Abstract
Older Latinos have low rates of physical activity (PA) and poor health outcomes. Focus groups and a single group, pre–post 3-month pilot of a culturally appropriate Latin dance program (BAILAMOS©) were conducted among older, inactive Latinos with a self-identified mobility limitation. Nine themes emerged from focus groups, including cultural influences of dance and barriers to dancing. In the pilot intervention, 9 of 13 older Latinos completed the program and attended 85% of the sessions. Self-reported PA significantly increased (p < .05; d = 1.39), but daily accelerometer-assessed PA decreased for moderate-intensity PA (p = .18; d = 0.29). Participants reported significantly greater enjoyment of PA (p < .05; d = 0.61) and improved physical quality of life (p = .23; d = 0.31). Improvements in mobility (p = .15; d = 0.56) and aspects of cognition were demonstrated. The BAILAMOS program appears to be feasible, culturally appropriate, and has the potential to improve psychosocial determinants of PA, self-reported PA, and aspects of health.
The health benefits associated with physical activity (PA) in the older adult population have been documented (Chodzko-Zajko et al., 2009). However, there is indication of racial/ethnic disparities in PA engagement as Latinos aged 55 to 74 are 46% to 50% less likely to engage in leisure time PA (LTPA) than older non-Latino Whites of comparable age (Marquez, Neighbors, & Bustamante, 2010). Dancing and walking have been cited as the only age-appropriate PA for older Latina women (Cromwell & Berg, 2006), and dance has long been an important form of socialization and leisure in Latin cultures (Delgado & Munoz, 1997; Lewis, 1994). However, unsafe neighborhoods and unfavorable weather conditions are significant barriers to walking for many older adults (Belza et al., 2004). As a result, we developed a culturally appropriate Latin dance program for this underserved population, given that dance challenges individuals physically and cognitively. Preliminary research indicates that dance has been associated with cognitive improvements (Kattenstroth, Kalisch, Holt, Tegenthoff, & Dinse, 2013) and can improve older adults’ lower extremity function (Keogh, Kilding, Pidgeon, Ashley, & Gillis, 2009). To our knowledge, no dance intervention studies have been conducted with older Latinos. Latin dance offers a unique dimension not found in traditional aerobic exercise, namely, interpersonal communication and reciprocity. Thus, we conducted focus groups to assess the receptivity for dancing among older Latinos; and tested the feasibility and impact of the BAILAMOS© (Balance and Activity In Latinos, Addressing Mobility in Older Adults) dance program, developed in collaboration with a Latin dance instructor.
Method
Phase 1: Focus Groups
Design and Sample
Two focus groups were conducted to examine older, urban Latinos’ perceptions and attitudes toward dance and PA prior to development of a dance intervention. Inclusion criteria were the following: (a) age ≥55 years, (b) English-speaking Latino/Hispanic, (c) mobility limitation (Weiss, Fried, & Bandeen-Roche, 2007), and (d) sedentarism (<2 day/week of aerobic exercise). Exclusion criteria included dependence on assistive device to ambulate and severe clinical conditions (e.g., recent bone fracture). Participants were recruited using established relationships in the area developed by the principal investigator, including presentations at the study site, senior housing facilities, a church, and coalition meetings; and through a health centers/clinic, word of mouth, flyers in mailboxes of senior housing facilities, and an ad in the neighborhood newspaper. Study approval was obtained through the University of Illinois at Chicago Institutional Review Board.
Data Collection and Analysis
Stratified by gender, two focus groups were conducted, with facilitation by the principal investigator who has previous research experience with older Latinos. Focus groups ranged from 65 to 75 minutes in length. After providing informed consent, a brief survey was administered to elicit demographic information. Participants were financially compensated. Focus groups were audio-taped and transcribed.
Computer software (i.e., Atlas/ti) was used across analyses procedures. Two bicultural research assistants coded the transcripts. No older adults were involved in data analysis. An iterative process was used for code development where initial codes were supplemented with emergent codes as they surfaced during data analysis. This was followed by discussions with the two research assistants and the principal investigator to resolve disagreements and reach consensus on appropriate use of codes. The percentage of agreement in coding was 70% across reviewers, with 100% agreement reached through face-to-face discussion of the final codes assigned. Identification and organization of common themes was accomplished through close reading of the transcripts and after segmentation and coding of the textual documents.
