Secondary Logo

Journal Logo

Physical Activity, Sport Participation, and Suicidal Behavior: U.S. High School Students

BROWN, DAVID R.1; GALUSKA, DEBORAH A.1; ZHANG, JIAN1; EATON, DANICE K.2; FULTON, JANET E.1; LOWRY, RICHARD2; MAYNARD, L. MICHELE1

Medicine & Science in Sports & Exercise: December 2007 - Volume 39 - Issue 12 - p 2248-2257
doi: 10.1249/mss.0b013e31815793a3
APPLIED SCIENCES: Psychobiology and Behavioral Strategies
Free

Purpose: To evaluate the associations of physical activity and sports team participation with suicidal behavior among U.S. high school students.

Methods: Data were from the 2003 Youth Risk Behavior Survey (N = 10,530 respondents). Exposure variables included physical activity (inactive, insufficient, moderately intensive, regular vigorously intensive, and frequent vigorously intensive) and sports team participation. Outcome variables were suicide ideation (seriously considering and/or planning suicide) and suicide attempts. Hierarchical logistic regressions were run, controlling for age, race, smoking, alcohol use, drug use, geographic region, unhealthy weight-control practices, and body mass index/weight perceptions.

Results: Compared with inactive students or sports team nonparticipants, the odds of suicide ideation were lower among boys reporting frequent vigorous-intensity physical activity (adjusted odds ratio (AOR) = 0.48; 95% confidence interval (CI) = 0.29, 0.79) and sports team participation, respectively (AOR = 0.65; 95% CI = 0.48, 0.86). The odds of suicide attempts were also lower among frequently vigorously active boys (AOR = 0.44; 95% CI = 0.21, 0.96) and sports team participants (AOR = 0.61; 95% CI = 0.40, 0.93). The odds of suicide attempts were lower for regular vigorously active girls compared with inactive girls (AOR = 0.67; 95% CI = 0.45, 0.99) and sports team participants compared with nonparticipants (AOR = 0.73; 95% CI = 0.57, 0.94). Associations with one exposure variable generally weakened when adjustment was made for the other exposure variable, or for feeling sad and hopeless.

Conclusions: The association of physical activity and sports team participation with suicide ideation and suicide attempts varied by sex. Further research is needed to clarify these different associations.

1Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Nutrition, Physical Activity, Obesity; and 2Division of Adolescent and School Health, Atlanta, GA

Address for correspondence: David R. Brown, Ph.D., Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Health Promotion, Division of Nutrition, Physical Activity, and Obesity, Physical Activity and Health Branch, Mailstop K-46, 4770 Buford Hwy. N.E., Atlanta, GA 30341-3724; E-mail: DBrown@cdc.gov.

Submitted for publication March 2007.

Accepted for publication August 2007.

In the United States, suicide among youth is a major public health problem (9). According to the U.S. Centers for Disease Control and Prevention's Web-based Injury Statistics Query and Report System (Leading Causes of Death) (http://www.cdc.gov/ncipc), suicide was the third-leading cause of death among persons 10-14 and 15-24 yr of age in 2004. The magnitude of the problem has lead to a Healthy People 2010 national objective, 18-2, to reduce the rate of suicide attempts by adolescents (35). Although predicting who will attempt suicide, and thus preventing suicide, remains elusive, several risk factors that may place persons at risk for attempting suicide are potentially modifiable. These include lifestyle behaviors such as smoking (4,17,18,22), alcohol use (2,10,17,18,22), and drug use (2,17,18,22). It is possible that physical inactivity, like other modifiable lifestyle behaviors, is related to suicidal behaviors.

Numerous hypothetical mechanisms have been given as to why participation in physical activity or sports may be linked to reduced odds of suicide ideation or attempts (5). Some explanations propose that active or athletic lifestyles are indirectly and not causally related with suicide ideation or attempts. For example, researchers have proposed that the interpersonal support that sports participants receive from coaches, teammates, parents, and friends may provide athletes with a therapeutic support base that reduces the risk of suicide during difficult times (6,28,34). In addition, persons who engage in sports may also adopt other positive lifestyle behaviors, and this constellation of behaviors may predispose against suicidal behaviors (28).

