Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
August 2004

The Relation Between Physical Activity and Mental Health Among Hispanic and Non-Hispanic White Adolescents

Author Affiliations

From the School of Public Health, Division of Epidemiology, University of Minnesota, Minneapolis (Ms Brosnahan and Drs Steffen, Lytle, and Patterson); and the Corpus Christi-Nueces County Public Health District, Corpus Christi, Tex (Dr Boostrom).


Copyright 2004 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2004

Arch Pediatr Adolesc Med. 2004;158(8):818-823. doi:10.1001/archpedi.158.8.818

Objective  To assess the relation of physical activity (PA) with feelings of sadness and suicidal thoughts and behaviors among Hispanic and non-Hispanic white adolescent boys and girls.

Design  Cross-sectional study using a modified 2001 Youth Risk Behavior Survey.

Participants  One thousand eight hundred seventy Hispanic and non-Hispanic white adolescents, aged 14 to 18 years, attending high school in Nueces County, Texas.

Main Outcome Measure  Logistic regression analysis was used to assess the relation between PA, including moderate and vigorous PAs, strength and toning, total PA, physical education class, and participation in team sports, and the dependent variables feelings of sadness and considering, planning, and attempting suicide.

Results  More boys reported participating in PA than girls (P<.001), and more girls than boys reported feelings of sadness and considering and planning suicide (P<.001). Greater attendance in physical education class was inversely related to feelings of sadness (odds ratio [OR], 0.80 [95% confidence interval (CI), 0.68-0.94]); participation in more total PA sessions per week was associated with a lower risk of considering suicide (OR, 0.72 [95% CI, 0.65-0.79]); and higher levels of vigorous PA (OR, 0.73 [95% CI, 0.57-0.93]), total PA (OR, 0.65 [95% CI, 0.48-0.87]), and strength and toning activity (OR, 0.64 [95% CI, 0.42-0.99]) were associated with a lower risk of planning suicide.

Conclusions  These findings are consistent with a beneficial effect of PA on feelings of sadness and suicidal behaviors in Hispanic and non-Hispanic white boys and girls. Physical activity may be considered as part of an intervention strategy to improve adolescent health as a whole.

Mental health disturbances, including nonclinical depression, anxiety traits, feelings of sadness, or temporary mood disorders, are associated with morbidity and mortality in the adolescent population.14 The prevalence of mental health disturbances among children and adolescents remains significantly high.14 The 2001 Centers for Disease Control and Prevention Youth Risk Behavior Survey1 (YRBS) indicates that 28.3% of US high school students felt sad or hopeless enough to discontinue doing some usual activities. Another study reported that 21% of US children aged 9 to 17 years possess at least 1 diagnosable mental or addictive disorder associated with minimum functional impairment.2 Approximately 5% of all 9- to 17-year-olds have been diagnosed as having fully developed chronic depression, which is associated with extreme functional impairment.2,3 Furthermore, the use of prescription psychotropic medication, particularly selective serotonin reuptake inhibitors and other antidepressant drugs, progressively increased between 1995 and 1999 by 62% and 195%, respectively, among employer-insured children and adolescents.4

Diagnosed mood disorders significantly increase the risk of suicide.5 Among students who completed the YRBS, 19% seriously considered attempting suicide, 14.8% made a suicide plan, and 8.8% attempted suicide.1 The age-specific mortality rate for suicide among 15- to 19-year-olds in 2000 was 8.2 per 100 000,6 which, although lower than in 1990, remains the third leading cause of death in this age group.7

The detrimental consequences arising from mental health disorders and suicide strongly imply a need to identify strategies to reduce this public health problem. Participation in physical activity (PA) has been associated with positive mood,8 greater self-esteem,9,10 and greater physical and psychological well-being.1114 Academic grade performance is positively associated with participation in PA15,16 and may also enhance self-esteem. Activity may imbue adolescents with energy, as sedentary individuals report fatigue, perhaps a consequence of poor physical fitness.11 Additionally, participation in PA and team sports activities may provide adolescents with a social network that tends to support and protect them from depression.12 Therefore, PA has been suggested as a potential protective agent that may reduce the prevalence of mental health disturbances and suicidal behaviors among adults and adolescents.13,14,1719 Prior studies in adolescents have not, however, examined these relations in large samples of adolescents or in Hispanic youth. The present study explores the hypothesis that PA, physical education class (PE), and participation in team sports are inversely associated with feelings of sadness, suicidal thoughts, and suicidal behaviors among a large sample of Hispanic and non-Hispanic white girls and boys living in Nueces County, Texas.