Phase 2: Pilot Intervention
Design and Sample
A single group, pre–post 3-month pilot test of BAILAMOS was conducted. Inclusion criteria and recruitment strategies were the same as for focus groups, with addition of (a) ability to speak Spanish or English, (b) adequate cognitive functioning (answered ≤3 wrong on the 20-item Mini Mental State Examination; Folstein, Folstein, & McHugh, 1975), (c) danced <2 times/month over previous 12 months, and (d) physician approval if deemed at risk for adverse events. A total of thirteen participants were enrolled in the study.
Measures
Measures were available in either English or Spanish, and participants had their choice.
Background measures
A brief survey elicited demographic information and participants’ health history and status. Body mass index (BMI; kg m−2) was calculated from height and weight.
Dance logs
Participants recorded the number of minutes danced, ratings of perceived exertion, how they were feeling, and enjoyment of the session.
Psychosocial measures
Psychosocial constructs were measured using the Physical Activity Self-Efficacy Scale (McAuley, 1993), available in English and in Spanish (Marquez & McAuley, 2006), and a modified version of the Physical Activity Enjoyment Scale (Kendzierski & DeCarlo, 1991), original available in English and in Spanish (de Gracia & Marco, 2000), where having participants use a Likert-type scale ranging from 1 to 7 in response to questions such as “Do you enjoy (1) or hate (7) physical activity?” was changed to asking participants to answer dichotomously if they enjoy or hate it. If they said they enjoy it, then they were asked to give a number from 1 to 3, with a lower number indicating more enjoyment. Similarly, if they said they hate it, then they were asked to give a number from 5 to 7 with a higher number indicating hating it more. Quality of life was measured with the 12-item Short Form Survey (Ware, Kosinski, & Keller, 1996), available in English and in Spanish (Gandek et al., 1998).
Physical activity
The CHAMPS Physical Activity Questionnaire for Older Adults (Stewart et al., 2001), previously used in Spanish (Rosario, Vazquez, Cruz, & Ortiz, 2008), was used to assesses self-reported lifestyle PA; and the ActiGraph Model GT3X accelerometers (The Actigraph, Pensacola, FL), small and lightweight triaxial accelerometers, were employed. Time spent in light, moderate, or vigorous intensity PA using cut-points for older adults were assessed (Miller, Strath, Swartz, & Cashin, 2010). Data were included in analysis if the accelerometer displayed at least 10 hours of data in a 24-hour period on at least 3 days.
Physical function performance
The three tests of the Short Physical Performance Battery (SPPB) (tests of balance, gait, and lower body strength/endurance) were employed (Guralnik et al., 1994). Also, mobility was assessed with the 400-meter walk, and participants were instructed to walk the 400-meter course at a comfortable and maintainable pace. Total time to complete the walk was recorded.
Cognitive function
We employed tests that assess functions likely to be related to aging and PA (Kramer, Erickson, & Colcombe, 2006) and that have been shown to be valid across ethnic and socioeconomic backgrounds (Krueger, Wilson, Bennett, & Aggarwal, 2009). The cognitive function tests included the East Boston Memory Test (EBMT; Albert et al., 1991), the Stroop Neuropsychological Screening Test (Trenerry, Crosson, DeBoe, & Leber, 1989), the Numbers Comparison test (Ekstron, French, Harmen, & Kermen, 1976), and a measure of verbal fluency (Welsh et al., 1994).
Function and disability
Function and disability were measured using the 15-item function subscale and the 8-item disability subscale of the Abbreviated Late Life Function and Disability Instrument (LL-FDI) (McAuley, Konopack, Motl, Rosengren, & Morris, 2005), available in English and in Spanish (Abizanda et al., 2011).
Procedures
Following screening, participants were scheduled for a baseline testing session. After acquiring informed consent, questionnaires and measures were administered by trained bilingual staff.
