go4cognition: Combined Physiological and Cognitive Intervention in Mild Cognitive Impairment

Abstract

Background:

While cognitive interventions in mild cognitive impairment (MCI) show positive effects on cognitive performance, physical activity appear to slow down cognitive decline, suggesting a relationship between both factors. However, previous combined programs that have shown significant improvement in cognitive function in MCI have typically trained cognition and physical functioning separately.

Objective:

This project aimed at evaluating two group interventions combining the stimulation of physical and cognitive domains in individuals with MCI: Simultaneous stimulation of physical and cognitive skills in comparison to a standardized training, which stimulates cognitive and physical functions separately.

Methods:

The study was designed as a randomized controlled trial. The first group was trained on the SpeedCourt® system while the second group completed the standardized Fitfor100 program. Training was completed by a total of 39 subjects with diagnosed MCI as determined by the CERAD (SpeedCourt®: 24 subjects, Fitfor100:15 individuals).

Results:

There were significant improvements of physical factors (e.g., hand strength and balance) in both groups. Improvement in the CERAD total score allowed for a post interventional classification of all participants into non-MCI and MCI. This effect persisted over a period of three months. Both forms of intervention were found to be effective in improving various cognitive functions which persisted for a period of three months.

Conclusion:

Both evaluated non-pharmacological, multicomponent interventions, which combined physical and cognitive training in a social setting showed improvement of cognitive functions leading to a persistent classification of former MCI patients in non-MCI patients.

Keywords

Cognitive training mild cognitive impairment multi- component intervention physiological intervention simultaneous stimulation of physical and cognitive abilities

INTRODUCTION

Aging is often associated with a decline in cognitive functions such as memory, attention, visuospatial abilities, and executive functions [1]. Mild cognitive impairment (MCI) is defined as the intermediate state between normal cognitive aging and dementia [2] and describes a status of a subjective cognitive decline in combination with objective memory impairment beyond normal age-related changes, whereas daily activities are not directly affected [3]. In a recent review, Petersen et al. (2018) suggested an MCI prevalence of 6.7% among individuals aged 60 to 64 years and 25.2% in those aged between 80 to 84 years [4]. Individuals with MCI have a 5–10% higher risk of developing dementia [5, 6]. Findings from a recent meta-analysis of 13 clinical studies involving a total sample of 4,301 individuals suggested an annual conversion rate from MCI to dementia of 9.6%, and a conversion rate of 39.2% over the natural observation period [7]. Proactive programs, which aim to delay the consequences of dementia and improve the wellbeing of people with MCI and their caregivers, are needed. Due to demographic changes, the incidence of dementia is expected to increase significantly in the future, this phenomenon will significantly affect the health care system worldwide, especially with regard to the medical, social, and economic costs of older people and their families [8]. Therefore, interventions that aim to improve cognitive functions are of particular interest. Modifiable lifestyle factors such as alcohol consumption, diet, physical exercise, and mental activity might decrease the risk of developing dementia [9 –13]. Lifelong socially and cognitively stimulating experiences, such as education and leisure, seem to be associated with a reduced risk of cognitive decline and dementia [14 –18]. Cognitive interventions designed for older adults with MCI showed benefit in global cognitive functioning, memory (short-term, working, long-term, verbal, and visual), attention, and psychomotor learning [19 –24]. Cognitive training (CT) refers to the repeated practice on a series of standard tasks targeting specific cognitive functions such as attention, memory, and problem solving, as well as the learning and practice of mnemonic strategies and skills [19, 25]. Training based improvement or maintenance of cognitive abilities [25 –27] are the objectives of CT [28 –32]. CT stimulates neuroprotective mechanisms based on synaptic plasticity [33]. Clinical trials on the effect of memory training in individuals with MCI have shown increased activation and connectivity in widespread frontal, temporal, and occipital brain regions [24] including the hippocampus as a key memory structure [34]. Physical activity serves to protect brain functions and can even lead to neurogenesis over the entire lifespan of adults [35]. Exercises in older adults at risk of AD, due to the diagnosis of MCI, may promote stable cognitive function and increase brain volume [35]. However, Nyberg et al. (2021) could show that the level of education has no effect on the rate of cognitive and on brain volume decline [36]. Epidemiological data proposes an association between a lower risk for dementia, moderate exercise, and physical activity, such as walking [37]. Physically active individuals with MCI seem to show slower cognitive decline [6 , 38–49]. The relationship between cognitive and physical functions in individuals with MCI has been supported by several cross-sectional studies [50]. Due to the complexity of cognitive impairment, single intervention trials, which train a specific function, may be too simplistic [27]. Recent randomized controlled trials (RCT) evaluating the effect of multidomain intervention, combining cognitive and aerobic training sessions in older individuals with MCI, indicated significantly increased cognitive functions in combination with parahippocampal cerebral blood flow [51], as well as a decrease in pre- to post-training activity of the precuneus/posterior cingulate cortex in delta, theta, and beta bands [52]. In previous combined programs cognition and physique were usually trained separately. A simultaneous stimulation might additively effect cognition and save resources. The aim of the “go4cognition” research project is to evaluate two combined cognitive and physical approaches with different foci: A new intervention that simultaneously stimulates physical functions while training cognitive functions of the participants with MCI in comparison to a well standardized program, which trained physical and cognitive functions separately. Using the CERAD as a standard diagnostic tool with pre-defined cut-off values for dementia and MCI, we were interested in the direct and midterm effect of this intervention with respect to possible changes in these classifications. We were further interested in evaluating the impact of this classification on other cognitive domains.

METHODS

Participants

The current study included elderly subjects aged between 65 and 86 years with MCI as determined by a standard neuropsychological examination using the CERAD [53, 54] with a total score > 85.1 as cut-off. This cut-off total score was also used for dividing participants into MCI and non-MCI in further analysis. Exclusion criteria for the study were dementia (CERAD total score < 68.5), clinical evidence for depression (Beck Depression Inventory II > 20) [55, 56] as well as severe neurological, cardiological, or orthopedic problems barring or limiting participation to the aerobic physical training program at the time of enrollment. Table 1 illustrates the included sample characteristics which were matched for age, years of education, and CERAD total score. All participants gave written consent prior to participation. The study was conducted in accordance with the current version of the Declaration of Helsinki of the World Medical Association. It was approved by the Ethics Committee of the medical association of North Rhine-Westfalia (No. 2019006). Subjects have been recruited from the general population through newspaper articles or the network of senior centers. 40 individuals fulfilled the previously defined neuropsychological inclusion criteria of MCI. Table 2 illustrates the included sample and its cognitive status before intervention. The two groups did not differ in any of these parameters. Five participants declined participation before the start of the training or during the first training week and were therefore excluded. Furthermore, one participant had to be excluded due to cardiologic pathologies. Five participants could be included again in a cross-over like design after a “washout phase”. The training was completed by a total of n = 39 subjects (SpeedCourt®: 10 women and 14 men; Fitfor100:8 women and 7 men) between the ages of 65 and 86 (SpeedCourt®: M = 74.83, SD = 6.77; Fitfor100: M = 76.33, SD = 7.22). On average, the participants completed 13 years of education (SpeedCourt®: M = 13.21, SD = 4.03; Fitfor100: M = 12.47, SD = 3.77). Here, n = 24 subjects participated in the training of the intervention I group (SpeedCourt®) and n = 15 subjects in the intervention II (Fitfor100) group. 53.8% of the participants stated that they were regularly active in sports while 38.5% stated that they actively managed their daily lives themselves (e.g., shopping, gardening). 7.7% stated that they were rather inactive. Furthermore, 69.2% reported being cognitively active (sudoku, crossword puzzles, drama, active choir participation). 30.8% reported being inactive in this regard.

