Elsevier

Geriatric Nursing

Volume 41, Issue 4, July–August 2020, Pages 421-428
Geriatric Nursing

Feature Article
Impact of a supervised multicomponent physical exercise program on cognitive functions in patients with type 2 diabetes

Highlights

•

We examined the impact of a multicomponent 32-week exercise program on cognitive functions in participants with type 2 diabetes.

•

The exercise program did not improve the cognitive function of the participants.

•

Future studies should examine the effectiveness of incorporating dual-task activities (merging physical and cognitive stimulation) in the exercise program.

Abstract

This study evaluated the impact of a multicomponent exercise program on cognitive functions in participants with Type 2 Diabetes. Participants (n = 70, 65.6 ± 5.9 years) engaged in the program (75 min per session; 3 x week) for 32 weeks. A battery of cognitive tests was performed at baseline and study completion. Two groups were formed according to their attendance rate (low and high attendance), and statistical comparisons were computed on their changes in cognitive performance. Such changes were also associated with the attendance rate for all participants. Results showed no significant differences between groups in their change scores, although there were some within-group differences in both groups. Correlation analysis showed that the attendance rate was not associated with cognitive performance changes, except for one variable. As the exercise program did not improve cognitive function, we discuss the potential of future interventions to incorporate dual-task activities merging physical and cognitive stimulation.

Keywords

Type 2 diabetes
Cognition
Multicomponent physical exercise

Introduction

Due to factors like the growth and aging of the population and the high prevalence of sedentary lifestyles, overweight and obesity, there is an increasing incidence and prevalence of people with Type 2 Diabetes Mellitus (T2DM).1 It is known that the T2DM can lead to severe micro-and macrovascular complications (ischemic heart disease and stroke are the leading causes of diabetes-related morbidity)2 and to an approximately two-fold increased mortality rate compared to the general population.3 On the other hand, accumulated evidence shows that T2DM is a risk factor for cognitive decline4 and it was reported that people with T2DM showed an almost twofold increase in the risk of dementia compared to non-T2DM controls.5

The main cognitive functions affected in this population seem to be memory, executive functioning, and speed of information processing.6 Such cognitive decline could jeopardize the appropriate management of T2DM, as patients must commit to lifelong self-care tasks (e.g., medication, checking blood glucose, and diet regime) requiring complex cognitive functioning.7 In particular, impairment in executive functioning – a domain of cognitive function crucial for planning, organizing and performing activities8 – can negatively affect the self-care capacity, and, in consequence, increasing the risk of poor glycemic control.9 The mechanisms underlying cognitive impairment in T2DM are probably multifactorial. Poor glycemic control, inflammatory mediators, rheological factors and dysregulation of the hypothalamic–pituitary–adrenal axis are pointed as moderators of the association between T2DM and cognitive decline.10

Considering the high prevalence of T2DM, public health strategies should be implemented at the primary, secondary and tertiary levels of prevention for ameliorating its negative effects in cognitive functioning. The primary level of prevention focuses on individuals with no evidence of symptoms or who are identified as having minimal but detectable signs or symptoms of cognitive impairment.12 In this sense, the control of the disease and the regular cognitive screening are fundamental preventive measures to reduce its consequences.11 At the secondary level, through early detection, T2DM patients with cognitive impairment should have access to rehabilitation measures to prevent from progressing to overt clinical symptoms. Finally, the tertiary prevention seeks to reduce disability, prevent relapses and recurrences of the illness.12

Physical exercise (PE) is recognized as an effective non-pharmacological therapeutic strategy to improve insulin action, glycemic control and reduce risk factors for cardiovascular disease.11 Furthermore, PE seems to be an efficient tool to promote brain health and cognitive function in normal and pathological conditions, protecting against cognitive impairment and/or degenerative diseases.12 The physiological mechanisms underlying the positive effect of PE on the brain includes the release of growth and neurotrophic factors involved in angiogenesis, synaptogenesis, and neurogenesis,13 increased cerebral blood flow, stimulation of neurotransmitters, higher levels of arousal, and increased mood and self-perception of competence.14

