The effect of exercise on mood in older, Mexican-American women.

Abstract: This experiment was designed to determine if a single bout of exercise was sufficient to improve the mood of older (60- 79 yrs.), Mexican-American women. Participants (n=42) were drawn from a central Texas senior center and randomly assigned to either 20 win of treadmill walking at 70% of HRR or 20 win of a group discussion regarding health-related issues. The primary dependent variables were drawn from Spanish translations of the PANAS and the EFI that were assessed prior to and immediately and 30 win following the experimental conditions. Results indicated that all women, regardless of condition, experienced a significant linear increase in positive affect, as measured by the PANAS, and revitalization and tranquility, as measured by the EFI. Thus, although a single bout of exercise is sufficient to improve the mood of this population, it appears to be no different from a period of social interaction.

**********

It is clear that maintaining a physically active lifestyle is an important component of physical and mental health. Regular physical activity is associated with lower mortality rates, decreased risk of cardiovascular disease, prevention or delay of high blood pressure, decreased risk of colon cancer, lower risk of non-insulin-dependent diabetes mellitus, improved mood and psychological well-being (USDHHS, 2000). Unfortunately, most Americans do not participate in sufficient levels of physical activity. Specifically, national data indicate that 39% of Americans do not regularly participate in regular leisure-time physical activity and only 32% regularly participate in sufficient levels of physical activity to promote health and fitness (USDHHS, 2000). The data are worse for Americans of Hispanic and Latino heritage where 54% of Hispanic Americans are wholly sedentary and only 24% participate in regular health promoting physical activity. This relative inactivity for Hispanic Americans provides some explanation of why the incidence of obesity and non-insulin-dependent diabetes is particularly high within this population (USDHHS, 2000). Additionally, recent census data indicate that the Hispanic-American population will become the largest minority group within the next twenty years (CDC, 2000). Thus, it is imperative that research be directed to those factors that might motivate a physically active lifestyle in Hispanic-Americans.

One of the most consistently cited sources of motivation for exercise is its ability to regulate mood on a daily basis (Thayer, Newman, & McClain, 1994). In fact, it has long been recognized that single (acute) bouts of exercise are sufficient to improve mood (Morgan, 1985). These improvements have been most consistently demonstrated as a reduction in negative psychological states, such as anxiety (Petruzzello, Landers, Hatfield & Salazar, 1991), negative affect (Bartholomew & Linder, 1998) and depressive mood (North, McCullagh, & Tran, 1990). More recently, attention has shifted to include an examination of positive psychological states such as revitalization, psychological well-being and positive affect. Although not as consistent as the reduction in negative states, acute exercise has been shown to increase these positive states in both younger (Bartholomew & Miller, 2002) and older samples (McAuley, Shaffer, & Rudolph, 1995). However, to date, there exists no published studies that have examined these effects within a sample of Mexican Americans. Given the drastic difference in levels of self-reported physical activity, it is not clear that the results of non-Hispanic white samples will generalize to the Mexican American population, particularly older Mexican Americans.

One of challenges of studying any ethnic minority group is the lack of valid instrumentation to assess the dependent variables of interest. For example, although activity-specific scales have been created to assess post exercise psychological states (Gauvin & Rejeski, 1993, McAuley & Courneya, 1994), neither of these has been translated for use by Spanish speakers. In addition, although there exist Spanish versions of general affective scales (Joiner, et al., 1997), because nomenclature varies widely by cultural sub-groups, a scale developed for one Hispanic sub-culture may not be valid for use in other Spanish speaking groups. Thus, it is not surprising that research has not yet targeted the Mexican American population to examine psychological responses to exercise. This study was designed to fill this void. Specifically, Spanish translations of two scales, the Exercise-Induced Feeling States Inventory (Gauvin & Rejeski, 1993) and the Positive Affect--Negative Affect Schedule (Watson, Clark, & Tellegen, 1988), were used in conjunction with 20 minutes of moderate intensity treadmill exercise. Participants were older, Mexican American women recruited from community centers. We hypothesized that the exercise would result in improved mood as assessed by the PANAS and improved feeling state as assessed by the EFI, and that this improvement would be relatively greater than that observed following the no-exercise, comparison social-interaction session.

