47 to 0.07, p = 0.151, I2 = 31.2%) (Figs. 2 3).

Fig. 2

Forest plot showing the effect sizes (Hedge’s g) of in-hospital physical exercise and early rehabilitation interventions on the incidence of delirium and cognitive function at discharge and post-discharge.

Fig. 3

This figure summarizes evidence about the effectiveness of exercise and physical rehabilitation interventions on the incidence of delirium and cognition in acutely hospitalized older adults.

Incidence of delirium

Five studies used the Confusion Assessment Method test [17 , 34] to examine the effects of an intervention that included physical rehabilitation on delirium at discharge. Moreover, Pitkälä et al. (2006) [31] also included the Diagnostic and Statistical Manual of Mental Disorders and Clinical Dementia Rating Scale for delirium assessment at admission and the Memorial Delirium Assessment Scale for cognition changes during hospitalization. Gorski et al. (2017) [35] did not detail how the delirium evaluation was performed. Our results show that no differences were obtained between the groups for the incidence of delirium at discharge (Hedge’s g = –0.07, 95% CI –0.35 to 0.21, p = 0.623, I2 = 59.3%) (Figs. 2 3).

3.5.4 Functional independence and physical performance

The effectiveness of a structured exercise program on functional abilities assessed by the Barthel Index and physical performance using the Short Physical Performance Battery was examined by an RCT [17]. Regarding the studies that included physical rehabilitation as part of a multidomain intervention, five studies [28 , 34] analyzed the effects of the intervention on functional capacity at discharge or postdischarge. Two of those studies [28, 34] used the Katz Index for monitoring changes in ADLs during hospitalization, and three trials [30 –32] examined functional independence by the Barthel Index at discharge or at 3- and 6 months postdischarge. Additionally, two studies [28, 32] examined the effects of interventions on instrumental ADLs assessed by the Lawton-Brody scale during acute hospitalization. Exercise or physical therapy interventions seemed to provide potential benefits for functional capacity during hospitalization [17, 28].

Psychological measures

Three trials assessed intervention effects on well-being status using the Philadelphia Geriatric Center Morale Scale, HUI-3, EuroQol –visual analog and EuroQol 5-Dimension scale at discharge [17] or at 3 months postdischarge [29, 30]. Physical exercise could be considered an effective treatment for improving health-related quality of life during hospitalization, and physical rehabilitation, as part of multidomain intervention, seemed to improve the patients’ well-being status after discharge. Moreover, three studies examined the effects of physical exercise or physical therapy on depression and anxiety assessed by the Geriatric Depression Scale at hospital discharge [17, 28] and 3- and 6 months postdischarge [31]. Exercise and early rehabilitation might have potential benefits for mental disorders during hospitalization, but no significant gains were observed in the intervention group during follow-up after the hospital stay.

3.5.6 Length of hospital stay

Seven studies [17 , 35] examined the differences between groups in the length of hospital stay in the ACE unit, and another study [31] assessed the number of days spent in the hospital during the follow-up year. Only one trial [30] observed a beneficial effect of the intervention for reducing the number of hospitalization days. Furthermore, Pitkälä et al. (2006) [31] observed that patients in the intervention group spent more days in the ACE unit than older adults in the usual-care group during the follow-up year.

Transfer after discharge

Six trials [17 , 28–32] evaluated the effects of the in-hospital intervention on transfer to long-term care institutions after discharge. Only one study [28] observed differences between the groups in the prevalence of patient transfer to long-term care institutions at 3 months postdischarge, showing a decreased number of older patients in the intervention group.

Falls

Two trials [17, 35] examined the effectiveness of the intervention on reducing the incidence of falls during hospitalization in older patients admitted to an ACE unit, and no differences were observed between the groups in either study.

Readmissions to the hospital

Four studies [17 , 28–30] assessed the number of older patients who were readmitted to the hospital after the hospital stay, and physical rehabilitation did not seem to reduce the incidence of readmissions compared to the usual-care group.