Dance program
BAILAMOS includes Merengue, Cha Cha Cha, Bachata, and Salsa and integrates principles of social cognitive theory (Bandura, 1997) to boost program retention and maximize PA outside the program (McAuley, Jerome, Elavsky, Marquez, & Ramsey, 2003; McAuley, Jerome, Marquez, Elavsky, & Blissmer, 2003). Self-efficacy was targeted by setting the sequencing of dance styles from the simplest style (i.e., Merengue) through the most difficult style (i.e., Salsa) to enhance mastery experiences. Additionally, participants were exposed to social modeling by being in a class with other individuals of similar age, function, and demographic characteristics (including the instructor). Finally, they received consistent verbal persuasion from the dance instructor and fellow participants and were informed of the physiological experiences they should anticipate from engaging in PA (e.g., sweating, increased heart rate).
The principal investigator and a professional dance instructor codeveloped an extensive BAILAMOS Dance Manual and class-by-class schedule. The 13-week program included one orientation week followed by 12 weeks of dance sessions. Classes met in a senior center community room twice per week for 1 hour per session in the morning. Each session included warm-up, stretching, steps for singles and couples dancing, cool-down, and dance logs. Fidelity checks were completed in Weeks 3, 6, 9, and 12 of the program to ensure that the proper procedures and protocol were being followed.
Data Analysis
We used paired-sample t tests to examine pre–post changes in outcomes as a result of the intervention, with effect sizes (Cohen’s d) computed given the small sample size.
Results
Phase 1: Focus Groups
Six men and six women participated. On average, both males and females were in their 70s (M = 75.08), overweight (M BMI = 28.83), and had low levels of education (75% less than high school graduate) and income (58% household income below $30,000). A total of nine themes emerged from the focus groups.
Culture
Both groups discussed the influence of culture on their interest in dancing. “Culture” included subthemes of the current acceptability of girls/women engaging in exercise and how girls/women did not have the opportunity to exercise or play sports when they were younger. It was noted that dancing is a part of life in Latin culture, and it was clear that dancing frequently begins when Latinos are young (“I was 9 years old when my brothers teach me how to dance polkas and boleros and rancheros”).
Favorite Aspects of Dancing
Participants liked that dancing is a total body movement (“Because you move a lot and your whole body is moving your legs, your arms, your head, everything on your body moves”), is a social activity (“People you knew come all the time and you didn’t have to have a partner”), and includes music (“There’s some kind of rhythm not only for the body but your brain in that you hear that music. It sends a message to the body”).
Barriers to Dancing
General barriers were mentioned, as one man noted, “What person our age wants to go out there at 8:00 o’clock at night and dance? I’m getting ready to go to bed at that hour.” Barrier subthemes included personal and safety barriers and the lack of availability and access to dance programs. There were comments that dance programs and facilities are not available (“If we had someplace we could go dance at least once a week or a month that would be great but there isn’t anything”) or accessible because of cost (“Those kinds of things would have to be offered to the seniors without having to pay because really when it comes to money a lot of them really can’t afford it”).
Dancing Versus Other Types of Physical Activity
Overall, participants perceived dancing to accrue more health benefits than other types of PA. For example, one woman noted, “When people swim they don’t move as much as we do when we’re dancing and they don’t sweat the way we do. When you’re dancing and you’re sweating you’re losing a lot of calories.”
Remaining Themes
Other important themes emerged, including that participants were generally aware of the benefits of being physically active. One man stated, “It’s the best medicine in the first place, exercise.” They also identified both cognitive and physical health benefits associated with dancing (“improves your ability to move around and it improves your thinking”).
Phase 2: Pilot Intervention
Feasibility
Table 1 presents descriptive data. Nine of 13 older Latinos completed the program for a 69% retention rate, with attrition attributed to factors unrelated to the program (i.e., home injury, change in work schedule, new grandchild caregiving duties, and left the country). There was no statistical difference in age, BMI, years in the United States, time to complete 400-meter walk, and SPPB scores (all ps > .05) between those who completed the program and those who did not.
Pilot Intervention Sample Descriptives.
Participants who completed the program attended 85% of the sessions. Reasons given for not attending classes included doctor appointments, vacation, and family responsibilities.
Pre–Post Intervention Outcomes
Table 2 shows pre–post intervention changes in PA and physical function. Minutes of self-reported moderate/vigorous LTPA increased considerably (p = .10; d = 1.51). Minutes of self-reported lifestyle PA significantly increased (p < .05; d = 1.39). Accelerometer-assessed PA decreased from pre- to postintervention for moderate intensity PA (p = .18; d = 0.29).