It is important to note that not all studies showed that PE had positive effects on the cognitive abilities of patients with T2DM. For instance, a recent systematic review concluded that there was no consistent evidence that PE contributes to a better cognitive function in adults with prediabetes or with T2DM.15 The authors reported that the exercise interventions have been mainly focused on aerobic activities and that although some studies suggested that PE could improve executive function and delay memory and global cognitive losses, the current evidence is still scarce. The authors proposed that future investigations should also examine the impact of resistance training programs, due to its potential clinical effects on cognition.15

There is evidence that different types of PE may have specific effects on brain and cognition.16 It has been hypothesized that a combination of different types of exercise can provide the greatest neurological effects.13 Such types of multicomponent exercise interventions (MCEI) include activities that target two or more physical fitness components (i.e., strength, aerobic, balance, and flexibility exercise).8 Some studies reported higher benefits on cognition for this type of intervention in healthy older adults, especially when involving a combination of both aerobic and resistance exercises.17,18

We found two randomized controlled trials that examined the effects of a MCEI in the cognitive function of people with T2DM. One of these studies showed that a 6-month progressive aerobic and resistance training program improved several cognitive tests (e.g., Color and Word Test, Trail Making A and B Tests) and a global cognitive score.19 In contrast, the other randomized controlled trial found a negative effect on the cognitive performance of people with T2DM after a 24-week program that combined aerobic and resistance exercise and lifestyle intervention. In this case, the authors emphasized the need for more research to examine the potential benefits of exercise in the cognitive health of T2DM patients.20

In summary, there is a scarcity of studies on the impact of PE in the cognitive functioning of people with T2DM, and the existing ones showed inconsistent findings. For promoting the functional status and quality of life of T2DM participants, it is important to identify the optimal type and dose of exercise that can delay or prevent cognitive functioning in this high-risk group. Therefore, the present study aimed to analyze the impact of participation in a 32-week multicomponent physical exercise program on the cognitive function of participants with T2DM. We hypothesize that executive functions, information processing speed, visuospatial ability, and reaction time would benefit from the multicomponent physical exercise program.

Methods

Study design and participants

This is a quasi-experimental study with a multicenter intervention conducted in three cities from Portugal (Évora, Maia, and Vila Real). People with T2DM participated in a MCEI during 32 weeks, three sessions per week with a duration of 75 min per session. Participants were assessed in several cognitive abilities (e.g., visual search, mental flexibility, working memory, inhibition and resistance to interference, and information processing speed) at the beginning and at the end of the exercise program.

During diabetology consultations of community health centers and local hospitals, patients with T2DM were informed by their medical doctors about the general study characteristics (e.g., objectives, duration and weekly frequency). For those patients interested, the medical doctors carried out a screening for relative or absolute contraindications to exercise (inclusion criteria 1 to 11; Table 1), and a total of eighty-eight T2DM patients (53 men and 35 women) were considered able to participate in the exercise program. During this period, members of the research team also met with the volunteers to clarify doubts. Written informed consent was obtained for all volunteers.

Table 1. Inclusion and exclusion criteria.

INCLUSION CRITERIA
•

Age between 50 and 80 years;

•

Diagnosed with DMT2 for at least one year;

•

HbA1c less than 10%;

•

Medication stabilized for at least 3 months;

•

Not having started insulin therapy in the last 6 months;

•

Complications related to diabetes controlled (namely diabetic foot, retinopathy and diabetic nephropathy);

•

Without severe cardiovascular, pulmonary or musculoskeletal symptoms and pathology;

•

Being independent and without changes in gait or balance;

•

Non-smoking;

•

Without participation in supervised exercise programs in the last 6 months;

•

Ability to participate in a long-term structured exercise;

•

Normal cognitive status*.

EXCLUSION CRITERIA
•

During the period study, participant started another supervised physical exercise program;

•

Failed the study protocol assessment.