Table 1

Demographic data (mean, standard deviation, and range) and CERAD total score from both groups included in this study

	SpeedCourt®			Fitfor100
	M	SD	Range	M	SD	Range
Age	74.83	6.77	65.00–85.00	76.33	7.22	65.00–86.00
Years of education	13.21	4.03	8–18	12.47	3.77	8–18
CERAD total score	77.66	6.09	68.5–85.04	80.14	4.405	69.33–84.43

CERAD, Consortium to Establish a Registry for Alzheimer’s Disease.

Table 2

Cognitive status as assessed by standard neuropsychological tests of all participants before intervention represented as percentile ranks (mean, standard deviation, and range)

	M	SD	Range
RWT (lexical)	61.050	25.822	10.00–98.00
RWT (categorial)	41.560	17.399	10.00–83.00
RWT (alternate categorical)	41.280	29.699	2.00–95.00
RWT (alternate lexical)	34.769	25.822	1.00–98.00
VLMT (learning)	20.756	17.668	4.00–85.00
VLMT (interference)	25.205	25.517	4.00–96.00
VLMT (loss after interference)	30.667	26.871	4.00–96.00
VLMT (direct recall)	14.962	16.488	4.00–67.50
VLMT (delayed recall)	16.936	15.700	4.00–57.50
VLMT (loss)	24.330	19.891	4.00–90.00
ROCF	15.790	17.119	0.00–90.00
TAP (alertness - phasic alertness)	49,333	34.811	1.00–99.00
TAP (divided - omissions)	26.872	25.255	1.00–90.00
TAP (divided - commissions)	49.385	35.118	4.00–100.00
TAP (Go/NoGo - RT)	37.385	31.919	1.00–98.00
TAP (Go/NoGo - commissions)	30.256	22.446	1.00–63.00
TAP (flexibility - RT)	28.385	29.315	1.00–96.00
TAP (flexibility - commissions)	36.564	34.421	1.00–87.00

M, mean; ROCF, Rey Osterrieth complex figure; RT, reaction time; RWT, Regensburger Wortflüssigkeits-Test; VLMT, Verbal Learning and Memory Test; WMS, Wechsler Memory Scale - Fourth Edition; SD, standard deviation; TAP, Test of Attentional Performance.

Study design

To determine cognitive changes related to the training intervention, the study was conducted as a randomized controlled design [57]. Participant recruitment was performed using a two-phase scheme. A first screening phase was performed to select potentially eligible subjects from the general population. These potentially eligible participants underwent a second phase of clinical confirmation, which provided the cognitive and physiological status. The study design included screening measures, pre-intervention tests, the six-week intervention period, post-intervention tests, and a three-month follow-up evaluation of the cognitive status. Baseline measures took place one week to four weeks before the pre-intervention tests. The post-intervention tests were carried out within seven days after completion of the intervention.

Randomization

Subjects were randomly assigned either to the SpeedCourt® or to the Fitfor100 intervention group matched for gender in a 2:1 ratio due to some logistic problems, considering a computer-generated simple randomization algorithm. Randomization was performed after baseline assessment. The intervention was performed at the “Athletikum Rhein-Ruhr” located at the BG Klinikum Duisburg. The “Fitfor100” (Deutsches Institut für angewandte Sportgerontologie e.V.) program of the intervention II group was performed at the “Gute Hoffung” Oberhausen. Due to transport issues some participants had to be re-assigned to the intervention I or intervention II group, respectively.

MCI assessment

MCI assessment was performed using the CERAD, “Consortium to Establish a Registry for Alzheimer’s Disease” [53, 54]. The total score has been calculated by using the demographic correction regression formula by Chandler et al. (2005) (raw score –(–0.324 · age+0.897·education–2.858·gender)) [58]. The raw score was determined by adding the following test results: Verbal Fluency, Boston Naming Test, Learn word list, Construction exercises, Word list recognition, and Word list Discrimination. The demographically corrected CERAD total score has been shown to be accurate in differentiating independent samples of normal controls, MCI, and AD subjects (normal controls and MCI: AUC 0.823, MCI and AD: AUC 0.882) [58]. CERAD total scores showed a high test–retest reliability (r = 0.95) [58]. This assessment was performed three times: before the intervention, directly after the intervention, and three months after the intervention.

Assessment of physical parameters

All physical assessments were conducted at the sports-biomechanic laboratory “Athletikum-Rhein-Ruhr” at the BG Klinikum Duisburg, Clinic for Arthroscopic Surgery, Sports Traumatology and Sports medicine. The following parameters were assessed before and after training: As an indicator of the global strength capacities, hand grip strength of both hands was measured using the Jamar Jackson hand grip dynamometer. To assess postural control in a static and dynamic situation the Tandem Stand was used, as a marker for the risk of falling. The Timed Up and Go-test (TUG) measures the time that a participant needs to stand up from a chair with armrests, walk 3 m, turn, walk back, and sit down. The time (in seconds) needed between chair raise and sitting down is recorded. Additionally, stabilometric measurements were recorded on an instrumented treadmill (HP Cosmos Quasar, h/p/cosmos sports & medical GmbH, Nussdorf-Traunstein, Germany) equipped with a capacitance-based pressure distribution measurement platform (FDM-THQ 2i, zebris Medical GmbH, Weitnau, Germany). The test protocol also included one bipedal stance for 30s with open eyes (center of Pressure, open eyes (COPo)) followed by a self-selected resting period and a unipedal stance for at least 20s for each leg. Resting time between single leg stances was 30s. Stabilometric measurements derived from the COP signal were the average velocity of the COP (COP speed), path length of the COP over a 30-s period (COP length) and 95% confidence ellipse around the COP trajectory (COP area). This assessment was performed two times, before the intervention and after the intervention.

Assessment of cognitive functions

The computer-based German Test of Attentional Performance (TAP 2.3.1) [59] was used to evaluate different forms of attention performance (Alertness, Go/NoGo, Divided Attention, and Flexibility) before, directly after intervention, and at three-month follow-up. Visual-spatial memory was assessed using the Rey Osterrieth complex figure (ROCF) [60]. Verbal learning and memory capacity were investigated with the Verbal Learning and Memory Test (VLMT) [61]. To evaluate the capacity of verbal and visual short-term and working memory the subtests digit-span and block-design of the Wechsler Memory Scale - Fourth Edition [62] were used. Executive functions were further assessed by using the German version of a verbal fluency task (Regensburger Wortflüssigkeits-Test (RWT)) [63]. Additionally, the Beck Depression Inventory II (Cronbach’s α= 0.90) was used to assess possible symptoms of depression [55, 56]. This assessment was performed three times: before the intervention, directly after the intervention, and three months after finishing the intervention.

SpeedCourt® (Intervention group I)

Training sessions were conducted on the SpeedCourt® system (Global Speed, Hemsbach, Germany). The SpeedCourt® is an interactive training device that can be used in professional athletic training for agility-like, change of direction movements [64] and in rehabilitation settings [65]. It consists of a rectangular mat (5.5 m×5.5 m) fitted with contact plates that registers distance covered (meters), contact time, total session time (seconds) and contact errors. Testing and Training protocols can be visualized on a screen positioned in front of participants. Foot contact with the target field elicits auditory feedback as well as the presentation of the new target field (Fig. 1). Training was conducted over a period of six weeks, twice a week for approximately 30 min. While in general no time limit was set for the completion of the training session, training sessions did not exceed 1500 m to avoid physically overloading of the elderly participants. Participants were asked to wear comfortably clothes and firm footwear during test and training. They were positioned in the center field of the SpeedCourt® in front of the screen in a standing position. The screen depicted an image of the SpeedCourt® and assigned numbers to each contact field (1–8) in clockwise fashion. The participants should memorize this basic allocation of numbers to the fields for 60 s; these numbers were hidden during the subsequent test. The test protocol included two training and six testing levels each presenting a random sequence of numbers via the screen. Participants were asked to walk the sequence by triggering the contact field on the ground as fast as possible, returning to the court’s center field after each triggered field. Each participant had to complete the training sessions once prior the very first training session. During familiarization with the test procedure a sequence of maximum two numbers had to be memorized for each training trial. During the test protocol, each level consisted of ten different sequences of numbers that were displayed in random order until all numbers of that level were triggered by foot contact. Contacts per sequence amounted to 30 in test-level 1 and increased to 80 in level 6. To control physical exertion heart rate was monitored continuously during each session. Participants advanced to the next level if they correctly triggered at least 75% of the displayed fields, below a quota of 50%, participant fell back to the previous level.

