Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Feb 1.
Published in final edited form as: J Appl Gerontol. 2021 May 3;41(2):496–505. doi: 10.1177/07334648211010580

Cardio-Dance Exercise to Improve Cognition and Mood in Older African Americans: A Propensity-Matched Cohort Study

Bernadette A Fausto 1, Solaleh Azimipour 1, Lisa Charles 2, Christina Yarborough 1, Keyla Grullon 1, Emily Hokett 3, Paul R Duberstein 4, Mark A Gluck 1
PMCID: PMC8563498  NIHMSID: NIHMS1689194  PMID: 33938312

Abstract

The current study sought to determine the influence of initial sleep quality and body mass index on the cognitive and mood outcomes of a community-based cardio-dance exercise program. Thirty-two older African Americans who participated in a five-month cardio-dance exercise program were propensity-matched to 32 no-contact controls. Participants completed neuropsychological tests of attention, executive function, and memory and a self-reported depression measure at baseline and post-test. Among exercise participants, we observed significant improvements in depression (baseline=6.16±5.54, post-test=4.66±4.89, ηp2=.12, p=.009) and attention (baseline=40.53±14.01, post-test=36.63±13.29, ηp2=.12, p=.009) relative to controls. Improvements in executive function and attention were most pronounced among exercise participants with poor sleep quality (baseline=7.71±1.25, post-test=8.29±2.06, ηp2=.41, p=.04) and with obesity (baseline=38.05±12.78, post-test=35.67±13.82, ηp2 = .30, p=.001), respectively. This study provides novel evidence that exercise has the potential to improve depression in older African Americans. For those with poor sleep quality or obesity, exercise can also improve some cognitive outcomes.

Keywords: Cognition, Depression, African Americans, Exercise, Community, Sleep Deficiencies, Obesity

Introduction

Overview

Pathological cognitive decline (including Alzheimer’s disease [AD]) and depression are burgeoning, significant public health challenges (Alzheimer’s Association, 2019; World Health Organization, 2017). African Americans have disproportionately higher prevalence of AD and a more persistent and disabling course of depression as compared to non-Hispanic white Americans (Demirovic et al., 2003; Dunlop, Song, Lyons, Manheim, & Chang, 2003; Williams et al., 2007). Although the reasons for these health disparities are unknown, several health conditions, such as sleep deficiencies and obesity, are also more prevalent among African Americans and have been independently linked to cognitive and mood dysfunction (Ruiter, DeCoster, Jacobs, & Lichstein, 2010; E. Smith, Hay, Campbell, & Trollor, 2011; Tsuno, Besset, & Ritchie, 2005). Fortunately, regular aerobic exercise can improve cognition and mood in older adults and thus, may help mitigate racial disparities in AD and depression (National Academies of Sciences, 2017). However, little is known about who is likely to benefit the most from exercise interventions. By elucidating these subgroups, researchers and health professionals will be better positioned to define and target specific populations in efforts to prevent AD and depression in older African Americans.

Cognitive and Mood-Enhancing Benefits of Exercise

Regular aerobic exercise and physical activity are linked to cognitive and mood changes in older adulthood (e.g., Endeshaw & Goldstein, 2020; Erickson et al., 2011). Several randomized controlled trials of aerobic exercise training have shown restorative effects in the attention (Angevaren, Aufdemkampe, Verhaar, Aleman, & Vanhees, 2008; Langlois et al., 2013), executive function (Langlois et al., 2013), and memory (Erickson et al., 2011; Klusmann et al., 2010) domains (for a meta-analysis, see P. J. Smith et al., 2010). Exercise programs have also shown mood improvements on a range of psychological well-being outcomes in older adults (Cohen’s d = 0.19; Netz, Wu, Becker, & Tenenbaum, 2005) and in older adults with clinically significant depression, including those with treatment-resistant depression (Mura & Carta, 2013). However, few studies have examined the effectiveness of exercise interventions to improve mood exclusively in older African Americans (but see Nicolaidis, McKeever, & Meucci, 2013). Notably, African Americans have lower antidepressant medication use (Brody & Gu, 2020) and adherence (Kales et al., 2013), and are more receptive to non-pharmacological interventions for depression as compared to non-Hispanic whites (Dwight-Johnson, Sherbourne, Liao, & Wells, 2000). These differences in depression treatment may be due to medical mistrust stemming from historical exploitation of African Americans in medicine (Whaley, 2001) and cultural stigmatization of mental illness (Williams & Williams-Morris, 2000). Therefore, in addition to reducing stigma about mental illness and increasing accessibility, awareness, and use of mental health services in African American communities (Conner, Koeske, & Brown, 2009), developing tailored non-pharmacological approaches (e.g., exercise programs) to promote mental health for African Americans deserve further study.

Influences of Sleep Quality and Obesity on Exercise Outcomes

Individual health concerns, such as sleep deficiencies and obesity–both of which are more prevalent in African Americans than in the general population (Krueger & Reither, 2015; Sheehan, Frochen, Walsemann, & Ailshire, 2019)–may preclude regular exercise. Encouragingly, individuals with such health characteristics may be particularly responsive to exercise interventions. African Americans are more likely to have sleep deficiencies (e.g., poor sleep quality) even in the absence of a diagnosed sleep disorder (Jackson, Walker, Brown, Das, & Jones, 2020; Ruiter, DeCoster, Jacobs, & Lichstein, 2011). Adults with sleep deficiencies have poorer cognition (Dzierzewski, Dautovich, & Ravyts, 2018) and have greater odds of worsening depressive symptomology (Maglione et al., 2014). Similarly, obesity has been independently linked to cognitive (Arvanitakis, Capuano, Bennett, & Barnes, 2018) and mood dysfunction (Mansur, Brietzke, & McIntyre, 2015). Multi-component lifestyle interventions (e.g., combining exercise, diet, and behavioral techniques administered by health professionals) that target individuals with sleep complaints and obesity show promise in reducing risk of multiple morbidities in older adults (Andrieu et al., 2017). However, such lifestyle intervention programs are often labor-intensive, expensive, and may be difficult to employ in community settings with limited resources. Given the benefits of exercise, community-based exercise programs may be a practical, accessible method to improve cognitive and mood outcomes, particularly in subpopulations disproportionately affected by sleep deficiencies and obesity.

