Twelve percent of older Latinos in the U.S. are currently diagnosed with Alzheimer’s Disease (AD), and it is estimated that the number of Latinos with AD will increase by 832% by 2060 (Wu et al., 2016). Currently, there is no cure for AD; however, evidence suggests that protective factors for AD include regular physical activity (PA) (Blondell et al., 2014), cognitively and mentally stimulating leisure activities (Wilson et al., 2012), social engagement, and having a rich social network (Marioni et al., 2015). Latin dance is a particularly promising PA modality that targets these factors and is a culturally acceptable type of PA for middle-aged and older Latinos (Valenzuela et al., 2013; Marquez et al., 2014a).

Latin dance has long been part of the history, socialization, and culture of Latinos (Gonzalez et al., 1998). Dance as a PA modality for Latinos may evoke positive emotional responses that encourage PA participation in a population with low levels of PA (Arredondo et al., 2015). Latinos have cited dance as an important and desirable component to community-based programs (Larsen et al., 2015; Predovan et al., 2019). Furthermore, engaging in dance has been shown to improve or maintain cognition (Hamacher et al., 2015; Hwang and Braun, 2015; Merom et al., 2016; Predovan et al., 2019; Muiños and Ballesteros, 2021), partly due to the learning component and coordination of dancing (Voelcker-Rehage and Niemann, 2013; Rehfeld et al., 2017), the multisensory demands of the activity (Thøgersen-Ntoumani et al., 2018), the timing, synchronization and sequences of moves, and the energy expenditure associated with movement (Keogh et al., 2009). Randomized controlled trials (RCTs) have examined changes in cognitive performance among several types of dance styles including, ballroom dancing (Lazarou et al., 2017), Cha-Cha (Kim et al., 2011), and a blend of styles (e.g., line dance, jazz dance, rock’n roll, Latin, and square dance) (Hamacher et al., 2015) and have demonstrated changes in global cognition, executive function, episodic and working memory, and attention. Despite the benefits and appeal of dance, PA programs and interventions rarely implement dance programs specifically for Latinos, a historically excluded population at high risk of chronic disease, mobility disability (Angel et al., 2015), and cognitive impairment (Wu et al., 2016).

Given the need to address health inequities in Latinos, Marquez and colleagues created a Spanish-language, Latin dance program (BAILAMOS™—Balance and Activity In Latinos, Addressing Mobility in Older Adults) (Marquez et al., 2014a) for middle-aged and older Latinos to increase PA, thereby reducing risk of chronic diseases, mobility disability, and cognitive impairment. An initial pilot study with twelve participants demonstrated that self-reported PA increased significantly and small positive effects for executive function and speed of processing were seen in participants, all of whom received the Latin dance program (Marquez et al., 2014a). Following that small pilot, a two-group pilot trial was conducted to examine differences in cognitive performance among 57 older Latinos randomized to either the BAILAMOS™ dance program or a health education program (Marquez et al., 2017). Findings showed greater improvement in episodic memory in the dance group compared to the health education group. Balbim et al. (2021) examined the impact of the BAILAMOS™ dance program on physical activity, brain functional connectivity, and cognitive performance among 10 participants who received the program. Participation in the program led to an increase in self-reported moderate leisure-time PA, increases in brain functional connectivity, but no statistically significant changes in cognitive performance. These studies demonstrated promising findings for increasing PA, cognition, and brain function; however, these studies were limited by small sample sizes and short follow-up periods, making advanced analytical methods difficult to conduct. Therefore, the purpose of the present study was to examine the secondary outcome of changes in cognitive performance in a large sample of middle-aged and older Latinos participating in the BAILAMOS™ dance program or a health education control group at 4-months and 8-months post-intervention and compare the mediating effects of types of PA (i.e., minutes/week of total PA, minutes/week of total leisure PA, and minutes/week of moderate-to-vigorous PA), and between group assignment and change in cognitive domains.

Participants were on average 64.9 (SD = 7.09) years of age with an average BMI of 31.14 (SD = 4.78). The majority of participants were female (84.4%), with 8.38 (SD = 4.01) years of education, and 48.9% were married. Age was statistically different (0.050) as the significance level was defined as <0.05 and was controlled when conducting hypotheses tests on any baseline or demographic variables including baseline total PA, total leisure PA, and moderate-to-vigorous PA. For cognitive performance scores, there were significant between-group differences for Trail Making Test A and Stroop CW (Table 1).

A series of repeated measures ANCOVA were performed to test between group differences for each cognitive domain with age as a covariate (Table 2). There was a significant group difference in working memory between the two groups [Wilk’s λ = 0.98, F_(2,328) = 3.13, p = 0.045]. Follow-up univariate tests found no significant between-group differences at month 4 [F_(1,328) = 0.392, p = 0.532, d = 0.13], but the experimental dance group demonstrated a significant improvement in working memory at month 8 of the study [F_(1,328) = 5.79, p = 0.017, d = 0.20]. Multivariate analyses that tested change in executive functioning [Wilk’s λ = 0.99, F_(2,328) = 0.229, p = 0.80] and episodic memory [Wilk’s λ = 0.99, F_(2,328) = 0.241, p = 0.78] were not significant between the two groups at 4 and 8 months.