This study analyzed data from the Healthy Youth/Healthy Adults Study, a cross-sectional study conducted by the Corpus Christi-Nueces County Health District in Corpus-Christi, Tex, in collaboration with the University of Minnesota School of Public Health in Minneapolis. The primary goals of the Healthy Youth/Healthy Adults Study included determining the prevalence of overweight and obesity among adolescents attending high school in Nueces County and administering a modified YRBS questionnaire to determine the prevalence of risky health behaviors among these adolescents. The study protocol was approved by the Corpus Christi-Nueces County Health District internal review board and the University of Minnesota human subjects committee. Consent was obtained from all children and their parents/guardians.


The target population was students in Nueces County high schools. A weighted sampling scheme was implemented to include a larger sample from smaller, rural schools. Twelve of 25 schools in Nueces County were randomly selected to participate in the study. A convenience sample of classrooms was selected from courses required for 9th and 10th grade students.

The study sample was 1870 students who completed the survey of 2000, mostly 9th and 10th graders in these classes (93.5% response rate), including 1344 Hispanic students (71.9%), 406 non-Hispanic white students (21.7%), 90 African American students (4.8%), 17 Pacific Islander students (0.9%), 9 Asian students (0.5%), and 4 American Indian students (0.2%). Because of the small numbers of other race groups (n = 120), only Hispanic and non-Hispanic white students were analyzed. Other exclusions were sex not reported (n = 4), surveys incomplete (n = 26), and height and weight measurements missing (n = 329); 1391 students (74.3%) were included in the analysis.

Body Mass Index

Weight was measured without shoes to the nearest quarter pound using a physician balance scale. Height was measured to the nearest quarter inch using a physician portable scale stadiometer. Body mass index was calculated as weight in kilograms/height in meters squared.

Youth Risk Behavior Survey

Trained personnel administered a modified 2001 YRBS self-report instrument to students during regular school hours. A subset of YRBS data was used in the analyses, including demographic information (age, sex, and race/ethnicity), grades achieved in school (A-F), and smoking and drinking habits in the past 30 days (yes or no). Students were asked 5 questions about participation in PA and 4 questions about feelings of sadness and suicide behaviors (Figure 1).

Image not available

Physical activity and mental health questions from the modified 2001 Youth Risk Behavior Survey.

Physical Activity

Frequency of moderate and vigorous PAs, strength and toning, participation in PE, and participation on sports teams were reported. Moderate PA, total PA, PE, and sports teams were grouped dichotomously, while vigorous PA and strength and toning activity were grouped into tertiles.

The variable "total PA sessions" was created to differentiate students who are active in some form on at least 6 days of the week vs students who are not. This variable combined moderate, vigorous, and strength and toning activity sessions for 1 week and was then grouped into "0 to 5 sessions" and "6 to 21 sessions." The rationale for dichotomizing moderate and total PA was for statistical power, as few students reported moderate activity more than 2 days per week and few students reported more than 10 sessions of total PA per week.

Mental Health

Responses to the feelings of sadness and suicidal thoughts about considering or planning suicide were answered yes or no. However, responses to the last suicide question about attempting to commit suicide were grouped into 0, 1, 2 to 3, 4 to 5, or ≥6 times. Because few students answered with multiple suicide attempts, subjects were divided into "ever attempted" and "never attempted."