Purpose of the Current Study

Here, we examined the moderating influence of baseline sleep quality and body mass index (BMI) on the cognitive and mood outcomes of a community-based cardio-dance exercise program among older African Americans.

Method

Participants

Between March 2016 and July 2017, older African Americans from the Greater Newark, NJ area were recruited via a university-community partnership known as the Aging & Brain Health Alliance (for more details on recruitment, see Gluck, Shaw, & Hill, 2018 and refer to Supplemental Table 1). Established in 2006, the partnership comprises a growing network of university investigators; health organizations; and community members and leaders from local churches, public and subsidized housing management, senior centers, recreation centers, fitness companies, and other community organizations. Recently, the partnership has focused on bringing health value to greater Newark residents by offering free exercise classes within housing sites, senior centers, and churches. This exercise programming has also facilitated recruitment of older African Americans to participate in health-related research protocols, including the present community-based cardio-dance exercise study.

To participate in this study, candidates were required to be 60 years or older and have a Mini-Mental State Examination (MMSE) score of 24 or above, indicating no/low likelihood of cognitive impairment. Exclusion criteria were: a diagnosis of a neurological disorder or traumatic brain injury; diagnosis of any sleep disorders as the focus was on normative sleep quality; excessive alcohol and/or recreational drug use; currently taking medications typically prescribed for dementia (e.g., Razadyne, Namenda); planning to undergo a medical procedure which requires general anesthesia during the study period or had a medical procedure in the last three months requiring general anesthesia; and inability to see a computer screen at normal viewing distance.

Ninety-seven participants were screened, eleven of whom were excluded due to age (n=10) or low MMSE score (n=1). The remaining 86 participants self-selected to be enrolled in the exercise intervention (n = 38) or in the no-contact control group (n = 48). This non-randomized trial was considered formative research to achieve the following: 1) establish and maintain mutual trust and understanding between university investigators and community members, leaders, and partners of the Aging & Brain Health Alliance; and 2) to inform future randomized clinical trial designs for exercise to improve health outcomes in African American communities. Written informed consent was obtained from all participants, and the protocol was approved by the Rutgers University institutional review board.

Procedure

Participants were telephone screened for initial eligibility. Potentially eligible candidates were invited to attend a baseline lab visit to confirm eligibility and undergo a two-hour testing battery including cognitive testing, physical performance measures, and health, lifestyle, and mood questionnaires (see Measures section). After baseline testing, participants elected to be in the exercise group or the no-contact control group. Exercise group participants attended twice-weekly, 60-minute cardio-dance exercise classes while the control group received no contact from study staff over the ensuing five months. After the intervention period (or equivalent no contact period), participants returned to the lab to repeat the testing battery.

Measures

The following standardized and widely used measures of mental status, literacy, cognition, and mood were administered: Mini-Mental State Examination (MMSE; Folstein, Folstein, & McHugh, 1975); North American Adult Reading Test (NAART; Uttl, 2002); WAIS-IV Digit Span subtest (Wechsler, 2008); Trail-Making Test- Part A & B (TMT-A and TMT-B; Reitan, 1979); Rey Auditory Verbal Learning Test (RAVLT; Rey, 1964); Beck Depression Inventory-II (BDI-II; Beck, Steer, & Brown, 1996). For more details on each measure, see Appendix.

Sleep Quality.

Baseline sleep quality was examined as a potential moderator of intervention effects on cognitive performance and mood. The variable is self-reported using a single item, “Rate your usual sleep quality” with response choices on a Likert scale from 1 = “very poor” to 5 = “very good”. Due to the low frequency of individuals reporting very poor and very good sleep quality, very poor and poor categories and good and very good categories were collapsed, resulting in three dummy-coded categories of sleep quality: 0 = “very poor and poor sleep quality”, 1 = “satisfactory sleep quality”, and 2 = “good and very good sleep quality.”

Body Mass Index (BMI).

A potential moderator of exercise-related cognitive and mood effects, baseline BMI is computed as the ratio of objectively-measured weight to height and expressed as kg/m2 (World Health Organization, 2006). Participants were classified into three BMI groups: normal (18.5 to 24.9 kg/m2), overweight (25 kg/m2 to 29.9 kg/m2), obese: ≥ 30 kg/m2.

Other Anthropomorphic Measurements.

In addition to BMI, participants’ resting blood pressure (both systolic and diastolic) and resting heart rate in a sitting position were recorded. Participants were asked to refrain from drinking caffeinated beverages before their testing visits (i.e., no more than one cup of caffeinated coffee within 30 minutes of testing start) to minimize measurement error (American Heart Association, 2018).

Physical Performance Measures.

The Six-Minute Walk Test (Steffen, Hacker, & Mollinger, 2002), Short Physical Performance Battery (Studenski et al., 2003), and Timed Up & Go (Steffen et al., 2002) were also administered to characterize participants’ aerobic fitness and physical function. For more details, see Appendix.

Intervention

The cardio-dance exercise intervention was coordinated by a local fitness and wellness consultancy and implemented by certified group exercise instructors. Two administrative assistants accompanied each class. Participants attended exercise classes two times a week (approximately 60 minutes per session) over five months for a total of 40 sessions in groups of up to 15 participants. These exercise classes involved learning an evolving series of cardio-dance routines and step sequences set to American popular and Latin music. Dance routines included steps from hip hop, merengue, samba, cumbia, and salsa. Each session started with 10 minutes of warm-up followed by 45 minutes of cardio-dance exercise and five minutes of cool down. Each session included a planned break at the midpoint, but participants were allowed to take as many breaks as needed.