Mortality

Three studies [28 , 35] examined the prevalence of in-hospital mortality, and five studies [17 , 29–31] assessed the mortality rate during the follow-up period after discharge in hospitalized older patients. Exercise and physical rehabilitation did not modify the vital trajectory of acutely hospitalized older adults during and after hospitalization. However, a secondary analysis of an RCT [26] showed that older patients who were adverse responders to the in-hospital exercise intervention and received usual hospital care had a higher mortality rate during the follow-up year than older adults who improved or maintained their functional capacity during the hospital stay.

DISCUSSION

Although our findings must be interpreted carefully due to the limited number of RCT assessed, inpatient physical rehabilitation programs seemed to be effective for improving patients’ cognitive function after hospitalization in older adults admitted for an acute medical condition compared to usual hospital care. In contrast, no differences were observed between the groups for the incidence of delirium. However, no meta-analytic evidence was obtained for functional capacity, psychological measures, well-being status, and other secondary endpoints, such as length of hospital stay, transfer after discharge, fall occurrence, readmissions, and mortality.

Acute hospitalization is frequently a sentinel event leading to short- and long-term disability in older medical patients [36]. In addition to functional decline, hospital stay is also associated with the development of cognitive impairment and an increased risk of dementia [3]. Moreover, previous research has suggested that cognitively impaired older patients are at an even higher likelihood of suffering hazards of hospitalization than patients with no cognitive decline [37]. Physical exercise has been proposed as a key treatment to prevent or reverse iatrogenic negative consequences of hospitalization in acutely hospitalized older patients. Indeed, recent meta-analytic evidence has demonstrated that in-hospital supervised exercise interventions appear to be safe and effective for enhancing functional independence and physical performance during and after hospitalization in older adults admitted to an ACE unit [2]. Although a unique RCT, which included a tailored supervised multicomponent exercise program, showed potential benefits on multiple cognitive domains, such as global cognitive function, executive function and verbal fluency [17 , 27], our results revealed that no differences were observed in cognitive changes between the exercise or multidomain interventions that considered early rehabilitation and usual hospital care in older patients during hospitalization. Furthermore, our finding that in-hospital physical rehabilitation interventions significantly enhance cognition after discharge (i.e., at 3 months postdischarge) over usual care in older patients admitted to an ACE unit is potentially relevant, given that cognitive functioning tends to decline substantially after hospitalization, even after controlling for acute illness severity and prehospital cognitive impairment in older adults [38]. Thus, this review highlights the importance of early rehabilitation protocols for avoiding the cognitive deterioration that usually occurs in this population after discharge.

Exercise benefits on the brain and cognition have been widely investigated in the literature [39, 40]. A growing number of randomized trials have examined whether physical exercise has a beneficial effect on perception and cognition in older adults, but further research is needed to determine which exercise modality is the best for obtaining cognitive gains and to explore physiological and neuromuscular changes to understand the relationship between exercise and cognition in older patients [41]. Although hospitalization per se seems to contribute to neurocognitive brain changes by promoting systemic inflammation [7, 8] and short-term exercise training appears to induce an anti-inflammatory response in acutely hospitalized older patients [42], no cognitive differences were obtained between the physical exercise intervention and usual care during hospitalization in this study. Furthermore, to the authors’ knowledge, this is the first meta-analysis to point toward the beneficial effects of inpatient physical exercise programs, including multidomain interventions that included physical rehabilitation, for improving cognitive function in acutely hospitalized older adults after hospital discharge.