METHODS

Sample. Forty-two women were recruited from three senior centers in a Central Texas city. Participants were self-identified Mexican-American women aged 60 to 79 years, living independently in the community. They reported their health as very good to excellent. The women had a resting blood pressure below 160/90 mm/hg and thus were no greater than mildly hypertensive, and were not taking medications known to affect their heart rate. Letters from their physicians indicated that there were no known medical contraindications to their participation in the study. All of the women were physically active, indicating that they exercised at least 20 minutes, three times weekly, and had done so for 6 months or more.

The women were assigned to the treatment or comparison group, using a table of random numbers. One woman in the comparison group insisted on walking on the treadmill so her data were excluded from the study. In addition, one woman failed to complete multiple questionnaires and was excluded from the analysis. These deletions left 21 women in the treatment group and 19 in the comparison group.

The sample size was estimated from a power analysis. Effect sizes of 1.34 for the PANAS and .91 for the EFI (Gauvin, et al., 1996) were reported in studies of young non-Hispanic Anglo populations. Although no studies were found that used these instruments with older Mexican Americans, we estimated a large effect size and power of. 90. Although a sample of 30 would yield adequate power to detect findings, 42 women were recruited to allow for attrition.

Instruments. The Positive Affect-Negative Affect Schedule (PANAS) measured positive and negative affect. This 20-item questionnaire, developed by Watson, Clark, and Telegen (1988), is scored on a 5-point Likert-type scale anchored with "very slightly or not at all" and "extremely" to indicate the extent to which items reflect respondents' feelings at that time. The instrument has 10 positive and 10 negative items, which are summed within each sub-scale to determine score for positive and negative affect.

The PANAS items were translated into Spanish by a bilingual Hispanic research assistant and back-translated by a bi-lingual Mexican American woman who had never seen the PANAS. All of the back-translated items corresponded to the original English items (Laffrey, Bartholomew, Lee, & Hernandez-Spina, 2000).

Within the physical activity domain, the PANAS has demonstrated acceptable construct validity and internal consistency with average alpha coefficients above .80 for both sub-scales (Crocker, 1997). In our study, the alpha coefficients were .77 for the positive and .93 for the negative subscale.

The Spanish-English Exercise-Induced Feeling State Inventory (EFI) measured participants' feelings relative to exercise. Gauvin and Rajeski (1993) developed the EFI to measure exercise-induced feeling states in 4 domains (revitalization, positive engagement, tranquility, and fatigue). Each domain is represented by 3 items within the sub-scales. The EFI has been used extensively in exercise studies and the alpha coefficients for the four sub-scales have been consistently above .70. The EFI items were adapted to increase their cultural relevancy for the Mexican American participants using the same methods as for the PANAS. The adapted scales have 3 items for each of the four subscales (Laffrey et al., 2000), but the wording in some of the items was modified. Alpha coefficients among the participants in this study were: .86 (tranquility), .68 (positive engagement), .76 (revitalization), and .83 (fatigue).

Heart rate was measured with a wireless Polar Accurex Plus Heart rate monitor. The Karvonen method was used to estimate heart rate reserve: 70% heart rate reserve = (220 - age - resting heart rate) X (0.7) + resting heart rate.

Procedure. Because most activities in senior centers, including exercise, are completed in small groups, women were seen in groups of 2 to 6 for both the exercise and the comparison condition. After the study was explained and informed consent was provided, the women were randomly assigned to either the exercise or comparison group.

Exercise Group. The women were asked to sit quietly for 5 minutes to record their resting heart rate. The Spanish-English versions of the PANAS, EFI, and personal questionnaire form were administered by interview. The women then walked on a treadmill for 20 minutes at 70% of their heart rate reserve. This moderate intensity exercise was selected became it falls within the Surgeon General and the American College of Sports Medicine recommendations of daily levels of physical activity (USHHS, 1996; ACSM, 2000). Immediately after completing the exercise, the PANAS and EFI were again administered to the women by interview. They were asked to sit and talk with the investigators about healthy nutrition and regular health examinations and mammograms for 30 minutes after which the PANAS and EFI were again administered by interview. The women were given a gift of $25.00 to thank them for participating in the intervention.

Comparison Group. The three test administrations were identical to the intervention group but they did not walk on the treadmill. During the time between testing, they sat at a table and talked about health relate matters, such as healthy eating and regular health examinations and mammograms.

Data Analysis. Data were analyzed via a 2 (group) x 3 (Time) doubly-multivariate ANOVA with repeated measures on the second factor. Significant results were decomposed via separate ANOVAs with repeated measures on the second factor. Post hoc comparisons were conducted from pre to each post measures.