The present study examined whether participation in the BAILAMOS™ dance program improved cognitive performance compared to a health education control group and investigated which PA category predicted cognitive changes among middle-aged and older Latinos over eight months. The study showed that participants who were randomized to the 4-month Latin dance program followed by a 4-month maintenance phase improved in working memory compared to participants in the health education control group. Findings indicate that improvements in working memory were mediated by participation in total PA and total leisure PA at 8 months only. While effect sizes (Cohen’s d) were modest in this study, small effect sizes are typically found in RCTs in older adults (Erickson et al., 2019). These findings support the current literature on the influence of leisure PA and total PA on cognitive performance and highlight the importance of culturally appropriate PA modalities for middle-aged and older Latinos.

The between-groups analysis between the two groups at 4-months found no differences in any of the cognitive measures, which is similar to findings from a small pilot examining the BAILAMOS™ dance program in older Latinos (Balbim et al., 2021). However, our study found improved working memory at month 8 in the dance group. This finding is in line with the literature demonstrating that longer dance interventions (>4 months) are needed to detect changes in cognitive performance. For example, Fausto and colleagues found that African American older adults who received a 5-month cardio-dance program consisting of dance routines and steps from hip-hop, merengue, samba, cumbia, and salsa improved attention compared to a propensity-matched control group (Fausto et al., 2021). Another study by Lazarou et al. (2017) reported that participants in a 10-month international ballroom dancing program improved in global cognition, reaction time, visuospatial skills, selective attention, and attentional switching. Our study found an improvement in working memory, and this finding is especially pertinent because working memory decline has been recognized as one of the primary contributing factors to cognitive impairment in older adults (Park et al., 2002). Therefore, Latin dance could be an important strategy to improve or maintain working memory among middle-aged and older Latinos, a population at high risk of cognitive decline.

Our findings demonstrated that changes in working memory at 8 months were mediated by total PA and total leisure PA. This finding suggests that PA categories may function as a mechanism to improve working memory. Studies have suggested that different types of PA may determine the extent to which PA impacts cognition (Quigley et al., 2020). For example, Quigley et al. (2020) concluded that single PA modes are effective in inducing improvements in cognition among older adults, but combined PA modes may offer magnified cognitive benefits. The advantage of dancing is that it is a PA mode that embeds aerobic PA and neuromotor PA (i.e., balance and proprioceptive training). Furthermore, participants in the Latin dance group were provided with a PA modality that facilitated an increase in their total PA and functioned as a leisure type of PA.

Changes in working memory at 8 months may have been due to several reasons. First, dancing poses a high demand on attention and memory for individuals. Participants initially learned four dance styles. During the 4-month maintenance phase, four additional dance moves were added to each dance style. Participants were able to practice what they learned during the maintenance phase, therefore the process of recalling steps and learning new dance steps may be due to the working memory process known as chunking information (Tulving and Craik, 2000). Participants in the dance intervention did this through the process of combining dance steps into sequences. For example, participants learned the side basic, back basic, and backside basic dance steps independently, which then were chunked into what is known to be the salsa dance style. Similar to the way the brain manipulates information into chunks, here we see the same process happening with the learning of different Latin dance styles, which helps to explain the improvement in working memory scores. Second, participants continued the dance program for an additional 4 months (maintenance phase). Indigenous leaders in the treatment group received training on how to continue the dance program without a professional instructor during the maintenance phase while the health education group stopped meeting. Therefore, with the train the trainer model, participants were provided the tools and training necessary to continue leading the program. Furthermore, dance participants may have felt empowered to implement strategies learned to overcome PA barriers, strengthened social ties, and continue engaging in a cognitively stimulating activity (e.g., dance). This study highlights the importance of implementing dance as a PA modality to increase PA, thereby, improving working memory and underscores how imperative it is to provide tools and training to create sustainable community programs led by community members.

Our study had several strengths including a large sample of community dwelling older, Spanish-speaking Latinos, a group that is rarely included in large RCTs, but has a high level of need for health-related interventions. The study also included a culturally relevant PA intervention (i.e., dance) with a 4-month maintenance phase embedded within the program. The study also had several limitations. First, PA was self-reported; thus, PA may be over- or under-estimated. Second, participants resided in Chicago and a majority identified as Mexican American, thereby limiting generalizability of findings to other geographic areas and other Latino subgroups. For the cognitive outcomes, improvements in cognitive performance should be interpreted with caution considering potential practice effects; however, the small changes observed in the control group suggest that practice effects, if any, are small. Also, no further follow-up was conducted to determine whether improvement in working memory were maintained. Lastly, the a priori sample size calculation of the trial was not performed specifically for the mediation analyses, but rather based on the primary outcome of changes in PA.

In sum, this study underscores the importance of creating and delivering culturally relevant, community-based Latin dance programs that promote total PA and function as a leisure time PA for Latinos. Furthermore, implementing maintenance phases to create more sustainable programs and empowering community members to lead programs is critical to create a culture of health. Future studies may consider including longer follow up periods to determine the duration of dance intervention effect on cognition.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by University of Illinois at Chicago. The patients/participants provided their written informed consent to participate in this study.

SA wrote the manuscript with feedback and guidance from NK, GB, and DM. NK conducted the statistical analysis. RW, JW, SH, DB, MB, EM, PV, IM, and TW provided critical feedback. All authors contributed to the final version of the manuscript.

The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Center for Advancing Translational Sciences or the National Institutes of Health.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.