Stata 7.0 (Stata Corporation, College Station, Tex) was used, taking the 2-stage sampling design into account. Two-tailed t tests were used to determine variation in the means of continuous variables. The Mantel-Haenszel χ2 statistic was used to test for variation in the distributions of categorical variables. Logistic regression analysis was used to determine the relation between each PA variable and risk of feelings of sadness, considering suicide, planning suicide, or attempting suicide. Models were adjusted for confounding factors including age, sex, ethnicity, body mass index, school, classroom, smoking, drinking, and grades in school. The variables smoking, drinking, and grades in school were included in the regression models because previous studies have shown correlations with PA, feelings of sadness, and depression.10,12,15,16,20,21 Interactions between PA, sex, and ethnicity were tested.


Students' ages ranged from 14 to 18 years (mean, 15.4 years), with no difference by sex or race (Table 1). Seven hundred thirty-one subjects (53%) in the study population were male, and 1077 (77%) were Hispanic. Hispanic students had greater body mass index and received lower grades in school than did non-Hispanic white students (P<.01). More non-Hispanic white boys reported drinking alcohol than Hispanic boys (91[56%] of 165 vs 243 [44%] of 566; P = .007), while more non-Hispanic white girls reported smoking cigarettes than Hispanic girls (52 [35%] of 149 vs 125 [25%] of 511; P = .01). Four hundred fifty-nine students (about 33%) reported using marijuana, cocaine, or inhalants, which was not different across sex or ethnicity (data not shown).

Table 1. 
Image not available
Demographic Characteristics and Physical Activity Habits Among Adolescents According to Sex and Ethnicity*

In Table 1, more boys than girls reported engaging in the highest levels of moderate PA (304 [44%] of 731 vs 205 [33%] of 660; P<.001), vigorous PA (328 [47%] of 731 vs 214 [34%] of 660; P<.001), strength and toning (228 [33%] of 731 vs 135 [21%] of 660; P<.001), and total PA (528 [72%] of 731 vs 386 [58%] of 660; P<.001), as well as participating in more days of PE each week (309 [45%] of 731 vs 235 [38%] of 660; P = .006) and team sports (415 [60%] of 731 vs 316 [50%] of 660; P<.001). Non-Hispanic white girls reported greater prevalence of frequent moderate PA than Hispanic girls (66 [44%] of 149 vs 139 [29%] of 511; P<.001). The frequency of vigorous PA among non-Hispanic white and Hispanic boys and girls was not different. More non-Hispanic white girls than Hispanic girls reported participating in team sports (88 [60%] of 149 vs 228 [48%] of 511; P = .01); there was no ethnic difference in boys.

In Table 2, more Hispanic girls reported feeling sad or hopeless than non-Hispanic white girls (205 [40%] of 511 vs 48 [32%] of 149; P = .08), and more non-Hispanic white boys reported considering suicide than Hispanic boys (28 [17%] of 165 vs 58 [10%] of 566; P = .02). Otherwise, there were no other racial differences among girls or boys for these characteristics. Differences did exist between sexes, as more girls than boys reported feeling sad or hopeless (253 [38%] of 660 vs 164 [23%] of 731; P<.001), considering suicide (153 [23%] of 660 vs 86 [12%] of 731; P<.001), and planning suicide (109 [17%] of 660 vs 69 [10%] of 731; P<.001). There was no difference between sexes for attempting suicide.

Table 2. 
Image not available
Feelings of Sadness and Suicidal Behaviors Among Adolescents According to Sex and Ethnicity*

No significant interactions were found between PA variables and sex or ethnicity. As shown in Table 3, students who participated in 3 to 5 days per week of PE were less likely to feel sad than those participating in PE 0 to 2 days per week, after controlling for sex, ethnicity, age, school, classroom, body mass index, smoking, drinking, and grades in school (odds ratio [OR], 0.80 [95% confidence interval (CI), 0.68-0.94]). Students engaging in more moderate PA, strength and toning activity, and total PA and who were on at least 1 sports team had a lower risk of sadness, although this was not statistically significant. There was no association between vigorous PA and sadness.