RESULTS

Descriptive findings. The women ranged from 60 to 79 years of age (mean age = 68; SD = 5.9 years); nine were born in Mexico and 32 in the United States. Twenty were married, and 22 were separated, divorced or widowed. Years of schooling ranged from 1 to 16 years (mean = 9.8 years; SD = 3.5 years). Twenty-five of the women rated their annual income as adequate while 17 rated income as inadequate to meet their needs.

Mood Findings. The MANOVA revealed a significant main effect for Time, Wilks' Lambda F (12, 27) = 2.73, [p.bar] < .05, across the 6 dependent variables. Neither the main effect for Group, [F.bar] (6, 33) = 1.21,12 = .33, nor the Time x Group interaction was significant, Wilks' Lambda [F.bar] (12, 27) = 1.75, [p.bar] = .11. Follow-up univariate ANOVAs on each of the sub-scales indicated that the significant multivariate effect of Time was due to significant main effect changes over Time for Positive Affect, [F.bar] (2, 59) = 5.34, [p.bar] < .05; Revitalization, [F.bar] (2, 59) = 8.22, [p.bar] < .05; Exhaustion, [F.bar] (2, 59) = 3.96, [p.bar] < .05; and Tranquility, [F.bar] (2, 59) = 5.51, [p.bar] < .05. Post hoc contrasts revealed significant linear trends for Positive Affect, [F.bar] (1, 38) = 8.38, [p.bar] < .05; Revitalization, F (1, 38) = 3.64,12 < .05; and Tranquility, [F.bar](1,38) = 8.26, [p.bar] < .05; in which scores increased from baseline to each follow-up time point for both groups. In addition, Exhaustion demonstrated a significant quadratic trend, [F.bar] (1, 38) = 6.53, [p.bar] < .05, in which scores increased at the first follow-up time point, and were reduced at the final assessment.

DISCUSSION

This experiment was designed to determine the effect of an acute bout of exercise on the mood of older, Mexican-American women. Results indicated that a 20-minute bout of treadmill exercise, completed in small groups, was sufficient to increase positive affect, revitalization and tranquility in these women. No differences were found for negative affect. However, where significant improvements in mood were reported, the magnitude of these changes was no different from the improvements reported following a comparison condition of small group social interaction. Thus, the completion of 20 min of exercise did not provide a unique improvement in mood beyond the benefits of social interaction.

The failure to find differences relative to the control condition agrees with earlier research. Specifically, numerous studies have shown that exercise is no better than a period of quiet rest in its impact on mood (Morgan, 1985, Petruzzello, et al., 1991), which has led Morgan to suggest that exercise may serve to improve mood as it offers a time out from considering life's worries (Morgan & O'Conner, 1989). Thus, rather than representing a unique effect of exercise, the apparent benefit of exercise may result from the opportunity for social interaction as a distracting event.

In addition, both the exercise and comparison groups were regular exercisers who were participating in a study concerning their feelings about exercising. Therefore, the responses to the comparison condition may have resulted from demand characteristics that included the feelings and mood that they associated with exercising rather than feelings specific to the moment. According to Becker, Roberts and Voelmeck (2003), simply responding to a questionnaire is a treatment and can partially explain similarities between treatment and comparison groups. However, similar procedures have shown differences between exercise and control conditions (Bartholomew & Linder, 1998; Bartholomew & Miller, 2002). It is, therefore, more likely that the degree of social interaction resulted in the enhanced positive feelings of the comparison participants. For example, mood has been significantly greater following exercise in a socially enriched environment relative to a bland environment (Turner, Rejeski & Brawley, 1997) and with more cohesive group exercise settings (Spink & Carron, 1993). This may be particularly true for this population. Emerging data suggests that the health of older, Mexican Americans is enhanced by living in a relatively homogenous neighborhood that provides ample opportunities for social interaction with other Mexican Americans (Patel, Eschbach, Rudkin, Peek & Markides, 2003). As both experimental conditions provided the opportunity to interact with similar-aged, Mexican-American women, it is not surprising that both groups responded with improved mood.

However, these interpretations do not diminish the importance of the primary result, that these older, Mexican-American women did report mood enhancement following exercise. This indicates that women of this ethnicity and age are just as likely to experience enhanced mood in conjunction with exercise as are their younger and non-Hispanic Anglo counterparts.