⁎

According to the Portuguese version of the Mini-Mental State Examination (MMSE).37

The flow of participants through each stage and details of dropouts are shown in Fig. 1. From the eighty-eight participants with medical clearance to participate in the MCEI, five participants were subsequently identified with cognitive deficit (12th inclusion criteria) by specialists of the research team, so they were removed from the study. The final group was constituted by 70 participants (Table 2), since 3 and 10 patients dropped from the study before performing the cognitive pre-tests or during the intervention period, respectively.

Fig. 1
  1. Download : Download high-res image (399KB)
  2. Download : Download full-size image

Fig. 1. Flow of participants in the study.

Table 2. Demographics characteristics of the participants.

Low attendance (N = 24)min - maxMedium attendance (N = 21)min - maxHigh attendance (N = 25)min - maxTotal
Age (years)66.3 ± 5.957 - 7864.76 ± 6.4655 - 7765.68 ± 5.5557 - 7765,61 ± 5.89
MMSE (points)28 ± 1.9623 - 3027.55 ± 1.8224 - 3028.24 ± 1.3326 - 3027.96 ± 1.71
Education (years)9.33 ± 4.91⁎3 - 177.67 ± 4.734 - 185.68 ± 2.794 - 137.53 ± 4.42
BMI (kg/m2)29.71 ± 5.5821 - 44.630.79 ± 3.721 - 39.130.22 ± 4.4622.2 - 42.330.22 ± 4.64
Gender (% woman)58.3⁎–33.3–24–38.6

Note: BMI= Body Mass Index; MMSE= Mini-Mental State Examination.

⁎

p < 0.05 for comparison between groups.

This study was approved by the Ethics Committees of the University of Évora and the Northern Regional Health Administration and was conducted in accordance with the Helsinki Declaration.21

Procedures

The cognitive testing was performed individually at pre- and post-intervention. Experts with academic degrees in social sciences conducted the participants' assessment. Before the assessment sessions, the test protocols were carefully revised in a meeting with all the experts. The participants’ cognitive evaluation was carried out in a single session (∼2 h) in a quiet room (at the local university institution) and the tests were performed in the same order by all participants. The same experts conducted the cognitive assessment at all times.

Outcome measures

•

Trail Making Test (TMT). This test was used to assess visual search, mental flexibility, and information processing speed. The participant used a pencil to connect 25 encircled numbers in ascending order (①-②-③…; Part A) or alternating in ascending order between numbers and letters distributed across a page (①-ⓐ-②-ⓑ-③-ⓒ…; Part B). Participants were instructed to complete each part as quickly and accurately as possible and the time (seconds) was recorded.22

•

Stroop color and word test. This task was used to assess inhibition and resistance to interference. It comprised three tasks of 45 s each. In the first task (word), participants were required to read the words (of 4 colors), while in the second task (color), participants named the color of Xs. In the last task (word-color), participants were asked to name the color of the ink for each word (the color word appears in an incongruent color). Each task used an A4 sheet (210 × 297 mm), consisting of 5 columns of 20 stimuli. The number of correct responses was recorded for each task. An interference score was calculated according to the test manual.23

•

Digit Span Test. For evaluating working memory, the Digit Span subtest of the Wechsler Adult Intelligence Scale-III (WAIS III) was administered. The test consisted of 2 tasks. In the Digit Span Forward task, the participants were required to memorize a number sequence and repeat it verbally in the same order. In the Digit Span Backward task, participants were instructed to repeat the sequence of numbers in reverse order. The number of items in each series was progressively increased. A separate score was considered for each Digit Span task and a total score was calculated from the sum of both scores.24

•

Phonemic and semantic verbal fluency test. This test was used to evaluate non-motor processing speed, language production, and executive functions. In the semantic fluency task, participants were asked to name as many different animal names as possible in 60 s. In the phonemic fluency task, participants were asked to generate as many words as possible, beginning with a specific letter. A 60-s trial was given for each letter (“M”, “R” and “P”). The semantic fluency score corresponded to the total number of animals mentioned and the phonemic fluency score was the sum of the correct words in the three trials.25