In addition to cognitive function, the role of exercise and early rehabilitation on the occurrence of incident delirium was also examined in this review. In line with the cognitive findings obtained during hospitalization, exercise interventions do not appear to provide potential benefits on the incidence of delirium in older medical patients admitted to an ACE unit. Delirium is an acute change in cognition that complicates the stay of up to 56% of hospitalized older adults and promotes loss of functional independence, psychological distress, and sustained cognitive dysfunction, accelerating cognitive decline in patients with dementia [43 –46]. Currently, the best management strategies to treat and prevent delirium are multidomain interventions that focus on treating precipitating conditions, medication reviews, managing distress, mitigating complications, and maintaining engagement in environmental issues [47]. Because delirium is an independent predictor of persistent poor cognition and functional status during the year after hospitalization in the elderly population, future studies should investigate the physiological mechanisms underlying the syndrome and explore new strategies, including other in-hospital exercise interventions, to prevent the incidence of delirium in this population.

Some limitations of our review should be acknowledged. First, the limited number of available trials and the heterogeneity with respect to the exercise training protocols (e.g., mobility exercises, multicomponent interventions) and multidomain interventions (e.g., prepared hospital environment, social support, nutritional interventions) made it difficult to assess the inherent effects of physical exercise on the incidence of delirium and cognitive impairment in acutely hospitalized older adults. Non-randomized controlled trials were also included in the review, increasing the potential risk of bias of the results obtained. Additionally, the lack of details about the intervention protocols described (i.e., duration, frequency, intensity of the exercises) made it complicated to show robust evidence. Second, no subgroup analysis was performed due to the small number of studies included in the review. Third, the wide variability of cognitive tests used for measuring different cognitive domains, such as executive function, attention, and verbal fluency, could be one of the principal reasons for the large variations in the cognitive benefits between the trials. Finally, changes in the incidence of delirium and cognitive function were evaluated at different time points of time across trials (e.g., at admission, at discharge, and at 3 and 6 months after discharge), hindering us from drawing consistent conclusions about the effectiveness of in-hospital exercise interventions. It is important to note that all the abovementioned limitations were impossible for the authors of the present study to control.

Our study, nevertheless, has several strengths. We implemented a rigorous process to adhere to the recommended practices in systematic reviews, which included duplicate and independent procedures in all review stages (except meta-analyses) as well as taking extra care to avoid duplicating data generated by common study samples (unit-of-analysis error). To our knowledge, this is the first meta-analysis to analyze the effects of in-hospital exercise and physical rehabilitation interventions on incident delirium and cognitive impairment in older medical patients admitted to ACE units. Recent evidence has demonstrated that in-hospital exercise interventions appear to be safe and effective for improving functional independence and physical performance in this population [2], but no meta-analytic evidence has previously considered the role of exercise or early rehabilitation on cognitive function in acutely hospitalized older adults. Moreover, we established strict inclusion/exclusion criteria to focus on a specific segment of the population with medical conditions (i.e., acutely hospitalized older medical patients).

4.1 Conclusions

In-hospital physical exercise and early rehabilitation programs seem to be effective interventions for enhancing cognitive function after discharge in older patients hospitalized for an acute medical condition. However, no potential benefits were obtained over usual hospital care for the incidence of delirium. Future studies should explore the physiological mechanisms underlying cognitive changes during hospitalization and new physical exercise strategies (e.g., higher intensities of training) to revert delirium and prevent the cognitive impairment that frequently occurs in acutely hospitalized older adults.

Footnotes

DISCLOSURE STATEMENT

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Boyd

, Landefeld

, Counsell

, Palmer

, Fortinsky

, Kresevic

, Burant

, Covinsky

(2008) Recovery of activities of daily living in older adults after hospitalization for acute medical illness. J Am Geriatr Soc 56, 2171–2179.

Valenzuela

, Morales

, Castillo-García

, Mayordomo-Cava

, García-Hermoso

, Izquierdo

, Serra-Rexach

, Lucia

(2020) Effects of exercise interventions on the functional status of acutely hospitalised older adults: A systematic review and meta-analysis. Ageing Res Rev 61, 101076.