Thus, it does not appear that the relatively high incidence of sedentary lifestyles among older Mexican Americans is due to a negative emotional state that occurs in conjunction with exercise. In fact, the magnitude of these effects is quite similar to those reported by younger, non-Hispanic Anglo populations (Bartholomew & Miller, 2002). Thus, research should target other aspects of the exercise experience, e.g. self-efficacy, perceived barriers etc. that might serve to explain the demographic differences in participation rates for physical activity.

It is important to point out that these results are limited to those situations in which older Mexican-American women exercise in small groups and can experience some social interaction during their exercise experience. Obviously, many Mexican-American women choose to exercise in isolation (Laffrey, 2000); and it is not clear from these data if this population would experience the same resultant mood following individual exercise as if they had completed group exercise. Future research is required to test this and other possible moderators of the effect. These data are also limited to women who are presently involved in physical activity. Given the low rates of activity in this population (USDHHS, 2000), these women constitute a relative minority of older Mexican-American women. Specifically, these women are likely more motivated to exercise than most women of their age and ethnicity, which calls into question the ability of these effects to hold for less motivated women. Future research must, therefore, be done with sedentary women before these data can be generalized.

In sum, health educators would be wise to consider implementing small group exercise when working with older, Mexican-American women. Earlier data suggest that social interaction with other Mexican Americans is important to the health of this population (Patel, et al., 2003). In addition, regular exercise carries with it a reliable reduction in all-cause mortality, CVD, diabetes, etc (USHHS, 2000). Finally, data from the present experiment suggests that each bout of group exercise will also bring about an improvement in mood. Thus, the use of small group exercise is a strategy that allows health educators to positively benefit this population along a host of dimensions.

REFERENCES

American College of Sports Medicine Position Stand (1998). The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness, and flexibility in healthy adults. Medicine and Science in Sports and Exercise, 30, 975-991.

Bartholomew, J. B., & Linder, D. E. (1998). State anxiety following resistance exercise: The role of gender and exercise intensity. Journal of Behavioral Medicine, 21, 205-219.

Bartholomew, J. B., & Miller, B. M. (2002). Affective responses to an aerobic dance class: The impact of perceived performance. Research Quarterly for Exercise and Sport, 73, 301-309.

Becker, H., Roberts, G., & Voelmeck, W. (2003). Explanations for improvement in both experimental and control groups. Western Journal of Nursing Research, 25, 746-755.

Centers for Disease Control and Prevention (CDC). (2000). Behavioral Risk Factor Surveillance System Survey Data. Atlanta, Georgia: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention.

Crocker, P. R. E. (1997). A confirmatory factor analysis of the Positive Affect Negative Affect Schedule (PANAS) with a youth sport sample. Journal of Sport & Exercise Psychology, 19: 91-97.

Gauvin, L., & Rejeski, W. J. (1993). The Exercise-Induced Feeling Inventory: Development and initial validation. Journal of Sport & Exercise Psychology, 15, 403-423.

Joiner, T. E., Sandin, B., Chorot, P. Lostao, L. & Marquina, G. (1997). Development and factor analytic validity of the SPANAS among women in Spain: (More) cross-cultural convergence in the structure of mood. Journal of Personality Assessment, 68, 600-615.

Laffrey, S.C. (2000). Physical activity among older Mexican American women. Research in Nursing and Health, 23, 383-392.

Laffrey, S.C., Bartholomew, J.B., Lee, Y.S., & & Hernandez-Spina, E. (2000). Development of instruments to measure exercise-related feelings of Mexican American adults. Presented at the Research Conference on Improving the Health of Women and Underserved Populations. Center for Health Promotion Research, The University of Texas at Austin.

McAuley, E., & Courneya, K. S. (1994). The Subjective Exercise Experiences Scale (SEES): Development and Preliminary Validation. Journal of Sport and Exercise Psychology, 16, 163-177.

McAuley E., Shaffer S. M. & Rudolph D. (1995). Affective responses to acute exercise in elderly impaired males: the moderating effects of self-efficacy and age. International Journal of Aging and Human Development, 41, 13-27.

Morgan, W. P. (1985). Affective beneficence of vigorous physical activity. Medicine and Science in Sports and Exercise, 17, 94-100.

Morgan, W. P. & O'Conner, P. (1989). Psychological models. In R. K. Dishman (Ed.), Exercise Adherence: Its Impact on Public Health (pp. 91-121). Champaign, IL: Human Kinetics.

North, C. T., McCullagh, P. & Tran, Z. V. (1990). The effect of exercise on depression. Exercise and Sports Sciences Reviews, 18, 379-415.