Pilot Intervention Physical Activity, Physical Function, and Psychosocial and Cognitive Function Measures.
Note. CHAMPS = CHAMPS Physical Activity Questionnaire for Older Adults; LL-FDI = Late Life Function and Disability Instrument; MV = moderate/vigorous; PA = physical activity; SPPB = Short Physical Performance Battery; QoL = quality of life.
p < .05.
Objectively there was a moderate effect of the dance program on mobility, as time to complete the 400-meter walk decreased (p = .15; d = 0.56). Scores on the disability component of the LL-FDI demonstrated small to moderate improvements (d = 0.16-0.41), whereas scores on the function component demonstrated small decreases (d = 0.06-0.18).
As can be seen in Table 2, participants reported significantly greater enjoyment of PA after the program (p < .05; d = 0.61) and improved physical quality of life (p = .23; d = 0.31) but not mental quality of life. PA-related self-efficacy decreased over time (p = .13; d = 0.89). Small positive effects were seen for the Stroop Neuropsychological Screening (p = .54 and .65; d = 0.09 and 0.11) and Numbers Comparison tests (p = .65; d = 0.12), whereas small negative effects were demonstrated for the Verbal Fluency test (p = .23; d = 0.24).
Discussion
Behavioral interventions targeting early stages of the disablement process have the potential to prevent or delay disability (Popa, Reynolds, & Small, 2009). Our qualitative assessment and pre–post intervention study provides preliminary support for the BAILAMOS dance program. Focus group discussions centered on culture and various factors related to dancing. Many women were restricted from engaging in exercise or sports as children, but dancing was a form of PA to which they were introduced at a young age. Consequently, a dance program in later life allows them to participate in an activity performed in early life. Overall dancing is an activity that many had an interest in; however, many barriers were identified (e.g., lack of access). The BAILAMOS dance program helps overcome many of those barriers, by providing a structured group activity and enabling access and availability of a program indoors, so as to avoid lack of personal safety and inclement weather when walking outdoors.
With low attrition and high session attendance, the BAILAMOS dance program proved feasible. In a postintervention focus group, participants reported enjoying learning how to dance, being motivated to be active, social aspects, and the instructor. Moreover, at the end of the program participants self-reported significantly more lifestyle PA and greater enjoyment of PA compared with baseline. Participation in BAILAMOS resulted in greater mobility, as indicated by the change in 400-meter walk. Small improvements in two sets of cognitive tests, the Stroop Neuropsychological Screening and Numbers Comparison, were expected as they test aspects of executive function.
Contrary to hypotheses, objectively assessed PA decreased after the program. It is not unusual for self-reported PA to differ substantially from accelerometer-based data (Prince et al., 2008). It is possible that since the CHAMPS (subjective assessment of PA) asked about PA over the past 4 weeks, whereas the accelerometer was only worn for 1 week, the discrepancy is partially explained. Also, it is possible that as a result of the program, participants were more aware of any PA they were engaging in, and thus reported doing more.
Study limitations should be considered when interpreting the findings. Focus groups were conducted in English, yet the pilot intervention was conducted in both Spanish and English. Also, small sample sizes were included in both phases of the work. Future qualitative studies examining perceptions of dance as a form of PA in Latinos will benefit from larger samples that include diverse Latino subgroups. Also, future studies should examine the cost of conducting such a program. Despite the small sample size of this pilot intervention, several outcomes were positively affected, and these findings combined with strong program retention, participation, and satisfaction underlie the promise of the BAILAMOS program.
The number of older Latinos will increase by 224% by 2030 (Greenberg, 2009). Poor health and low LTPA levels affect older Latinos for many reasons, and culturally appropriate PA that will be adopted and adhered to are essential to promote health in this population. Initial results are promising for the BAILAMOS program to help fill the gap of culturally appropriate PA programs.
Footnotes
Acknowledgements
We acknowledge the contributions and assistance of Jeanine Campa, Jennifer Galvan, Sarah Janicek, Miguel Mendez, Julia Lopez, and Villa Guadalupe Senior Services, Inc, and thank all contributors from South Chicago.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Department of Kinesiology and Nutrition at the University of Illinois at Chicago. Rosalba Hernandez was supported by the National Institute of Health (NIH) under Award Number 5T32 HL069771-10 (M. Daviglus, principal investigator).