During the 45-minute cardio-dance exercise portion, participants were encouraged to maintain a Borg 10-Point Rating of Perceived Exertion (RPE; Borg, 1982) of 4 to 6 (“somewhat hard” to “hard”) and achieve an intensity of 65–80% of their individual heart rate (HR) reserve (for male participants, calculated as 220 – age = maximum HR [max HR]; for women, calculated as 206 – .88 × [age of participant] = max HR as the traditional male-specific calculations overestimate the maximum HR for age in women; Max HR × .65 = target lower bound HR; Max HR × .80 = target upper bound HR) (Fox III, Naughton, & Haskell, 1971; Gulati et al., 2010). During the planned break, the administrative assistants logged HRs and RPEs for each participant, ensuring both were in the target intensity zone (see Appendix for additional details). Attendance records were maintained for each participant. Instructors and administrative assistants were trained to record and report any adverse events.

Propensity-Score Matching Procedure

Because the participants were not randomly assigned to groups, propensity-score matching was used to account for confounding and self-selection biases (D’Agostino Jr., 1998). Each participant in the exercise group was matched to one control using a 1:1 propensity score matching procedure without replacement and a matching tolerance set to 0.10. The following baseline variables were entered as predictors of group membership (0=control group, 1=exercise group): age, sex, education, MMSE, NAART, Digit Span Total, TMT-A, TMT-B, RAVLT learning total, RAVLT short delay recall, RAVLT long delay recall, resting systolic and diastolic blood pressure, resting heart rate, BDI-II, sleep quality, BMI, VO2max estimate (derived from their Six-Minute Walk Test performance; see Appendix), Short Physical Performance Battery, and Timed Up & Go. A propensity score was computed from the logistic regression equation for each participant, which indicated the probability that a participant would be in the exercise group (D’Agostino Jr., 1998).

Statistical Analyses

Contingent on variable scaling, parametric and non-parametric tests were conducted to compare the exercise group and the propensity-matched control group on baseline demographics, cognition, depressive symptomology, physical performance, and anthropomorphic measurements.

Repeated measures analysis of covariance tested for interactions of group × time, sleep quality × group × time and BMI × group × time on one mood outcome (BDI-II) and eight cognitive outcomes: 1) Digit Span Forward, 2) Digit Span Backward, 3) Digit Span Sequencing, 4) TMT-A, 5) TMT-B, 6) RAVLT Learning Total, 7) RAVLT Short Delay, and 8) RAVLT Long Delay. Partial eta-square effect sizes (ηp2) were categorized as small (0.01), medium (0.06), and large (0.14), respectively (Cohen, 2013).

The Benjamini-Hochberg procedure was applied to control the false discovery rate (FDR) for multiple hypotheses testing (Benjamini & Hochberg, 1995). Statistical analyses were performed using IBM SPSS® Statistics for Mac, version 26 (IBM Corp., Armonk, N.Y., USA).

Power Analysis

A power analysis using G*Power software version 3.1.9.6 (Faul, Erdfelder, Buchner, & Lang, 2009) determined that a sample size of 64 has at least 95% power to detect a medium effect size (Cohen’s f2 = 0.25) for a significant group (exercise vs. control) × time (baseline, post-test) interaction (α = .05, two-tailed). The sample size of 64 also has more than 80% power to detect a medium effect size (Cohen’s f2 = 0.25) for a significant three-way interaction with five groups (three sleep quality groups [poor, satisfactory, good] or three BMI groups [normal: 18.5 to 24.9 kg/m2, overweight: 25 kg/m2 to 29.9 kg/m2, obese: ≥ 30 kg/m2 and two intervention groups [exercise, control]) and two measurements (baseline, post-test) with α = .05 (two-tailed).

Results

Eighty-six participants were subjected to propensity-score matching. The procedure yielded an analytic N of 64 participants (n = 32 exercise participants; n = 32 propensity-matched controls). The 22 excluded participants did not differ from the 64 included participants on demographic, cognitive, physical, or anthropomorphic measures, ps > .05, except for MMSE, p = .03, and resting heart rate, p = .03 (see Supplemental Table 2). There were no changes in sleep quality or BMI from baseline to post-test for either intervention group, ps > .05. There were no adverse events.

Baseline Characteristics

Descriptive statistics of the analytic sample (N = 64) by exercise and control group are reported in Table 1. The exercise group had higher (p = .04) MMSE scores (M = 28.44, SD = 1.34) as compared to the control group (M = 27.72, SD = 1.40) so we controlled for this variable in all outcome analyses. There were no other baseline differences between groups on demographic, cognitive, physical, or anthropomorphic measures (ps > .05). On average, the exercise group completed 33.70 (SD = 7.50) of 40 total exercise sessions.

Table 1.

Summary Statistics for Analytic Sample by Group (N = 64).

Measure Cardio-Dance Exercise Group (n = 32) Propensity-Matched Control Group (n = 32)
Baseline Post-Test Baseline Post-Test
M (n) SD (%) M (n) SD (%) M (n) SD (%) SD (%) SD (%)
Age (years) 69.50 7.18 -- -- 68.31 6.24 -- --
Sex (female) (29) (90.63) -- -- (23) (71.88) -- --
Education (years) 14.05 1.94 -- -- 13.84 2.22 -- --
NAART (errors) 35.47 11.06 -- -- 36.94 9.46 -- --
MMSE (24–30 points)* 28.44 1.34 28.09 1.30 27.72 1.40 27.97 1.60
Digit Span Forward 9.38 1.74 9.56 1.78 9.53 2.49 9.38 1.91
Digit Span Backward 7.59 1.39 7.91 1.67 7.81 2.02 7.97 1.68
Digit Span Sequencing 7.22 1.39 6.91 1.35 6.88 1.91 7.00 1.84
Trail-Making Test – Part A (s) 40.53 14.01 36.63 13.29 38.47 14.23 40.84 16.92
Trail-Making Test – Part B (s) 138.25 53.60 122.31 72.05 150.00 82.62 115.50 41.65
RAVLT – Learning Total 43.66 7.56 47.19 10.31 41.31 9.64 43.78 10.02
RAVLT – Short Delay 8.16 3.10 9.56 3.11 7.28 2.97 8.41 3.28
RAVLT – Long Delay 8.22 3.14 9.19 3.57 7.19 3.56 8.06 3.22
Beck Depression Inventory-II 6.16 5.54 4.66 4.89 6.53 3.95 7.03 5.47
Body Mass Index (log10) 1.51 0.08 1.48 0.08 1.48 0.22 1.46 0.09
Sleep Quality
Very Poor/Poor (7) (21.90) (9) (28.10) (4) (12.50) (6) (18.80)
Satisfactory (15) (46.90) (11) (34.40) (4) (43.80) (14) (43.80)
Good/Very Good (10) (31.30) (12) (37.50) (14) (43.80) (12) (37.50)
Resting Systolic Blood Pressure (mmHg) 139.22 20.16 135.81 19.53 143.09 17.27 139.66 15.75
Resting Diastolic Blood Pressure (mmHg) 77.53 11.89 76.19 18.49 80.84 11.15 80.03 8.79
Resting Heart Rate (bpm) 68.41 12.80 70.59 11.43 68.72 8.31 70.22 10.34
VO2max Estimate 15.03 2.69 15.46 1.94 14.53 2.58 15.23 2.57
Short Physical Performance Battery 9.44 1.98 9.78 1.31 9.63 1.86 9.66 1.43
Timed Up and Go 10.32 2.27 10.74 2.33 10.51 2.35 10.87 2.73
Open in a new tab