Petal, K.V., Eschbach, K., Rudkin, L.L., Peek, M.K., & Markides, K.S. (2003). Neighborhood context and self-rated health in older Mexican Americans. Annuls of Epidemiology, 13, 620-628.

Petruzzello, S. J., Landers, D. M., Hatfield, B. D., Kubitz, K. A., & Salazar, W. (1991). A Meta-Analysis on the Anxiety-Reducing Effects of Acute and Chronic Exercise: Outcomes and Mechanisms. Sports Medicine, 11, 143-182.

Spink, K. S., & Carron, A. V. (1993). The effects of team building on the adherence pattern of female exercise participants. Journal of Sport and Exercise Psychology, 15, 39-49.

Thayer, R. E., Newman J. R. & McClain T. M. (1994). Self-regulation of mood: strategies for changing a bad mood, raising energy, and reducing tension. Journal of Personality and Social Psychology, 67, 910-25.

Turner, E. E., Rejeski, W.J. & Brawley, L. R. (1997). Psychological benefits of physical activity are influenced by the social environment. Journal of Sport and Exercise Psychology, 19, 119-130.

U. S. Department of Health and Human Services (1996). Physical Activity and Health: A Report to the Surgeon General, (pp. 81-172). Atlanta GA: U.S. Department of Health and Human Services, Centers for Disease Control and Prevention, National Centers for Chronic Disease Prevention and Health Promotion.

U.S. Department of Health and Human Services (2000). Healthy People 2010: Understanding and Improving Health. 2nd ed. Washington, DC: U.S. Government Printing Office, November 2000.

Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and validation of brief measures of positive and negative affect: The PANAS scales. Journal of Personality and Social Psychology, 54, 1063-1070.

John B. Bartholomew, PhD

Shirley C. Laffrey, PhD, MPH, RN

Marcus Kilpatrick, PhD

Robert J. Spina, PhD

John B. Bartholomew, Ph.D., is an Associate Professor in the Department of Kinesiology and Health Education at The University of Texas at Austin. Shirley C. Laffery, Ph.D., MPH, R.N., is an Associate Professor in the School of Nursing at The University of Texas at Austin. Marcus Kilpatrick, Ph.D. is an Assistant Professor in the School of PE, Wellness and Sport Studies at the University of South Florida. Robert J. Spina is a Professor in the Department of Kinesiology at San Francisco State University. Address all correspondence to John B. Bartholomew, Ph.D., Department of Kinesiology and Health Education, 222 Bellmont Hall, Austin, TX 78712, PHONE: [phone], FAX: [phone], E-MAIL: [e-mail]

Table 1. Psychological States: means and standard deviations

                        Baseline     Post-test 1    Post-test 2

Positive Affect
  Exercise            40.38 (5.75)   41.24 (6.56)   41.38 (6.65)
  Comparison          40.53 (8.44)   42.47 (6.53)   44.37 (5.64)
  Total               40.45 (7.06)   41.83 (6.54)   42.80 (6.30)

Negative Affect
  Exercise            11.95 (2.91)   10.47 (0.75)   10.57 (1.78)
  Comparison          13.34 (9.37)   13.74 (6.76)   13.05 (5.84)
  Total               12.80 (6.76)   12.03 (4.91)   11.75 (4.35)

Revitalization
  Exercise            3.03 (0.91)    3.31 (0.71)    3.41 (0.63)
  Comparison          3.23 (0.68)    3.44 (0.68)    3.70 (0.38)
  Total               3.13 (0.80)    3.37 (0.69)    3.55 (0.54)

Tranquility
  Exercise            3.30 (0.94)    3.56 (0.55)    3.75 (0.31)
  Comparison          3.56 (0.54)    3.49 (0.65)    3.79 (0.32)
  Total               3.43 (0.78)    3.53 (0.59)    3.77 (0.31)

Positive Engagement
  Exercise            3.27 (0.73)    3.40 (0.58)    3.60 (0.55)
  Comparison          3.54 (0.55)    3.53 (0.54)    3.49 (0.64)
  Total               3.40 (0.65)    3.46 (0.56)    3.55 (0.59)

Physical Exhaustion
  Exercise            0.65 (0.86)    0.70 (0.94)    0.51 (0.76)
  Comparison          1.02 (1.26)    1.39 (0.99)    0.70 (0.88)
  Total               0.82 (1.07)    1.02 (1.01)    0.60 (0.81)