Table 3. 
Image not available
Adjusted Odds Ratios of the Risk of Sadness, Considering Suicide, and Planning Suicide Among Hispanic and Non-Hispanic White Adolescents Participating in Physical Activity*

Students engaging in at least 6 sessions of total PA per week were significantly less likely to consider suicide than those engaging in fewer than 6 sessions of total PA per week (OR, 0.72 [95% CI, 0.65-0.79]) (Table 3). Participation in PE 3 to 5 days per week and playing on at least 1 sports team in the past year was associated with a lower risk of considering suicide (ORs, 0.75 and 0.75, respectively); however, these associations were not significant as the 95% confidence intervals included 1. Higher levels of moderate, vigorous, and strength and toning activity were also associated with a nonsignificant, lower risk of considering suicide.

Students were significantly less likely to plan suicide if they engaged in the highest level of vigorous PA (OR, 0.73 [95% CI, 0.57-0.93]), strength and toning activity (OR, 0.64 [95% CI, 0.42-0.99]), and total PA (OR, 0.65 [95% CI, 0.48-0.87]) compared with those in the lowest activity levels (Table 3). Likewise, greater participation in PE and playing on at least 1 sports team in the past year were each associated with a lower, albeit nonsignificant, risk of planning suicide. There was no association found between any of the PA measures and attempting suicide (data not shown).


Physical activity behaviors were inversely associated with feelings of sadness and considering and planning suicide among Hispanic and non-Hispanic white students attending high school in Nueces County. High school students attending more PE sessions per week were less likely to report feelings of sadness. Participation in a greater number of total PA sessions was significantly associated with a lower risk of considering suicide, and higher levels of vigorous PA, strength and toning activity, and total PA were all associated with a decreased risk of planning suicide.

Much evidence supports an inverse association between PA and mental health behaviors in adults and adolescents, including greater self-esteem and psychological well-being.814,18,19,2225 One cross-sectional study found that adolescents who participate in sports 2 to 3 times per week and belong to a sports club or adolescents who participate in sports every day were less likely to attempt suicide than those who participate in sports once a week or never.10 Another study conducted among high school students demonstrated associations between suicide risk and solitary activities, including watching television alone, time spent alone other than watching television, and having nothing to do for more than 1 hour; however, no association between athletic involvement and risk for suicide was found.22 Moderate sport involvement was associated with lower depression scores,23 and vigorous sport and recreational activity displayed an independent but weak association with positive emotional well-being.24

We observed no relation between PAs and risk of attempting suicide in Hispanic and non-Hispanic white adolescents. However, Unger25 reported an inverse association between participation in PA, along with sports team participation, and suicidal thoughts and behaviors among 9th through 12th grade boys but not girls. The study revealed that girls who exercised 6 to 7 days per week and did not participate in team sports were at the greatest risk of suicidal behaviors, while girls who participated in team sports in addition to exercising most frequently were also at a higher risk of reporting suicidal behaviors. The author of this study suggested that these results may be owing to a widespread perception of overweight and negative body image among adolescent girls who then exercise excessively to lose weight. Girls with this perception may then have poor self-esteem that may lead to depression or suicidal behaviors. Brown and Blanton17 also showed that college-aged women who were vigorously and moderately active were more likely to report suicidal behaviors than women who were not active. The reason for the lack of association between PA and attempted suicide in the current study is unclear. It is possible that the ethnic distribution of this study population may be a factor, because PA and suicide attempts may not have the same meaning in Hispanic girls or boys as they do in non-Hispanic white girls or boys or in populations with different ethnic distributions; the lack of ethnic differences in the mental health associations in this study may have resulted from the small number of non-Hispanic white students. National survey data show lower rates of depression and suicide in Mexican Americans than in non-Hispanic white adults.26

Another important finding of this study was the association between PE participation and a lower risk of sadness. This is of particular public health interest given that many schools are currently cutting back on PE funding, and many schools do not require PE after the 8th grade. To date, no other studies have been published that show an association between PE and feelings of sadness and suicidal behaviors, especially among Hispanic adolescents. The results of this study suggest that PE may be beneficial in decreasing the risk of sadness and suicidal behaviors among 9th and 10th grade students. Conversely, adolescents with fewer mental health problems may be more likely to seek out and engage in various forms of PA. However, causality cannot be inferred because temporality between exposure and outcome cannot be determined in this cross-sectional study design. Nevertheless, intervention studies have shown that participation in PA reduces anxiety and depression in adolescents.8,1214 This study was also not designed to measure the quality of the PE lessons. One cannot overlook the fact that the content of PE lessons has been shown to increase PA levels27 and may affect mental health outcomes as well. Additionally, although the state of Texas requires that students complete 3 semesters of PE meeting 5 days per week prior to high school graduation,28 we cannot differentiate in this study between the students who were enrolled in PE to fulfill a requirement or enrolled because they voluntarily elected to take it.