Note.

*

Significant baseline differences at ps < .05 by group as indicated by independent samples t-tests, Mann-Whitney U tests, or chi-square tests for independence, contingent on variable scaling. NAART=North American Adult Reading Test, an assessment of literacy level; MMSE = Mini-Mental State Examination, a measure of gross mental status; Digit Span Forward, Backward & Sequencing, a subtest of the Wechsler Adult Intelligence Scale-Fourth Edition (WAIS-IV) used to assess attention, executive function, and working memory; Trail-Making Test – Part A, a measure of visual attention and processing speed; Trail-Making Test – Part B, a measure of executive function including set shifting and mental flexibility; RAVLT=Rey Auditory Verbal Learning Test, a word learning test to assess episodic memory; Beck Depression Inventory-II, measure of depressive symptomology; VO2max Estimate, a measure of aerobic fitness derived from the Six-Minute Walk Test; Short Physical Performance Battery, a measure of general physical performance; Timed Up and Go, a measure of physical function related to fall risk and balance.

Cognitive Outcomes

Digit Span.

For Digit Span Forward and Digit Span Sequencing subtests, there were no significant interactions or main effects. For Digit Span Backward, there was a significant sleep quality × group × time interaction, F(2, 53) = 6.42, p = .003, FDR-corrected pcritical = .009, ηp2 = .20 (see Figure 1).

Figure 1.

Figure 1.

Open in a new tab

Significant Sleep Quality × Group × Time interaction on Digit Span Backward performance, controlling for baseline Mini-Mental State Examination score: The exercise group with poor sleep quality remained relatively stable (blue line in Panel A) while the control group with poor sleep quality declined on Digit Span Backward performance (blue line in Panel B).

a) Exercise group: Digit Span Backward scores at baseline and post-test stratified by baseline sleep quality.

b) Control group: Digit Span Backward scores at baseline and post-test stratified by baseline sleep quality.

Higher scores indicate better performance. Error bars represent +/− 2 SEs.

The significant three-way interaction for Digit Span Backward was followed by simple effects mixed ANCOVAs at each sleep quality level, controlling for baseline MMSE, with intervention group as the between-subjects factor and time as the within-subjects factor. Among individuals with poor sleep quality, there was a significant group × time interaction, F(1, 8) = 9.36, p = .04, FDR-corrected pcritical = .06, ηp2 = .41. This interaction was driven by a decline in Digit Span Backward performance from baseline 10.50 (SD = 2.52) to post-test 8.00 (SD = 1.63) observed in the control group with poor sleep quality. The exercise group remained largely stable in Digit Span Backward performance with mean scores of 7.71 (SD = 1.25) at baseline and 8.29 (SD = 2.06) at post-test. ANCOVAs for satisfactory and good sleep quality levels did not show any significant interactions or main effects, ps > .05.

Trail-Making Test.

For TMT-A, there was a significant group × time interaction, F(1, 53) = 7.37, p = .009, FDR-corrected pcritical = .03, ηp2 = .12. Whereas the exercise group improved on TMT-A performance from baseline (M = 40.53, SD = 14.01) to post-test (M = 36.63, SD = 13.29), the control group remained relatively stable (see Figure 2a).

Figure 2.

Figure 2.

Open in a new tab

(A) Significant Group × Time interaction on Trail-Making Test—A performance, controlling for baseline Mini-Mental State Examination score: The exercise group improved on TMT-A performance from baseline to post-test (red line in Panel A) while the control group remained relatively stable (blue line in Panel A).

(B and C) Significant BMI × Group × Time interaction on Trail-Making Test—A performance, controlling for baseline Mini-Mental State Examination score. In the normal BMI group (blue lines in Panels B and C), whereas the exercise group improved on Trail-Making Test—A performance, the control group remained stable. In the obese BMI group (green lines in Panels B and C), whereas the exercise group remained stable, the control group worsened on Trail-Making Test—A performance.

Lower scores indicate better performance. Error bars represent +/− 2 SEs.

There was also a significant BMI × group × time interaction for TMT-A performance, F(2, 53) = 3.20, p = .04, FDR-corrected pcritical = .04, ηp2 = .11 (see Figures 2b and 2c). The significant BMI × group × time interaction was followed by simple effects mixed ANCOVAs at each BMI level with intervention group as the between-subjects factor, time as the within-subjects factor, and MMSE as a covariate. Among participants in the normal BMI group, there was a significant group × time interaction, F(1, 10) = 8.99, p = .01, FDR-corrected pcritical = .02, ηp2 = .47. Whereas the control group demonstrated no change in TMT-A performance, the exercise group improved from baseline (M = 69.50, SD = 2.12) to post-test (M = 47.00, SD = 7.07). Among participants in the obese BMI group, there was a significant group × time interaction, F(1, 29) = 12.30, p = .001, FDR-corrected pcritical = .04, ηp2 = .30. While the control group demonstrated worse TMT-A performance from baseline (M = 38.82, SD = 15.34) to post-test (M = 47.82, SD = 17.33), the exercise group showed small improvements over time (baseline: M = 38.05, SD = 12.78; post-test: M = 35.67, SD = 13.82). The follow-up ANCOVA for the overweight BMI group did not show any interactions or main effects.