•

Reaction Time. Simple and choice reaction time in single and dual-task conditions were evaluated with the Deary-Liewald reaction time task.26 In the simple reaction time task (SRT), the participant had to press a key as quickly as possible in response to a single stimulus (a cross appears inside of a square on the computer screen). On the choice reaction time task (CRT), there were four stimuli and participants had to press the button that corresponded to the correct response. The SRT and CRT were also performed in dual-task conditions (starting from zero, the participant counts forward in twos). All reaction time tasks involved 8 practice trials and 20 test trials. The inter-stimulus interval ranged between 2 and 5 s and was randomized within these boundaries.27 The median reaction time (milliseconds) was computed for each participant in each task.

Multicomponent physical exercise intervention

The exercise intervention was the Portuguese community-based exercise program “Diabetes em Movimento”, which was developed for middle-aged and older adults with T2DM.28 Participants engaged in the exercise program for 32 weeks, three times per week (on non-consecutive days) with 75 min per session. The sessions were conducted by physical exercise professionals (with a sport science degree) in groups of less than 30 participants. This program involves a combination of aerobic, resistance, agility/balance and flexibility exercises according to the following structure:

•

Warm-up (10 min). Continuous brisk walking on the sports hall.

•

Aerobic exercise (30 min). Continuous brisk walking and interval walking. In some parts of the session, participants carried an external load (± 0.75 kg), manipulate balls or walk through an obstacle circuit.

•

Resistance exercise (20 min). Each session included six exercises (for upper, lower limbs and torso) using diverse resistances (bodyweight, and bottles/dumbbells, and gymnastic balls). It was organized in a circuit mode (alternating upper limbs, lower limbs and torso) with no rest between exercises and a 1-minute rest between circuits. The number of circuits increased progressively from one to four with 20 and 30 repetitions for bilateral and unilateral exercise, respectively. Individual overload was promoted when the last repetitions of each exercise were performed without evident local muscle fatigue.

•

Agility/balance exercise (10 min). Traditional and pre-sport games performed with gymnastic balls in pairs or groups (e.g., two teams aligned; elements of each team try to transfer the balls, through passes, as quickly as possible between two boxes placed at opposite sides).

•

Flexibility exercise/cool-down (5 min). A sequence of static (held for 10 s) and dynamic (20 repetitions) stretching was performed with the support of a chair.

The exercise sessions were developed with low-cost materials such as chairs, 0.5 L water bottles with sand (± 0.75 kg), dumbbells (1, 2 or 3 kg), gymnastic balls, cones and flags, and colorful vests. The program consisted of five exercise session plans, each of them with different aerobic, resistance and agility/balance exercises, successively applied over time to induce stimuli variability. Exercise sessions were planned for moderate intensity (12–13 points on a subjective effort perception scale of 6–20 points) and were systematically monitored through Borg's rating of perceived exertion (RPE) scale29 as each participant recorded the overall intensity at the end of each session.30

Statistical analyses

For studying the effectiveness of the exercise program, we used comparative and correlational analysis of the cognitive outcomes. We compared the cognitive performance of two groups formed according to the percentile distribution of the participants’ attendance to the exercise sessions: ≤ 33.3th, low attendance group, and ≥ 66.6th percentile, high attendance group. According to the results of the Shapiro–Wilk test of normality, independent-sample t-test or the non-parametric Mann-Whitney U test were used to study differences at baseline between the two groups. Accordingly, the paired t-test or the Wilcoxon test was used to examine the score changes in the measured variables within the groups. To assess whether the ≤ 33.3th and ≥ 66.6th percentile groups showed differential changes after the exercise program, we used analyses of covariance (ANCOVA) in the change scores (i.e., post-test–pre-test/pre-test*100), with education level serving as a covariate. ANCOVA is robust to violations of normality of distributions, particularly when the group sizes are similar.31 Effect sizes are reported as partial eta-squared (ηp2), with cut-off values of 0.01, 0.06, and 0.14 for small, medium, and large effects, respectively.32

Finally, the non-parametric partial correlation was used to study the association between the participants’ attendance rate and the change scores in the cognitive tests, controlling for education. Cut-off values of 0.10, 0.30 and 0.50 represent small, moderate and high correlations, respectively.32 Statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) software version 22, and the level of significance was set at p < 0.05 for all analyses.