Ehlenbach

, Hough

, Crane

, Haneuse

, Carson

, Curtis

, Larson

(2010) Association between acute care and critical illness hospitalization and cognitive function in older adults. JAMA 303, 763–770.

Inouye

, Westendorp

RGJ

, Saczynski

(2014) Delirium in elderly people. Lancet 383, 911–922.

Young

, Inouye

(2007) Delirium in older people. BMJ 334, 842–846.

Marcantonio

(2018) Delirium in hospitalized older adults. N Engl J Med 378, 96–97.

Walker

, Gottesman

, Wu

, Knopman

, Mosley

Jr , Alonso

, Kucharska-Newton

, Brown

(2018) Association of hospitalization, critical illness, and infection with brain structure in older adults. J Am Geriatr Soc 66, 1919–1926.

Walker

, Hoogeveen

, Folsom

, Ballantyne

, Knopman

, Windham

, Jack

, Gottesman

(2017) Midlife systemic inflammatory markers are associated with late-life brain volume: The ARIC study. Neurology 89, 2262–2270.

Abbott

(2000) Inflammatory mediators and modulation of blood-brain barrier permeability. Cell Mol Neurobiol 20, 131–147.

10.

De Felice

, Ferreira

(2014) Inflammation, defective insulin signaling, and mitochondrial dysfunction as common molecular denominators connecting type 2 diabetes to Alzheimer disease. Diabetes 63, 2262–2272.

11.

Lucke

, van der Mast

, de Gelder

, Heim

, de Groot

, Mooijaart

, Blauw

(2018) The six-item cognitive impairment test is associated with adverse outcomes in acutely hospitalized older patients: A prospective cohort study. Dement Geriatr Cogn Disord Extra 8, 259–267.

12.

Witlox

, Eurelings

LSM

, de Jonghe

JFM

, Kalisvaart

, Eikelenboom

, van Gool

(2010) Delirium in elderly patients and the risk of postdischarge mortality, institutionalization, and dementia: A meta-analysis. JAMA 304, 443–451.

13.

Gill

, Allore

, Gahbauer

, Murphy

(2010) Change in disability after hospitalization or restricted activity in older persons. JAMA 304, 1919–1928.

14.

Covinsky

, Pierluissi

, Johnston

(2011) Hos-pitalization-associated disability: “She was probably able to ambulate, but I’m not sure.”. JAMA 306, 1782–1793.

15.

Brown

, Friedkin

, Inouye

(2004) Prevalence and outcomes of low mobility in hospitalized older patients. J Am Geriatr Soc 52, 1263–1270.

16.

Brown

, Redden

, Flood

, Allman

(2009) The underrecognized epidemic of low mobility during hospitalization of older adults. J Am Geriatr Soc 57, 1660–1665.

17.

Martínez-Velilla

, Casas-Herrero

, Zambom-Ferraresi

, Saez de Asteasu

, Lucia

, Galbete

, García-Baztan

, Alonso-Renedo

, González-Glaría

, Gonzalo-Lázaro

, Apezteguía Iraizoz

, Gutierrez-Valencia

, Rodríguez-Mañas

, Izquierdo

(2019) Effect of exercise intervention on functional decline in very elderly patients during acute hospitalization: A randomized clinical trial. JAMA Intern Med 179, 28–36.

18.

Izquierdo

, Merchant

, Morley

, Anker

, Aprahamian

, Arai

, Aubertin-Leheudre

, Bernabei

, Cadore

, Cesari

, Chen

, de Souto Barreto

, Duque

, Ferrucci

, Fielding

, García-Hermoso

, Gutiérrez-Robledo

, Harridge

SDR

, Kirk

, Kritchevsky

, Landi

, Lazarus

, Martin

, Marzetti

, Pahor

, Ramírez-Vélez

, Rodríguez-Mañas

, Rolland

, Ruiz

, Theou

, Villareal

, Waters

, Won Won

, Woo

, Vellas

, Fiatarone Singh

(2021) International Exercise Recommendations in Older Adults (ICFSR): Expert consensus guidelines. J Nutr Health Aging 25, 824–853.