For TMT-B, there was only a significant main effect of time, F(1, 53) = 4.13, p = .05, ηp2 = .07. There was a significant improvement in TMT-B performance for the overall sample, t(63) = 3.19, p = .002. There were no significant interactions.

RAVLT.

Because of the high multicollinearity between RAVLT learning total, RAVLT short delay, and RAVLT long delay (rs > .70; see Supplemental Table 3), these outcomes were subjected to repeated measures multivariate ANCOVA with baseline MMSE as a covariate. There were no significant interactions or main effects of group or time on RAVLT performance, ps > .05, FDR-corrected pscritical not significant.

Mood Outcome

BDI-II.

For BDI-II, there was a significant group × time interaction, F(1, 53) = 7.42, p = 009, FDR-corrected pcritical = .02, ηp2 = .12 (see Figure 3). Whereas the control group remained largely stable in depressive symptomology from baseline (M = 6.53, SD = 3.95) to post-test (M = 7.03, SD = 5.47), the exercise group showed a decrease from baseline (M = 6.16, SD = 5.54) to post-test (M = 4.66, SD = 4.89). There were no significant interactions by BMI or sleep quality.

Figure 3.

Figure 3.

Open in a new tab

Significant Group × Time interaction on Beck Depression Inventory-II, controlling for baseline Mini-Mental State Examination score: Whereas the control group remained largely stable (blue line), the exercise group showed a decrease in depressive symptomatology (red line).

Lower scores indicate lower depressive symptomology. Error bars represent +/− 2 SEs.

Discussion

The present study found that the cardio-dance exercise group demonstrated improvements in attention and reductions in depressive symptomology relative to the control group. Of greater significance, this study also found several novel interactions among sleep quality, BMI, and exercise in relation to cognitive performance. Specifically, poor sleeping participants in the control group demonstrated executive function decline, while those in the exercise group showed modest improvements. Additionally, participants who were classified as obese in the control group showed a decline in visual attention and processing speed, while those in the obese group in the exercise condition showed small improvements. Thus, exercise may be particularly neuroprotective for relatively healthy older African Americans with poor sleep quality and high BMIs.

The exercise-related improvements in cognition and mood are consistent with prior research (Langlois et al., 2013; Netz et al., 2005; Stephens, 1988), although the existing evidence for cognitive benefits is less robust (van Uffelen, Paw, Hopman-Rock, & van Mechelen, 2008; Vidoni et al., 2021). The favorable results here might be related to the group setting and the cardio-dance exercise format. Group workouts facilitate peer support and promote exercise adherence (Kritz, Thøgersen-Ntoumani, Mullan, Stathi, & Ntoumanis, 2020) and may thereby increase the salutary cognitive and mood effects of exercise. Further, unlike the more common solo aerobic exercise regimens designed to improve cardiorespiratory fitness (e.g., treadmill walking), dancing is both physically and cognitively demanding (Predovan, Julien, Esmail, & Bherer, 2019). Indeed, emerging research suggests exercise that promotes coordination learning (i.e., dance/movement) improves cognition and mobility in the absence of aerobic fitness improvements (Esmail et al., 2020; Sinha, Berg, Yassa, & Gluck, 2021). This study also provides empirical evidence that group-based cardio-dance exercise may be a feasible, effective, and appealing form of exercise for community-dwelling older African Americans. In a recent focus group study conducted among community-dwelling older African Americans, dance was the preferred form of exercise and the majority expressed a desire to exercise in groups (Gothe & Kendall, 2016). Future research should determine the ideal dose (i.e., duration, frequency, and intensity) of group-based cardio-dance exercise to optimize both cognitive and mood outcomes in this population.

This is the first study to examine whether initial sleep quality and BMI influence responsiveness to an exercise intervention in older African Americans. While exercise-induced sleep quality improvements and BMI reductions may mediate cognitive improvements in older adults (Stillman, Cohen, Lehman, & Erickson, 2016), our findings highlight the importance of assessing sleep quality and BMI as moderators of intervention response on cognitive outcomes. Considering that both poor sleep quality and obesity disproportionately affect African Americans and are often related to lower activity levels (Lentino, Purvis, Murphy, & Deuster, 2013), the combined influence of these factors could result in accelerated cognitive decline in this population. Given the mixed findings of exercise efficacy to improve cognitive outcomes (van Uffelen et al., 2008; Vidoni et al., 2021), more research is needed to identify additional lifestyle and health moderators of intervention response. For example, some evidence suggests individuals with more cardiovascular risk factors experience greater benefits from lifestyle interventions (Chhetri et al., 2018). Elucidating these subgroup effects may help researchers and health professionals target high-risk participants and patients who would likely benefit the most from exercise interventions.

Study limitations include lack of random assignment and an active control group. Effects of self-selection bias were mitigated by propensity-score matching methods (D’Agostino Jr., 1998). However, given the employment of a no-contact control group, the observed cognitive and mood effects might be attributed to both the effect of exercise and the social interaction afforded by the cardio-dance exercise classes in group format. Indeed, research suggests social support is a mediator of physical activity change (Murrock & Madigan, 2008). Although pulse self-palpation for HRs can be readily learned by older adults and are generally accurate (Jaakkola, Virtanen, Vasankari, Salminen, & Airaksinen, 2017), future studies should employ HR monitors to further ensure fidelity of the intervention. Finally, sleep quality was assessed with a single self-report item. Future research should aim to use both subjective and objective (e.g., actigraphy or polysomnography) measures of sleep quality to better assess the interrelation among sleep quality, exercise, and cognitive performance. Despite these limitations, formative, minimally-invasive, non-randomized trials may be helpful by familiarizing communities typically underrepresented in research for future randomized controlled trial protocols. Specifically, the present study could serve as a model for designing more rigorous randomized controlled trials employing dance-based exercise interventions among older African Americans.