Results

Overall, the participants attended to 61.3 ± 33.4% of the sessions and the subjects' subjective rating of perceived exertion mean was 13 ± 1.3 points. For the low attendance group, the adherence to the exercise program session was 29.1 ± 13.8% (rating of perceived exertion was 12.6 ± 1.4 points); for the high attendance group, the adherence was 92.3 ± 3.3% (rating of perceived exertion was 13.4 ± 1.1 points).

Table 3 shows the participants’ performance in the cognitive tests at baseline and after 32 weeks. At baseline, a significant difference between groups was found only in phonemic fluency total score (p = 0.013). ANCOVA showed that there were no significant differences between groups in the 32-week change scores. Within-group analysis showed that there was a significant improvement from pre- to post-test for the < 33.3th percentile group on TMT-A (−14.04%; p = 0.002), TMT-B (−4.71%, p = 0.039) and phonemic fluency total score (22.45%, p = 0.001). For the ≥ 66th percentile group, a significant improvement was found in the simple reaction time – median (−9.66%; p = 0.012).

Table 3. Neuropsychological test performance at baseline and at 32 weeks.

Cognitive variablesBaseline (Mean ± SD)32 Weeks (Mean ± SD)Adjusted means, M (95% CI)p‡ηp2
TMT – A (s)
 Low attendance71.02 ± 23.1658.76 ± 28.73b−14.04 (−25.32; −2.76)0.2020.035
 High attendance70.94 ± 30.3263.72 ± 19.14−3.39 (−14.42; 7.64)
TMT – B (s)
 Low attendance159.54 ± 92.03142.54 ± 90.73b−4.71 (−1.53; 9.12)0.6750.004
 High attendance200.23 ± 78.91181.47 ± 76.57−8.98 (−2.48; 4.52)
Stroop (word) (n)
 Low attendance74.3 ± 17.7975.48 ± 15.682.86 (−4.29; 10.00)0.7610.002
 High attendance67.32 ± 15.2169.6 ± 16.374.44 (−2.38; 11.26)
Stroop (color) (n)
 Low attendance54.13 ± 12.2352.87 ± 11.32−1.94 (−7.22; 3.35)0.1950.038
 High attendance51.13 ± 8.7850.92 ± 10.043.14 (−2.01; 8.23)
Stroop (word-color) (n)
 Low attendance28.17 ± 9.928.39 ± 9.326.04 (−6.18; 18.26)0.8430.001
 High attendance26.88 ± 9.1727.96 ± 9.167.82 (−4.11; 19.76)
Stroop (Interference) (n)
 Low attendance31.48 ± 6.6331.35 ± 6.410.33 (−5.64; 4.98)0.3150.023
 High attendance29.54 ± 5.3129.68 ± 6.053.61 (−1.58; 8.79)
Digit Span Test (forward) (n)
 Low attendance9.0 ± 3.329.08 ± 2.61.36 (−7.64; 10.34)0.7030.003
 High attendance8.12 ± 2.28.24 ± 2.313.86 (−4.93; 12.68)
Digit Span Test (backward) (n)
 Low attendance4.63 ± 24.54 ± 1.822.04 (−28.13; 32.20)0.2540.023
 High attendance4.64 ± 1.735.04 ± 1.8627.44 (−2.06; 56.94)
Digit Span Test (total score) (n)
 Low attendance13.63 ± 3.5213.63 ± 3.790.34 (−8.74; 9.42)0.4600.012
 High attendance12.76 ± 2.513.28 ± 3.685.27 (−3.61; 14.15)
Semantic fluency (n)
 Low attendance14.79 ± 4.1816.82 ± 4.9417.83 (0.51; 35.16)0.3510.020
 High attendance14.52 ± 4.915.2 ± 5.26.07 (−10.05; 22.18)
Phonemic fluency (n)
 Low attendance29.59 ± 11.2334.55 ± 12.38a22.45 (7.84; 37.05)0.2640.028
 High attendance20.84 ± 9.56†22.8 ± 9.2310.54 (−3.05; 24.13)
SRT Median (ms)
 Low attendance422.88 ± 185.10382.68 ± 152.47−5.59 (−15.39; 4.22)0.5670.008
 High attendance497.77 ± 231.88423.70 ± 190.15b−9.66 (−18.78; 0.54)
SRT DT Median (ms)
 Low attendance568.42 ± 145.29548.46 ± 113.87−2.69 (−13.95; 8.57)0.7760.002
 High attendance605.98 ± 206.79566.76 ± 163.01−0.36 (−11.08; 10.37)
CRT Correct Count (n)
 Low attendance38.54 ± 2.1738.86 ± 1.641.10 (−0.58; 2.78)0.6920.004
 High attendance39.16 ± 1.6039.48 ± 0.960.62 (−0.95; 2.18)
CRT Correct Median (ms)
 Low attendance733.42 ± 180.41711.18 ± 183.24−0.17 (−8.09; 7.75)0.4050.016
 High attendance776.62 ± 196.54811.86 ± 188.544.63 (−2.75; 12.00)
CRT DT Correct Count
 Low attendance38.08 ± 2.3438.50 ± 2.201.60 (−2.06; 5.26)0.9200.000
 High attendance38.48 ± 3.0238.88 ± 1.791.33 (−2.08; 4.74)
CRT DT Correct Median (ms)
 Low attendance847.04 ± 179.50805.09 ± 149.08−0.49 (−8.39; 7.41)0.5380.009
 High attendance870.04 ± 213.64897.00 ± 204.433.05 (−4.31; 10.40)