19.

Sáez de Asteasu

, Martínez-Velilla

, Zambom-Ferraresi

, Casas-Herrero

, Cadore

, Galbete

, Izquierdo

(2019) Assessing the impact of physical exercise on cognitive function in older medical patients during acute hospitalization: Secondary analysis of a randomized trial. PLoS Med 16, e1002852.

20.

Liberati

, Altman

, Tetzlaff

, Mulrow

, GØtzsche

, Ioannidis

, Clarke

, Devereaux

, Kleijnen

, Moher

(2009) The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: Explanation and elaboration. PLoS Med 6, e1000100.

21.

Furlan

, Pennick

, Bombardier

, van Tulder

, Editorial Board, Cochrane Back Review Group (2009) 2009 updated method guidelines for systematic reviews in the Cochrane Back Review Group. Spine 34, 1929–1941.

22.

Maher

, Sherrington

, Herbert

, Moseley

, Elkins

(2003) Reliability of the PEDro scale for rating quality of randomized controlled trials. Phys Ther 83, 713–721.

23.

Follmann

, Elliott

, Suh

, Cutler

(1992) Variance imputation for overviews of clinical trials with continuous response. J Clin Epidemiol 45, 769–773.

24.

Borenstein

, Hedges

, Higgins

JPT

, Rothstein

(2021) Introduction to meta-analysis. JohnWiley and Sons.

25.

Furuya-Kanamori

, Barendregt

, Borenstein

SAR

(2018) A new improved graphical and quantitative method for detecting bias in meta-analysis. Int J Evid Based Healthc 16, 195–203.

26.

Sáez de Asteasu

, Martínez-Velilla

, Zambom-Ferraresi

, Casas-Herrero

, Cadore

, Ramírez-Vélez

, Izquierdo

(2019) Inter-individual variability in response to exercise intervention or usual care in hospitalized older adults. J Cachexia Sarcopenia Muscle 10, 1266–1275.

27.

Sáez de Asteasu

, Martínez-Velilla

, Zambom-Ferraresi

, Ramírez-Vélez

, García-Hermoso

, Izqui-erdo

(2021) Cognitive function improvements mediate exercise intervention effects on physical performance in acutely hospitalized older adults. J Am Med Dir Assoc 22, 787–791.

28.

Landefeld

, Palmer

, Kresevic

, Fortinsky

, Kowal

(1995) A randomized trial of care in a hospital medical unit especially designed to improve the functional outcomes of acutely ill older patients. N Engl J Med 332, 1338–1344.

29.

Ekerstad

, Karlson

, Dahlin Ivanoff

, Landahl

, Andersson

, Heintz

, Husberg

, Alwin

(2017) Is the acute care of frail elderly patients in a comprehensive geriatric assessment unit superior to conventional acute medical care? Clin Interv Aging 12, 1–9.

30.

Asplund

, Gustafson

, Jacobsson

, Bucht

, Wahlin

, Peterson

, Blom

, Ängquist

(2000) Geriatric-based versus general wards for older acute medical patients: A randomized comparison of outcomes and use of resources. J Am Geriatr Soc 48, 1381–1388.

31.

Pitkälä

, Laurila

, Strandberg

, Tilvis

(2006) Multicomponent geriatric intervention for elderly inpatients with delirium: A randomized, controlled trial. J Gerontol A Biol Sci Med Sci 61, 176–181.

32.

Jeffs

, Berlowitz

, Grant

, Lawlor

, Graco

, de Morton

, Savige

, Lim

(2013) An enhanced exercise and cognitive programme does not appear to reduce incident delirium in hospitalised patients: A randomised controlled trial. BMJ Open 3, e002569.

33.