Fig. 1

Illustration of the SpeedCourt^® including the marked fields and the screen, showing the sequence of fields that have to be visited.

Fitfor100 (Intervention group II)

The Fitfor100 group completed the Fitfor100 program (https://www.ff100.de/), a standardized and well evaluated program consisting of exercise training suitable for senior citizens and places cognitive demands that are closely relate to activities of everyday live on the participants. In contrast to the SpeedCourt® approach, physical and cognitive exercises are not performed simultaneously. The aim of the intervention, which takes place twice a week for 1 h in a group-setting, is to improve everyday skills as climbing stairs and getting up from a chair. Various coordination exercises train balance skills and help to prevent falls. Strengthening exercises with weights build the basis. The range of exercise helps older adults and people suffering from dementia to be able to carry out everyday activities independently for as long as possible and to counteract the dwindling strength.

Data processing and statistical analysis

All data were analyzed using the software Statistical Package for the Social Sciences (SPSS, version 25.0; IBM Corp., Armonk, NY, USA; https://www-01.ibm.com/software/de/stats24/). It has been shown that analysis of variance with repeating measures is very robust against the violation of the assumption of normally distributed data [66] Therefore, all analyses were performed using ANOVAs. To control for better comparability and to reduce redundancy and complexity of the SpeedCourt® data, we performed a normalization by computing the average of all values, with X_max and X_min presenting the maximum and minimum values. Normalization was than accomplished by using the formula: X_n = ((X–X_min)/(X_max–X_min)) · 100. To perform the subgroups analysis, repeated measurements ANOVA was conducted including the results of HP (high performer, participants reaching the highest level) and LP (low performer, participants not reaching the highest level). For all COP related parameters, boxplot graphs were used to identify outliers in a first step, which were excludes from further analysis. Repeated measurements ANOVA was used to identify differences between intervention I and intervention II group over time including the factors Group (SpeedCourt® versus Fitfor100) and Time (pre versus post). Distribution assumptions were assessed by Mauchly’s test for sphericity and Shapiro- Wilk-Test for normal distribution. In case of significant difference potential effect sizes were estimated by ω².

Neuropsychological assessment (pre-intervention, post-intervention, and three-month follow-up analysis)

For assessing improvements in cognitive performance within and between groups, a mixed analysis of variance (ANOVA) was accomplished including the factor Group (SpeedCourt® versus Fitfor100) and Time (pre-intervention, post-intervention, three-month follow-up). To guarantee the requirements of an ANOVA— normal distribution of data and sphericity— a Kolmogorov-Smirnov and a Mauchly’s test were conducted beforehand. In case of lacking sphericity, a Greenhouse-Geisser respective Huynd-Feldt correction was applied to adjust the degrees of freedom. Boxplots were created to detect deviating data. To adjust outcomes for multiple comparisons, a Bonferroni correction was applied. Significant results were further evaluated using a Bonferroni Post Hoc test.

MCI status analysis

For further analysis, participants were subdivided after the intervention into MCI and non-MCI based on the cognitive status as assessed by the CERAD total score. Cognitive and physical profiles of these groups were further analyzed to characterize the cognitive profile of the two groups, gaining further insight in possible underlying modulating processes.

RESULTS

Physical parameters

Results of the Timed Up and Go (TUG) and the Tandem Stand (TS) tests showed no impairment of functional mobility with TUG scores below one point (equaling measured time under ten seconds) for all participants in the first examination. Furthermore, participants showed no impairments in this study at the level of static posture control as measured by TS test. Regardless of group, all participants improved in TUG and TS from pre- to post-testing even if not reaching significance. Analysis of variance revealed a statistically significant main effect for the factor Time in TUG (F(1, 37) = 5.15, p = 0.029, η_p² = 0.0013, n = 39). However, neither a significant main effect for the factor Group (F(1,37) = 0.206, p = 0.635, η_p² = 0.005) nor for the interaction was found in TUG (F(1,37) = 0.002, p = 0.969). As analysis of variance are robust against violation data of the TS test were analyzed with this approach including the factors Time (pre versus post) and Group (SpeedCourt® versus Fitfor100). Results yield evidence for significant changes for the factor Time (F(1,37) = 7.28, p = 0.01). Neither the main effect for the factor Group (F 1, 37) = 0.06, p = 0.795) nor the interaction reached significance (F(1, 37) = 0.58, p = 0.45). Analysis of variance for the hand grip strength revealed a statistically significant main effect for the factor Time (F(1, 37) = 5.35, p = 0.026; η_p² = 0.006). Neither the interaction (F(1, 37) = 0.943; p = 0.338; η_p² = 0.001) nor the main effect for the factor Group (F(1,37) = 0.187; p = 0.67, η_p² = 0.005) reached significance. In general, all participants, regardless of group affiliation improved in handgrip strength from pre- to post-testing. COP measurements: Further ANOVAs were performed using the factors Time (pre versus post) and Group (SpeedCourt® versus Fitfor100) including the variables COP area (COP-a, [mm²]) COP length (COP-l, [mm]) and COP speed (COP-s, [mm/s]), both in an open-eyes condition and in a closed-eyes condition. However, only the parameter COP- Area (closed eyes) showed a statistical significance regarding the factor Time (F(1, 29) = 6.05, p = 0.020; η_p² = 0.031) and the Time×Group interaction (F(1, 29) = 5.37, p = 0.028; η_p² = 0.027). The interaction term was further evaluated using paired t-tests. Results could show that the COP-Area was reduced from pre-intervention to post-intervention only for the intervention I group (t(17) = –2.97, p = 0.009 (CG: t(12) = 0.19, p = 0.85). The factor Group did not reach significance (F(1,29) = 0.31, p = 0.86). All other analysis yielded no evidence for significant difference (see Table 3).

Table 3

Results of the analysis of variance including the factor Time (pre versus post) and Group (high performer versus low performer) for the COP-parameters distance and velocity, separated for the conditions eyes open and eyes closed

	Condition	Time	Group	Interaction
Distance	Eyes open	F(1,29) = 0.10, p = 0.75	F(1,28) = 0.63, p = 0.43	F(1,28) = 1.84, p = 0.18
	Eyes closed	F(1,29) = 3.34, p = 0.78	F(1,29) = 0.11, p = 0.73	F(1,28) = 0.99, p = 0.33
Velocity	Eyes open	F(1,28) = 2.40, p = 0.13	F(1,28) = 0.63, p = 0.43	F(1,28) = 1.83, p = 0.18
	Eyes closed	F(1,29) = 3.50, p = 0.07	F(1,29) = 0.93, p = 0.34	F(1,28) = 0.93, p = 0.34