Note:

†

p < 0.05 between groups at baseline – Mann-Whitney U test.

a

p < 0.05 changes within the group by Paired T Test.

b

p < 0.05 changes within the group – Wilcoxon test.

‡

p-value based on ANCOVA; ηp2 partial eta-squared. CI= confidence interval; TMT = trail making test; SRT = simple reaction time; DT = dual task; CRT = choice reaction time.

Non-parametric partial correlations between the participants' attendance rate and the cognitive test scores are presented in Table 4. The higher attendance rate was positively associated with more time to complete the TMT– A (r = 0.273, p = 0.030). All other correlations were not statistically significant.

Table 4. Non-parametric partial correlation between attendance rate and change scores in the cognitive tests, controlling for years of education.

Cognitive variablesAttendance rate
TMT – A (s)0.273*
TMT – B (s)−0.160
Stroop (word) (n)−0.042
Stroop (color) (n)0.089
Stroop (word-color) (n)0.158
Stroop (Interference) (n)0.051
Digit Span Test (forward) (n)−0.031
Digit Span Test (backward) (n)0.055
Digit Span Test (total score) (n)−0.008
Semantic fluency (n)−0.113
Phonemic fluency (n)−0.167
SRT Median (ms)−0.198
SRT DT Median (ms)−0.075
CRT correct count (ms)−0.027
CRT correct Median (ms)0.125
CRT DT correct count−0.008
CRT DT correct Median (ms)0.048

Note:

⁎

p < 0.05; TMT = trail making test; SRT = simple reaction time; DT = dual task; CRT = choice reaction time.

Discussion

This study aimed to analyze the impact of a long-term MCEI on the cognitive functions on participants with T2DM. A sample of 70 individuals participated in the program consisting of a combination of aerobic training, strength, agility/balance and flexibility, which was structured in accordance to the international recommendations for physical activity and exercise for individuals with type 2 diabetes.33

Overall, the results showed that there was no significant difference in cognitive performance changes between the groups with higher and lower attendance rates to the MCEI. There were few positive changes within the groups from baseline to the 32-weeks, which, curiously, were more frequent for the group with low attendance rate (TMT-A, TMT-B, and phonemic fluency total score) than for the group with high attendance rate (simple reaction time). Also, the correlation analysis (which included all 70 participants) showed just one weak association between the performance in cognitive function and attendance rate. In this case, a higher number of presences in the exercise sessions was associated with worse performance in TMT-A.