Strengths of this study include the examination of several PA measures. The majority of studies on this topic focus mainly on sports involvement or vigorous PA as an indicator for PA behaviors; however, PA embraces much more than just sports involvement or vigorous activity. The present study addresses this by examining participation in moderate and vigorous PA, strength and toning activity, total PA, PE, and team sports. This study also provides insight into a predominantly Hispanic population (77%). To date, no other studies have looked at the associations of PA with sadness and suicide behaviors in a large Hispanic or other minority population. In addition, weight and height were not based on self-report but measured by trained staff.

The associations observed in the present study should be interpreted with caution. Because this study was cross-sectional, it is impossible to determine the temporal relationship between PA and feelings of sadness and suicidal behaviors. Also, because the responses were based on self-report, it is possible that some respondents may have overreported or underreported PA or sadness and suicidal behaviors in an effort to provide more "socially desirable" answers. Additionally, the YRBS instrument may not be as precise as other methods used to measure PA or mental health parameters, including pedometers, doubly labeled water, or the Centers for Epidemiologic Studies Depression Scale.29 Furthermore, the validity and reliability of the YRBS have not been verified in a predominantly Hispanic population. However, within the limitations of the study design, the results of this study reveal an interesting and potentially important relationship between PA behavior with sadness and suicidal intentions. It is possible that PAs often take place with other peers and thus increase social exposure and self-esteem. Other mechanisms by which PA may affect mental health or well-being include positive physiological responses from endorphin30 and/or monoamine31 production.


The results of the present study suggest that higher levels of PA is one factor that is associated with a decreased risk of sadness and suicidal intentions among Hispanic and non-Hispanic white adolescents living in Nueces County. Promotion of PA may be a useful part of intervention programs that include the reinforcement of other protective factors such as efficient social skills or a high self-esteem and aim at improving adolescents' health as a whole. Thus, future studies should examine the effects of a global health-promoting approach on the prevalence of sadness and suicidal thoughts and behaviors among adolescents.

What This Study Adds

Feelings of sadness and suicidal behaviors are components of mental health disturbances that contribute to morbidity and mortality in the adolescent population. Because feelings of sadness and suicidal behaviors among adolescents remain high, an important public health goal is to identify strategies or effective means to prevent and reduce the prevalence of these behaviors. This study reports the relationship of a variety of PAs with sadness and suicidal behaviors among a large sample of Hispanic and non-Hispanic white adolescents. The results suggest that PA may assist in reducing feelings of sadness and other mental health disturbances in a predominantly Hispanic population.

Back to top
Article Information

Correspondence: Lyn M. Steffen, PhD, MPH, 1300 S Second St, Suite 300, Minneapolis, MN 55454-1015 (

Accepted for publication April 14, 2004.

This study was funded by the Texas Department of Health, Austin.