High performer versus Low performer Sub-group analysis

The SpeedCourt® training intervention placed different physical demands on the participants. To evaluate possible dose-response effects of the physical part of the cognitive- physical training intervention of participants, they were grouped into high performer (HP) and low performer (LP) with HP including participants reaching the highest possible training level (80 contacts per level) during the course of the intervention and participants who did not met that criterion (LP). An ANOVA was performed including the factor Group (high performer versus low performer) and the factor Time (pre-intervention versus post-intervention). High performer showed a statistically significant improvement in cognitive performance determined by the CERAD total score, compared to low-performing participants (main effect factor Group) F(1, 22) = 4.915, p = 0.037, η_p² = 0.138. Results further provide evidence for a significant main effect for the factor Time F(1,22) = 20.962, p < 0.001, η_p² = 0.488. Additionally, a statistically significant interaction between Time and Group could be determined, i.e., F(1,22) = 4.940, p = 0.0337, η_p² = 0.183. The high performing group showed a significant improvement in performance F(1, 14) = 27.368, p < 0.001, η_p² = 0.662, while this was not evident for the low performing group F(1,8) = 2.852, p = 0.130, η_p² = 0.263. Regarding all measured SpeedCourt® parameters we could find a significantly improvement between both groups: HP and LP, p < 0.05. Paired t-tests revealed statistically significant improvement for the variables contact (t(1, 22) = –3.756, p = 0.001), distance (t(1,22) = –4.022, p < 0.001), duration (t(1,22) = –2.685, p = 0.014), percentage of correct movements (t(1,22) = –2.463, p = 0.022), and speed (t(1,22) = –2.956, p = 0.007), of those that participants in the high performer group as opposed to the low performer participants (LP).

Neuropsychological tests

All results of the statistical analysis using analysis of variance including the factor Group (SpeedCourt® versus Fitfor100) and Time (pre-intervention, post-intervention, and three-month follow-up) are summarized in Table 4. Significant effects of the factor Time, which lasted over three months, were found with respect to visual memory, as assessed by the ROCF F(2,66) = 4,484, p = 0.015, η_p² = 0.120, as well as in relation to verbal fluency (categorical: F(2, 68) = 46.575, p < 0.001, η_p² = 0.578, alternate categorical: F(2, 68) = 5, 840, p = 0.005, η_p² = 0.147, alternate lexical: F(2, 68) = 5059, p = 0.009, η_p² = 0.130), which was assessed using the RWT. Additionally, evaluating the effectiveness on response inhibition using the Go/NoGo-Task (TAP), a significant reduction of reaction time and number of commissions could be observed (reaction time: F(2,64) = 7.244, p = 0.001, η_p² = 0.185; commissions: F(2,64) = 7.235, p = 0.001, η_p² = 0.184; see Table 4). Interaction effects between group membership and time could not be found in any test. There were also no significant differences between the groups.

Table 4

Results of the Analysis of Variance including the factor Time (pre-intervention, post-intervention, three-month follow-up) and Group (SpeedCourt® versus Fitfor100)- for all included neuropsychological tests

	Time				group				time×group
Test	df	F	p	η _p ²	df	F	p	η _p ²	df	F	p	η _p ²
ROCF	2, 66	4.48	0.015	0.12	1, 33	0.06	0.807	0.00	2, 66	1.45	0.243	0.04
BDI	2, 68	0.61	0.549	0.02	1, 34	0.39	0.535	0.01	2, 68	0.22	0.801	0.01
RWT (lexical)	2, 68	2.29	0.109	0.06	1, 34	0.01	0.913	0.00	2, 68	0.54	0.588	0.02
RWT (categorical)	2, 68	46.58	<0.001	0.58	1, 34	0.29	0.595	0.01	2, 68	0.61	0.547	0.02
RWT (alternate categorical)	2, 68	5.84	0.005	0.15	1, 34	0.06	0.816	0.00	2, 68	1.22	0.301	0.04
RWT (alternate lexical)	2, 68	5.06	0.009	0.13	1, 34	0.06	0.816	0.00	2, 68	1.79	0.175	0.05
VLMT (learning)	2, 68	3.24	0.045	0.09	1, 34	0.01	0.907	0.00	2, 68	1.22	0.302	0.04
VLMT (interference)	2, 68	0.20	0.820	0.01	1, 34	0.12	0.729	0.00	2, 68	0.44	0.644	0.01
VLMT (loss after interference)	2, 68	2.83	0.066	0.08	1, 34	0.30	0.587	0.01	2, 68	0.03	0.966	0.00
VLMT (direct recall)	2, 68	1.17	0.315	0.03	1, 34	0.03	0.868	0.00	2, 68	0.41	0.662	0.01
VLMT (delayed recall)	2, 68	1.17	0.315	0.03	1, 34	0.03	0.868	0.00	2, 68	0.41	0.662	0.01
VLMT (loss)	2, 68	1.46	0.240	0.04	1, 34	0.45	0.508	0.01	2, 68	0.52	0.598	0.02
WMS (digit span)	2, 68	0.35	0.705	0.10	1, 34	0.03	0.868	0.00	2, 68	2.05	0.137	0.06
WMS (digit span - WM)	2, 68	1.96	0.150	0.06	1, 34	0.22	0.646	0.01	2, 68	1.40	0.253	0.04
WMS (block span)	2, 64	1.96	0.150	0.06	1, 32	0.83	0.370	0.03	2, 64	0.13	0.878	0.00
WMS (block span - WM)	2, 64	1.83	0.169	0.05	1, 32	3.07	0.089	0.09	2, 64	0.61	0.546	0.02
TAP (alertness - RT)	2, 64	1.12	0.334	0.03	1, 32	0.58	0.454	0.02	2, 64	0.44	0.588	0.65
TAP (divided - omissions)	2, 64	0.50	0.611	0.02	1, 32	2.31	0.139	0.07	2, 64	1.55	0.220	0.05
TAP (divided - commissions)	2, 64	1.92	0.156	0.06	1, 32	0.01	0.945	0.00	2, 64	0.51	0.605	0.02
TAP (Go/NoGo - RT)	2, 64	7.24	0.001	0.19	1, 32	1.05	0.312	0.03	2, 64	1.34	0.269	0.04
TAP (Go/NoGo - commissions)	2, 64	7.24	0.001	0.18	1, 32	2.01	0.166	0.06	2, 64	1.415	0.250	0.042
TAP (flexibility - RT)	2, 64	0.88	0.419	0.03	1, 32	0.14	0.707	0.00	2, 64	1.23	0.298	0.04
TAP (flexibility - commissions)	2, 64	0.58	0.565	0.02	1, 32	0.20	0.657	0.01	2, 64	1.28	0.286	0.04

ROCF, Rey Osterrieth complex figure; BDI, Beck Depression Inventory II; RT, reaction time; RWT, Regensburger Wortflüssigkeits-Test; TAP, Test of Attentional Performance; VLMT, Verbal Learning and Memory Test; WMS, Wechsler Memory Scale - Fourth Edition; WM, working memory.

CERAD

CERAD total score was assessed at three time points (pre-intervention, post-intervention, and three months after intervention). An ANOVA including the factor Group (SpeedCourt® versus Fitfor100) and Time (before intervention, after intervention, and three-month follow-up) was conducted. Results yield evidence for a significant main effect for the factor Time F(2, 68) = 27.27, p < 0.001, η_p² = 0.45 with regard to the cognitive status, as determined by the CERAD total score. The cognitive performance of both groups improved significantly and persisted also for a period of three months. There was no significant main effect of the factor Group, suggesting that intervention was not superior in one of the two groups (F(1, 34) = 1.14, p = 0.294, η_p² = 0.032). A statistically significant interaction between the factor Time and Group could not be found F(2, 68) = 0.082, p = 0.921, η_p² = 0.002.

Distribution of participants CERAD total score over time

The CERAD was applied at three time points: pre-intervention, post intervention, and three months after intervention. According to the cut-off value (85.1), participants were divided into MCI (below 85.2) and non-MCI (above 85.1), provided that this cut-off value was exceeded or not reached at both post-training (post, follow-up) time points, so that the two groups (MCI, non-MCI) included the same individuals at both post-training (post, three-month follow-up) time points. The distribution of the participants with respect to this criterion is illustrated in Fig. 2. A crosstabs analysis was used to examine association between categorical variables like cognitive status measured by CERAD total score over time (pre-intervention, post-intervention, and three-month follow-up). No expected cell frequencies were below 5. Results show a significant relation between the CERAD total score and time, χ²(2) = 33.805, p < 0.001, φ= 0.478.