Conclusion

Increasing exercise participation and physical activity is an important factor for maintaining cognitive and mental health in later life. Lack of exercise may accelerate cognitive decline among individuals with poor sleep quality or with BMIs in the obese range. While health professionals are trained to monitor patients’ weight, the quality of patients’ sleep is often overlooked in medical training (Gamaldo & Salas, 2008). Similarly, researchers, health professionals, and community leaders employing exercise interventions to improve cognitive health among older samples should especially target individuals with poorer sleep quality and higher BMI. Finally, broadly translating less conventional forms of exercise, such as dancing in groups, may help promote physical activity, cognitive health, and mental health among older African Americans, thereby improving public health in their communities.

Supplementary Material

Supplemental Materials

Acknowledgments

The authors would like to thank Delores Hammonds and Mildred Evans for their contributions to our community education, outreach, and recruitment efforts. Glenda Wright (New Jersey Association of Public and Subsidized Housing; A&BHA Director of Community Engagement for Public and Subsidized Housing) and Rev. Dr. Glenn Wilson (Pilgrim Baptist Church; A&BHA Director of Church Relations) for their leadership in community engagement.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by grants to MAG from the National Institutes of Health-National Institute on Aging (1R01AG053961), Office of Minority Health at HHS and the NJ Department of Health’s Office of Minority and Multicultural Health (MH-STT-15–001, OMMH21HDP002), and by support from the Chancellor’s and Provost’s offices at Rutgers University–Newark

Footnotes

Declaration of Conflicting Interests

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

IRB Approval Committee

Rutgers University Arts & Sciences IRB, New Brunswick, NJ

IRB Protocol/Human Subjects Approval Number

#Pro2020000127

Supplemental Material

Supplemental material for this article is available online.