The results found in the present study do not support previous evidence from the general physical activity literature that exercise has a positive effect on cognitive functioning.14,17,18 On the other hand, our results are aligned with the conclusions of a recent systematic review,15 according to which there is no consistent evidence that exercise interventions contribute to a better cognitive function in participants with T2DM. In particular, we found one study which reported that the cognitive performance of people with T2DM declined following a 24-week exercise program.20

Some reasons could account for the lack of effectiveness of the exercise program. In the present study, the MCEI was designed to provide moderate-intensity exercise (12–13 points on a scale of 6–20 points29), which was confirmed by the subjects' rating of perceived exertion (mean of 13 ± 1.3 points). This contrasts with a recent randomized controlled trial, which included higher levels of exercise intensity (for both the aerobic and resistance training components) that showed positive effects on the cognitive functioning of participants with T2DM.19 More studies are needed to verify if different levels of exercise intensity can have a differentiating effect in the cognitive functioning of T2DM patients. One should note, however, that vigorous-intensity exercise for T2DM patients requires a detailed pre-exercise medical clinical evaluation, including a cardiological stress test.11,34

Considering that the impact of PE on the brain and cognition varies as a participant´s age, we hypothesize that age heterogeneity of our sample (55–78 years) may have influenced the results. A highly-cited meta-analysis Colcombe and Kramer18 focused in a similar age range that our study, demonstrated significant discrepancies between the size effect of the PE on cognition when taking into consideration the different age groups (55–65 years, effect size=0.298; 66–70 years, effect size=0.693; 71–80 years, effect size=0.549).18 As older adults have greater cognitive limitations than middle-aged adults, they probably have more potential for cognitive improvement.35

The absence of significant improvement in cognitive outcomes in people with higher levels of adherence to the exercise program probably is also related to their baseline level of cognitive functioning. Thus, despite people with T2DM have a higher risk of cognitive impairment,36 the participants in the present study had normal general cognition level (evaluated by the MMSE).37 Furthermore, we have compared the participants’ level of performance in the cognitive tests at baseline with the normative values for the general population and concluded that they were quite normal. A previous study focused on participants with chronic systemic disorders, including diabetes mellitus, reported that there was no improvement in global cognitive functioning in people with normal cognitive status after their participation in a 3-month supervised resistance and aerobic exercise program.36 One should highlight, however, that the effectiveness of an intervention should not be measured only by the level of improvement of a number of cognitive variables. Hence, the maintenance of the cognitive status or the attenuation of expected cognitive losses in specific groups of the population (e.g., older adults, people with neurodegenerative disabilities) could also be an important goal in some contexts.

Epidemiological studies point to a higher prevalence of T2DM in the coming decades,38 and it is necessary to establish effective strategies to promote the general health of these patients. The preservation or improvement of cognitive functioning in people with T2DM is critical and translates into better self-care and performance of activities of daily living.7 Conversely, cognitive decline is an important factor for institutionalization and loss of quality of life.13 This situation can be aggravated in older people with T2DM due to their high risk of vascular brain damage and neurodegenerative change.39 In this context, it is important to test the potential of different forms of exercise in people with T2DM. Multimodal interventions could be a relevant alternative to more traditional forms of exercise focused on physical fitness components (like that of the current study) due to its potential for improving not only physical fitness but also cognitive functioning.40 Moreover, the nature of some of the activities of multimodal programs (e.g., learning tasks and collaborative tasks), could lead to an increase on the participants’ levels of satisfaction and motivation, which are crucial for maintaining their participation and boosting the program effects.41