Fig. 2

Distribution of participants based on the CERAD cut-off (81.5 separating MCI and non-MCI) for pre-intervention, post-intervention, and 3-month follow-up time points.

Further analysis based on the distinction into MCI and non-MCI were performed comparing the development in the TUG between those who achieved a CERAD total score > 85.1 (non-MCI) and those who showed a lower CERAD total score (MCI), the non-MCI group showed higher initial speed in combination with a continuous improvement, while the performance of the MCI group rather stagnant. A significant correlation between participants being categorized as non-MCI and the results of the TUG directly after intervention could be found (r = –0.418, p = 0.009).

The ability to learn new, non- associated verbal information was assessed using the VLMT. A significant correlation between the learning rate in trial one-five in the VLMT and the CERAD total score could be shown (after intervention: r = 0.607, p < 0.001); three-month follow-up: r = 0.569, p < 0.001) for all participants at both post-intervention time points, but not at baseline (r = 0.158, p = 0.337). For participants being classified as MCI a significant correlation could be shown directly after intervention (r = 0.771, p < 0.001) and at the follow-up examination after three months (r = 0.504, p < 0.001). For participants being classified as non-MCI a significant correlation between the learning rate in trial one-five in the VLMT and the CERAD total score could be shown at the follow-up examination (r = 0.567, p = 0.022). Again, the non-MCI group showed better baseline performance and improved constantly, while the MCI group’s performance improved only slightly. However, this performance difference at baseline did not meet significance. Further analysis using paired t-tests yielded evidence, for a significant performance improvement when comparing the performance before training with the performance three months after training t(15) = –3.570, p = 0.003 as well as if comparing the performance after training with the performance three months after the training t(15) = –2.155, p = 0.084 for the non-MCI group. This was not evident for the MCI group (comparing the performance before training with the performance three months after training: t(19) = 0.298, p = 0.769, comparing the performance after training with the performance three months after the training: t(19) = 0.025, p = 0.980) (see Fig. 3).

Fig. 3

Performance distribution of VLMT learning rate (trial 1-5) grouped by CERAD cut-off (81.5 to discriminate MCI from non-MCI) for pre-intervention, post-intervention, and 3-month follow-up time points.

DISCUSSION

The present study aimed at investigating the effect of two different training approaches for subjects diagnosed with MCI according to the cut of value defined by the CERAD. Results show an improvement for the physical performance parameter for all participants of the SpeedCourt® group. This reflects a training effect over time, but it also emphasizes that training leads to an improvement in these parameters in older subjects due to a regular training. There were significant physical improvements in terms of hand strength, TUG, and TS performance in both groups after training, suggesting that a secure gait and stance can be assumed, which has a positive effect on independence in every day live activities. The combination with the cognitive improvements was already shown in other studies and underlines the connection of cognitive and physical abilities, as well as the positive effect of a combined intervention [50]. However, it has to be mentioned that all participants started at a high physical level. Bamidis et al. (2015) indicate a possible dose-response effect regarding the role of training load both physical and cognitive [67]. To differentiate possible dosage effects our cohort was split into a high performer (HP) and low performer (LP) group with HP consisting of persons reaching the highest possible training level (80 contacts per level) during the course of the intervention and persons not meeting that criterion (LP). We hypothesize that higher walking speed and the longer distance covered might have led to an increased physical training load emphasizing the effect of the benefit of the cognitive part of the training intervention. This is supported by the higher CERAD Scores in the HP group which were outside the range of MCI. Additionally, current results yield evidence for an improvement in the CERAD total score leading to a post interventional classification of participants into non-MCI and MCI. Interestingly, this classification persisted up to three months after intervention and suggests that participants improved their cognitive status in a way that they did not meet the criteria of MCI afterward anymore. It should further be mentioned that this has been achieved after an intervention of only six weeks regardless of the program that was applied. Both forms of intervention were found to be effective in improving memory performance, word fluency, the ability to learn new, non-associated verbal information, and response inhibition lasting for a period of three months. Both interventions combined various factors that seem to have an impact on delaying cognitive decline [27]. While the intervention I group trained memory while being physically active, the intervention II-group focused on cognitively stimulating, aerobic training in a group setting separately. As both groups included a combined cognitive and physical approach with different foci, current results emphasize the effect of this combination. Our results furthermore support the relationship between cognitive and physical functions in subjects with MCI [50]. Results are in line with previous studies, investigation the effectiveness of combined physical and cognitive interventions in comparison to single versions [52]. Multicomponent treatments showed positive, lasting effects on cognition [68 –74], as well as on their implementation on everyday life [75 –77]. Six months of multicomponent therapy (including exercise, activities of daily living, cognitive and social orientated intervention parts) stabilized cognitive and activities of daily living abilities in 362 persons with MCI and moderate dementia, who live at home and regularly visit a day care center [78]. The present results support these findings. Improvements in cognitive abilities of participants with amnestic MCI after completing a multicomponent treatment have been shown to persist up to two years [69]. Therefore, these combined forms of intervention result in long-term active participation and improvement in quality of life in older adults with MCI [70]. In this line, the results of a recent randomized controlled trial, investigating the effectiveness of a 2-year multidomain intervention including diet-treatment components, exercise, cognitive training, and vascular risk monitoring, suggested an improvement in cognitive performance in 1,260 subjects aged 60 to 77 years after intervention [79]. Due to the large number of combined approaches, it is difficult to say which intervention components have proven to be effective and which combination would be useful and efficient. In the present study, two combinations of two to three components proved to be effective, which enables efficient multimodal therapy. It is recommended that some proven effective multicomponent treatments should be administered daily for more than one year [75 , 80], what seems impossible to realize in clinical practice. The results from the present study yield evidence for the effectiveness of two short, combined intervention programs, which effects lasted for several months.

A post hoc classification of neuropsychological and motor performance based on the CERAD total score yield evidence for a pronounced ability to learn new, non-associated verbal information as assessed shown by the VLMT and TUG improvements. As further changes in neuropsychological or motor aspects were not observed, these two parameters might reflect the critical factors which might further predict intervention related improvements. The possibility to change the MCI status might be closely linked to the possibility to improve memory performance as illustrated by the changes in the VLMT results. Interestingly, the TUG parameters reflect very basic motor performance but seem to have high predictive power. Using physical improvement to group participants into high and low performer, results suggest a close relation between high physical SpeedCourt® performance and changes of the CERAD total score. This result in combination with the aforementioned relation of TUG and memory improvement further support the close relation of physical and cognitive improvements in this intervention group. Further research might therefore focus on these modulating factors. It might be asked whether preserved physical abilities might reflect a predictor for interventional outcome. It might be discussed controversially whether the CERAD cut-off value for classifying MCI reflects enough ecological validity that can be seen in activity of daily living. On the other hand, improvements of VLMT scores concerning the ability to learn new, non-associated verbal information as classified by this cut-off value might reflect enough ecological validity that support subjects during daily living. This improvement further reflects changes in one of the critical cognitive domains affected by dementia and also MCI. It has to be mentioned and emphasized that this improvement persisted up to three months after intervention. This emphasizes the therapeutical effect of the used intervention independent of the CERAD cut-off value.