References

  1. Alzheimer’s Association. (2019). 2019 Alzheimer’s disease facts and figures. Alzheimer’s & Dementia, 15(3), 321–387. [Google Scholar]
  2. American Heart Association. (2018). Steps for accurate BP measurement. Retrieved from Dallas, TX:
  3. Andrieu S, Guyonnet S, Coley N, Cantet C, Bonnefoy M, Bordes S, . . . MAPT Study Group. (2017). Effect of long-term omega 3 polyunsaturated fatty acid supplementation with or without multidomain intervention on cognitive function in elderly adults with memory complaints (MAPT): a randomised, placebo-controlled trial. The Lancet Neurology, 16(5), 377–389. [DOI] [PubMed] [Google Scholar]
  4. Angevaren M, Aufdemkampe G, Verhaar HJJ, Aleman A, & Vanhees L (2008). Physical activity and enhanced fitness to improve cognitive function in older people without known cognitive impairment. Cochrane Database of Systematic Reviews, 3. doi: 10.1002/14651858.CD005381.pub3 [DOI] [PubMed] [Google Scholar]
  5. Arvanitakis Z, Capuano AW, Bennett DA, & Barnes LL (2018). Body mass index and decline in cognitive function in older black and white persons. The Journals of Gerontology: Series A, 73(2), 198–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Beck AT, Steer RA, & Brown G (1996). Manual for the Beck Depression Inventory-II. San Antonio, TX: Psychological Corporation. [Google Scholar]
  7. Benjamini Y, & Hochberg Y (1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistics Society, 57, 289–300. [Google Scholar]
  8. Borg GAV (1982). Psychophysical bases of perceived exertion. Medicine & Science in Sports & Exercise, 14, 377–381. [PubMed] [Google Scholar]
  9. Brody DJ, & Gu Q (2020). Antidepressant use among adults: United States, 2015–2018. Retrieved from Hyattsville, MD: [PubMed]
  10. Chhetri JK, de Souto Barreto P, Cantet C, Pothier K, Cesari M, Andrieu S, . . . Vellas B (2018). Effects of a 3-year multi-domain intervention with or without omega-3 supplementation on cognitive functions in older subjects with increased CAIDE dementia scores. Journal of Alzheimer’s Disease, 61(1), 71–78. doi: 10.3233/JAD-180209 [DOI] [PubMed] [Google Scholar]
  11. Cohen J (2013). Statistical power analysis for the behavioral sciences.: Academic Press. [Google Scholar]
  12. Conner KO, Koeske G, & Brown C (2009). Racial differences in attitudes toward professional mental health treatment: The mediating effect of stigma. Journal of Gerontological Social Work, 52(7), 695–712. [DOI] [PubMed] [Google Scholar]
  13. D’Agostino RB Jr. (1998). Propensity score methods for bias reduction in the comparison of a treatment to non-randomized control group. Statistics in Medicine, 17(19), 2265–2281. doi: [DOI] [PubMed] [Google Scholar]
  14. Demirovic J, Prineas R, Loewenstein D, Bean J, Duara R, Sevush S, & Szapocznik J (2003). Prevalence of dementia in three ethnic groups: The South Florida program on aging and health. Annals of Epidemiology, 13(6), 472–478. [DOI] [PubMed] [Google Scholar]
  15. Dunlop DD, Song J, Lyons JS, Manheim LM, & Chang RW (2003). Racial/ethnic differences in rates of depression among preretirement adults. American Journal of Public Health, 93(11), 1945–1952. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dwight-Johnson M, Sherbourne CD, Liao D, & Wells KB (2000). Treatment preferences among depressed primary care patients. Journal of General Internal Medicine, 15(8), 527–534. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Dzierzewski JM, Dautovich N, & Ravyts S (2018). Sleep and cognition in older adults. Sleep Medicine Clinics, 13(1), 93–106. doi: 10.1016/j.jsmc.2017.09.009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Endeshaw Y, & Goldstein F (2020). Association between physical exercise and cognitive function among community-dwelling older adults. Journal of Applied Gerontology. doi: 10.1177/0733464820952242 [DOI] [PubMed] [Google Scholar]
  19. Erickson KI, Voss MW, Prakash RS, Basak C, Szabo A, Chaddock L, . . . Kramer AF (2011). Exercise training increases size of hippocampus and improves memory. Proceedings of the National Academy of Sciences, 108(7), 3017–3022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Esmail A, Vrinceanu T, Lussier M, Predovan D, Berryman N, Houle J, . . . Bherer L (2020). Effects of dance/movement training vs. aerobic exercise training on cognition, physical fitness and quality of life in older adults: A randomized controlled trial. . Journal of Bodywork and Movement Therapies, 24(1), 212–220. [DOI] [PubMed] [Google Scholar]
  21. Faul F, Erdfelder E, Buchner A, & Lang AG (2009). Statistical power analyses using G* Power 3.1: Tests for correlation and regression analyses. Behavior Research Methods, 41(4), 1149–1160. [DOI] [PubMed] [Google Scholar]
  22. Folstein MF, Folstein SE, & McHugh PR (1975). “Mini-mental state”: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189–198. [DOI] [PubMed] [Google Scholar]
  23. Fox III SM, Naughton JP, & Haskell WL (1971). Physical activity and the prevention of coronary heart disease. Annals of Clinical Research, 3, 404–432. [PubMed] [Google Scholar]
  24. Gamaldo CE, & Salas RE (2008). Sleep medicine education: Are medical schools and residency programs napping on the job? Nature Clinical Practice Neurology, 4, 344–345. [DOI] [PubMed] [Google Scholar]
  25. Gluck MA, Shaw A, & Hill D (2018). Recruiting older African Americans to brain health and aging research through community engagement. Generations, 42(2), 78–82. [PMC free article] [PubMed] [Google Scholar]
  26. Gothe NP, & Kendall BJ (2016). Barriers, motivations, and preferences for physical activity among female African American older adults. Gerontology and Geriatric Medicine, 2, 1–8. doi: 10.1177/2333721416677399 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Gulati M, Shaw LJ, Thisted RA, Black HR, Bairey Merz CN, & Arnsdorf MF (2010). Heart rate response to exercise stress testing in asymptomatic women. Circulation, 122(2), 130–137. doi: 10.1161/CIRCULATIONAHA.110.939249 [DOI] [PubMed] [Google Scholar]
  28. Jaakkola J, Virtanen R, Vasankari T, Salminen M, & Airaksinen KJ (2017). Self-detection of atrial fibrillation in an aged population: Three-year follow-up of the LietoAF intervention study. BMC Geriatrics, 17(1), 1–8. doi: 10.1186/s12877-017-0607-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Jackson CL, Walker JR, Brown MK, Das R, & Jones NL (2020). A workshop report on the causes and consequences of sleep health disparities. Sleep. doi: 10.1093/sleep/zsaa037 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Kales HC, Nease DE Jr, Sirey JA, Zivin K, Kim HM, Kavanagh J, . . . Blow FC (2013). Racial differences in adherence to antidepressant treatment in later life. The American Journal of Geriatric Psychiatry, 21(10), 999–1009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Klusmann V, Evers A, Schwarzer R, Schlattmann P, Reischies FM, Heuser I, & Dimeo FC (2010). Complex mental and physical activity in older women and cognitive performance: A 6-month randomized controlled trial. Journals of Gerontology Series A: Biomedical Sciences and Medical Sciences, 65(6), 680–688. [DOI] [PubMed] [Google Scholar]
  32. Kritz M, Thøgersen-Ntoumani C, Mullan B, Stathi A, & Ntoumanis N (2020). It’s better together”: A nested longitudinal study examining the benefits of walking regularly with peers versus primarily alone in older adults. Journal of Aging and Physical Activity, 1, 1–11. doi: 10.1123/japa.2020-0091 [DOI] [PubMed] [Google Scholar]