Grunbaum  JAKann  LKinchen  SA  et al.  Youth risk behavior surveillance—United States, 2001. MMWR Surveill Summ. 2002;511- 64
Shaffer  DFisher  PDulcan  MK  et al.  The NIMH Diagnostic Interview Schedule for Children Version 2.3 (DISC 2.3), description, acceptability, prevalence rates, and performance in the MECA Study: Methods for the Epidemiology of Child and Adolescent Mental Disorders Study. J Am Acad Child Adolesc Psychiatry. 1996;35865- 877
Shaffer  DGould  MSFisher  P  et al.  Psychiatric diagnosis in child and adolescent suicide. Arch Gen Psychiatry. 1996;53339- 348
Shatin  DDrinkard  CR Ambulatory use of psychotropics by employer-insured children and adolescents in a national managed care organization. Ambul Pediatr. 2002;2111- 119
US Department of Health and Human Services, Mental health: a report of the surgeon general [Office of the Surgeon General Web site]. Available at: October 24, 2002
Centers for Disease Control and Prevention, National Center for Health Statistics, Health, United States, 2002. Trends in the health of Americans: mortality Available at: January 4, 2003
Hoyert  DLKochanek  KDMurphy  SL Deaths: final data for 1997. Natl Vital Stat Rep. 1999;471- 104
Williamson  DDewey  ASteinberg  H Mood change through physical exercise in nine- to ten-year-old children. Percept Mot Skills. 2001;93311- 316
Sonstroem  RJ Exercise and self-esteem. Exerc Sport Sci Rev. 1984;12123- 155
Ferron  CNarring  FCauderay  MMichaud  PA Sport activity in adolescence. Health Educ Res. 1999;14225- 233
Centers for Disease Control and Prevention, Physical Activity and Health: Report of the Surgeon General.  Atlanta, Ga US Dept of Health and Human Services, Centers for Disease Control and Prevention1996;
Kirkcaldy  BDShephard  RJSiefen  RG The relationship between physical activity and self-image and problem behavior among adolescents. Soc Psychiatry Psychiatr Epidemiol. 2002;37544- 550
Calfas  KTaylor  WC Effects of physical activity on psychological variables in adolescence. Pediatr Exerc Sci. 1994;6406- 423
Calfas  K The relationship between physical activity and the psychological well-being of youth. Rippe  JMed.Lifestyle Medicine. Malden, Mass Blackwell Science1999;967- 979
Daley  AJRyan  J Academic performance and participation in physical activity by secondary school adolescents. Percept Mot Skills. 2000;91531- 534
Pate  RRHeath  GWDowda  MTrost  SG Associations between physical activity and other health behaviors in a representative sample of US adolescents. Am J Public Health. 1996;861577- 1581
Brown  DRBlanton  CJ Physical activity, sports participation, and suicidal behavior among college students. Med Sci Sports Exerc. 2002;341087- 1096
Taylor  CBSallis  JFNeedle  R The relation of physical activity and exercise to mental health. Public Health Rep. 1985;100195- 202
Morgan  WPGoldston  SE Exercise and Mental Health.  Washington, DC Hemisphere Publishing Corp1987;
Albers  ABBiener  L The role of smoking and rebelliousness in the development of depressive symptoms among a cohort of Massachusetts adolescents. Prev Med. 2002;34625- 631
Rey  JMSawyer  MGRaphael  BPatton  GCLynskey  M Mental health of teenagers who use cannabis. Br J Psychiatry. 2002;180216- 221
Mazza  JJEggert  LL Activity involvement among suicidal and nonsuicidal high-risk and typical adolescents. Suicide Life Threat Behav. 2001;31265- 281
Sanders  CEField  TMDiego  MKaplan  M Moderate involvement in sports is related to lower depression levels among adolescents. Adolescence. 2000;35793- 797
Steptoe  AButler  N Sports participation and emotional well-being in adolescence. Lancet. 1996;3471789- 1792
Unger  J Physical activity, participation in team sports, and risk of suicidal behavior in adolescents. Am J Health Promot. 1997;1290- 93
Oquendo  MAEllis  SPGreenwald  SMalone  KMWeissman  MMMann  JJ Ethnic and sex difference in suicide rates relative to major depression in the United States. Am J Psychiatry. 2001;1581652- 1658
McKenzie  TLNader  PRStrikmiller  PK  et al.  School physical education. Prev Med. 1996;25423- 431
Texas Education Agency, Texas state graduation requirements. Available at: September 23, 2003
Garrison  CZAddy  CLJakson  KLMcKeown  REWaller  JL The CES-D as a screen for depression and other psychiatric disorders in adolescents. J Am Acad Child Adolesc Psychiatry. 1991;30636- 641
North  TCMcCullagh  PTran  ZR Effect of exercise on depression. Exerc Sport Sci Rev. 1990;18379- 415
Ransford  CP A role for amines in the antidepressant effect of exercise: a review. Med Sci Sports Exerc. 1982;141- 10