It should be discussed whether this effect might also be explained by a practice effect. This effect was only found in the non-MCI group. However, at each time point, a different parallel version of the VLMT has been used. As discussed by Sandersin et al. (2021), the author could not demonstrate a practice effect in their study, even if other studies demonstrated a clear practice effect [81]. The authors argue with a reduced practice effect due to retroactive interference of the (6-month follow up) with the exposure to baseline and follow up examination. It is not clear whether the practice effect might be time dependent as Duff and colleagues (2012) could demonstrate a practice effect for MCI patients within a time interval of 1 week [82]. So possible practice effect might be responsible for the shown effect of the non-MCI group. On the other hand, no intervention has been performed in the two mentioned studies which might favor to explain the shown effect by the therapeutical intervention. As no further analysis were performed as suggested by Machulda et al. (2013) and Sandersin et al. (2021), a combination of practice effect and the therapeutical effect of the used intervention might explain the shown effect best, as a separation of both effects cannot be realized [81, 83]. A contextual practice effect with respect to familiarity with such a test situation cannot also be completely excluded [83]. The development of possible strategies during the processing of the tasks set cannot be ruled out either. To prevent possible practice effects, parallel versions were used when available (VLMT, RWT) to collect data on performance at the three different time points. The tests used indicate retest stability (VLMT: r_tt = 0.68 –r_tt = 0.87; ROCF: r_tt = 0.76–r_tt = 0.89; RWT: r_tt = 0.72–r_tt = 0.89), which should allow for progression diagnosis. Practice effects in other cognitive functions were not detected in previous studies [83].

The group size of this study might reflect a limiting factor. However, it has to be mentioned that in total 173 subjects were assessed using the CERAD. Four fulfilled the criteria of dementia and 112 did not reach the criteria for MCI.

It has further to be mentioned that factors like starting a training with social contacts might have an impact on the performance of the participants. A further sham/placebo or placebo group would have provided further insight into the underlying processes and effects of the applied training but would have not been practicable. However, the results could demonstrate that even after the training has finished, the effect persisted for a period of three months without any contacts to the supervisor or the training group. Furthermore, the time between the second evaluation and the three-month follow-up was twice as long as the intervention. This finding suggests that even without any training related activities or social contacts, participants kept their cognitive level as defined by the CERAD cut-off. Additionally, a selection bias might be discussed as most participants had long years of education and reported regular sport activities and normal activities of daily living like shopping. But beside the fact that a selection bias might exist, which cannot be ruled out as no participant can be obliged to participate, cognitive functions were in general in the range of MCI and neuropsychological assessment yielded evidence for average attentional performance and borderline memory performance (see Table 2), suggesting that the included sample is representative for the general population.

CONCLUSION

In conclusion, the two presented non-pharmacological, multicomponent interventions, which combined physical and cognitive training in a social setting, showed to improve the cognitive functions of participants with MCI efficiently. Interestingly, this effect persisted over a period of three months.

Footnotes

ACKNOWLEDGMENTS

This work was supported by the Leitmarkt Agentur.NRW, the European Union and the state government of North Rhine-Westfalia, Germany.

Authors’ disclosures available online ().

References

Salthouse

(2003) Memory aging from 18 to 80. Alzheimer Dis Assoc Disord 17, 162–167.

Petersen

, Negash

(2008) Mild cognitive impairment: An overview. CNS Spectr 13, 45–53.

Petersen

, Smith

, Waring

, Ivnik

, Tangalos

, Kokmen

(1999) Mild cognitive impairment: Clinical characterization and outcome. Arch Neurol 56, 303–308.

Petersen

, Lopez

, Armstrong

, Getchius

, Ganguli

, Gloss

, Gronseth

, Marson

, Pringsheim

, Day

, Sager

, Stevens

, Rae-Grant

(2018) Practice guideline update summary: Mild cognitive impairment: Report of the guideline development, dissemination, and implementation subcommittee of the American academy of neurology. Neurology 90, 126–135.

Petersen

(2011) Clinical practice. Mild cognitive impairment. N Eng J Med 364, 2227–2234.

Teixeira

, Gobbi

, Corazza

, Stella

, Costa

, Gobbi

(2012) Non- pharmacological interventions on cognitive functions in older people with Mild Cognitive Impairment (MCI). Arch Gerontol Geriatr 54, 175–180.

Mitchell

, Shiri-Feshki

(2009) Rate of progression of mild cognitive impairment to dementia–meta-analysis of 41 robust inception cohort studies. Acta Psychiatr Scand 119, 252–265.

Clarkson

, Giebel

, Larbey

, Roe

, Challis

, Hughes

, Jolley

, Poland

, Russell

; Members of the HoSt-D (Home Support in Dementia) Programme Management Group (2016) A protocol for a systematic review of effective home support to people with dementia and their carers: Components and impacts. J Adv Nurs 72, 186–196.

Amoyal

, Fallon

(2012) Physical exercise and cognitive training clinical inter- ventions used in slowing degeneration associated with mild cognitive impairment: A review of the recent literature. Top Geriatr Rehabil 28, 208–216.

10.

Barnes

, Santos-Modesitt

, Poelke

, Kramer

, Castro

, Middleton

, Yaffe

(2013) The Mental Activity and eXercise (max) trial: A randomized controlled trial to enhance cognitive function in older adults. JAMA Intern Med 173, 797–804.

11.

de Bruijn

, Schrijvers

, de Groot

, Witteman

, Hofman

, Franco

, Koudstaal

, Ikram

(2013) The association between physical activity and dementia in an elderly population: The Rotterdam Study. Eur J Epidemiol 28, 277–283.

12.

Erickson

, Voss

, Prakash

, Basak

, Szabo

, Chaddock

, Kim

, Heo

, Alves

, White

, Wojcicki

, Mailey

, Vieira

, Martin

, Pence

, Woods

, McAuley

, Kramer

(2011) Exercise training increases size of hippocampus and improves memory.-. Proc Natl Acad Sci U S A 108, 3017–3022.

13.

Karp

, Paillard-Borg

, Wang

, Silverstein

, Winblad

, Fratiglioni

(2006) Mental, physical and social components in leisure activities equally contribute to decrease dementia risk. Dement Geriatr Cogn Disord 21, 65–73.

14.

Barnes

, Yaffe

(2011) The projected effect of risk factor reduction on Alzheimer’s disease prevalence. Lancet Neurol 10, 819–828.

15.

Hertzog

, Kramer

, Wilson

, Lindenberger

(2008) Enrichment effects on adult cognitive development: Can the functional capacity of older adults be preserved and enhanced? Psychol Sci Public Interest 9, 1–65.

16.

Valenzuela

, Sachdev

(2006) Brain reserve and cognitive decline: A non- parametric systematic review. Psychol Med 36, 1065–1073.

17.

Valenzuela

, Sachdev

(2006) Brain reserve and dementia: A systematic review. Psychol Med 36, 441–454.

18.

Verghese

, Lipton

, Katz

, Hall

, Derby

, Kuslansky

, Ambrose

, Sliwinski

, Buschke

(2003) Leisure activities and the risk of dementia in the elderly. N Engl J Med 348, 2508–2516.

19.

Belleville

, Gilbert

, Fontaine

, Gagnon

, Ménard

, Gauthier

(2006) Improvement of episodic memory in persons with mild cognitive impairment and healthy older adults: Evidence from a cognitive intervention program. Dement Geriatr Cogn Disord 22, 486–499.

20.

Cipriani

, Bianchetti

, Trabucchi

(2006) Outcomes of a computer-based cognitive rehabilitation program on Alzheimer’s disease patients compared with those on patients affected by mild cognitive impairment. Arch Gerontol Geriatr 43, 327–335.

21.

Gates

, Sachdev

, Singh

MAF

, Valenzuela

(2011) Cognitive and memory training in adults at risk of dementia: A systematic review. BMC Geriatr 11, 55.

22.

Jean

, Bergeron

MÈ

, Thivierge

, Simard

(2010) Cognitive intervention programs for individuals with mild cognitive impairment: Systematic review of the literature. Am J Geriatr Psychiatry 18, 281–296.