  33. Krueger PM, & Reither EN (2015). Mind the gap: race/ethnic and socioeconomic disparities in obesity. Current Diabetes Reports, 15(11). doi: 10.1007/s11892-015-0666-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Langlois F, Vu TTM, Chassé K, Dupuis G, Kergoat MJ, & Bherer L (2013). Benefits of physical exercise training on cognition and quality of life in frail older adults. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 68(3), 400–404. [DOI] [PubMed] [Google Scholar]
  35. Lentino CV, Purvis DL, Murphy KJ, & Deuster PA (2013). Sleep as a component of the performance triad: The importance of sleep in a military population. US Army Medical Department Journal. [PubMed] [Google Scholar]
  36. Maglione JE, Ancoli-Israel S, Peters KW, Paudel ML, Yaffe K, Ensrud KE, & Stone KL (2014). Subjective and objective sleep disturbance and longitudinal risk of depression in a cohort of older women. Sleep, 37(7), 1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Mansur RB, Brietzke E, & McIntyre RS (2015). Is there a “metabolic-mood syndrome”? A review of the relationship between obesity and mood disorders. Neuroscience & Biobehavioral Reviews, 52, 89–104. [DOI] [PubMed] [Google Scholar]
  38. Mura G, & Carta MG (2013). Physical activity in depressed elderly: A systematic review. Clinical Practice & Epidemiology in Mental Health, 9, 125–135. doi: 10.2174/1745017901309010125 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Murrock CJ, & Madigan E (2008). Self-efficacy and social support as mediators between culturally specific dance and lifestyle physical activity. Research and Theory for Nursing Practice, 22(3), 192–204. [PMC free article] [PubMed] [Google Scholar]
  40. National Academies of Sciences, E., and Medicine,. (2017). Preventing cognitive decline and dementia: A way forward. Retrieved from Washington, DC: [PubMed]
  41. Netz Y, Wu MJ, Becker BJ, & Tenenbaum G (2005). Physical activity and psychological well-being in advanced age: a meta-analysis of intervention studies. Psychology and Aging, 20(2), 272–284. [DOI] [PubMed] [Google Scholar]
  42. Nicolaidis C, McKeever C, & Meucci S (2013). A community-based wellness program to reduce depression in African-Americans: Results from a pilot-intervention. Progress in Community Health Partnerships: Research, Education, and Action, 7(2), 145–152. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Predovan D, Julien A, Esmail A, & Bherer L (2019). Effects of dancing on cognition in healthy older adults: A systematic review. Journal of Cognitive Enhancement, 3(2), 161–167. doi: 10.1007/s41465-018-0103-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Reitan RM (1979). Trail making test: TMT. Testzentrale. [Google Scholar]
  45. Rey A (1964). L’examen clinique en psychologie: Presses Universitaires de France. [Google Scholar]
  46. Ruiter ME, DeCoster J, Jacobs L, & Lichstein KL (2010). Sleep disorders in African Americans and Caucasian Americans: A meta-analysis. Behavioral Sleep Medicine, 8(4), 246–259. [DOI] [PubMed] [Google Scholar]
  47. Ruiter ME, DeCoster J, Jacobs L, & Lichstein KL (2011). Normal sleep in African-Americans and Caucasian-Americans: a meta-analysis. Sleep Medicine, 12(3), 209–214. [DOI] [PubMed] [Google Scholar]
  48. Sheehan CM, Frochen SE, Walsemann KM, & Ailshire JA (2019). Are US adults reporting less sleep?: Findings from sleep duration trends in the National Health Interview Survey, 2004–2017. Sleep, 42(2), 1–8. doi: 10.1093/sleep/zsy221 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Sinha N, Berg CN, Yassa MA, & Gluck MA (2021). Increased dynamic flexibility in the medial temporal lobe network following an exercise intervention mediates generalization of prior learning. Neurobiology of Learning and Memory, 177, 107340. doi: 10.1016/j.nlm.2020.107340 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Smith E, Hay P, Campbell L, & Trollor JN (2011). A review of the association between obesity and cognitive function across the lifespan: implications for novel approaches to prevention and treatment. Obesity Reviews, 12(9), 740–755. [DOI] [PubMed] [Google Scholar]
  51. Smith PJ, Blumenthal JA, Hoffman BM, Cooper H, Strauman TA, Welsh-Bohmer K, & Sherwood A (2010). Aerobic exercise and neurocognitive performance: a meta-analytic review of randomized controlled trials. Psychosomatic Medicine, 72(239–252). [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Steffen TM, Hacker TA, & Mollinger L (2002). Age-and gender-related test performance in community-dwelling elderly people: Six-Minute Walk Test, Berg Balance Scale, Timed Up & Go Test, and gait speeds. Physical Therapy, 82(2), 128–137. doi: 10.1093/ptj/82.2.128 [DOI] [PubMed] [Google Scholar]
  53. Stephens T (1988). Physical activity and mental health in the United States and Canada: evidence from four population surveys. Preventive Medicine, 17(1), 35–47. [DOI] [PubMed] [Google Scholar]
  54. Stillman CM, Cohen J, Lehman ME, & Erickson KI (2016). Mediators of physical activity on neurocognitive function: a review at multiple levels of analysis. Frontiers in Human Neuroscience, 10, 626. doi: 10.3389/fnhum.2016.00626 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Studenski S, Perera S, Wallace D, Chandler JM, Duncan PW, Rooney E, . . . Guralnik, J. (2003). Physical performance measures in the clinical setting. Journal of the American Geriatrics Society, 51(314–322). [DOI] [PubMed] [Google Scholar]
  56. Tsuno N, Besset A, & Ritchie K (2005). Sleep and depression. The Journal of Clinical Psychiatry, 66(10), 1254–1269. [DOI] [PubMed] [Google Scholar]
  57. Uttl B (2002). North American Adult Reading Test: Age norms, reliability, and validity. Journal of Clinical and Experimental Neuropsychology, 24(8), 1123–1137. doi: 10.1076/jcen.24.8.1123.8375 [DOI] [PubMed] [Google Scholar]
  58. van Uffelen JG, Paw MJCA, Hopman-Rock M, & van Mechelen W (2008). The effects of exercise on cognition in older adults with and without cognitive decline: a systematic review. Clinical Journal of Sport Medicine, 18(6), 486–500. [DOI] [PubMed] [Google Scholar]
  59. Vidoni ED, Morris JK, Watts A, Perry M, Clutton J, Van Sciver A, . . . Burns JM (2021). Effect of aerobic exercise on amyloid accumulation in preclinical Alzheimer’s: A 1-year randomized controlled trial. PloS One, 16(1), e0244893. doi: 10.1371/journal.pone.0244893 [DOI] [PMC free article] [PubMed] [Google Scholar]
  60. Wechsler D (2008). WAIS-IV: Administration and scoring manual. New York, NY: The Psychologial Corporation. [Google Scholar]
  61. Whaley AL (2001). Cultural mistrust and mental health services for African Americans: A review and meta-analysis. The Counseling Psychologist, 29(4), 513–531. [Google Scholar]
  62. Williams DR, Gonzalez HM, Neighbors H, Nesse R, Abelson JM, Sweetman J, & Jackson JS (2007). Prevalence and distribution of major depressive disorder in African Americans, Caribbean blacks, and non-Hispanic whites: results from the National Survey of American Life. Archives of General Psychiatry, 64(3), 305–315. [DOI] [PubMed] [Google Scholar]
  63. Williams DR, & Williams-Morris R (2000). Racism and mental health: The African American experience. Ethnicity & Health, 5(3–4), 243–268. [DOI] [PubMed] [Google Scholar]
  64. World Health Organization. (2006). Global database on body mass index. Retrieved from http://www.assessmentpsychology.com/icbmi.htm
  65. World Health Organization. (2017). Depression and other common mental disorders: Global health estimates. Retrieved from Geneva, Switzerland: [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Materials
NIHMS1689194-supplement-Supplemental_Materials.pdf (340KB, pdf)

RESOURCES