In this regard, a previous research with people with T2DM that used a traditional exercise program (e.g., aerobic and resistance training) and did not found improvements in the participants’ cognitive status, hypothesized that the use of activities such as dance and movement games, which require cognitive effort, could have resulted in greater benefits in cognition.36 Also, another report showed that executive functions benefit from simultaneous cognitive–physical training compared to exclusively physical multicomponent training in healthy older adults.40 It should be noted that executive function is one of the cognitive domains most affected by T2DM6 and it is of great relevance for daily live managing, including glycemic control.9 Results of a recent study found that an 8-week multimodal exercise intervention improved visual attention, executive function, and information processing speed in older nursing home residents.27 In addition to benefiting elderly with normal cognitive status, some studies have also found positive effects of this type of exercise in persons with cognitive impairment.42, 43, 44, 45 It was previously shown that the exercise program (“Diabetes em Movimento”) used on the current study had positive effects in functional fitness, glycemic control, and cardiovascular risk.30,46 However, despite its merits (including its low implementation costs and the possibility to engage a large number of participants per session), we did not find significant improvements in cognitive functioning due to participation in the program. Hence, considering recent evidence of a broad positive impact of multimodal interventions,27,47 it will be important to test in the future if the inclusion in exercise programs for T2DM patients of activities dependent on cognitive skills (eventually in dual-task situations) may result in positive changes on cognitive functioning.

Limitations and strengths

This study has some limitations that must be considered. It was a quasi-experimental design that did not include a traditional control group and, instead, we choose to compare the outcome results between two groups that differed in their attendance rate to the program. Obviously, this means that some of the participants from the low-level attendance group also engaged in some of the sessions, which could have influenced the results. Nevertheless, from an ethical point of view, this seems a good alternative to having a control group that does not offer the choice of engaging in a potential healthy program. Also, the present study had a relatively small sample size which limits its power to detect differences between the groups. Another limitation of this study was the lack of measurement of the patterns of physical activity. Thus, it is possible that even the participants of the low-level adherence group (≤ 33th percentile) increased their daily-life physical activity level during the study due to possible changes in their lifestyle.48 Finally, there was an evident heterogeneity in the age (55–78 years) of the participants.

The strength of this study included the fact that the sample was recruited from three cities of Portugal. Also, this study was a long-term intervention with a combination of four different exercise types, developed with low-cost material. Finally, the cognitive assessment included a broad range of tests targeting several cognitive abilities.

Conclusion

Considering the high prevalence of T2DM and its disabling effects, the prevention of associated complications is a great challenge for public health. For physical exercise acting as an effective alternative to treatment and lifestyle change in this population, it requires a methodology that guarantees safety, social support, and promotes motivation and satisfaction. Although in the present study the level of adherence to the 32-week multicomponent exercise program (focused on physical stimulation) did not improve the cognitive functioning of middle-aged and older participants with T2DM, this not invalidate its relevance. This is because the PE effects on the cognitive functioning in these participants have been not yet well established, and also because the baseline level of cognitive functioning of the participants of the present study was mostly positive.

Considering previous evidence of the positive impact of the community-based exercise program (“Diabetes em Movimento”) used in the current study on health and quality of life, and also of positive effects of multimodal interventions in cognitive functioning, it is important to examine in the future the impact of incorporating dual-task activities (merging physical and cognitive stimulation) in similar interventions.

CRediT authorship contribution statement

Nilton João Chantre Leite: Visualization, Conceptualization, Supervision, Writing - original draft, Data Curation, Formal analysis, Writing - review & editing. Romeu Duarte Carneiro Mendes: Project administrarion, Formal analysis, Writing - review & editing. Armando Manuel Mendonça Raimundo: Formal analysis, Writing - review & editing. Cristina Pinho: Conceptualization, Formal analysis, Writing - review & editing. João L. Viana: Supervision, Writing - review & editing. José Francisco Filipe Marmeleira: Conceptualization, Writing - original draft, Formal analysis, Writing - review & editing.

Declaration of Competing Interest

The authors declare that there is no conflict of interest.

Acknowledgments

The authors thank the participants and the physical exercise professionals for their contributions to this work.

References

View Abstract
© 2020 Elsevier Inc. All rights reserved.