23.

Rozzini

, Costardi

, Chilovi

, Franzoni

, Trabucchi

, Padovani

(2007) Efficacy of cognitive rehabilitation in patients with mild cognitive impairment treated with cholinesterase inhibitors. Int J Geriatr Psychiatry 22, 356–360.

24.

Simon

, Yokomizo

, Bottino

(2012) Cognitive intervention in amnestic mild cognitive impairment: A systematic review. Neurosci Biobehav Rev 36, 1163–1178.

25.

Clare

, Woods

(2003) Cognitive rehabilitation and cognitive training for early- stage Alzheimer’s disease and vascular dementia. Cochrane Database Syst Rev CD003260.

26.

Bahar-Fuchs

, Clare

, Woods

(2013) Cognitive training and cognitive rehabilitation for mild to moderate Alzheimer’s disease and vascular dementia. Cochrane Database Syst Rev 2013, CD003260.

27.

Lissek

, Suchan

(2021) Preventing dementia? Interventional approaches in mild cognitive impairment. Neurosci Biobehav Rev 122, 143–164.

28.

Anguera

, Boccanfuso

, Rintoul

, Al-Hashimi

, Faraji

, Janowich

, Kong

, Larraburo

, Rolle

, Johnston

, Gazzaley

(2013) Video game training enhances cognitive control in older adults. Nature 501, 97–101.

29.

Gagnon

, Belleville

(2012) Training of attentional control in mild cognitive impairment with executive deficits: Results from a double-blind randomised controlled study. Neuropsychol Rehabil 22, 809–835.

30.

Owen

, Hampshire

, Grahn

, Stenton

, Dajani

, Burns

, Howard

, Ballard

(2010) Putting brain training to the test. Nature 465, 775–778.

31.

Rizkalla

(2018) Cognitive training in the elderly: A randomized trial to evaluate the efficacy of a self-administered cognitive training program. Aging Ment Health 22, 1384–1394.

32.

Wolinsky

, Vander Weg

, Howren

, Jones

, Dotson

(2013) A randomized controlled trial of cognitive training using a visual speed of processing intervention in middle aged and older adults. PloS One 8, e61624.

33.

Valenzuela

, Sachdev

(2009) Can cognitive exercise prevent the onset of dementia? systematic review of randomized clinical trials with longitudinal follow-up. Am J Geriatr Psychiatry 17, 179–187.

34.

Rosen

, Sugiura

, Kramer

, Whitfield-Gabrieli

, Gabrieli

(2011) Cognitive training changes hippocampal function in mild cognitive impairment: A pilot study. J Alzheimers Dis 26, 349–357.

35.

Smith

, Nielson

, Woodard

, Seidenberg

, Rao

(2013) Physical activity and brain function in older adults at increased risk for Alzheimer’s disease. Brain Sci 3, 54–83.

36.

Nyberg

, Magnussen

, Lundquist

, Baaré

, Bartrés-Faz

, Bertram

, Boraxbekk

, Brandmaier

, Drevon

, Ebmeier

, Ghisletta

, Henson

, Junqué

, Kievit

, Kleemeyer

, Knights

, Kühn

, Lindenberger

, Penninx

BWJH

, Pudas

, Sørensen

, Vaqué-Alcázar

, Walhovd

, Fjell

(2021) Educational attainment does not influence brain aging. Proc Natl Acad Sci U S A 118, e2101644118.

37.

Larson

, Wang

, Bowen

, McCormick

, Teri

, Crane

, Kukull

(2006) Exercise is associated with reduced risk for incident dementia among persons 65 years of age and older. Ann Intern Med 144, 73–81.

38.

Baker

, Frank

, Foster-Schubert

, Green

, Wilkinson

, McTiernan

, Plymate

, Fishel

, Watson

, Cholerton

, Duncan

, Mehta

, Craft

(2010) Effects of aerobic exercise on mild cognitive impairment: A controlled trial. Arch Neurol 67, 71–79.

39.

, Doi

, Makizako

, Shimada

, Tsutsumimoto

, Hotta

, Nakakubo

, Park

, Suzuki

, (2015) Objectively measured physical activity, brain atrophy, and white matter lesions in older adults with mild cognitive impairment. Exp Gerontol 62, 1–6.

40.

Grande

, Vanacore

, Maggiore

, Cucumo

, Ghiretti

, Galimberti

, Scarpini

, Mariani

, Clerici

(2014) Physical activity reduces the risk of dementia in mild cognitive impairment subjects: A cohort study. J Alzheimers Dis 39, 833–839.

41.

Heyn

, Abreu

, Ottenbacher

(2004) The effects of exercise training on elderly persons with cognitive impairment and dementia: A meta-analysis. Arch Phys Med Rehabil 85, 1694–1704.

42.

Jedrziewski

, Ewbank

, Wang

, Trojanowski

(2010) Exercise and cognition: Results from the national long term care survey. Alzheimers Dement 6, 448–455.

43.

Lautenschlager

, Cox

, Flicker

, Foster

, van Bockxmeer

, Xiao

, Greenop

, Almeida

(2008) Effect of physical activity on cognitive function in older adults at risk for Alzheimer disease: A randomized trial. JAMA 300, 1027–1037.

44.

Makizako

, Liu-Ambrose

, Shimada

, Doi

, Park

, Tsutsumimoto

, Uemura

, Suzuki

, (2005) Moderate-intensity physical activity, hippocampal volume, and memory in older adults with mild cognitive impairment. J Gerontol A Biol Sci Med Sci 70, 480–486.

45.

Scherder

, Van Paasschen

, Deijen

, Van Der Knokke

, Orlebeke

, Burgers

, Devriese

, Swaab

, Sergeant

(2005) Physical activity and executive functions in the elderly with mild cognitive impairment. Aging Ment Health 9, 272–280.

46.

Tanigawa

, Takechi

, Arai

, Yamada

, Nishiguchi

, Aoyama

(2014) Effect of physical activity on memory function in older adults with mild Alzheimer’s disease and mild cognitive impairment. Geriatr Gerontol Int 14, 758–762.

47.

Teri

, Logsdon

, McCurry

(2008) Exercise interventions for dementia and cognitive impairment: The Seattle Protocols. J Nutr Health Aging 12, 391–394.

48.

Van Uffelen

, Chinapaw

, van Mechelen

, Hopman-Rock

(2008) Walking or vitamin B for cognition in older adults with mild cognitive impairment? A randomised controlled trial. Br J Sports Med 42, 344–351.

49.

, Kolanowski

, Strumpf

, Eslinger

(2006) Improving cognition and function through exercise intervention in Alzheimer’s disease. J Nurs Scholarsh 38, 358–365.

50.

Uemura

, Shimada

, Makizako

, Doi

, Yoshida

, Tsutsumimoto

, Anan

, Suzuki

, (2013) Cognitive function affects trainability for physical performance in exercise intervention among older adults with mild cognitive impairment. Clin Interv Aging 8, 97–102.

51.

Train the Brain Consortium (2017) Randomized trial on the effects of a combined physical/cognitive training in aged mci subjects: The train the brain study. Sci Rep 7, 39471.

52.

Styliadis

, Kartsidis

, Paraskevopoulos

, Ioannides

, Bamidis

(2015) Neuroplastic effects of combined computerized physical and cognitive training in elderly individuals at risk for dementia: An eloreta controlled study on resting states. Neural Plast 2015, 172192.

53.

Morris

, Heyman

, Mohs

, Hughes

, van Belle

, Fillenbaum

, Mellits

, Clark

(1989) The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part I. Clinical and neuropsychological assessment of Alzheimer’s disease. Neurology 39, 1159–1165.

54.