Izquierdo

, Casas-Herrero

, Zambom-Ferraresi

, Martinez-Velilla

, Alonso-Bouzón

, Rodríguez-Mañas

(2017) Multicomponent physical exercise program VIV-IFRAIL. Available at: http://www.vivifrail.com/resources/. Accessed April 3, 2021

34.

Inouye

, Bogardus

, Charpentier

, Leo-Summers

, Acampora

, Holford

, Cooney

Jr (1999) A multicomponent intervention to prevent delirium in hospitalized older patients. N Engl J Med 340, 669–676.

35.

Gorski

, Piotrowicz

, Rewiuk

, Halicka

, Kalwak

, Rybak

, Grodzicki

(2017) Nonpharmacological interventions targeted at delirium risk factors, delivered by trained volunteers (medical and psychology students), reduced need for antipsychotic medications and the length of hospital stay in aged patients admitted to an acute internal medicine ward: Pilot study. BioMed Res Int 2017, 1297164.

36.

Gill

, Allore

, Holford

, Guo

(2004) Hospitalization, restricted activity, and the development of disability among older persons. JAMA 292, 2115–2124.

37.

Provencher

, Sirois

, Ouellet

, Camden

, Neveu

, Allain-Boulé

, Emond

(2015) Decline in activities of daily living after a visit to a Canadian emergency department for minor injuries in independent older adults: Are frail older adults with cognitive impairment at greater risk? J Am Geriatr Soc 63, 860–868.

38.

Wilson

, Hebert

, Scherr

, Dong

, Leurgens

, Evans

(2012) Cognitive decline after hospitalization in a community population of older persons. Neurology 78, 950–956.

39.

Hillman

, Erickson

, Kramer

(2008) Be smart, exercise your heart: Exercise effects on brain and cognition. Nat Rev Neurosci 9, 58–65.

40.

Voss

, Nagamatsu

, Liu-Ambrose

, Kramer

(2011) Exercise, brain, and cognition across the life span. J Appl Physiol (1985) 111, 1505–1513.

41.

Sáez de Asteasu

, Martínez-Velilla

, Zambom-Ferraresi

, Casas-Herrero

, Izquierdo

(2017) Role of physical exercise on cognitive function in healthy older adults: A systematic review of randomized clinical trials. Ageing Res Rev 37, 117–134.

42.

Ramírez-Vélez

, Martínez-Velilla

, Fernández-Irigoyen

, Santamaría

, Palomino-Echeverría

, Izquierdo

(2020) Influence of short-term training on functional capacity and (anti-)inflammatory immune signalling in acute hospitalization. J Cachexia Sarcopenia Muscle 11, 1154–1157.

43.

Inouye

(1998) Delirium in hospitalized older patients: Recognition and risk factors. J Geriatr Psychiatry Neurol 11, 118–125; discussion 157-158.

44.

O’Malley

, Leonard

, Meagher

, O’Keeffe

(2008) The delirium experience: A review. J Psychosom Res 65, 223–228.

45.

O’Keeffe

, Lavan

(1997) The prognostic significance of delirium in older hospital patients. J Am Geriatr Soc 45, 174–178.

46.

Khachaturian

, Hayden

, Devlin

, Fleisher

, Lock

, Cunningham

, Oh

, Fong

, Fick

, Marcantonio

, Iyengar

, Rockwood

, Kuchel

, Eckenhoff

, MacLullich

AMJ

, Jones

, Davis

, D'Antonio

, Fargo

, Albert

, Williamson

, Ling

, Weiss

, Karlawish

, Petersen

, Blazer

, Khachaturian

, Inouye

(2020) International drive to illuminate delirium: A developing public health blueprint for action. Alzheimers Dement 16, 711–725.

47.

Wilson

, Mart

, Cunningham

, Shehabi

, Girard

, MacLullich

AMJ

, Slooter

AJC

, Ely

(2020) Delirium. Nat Rev Dis Primer 6, 90.