Mirra

, Heyman

, McKeel

, Sumi

, Crain

, Brownlee

, Vogel

, Hughes

, van Belle

, Berg

(1991) The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD): Part II. Standardization of the neuropathologic assessment of Alzheimer’s disease. Neurology 41, 479.

55.

Beck

, Steer

, Brown

(1996) volume 10. Pearson. Beck depression inventory (BDI-II).

56.

Kühner

, Bürger

, Keller

, Hautzinger

(2007) [Reliability and validity of the revised Beck Depression Inventory (BDI-II). Results from German samples.]. Nervenarzt 78, 651–656.

57.

Charness

, Gneezy

, Kuhn

(2012) Experimental methods: Between-subject and within-subject design. J Econ Behav Organ 81, 1–8.

58.

Chandler

, Lacritz

, Hynan

, Barnard

, Allen

, Deschner

, Weiner

, Cullum

(2005) A total score for the CERAD neuropsychological battery. Neurology 65, 102–106.

59.

Zimmermann

, Fimm

(2009) Testbatterie zur Aufmerksamkeitsprüfung (tap) version 2.2. psytest.

60.

Meyers

, Meyers

(1995) Rey complex figure test under four different administration procedures. Clin Neuropsychol 9, 63–67.

61.

Helmstaedter

, Lendt

, Lux

(2001) Verbaler Lern-und Merkfahigkeitstest: VLMT; Manual. Beltz-test.

62.

Wechsler

(2012) Wechsler preschool and primary scale of intelligence—fourth edition. The Psychological Corporation San Antonio, TX.

63.

Aschenbrenner

, Tucha

, Lange

(2000) Regensburger Wortflüssigkeits-Test: RWT. Hogrefe-Verlag Gmbh & Co, Göttingen Bern.

64.

Düking

, Born

, Sperlich

(2015) The Speedcourt: Reliability, usefulness and validity of a new method to determine change-of-direction speed. Int J Sports Physiol Perform 11, 130–134.

65.

Era

, Heikkinen

(1985) Postural sway during standing and unexpected disturbance of balance in random samples of men of different ages. J Gerontol 40, 287–295.

66.

Salkind

(2010) Encyclopedia of research design, Vol. 1. Sage.

67.

Bamidis

, Fissler

, Papageorgiou

, Zilidou

, Konstantinidis

, Billis

, Romanopoulou

, Karagianni

, Beratis

, Tsapanou

, Tsilikopoulou

, Grigoriadou

, Ladas

, Kyrillidou

, Tsolaki

, Frantzidis

, Sidiropoulos

, Siountas

, Matsi

, Papatriantafyllou

, Margioti

, Nika

, Schlee

, Elbert

, Tsolaki

, Vivas

, Kolassa

(2015) Gains in cognition through combined cognitive and physical training: The role of training dosage and severity of neurocognitive disorder. Front Aging Neurosci 7, 152.

68.

Buschert

, Friese

, Teipel

, Schneider

, Merensky

, Rujescu

, Möller

, Hampel

, Buerger

(2011) Effects of a newly developed cognitive intervention in amnestic mild cognitive impairment and mild Alzheimer’s disease: A pilot study. J Alzheimers Dis 25, 679–694.

69.

Buschert

, Teipel

, Jolk

, Hampel

, Rujescu

, Buerger

(2012) Long-term observation of a multicomponent cognitive intervention in mild cognitive impairment. J Clin Psychiatry 73, e1492–e1498.

70.

Dannhauser

, Cleverley

, Whitfield

, Fletcher

, Stevens

, Walker

(2014) A complex multimodal activity intervention to reduce the risk of dementia in mild cognitive impairment–thinkingfit: Pilot and feasibility study for a randomized controlled trial. BMC Psychiatry 14, 129.

71.

Hwang

, Choi

, Yoon

, Suh

, Lee

, Han

, Hong

(2012) The effect of cognitive training in patients with mild cognitive impairment and early Alzheimer’s disease: A preliminary study. J Clin Neurol 8, 190–197.

72.

Hyer

, Scott

, Lyles

, Dhabliwala

, McKenzie

(2014) Memory intervention: The value of a clinical holistic program for older adults with memory impairments. Aging Ment Health 18, 169–178.

73.

Suzuki

, Shimada

, Makizako

, Doi

, Tsutsumimoto

, Anan

, Uemura

, Lee

, Park

, (2012) Effects of multicomponent exercise on cognitive function in older adults with amnestic mild cognitive impairment: A randomized controlled trial. BMC Neurol 12, 128.

74.

Tsolaki

, Kounti

, Agogiatou

, Poptsi

, Bakoglidou

, Zafeiropoulou

, Soumbourou

, Nikolaidou

, Batsila

, Siambani

, Nakou

, Mouzakidis

, Tsiakiri

, Zafeiropoulos

, Karagiozi

, Messini

, Diamantidou

, Vasiloglou

(2011) Effectiveness of nonpharmacological approaches in patients with mild cognitive impairment. Neurodegener Dis 8, 138–145.

75.

Graessel

, Stemmer

, Eichenseer

, Pickel

, Donath

, Kornhuber

, Luttenberger

(2011) Non-pharmacological, multicomponent group therapy in patients with degenerative dementia: A 12-month randomized, controlled trial. BMC Med 9, 129.

76.

Luttenberger

, Donath

, Uter

, Graessel

(2012) Effects of multimodal non-drug therapy on dementia symptoms and need for care in nursing home residents with degenerative dementia: A randomized-controlled study with 6-month follow-up. J Am Geriatr Soc 60, 830–840.

77.

Nascimento

CMC

, Ayan

, Cancela

, Gobbi

LTB

, Gobbi

, Stella

(2014) Effect of a multimodal exercise program on sleep disturbances and instrumental activities of daily living performance on Parkinson’s and Alzheimer’s disease patients. Geriatr Gerontol Int 14, 259–266.

78.

Straubmeier

, Behrndt

, Seidl

, Özbe

, Luttenberger

, Graessel

(2017) Non-pharmacological treatment in people with cognitive impairment: Results from the randomized controlled German day care study.rztebl Int. Dtsch A 114, 815–821.

79.

Ngandu

, Lehtisalo

, Solomon

, Levälahti

, Ahtiluoto

, Antikainen

, Bäckman

, Hänninen

, Jula

, Laatikainen

, Lindström

, Mangialasche

, Paajanen

, Pajala

, Peltonen

, Rauramaa

, Stigsdotter-Neely

, Strandberg

, Tuomilehto

, Soininen

, Kivipelto

(2015) A 2 year multidomain intervention of diet, exercise, cognitive training, and vascular risk monitoring versus control to prevent cognitive decline in at-risk elderly people (finger): A randomised controlled trial. Lancet 385, 2255–2263.

80.

Luttenberger

, Hofner

, Graessel

(2012) Are the effects of a non-drug multimodal activation therapy of dementia sustainable? Follow-up study 10 months after completion of a randomised controlled trial. BMC Neurol 12, 151.

81.

Sanderson-Cimino

, Elman

, Tu

, Gross

, Panizzon

, Gustavson

, Bondi

, Edmonds

, Eglit

, Eppig

, Franz

, Jak

, Lyons

, Thomas

, Williams

, Kremen

; Alzheimer’s Disease Neuroimaging Initiative (2022) Cognitive practice effects delay diagnosis of MCI: Implications for clinical trials. Alzheimers Dement (N Y) 8, e12228.

82.

Duff

, Callister

, Dennett

, Tometich

(2012) Practice effects: A unique cognitive variable. Clin Neuropsychol 26, 1117–1127.

83.

Machulda

, Pankratz

, Christianson

, Ivnik

, Mielke

, Roberts

, Knopman

, Boeve

, Petersen

(2013) Practice effects and longitudinal cognitive change in normal aging vs. incident mild cognitive impairment and dementia in the mayo clinic study of aging. Clin Neuropsychol 27, 1247–1264.