People who are regularly physically active may also gravitate to or possess other positive lifestyle behaviors that reduce their risk for suicide. Physical activity is also associated with numerous mental health benefits (25), including enhanced emotional health (13,24,30), improved cognitive functioning (12,15), and better quality of life (7). These benefits could mediate a lower risk of suicide among physically active people than people who are sedentary. Quite possibly, biological mechanisms may also directly mediate the relationship between physical activity with suicidal behaviors. For example, deficient serotonergic functioning may play a role in suicide and suicide attempts (2,22,23), and mood improvements associated with physical activity may reflect increased levels of brain serotonin (11,13,24). Theoretically, physical activity may be inversely related to suicidal behavior as a result of neurobiological alterations that occur with physical activity.

Researchers have evaluated the association of physical activity and sports participation with suicidal behaviors, and a review of this literature summarizes findings (5). Some, but not all, studies show that physical activity and sports participation protect against suicidal behaviors (5). The lack of uniform findings may partly be attributable to different definitions of the dependent and independent variables used across the studies. The number and type of variables adjusted for in each study also differ widely, and many of the studies do not account for mental health variables, such as distress or depression, that may mediate the relationship between physical activity/sports participation and suicidal behaviors.

The purpose of this study was to extend the research evaluating the association of physical activity and sports team participation with suicide ideation and suicide attempts among U.S. high school students, using data from the National 2003 Youth Risk Behavior Survey (YRBS). We evaluated physical activity and sports team participation separately to determine the independent effects of these activity domains with suicide ideation and suicide attempts. Unlike previous research on this topic pertaining to students in the 9th-12th grades, we also controlled for self-reported feelings of sadness or hopelessness to evaluate whether this measure confounded, perhaps mediated, the associations of physical activity/sports team participation with suicide ideation/suicide attempts. In addition, we evaluated several levels of physical activity to determine whether a dose-response relationship existed between physical activity and suicide ideation and suicide attempts.

Back to Top | Article Outline

METHODS

Survey Design

Data from the 2003 YRBS were analyzed. The national school-based YRBS, which has been approved by an institutional review board at the Centers for Disease Control and Prevention (CDC) and is implemented by CDC, monitors the prevalence of priority health risk behaviors among youth. In 2003, a three-stage cluster-sample design was used to obtain a nationally representative sample of students in grades 9-12 in the 50 states and the District of Columbia. Following local procedures, parental permission was obtained before survey administration, and participation in the survey was voluntary. Students recorded their responses anonymously on computer-scannable answer sheets. Additional details on the sampling methodology are available elsewhere (8).

Back to Top | Article Outline

Study Sample

The response rate for the 2003 YRBS was 81% for schools and 83% for students, resulting in a 67% overall response rate. A total of 15,214 useable questionnaires were available for analysis. As a result of missing data (N = 4684) for variables in our analysis, including missing data for physical activity (N = 337), sports participation (N = 505), suicide ideation (N = 93), suicide attempts (N = 1787), and other variables (N = 1962), 10,530 students were included in the final study sample (69% of the total sample).

Back to Top | Article Outline

Measures

Outcome variables: suicide ideation and attempts.

Suicide ideation was defined as responding yes to either of the following two questions: "During the past 12 months, did you ever seriously consider attempting suicide?" or "During the past 12 months, did you make a plan about how you would attempt suicide?" Seventy-nine percent (girls 82%; boys 74%) of the study population we defined as reporting suicide ideation indicated that they had both seriously considered and made plans for suicide during the past 12 months. Therefore, we defined suicide ideation as seriously considering and/or planning suicide, because of the substantial overlap among students who endorsed both items. Suicide attempts were defined as responding one time or more to the question, "During the past 12 months, how many times did you actually attempt suicide?"

Back to Top | Article Outline

Exposure variables: physical activity and sports team participation.

Vigorous-intensity physical activity was assessed by the question, "On how many of the past 7d did you exercise or participate in physical activity for at least 20 min that made you sweat and breathe hard, such as basketball, soccer, running, swimming laps, fast bicycling, fast dancing, or similar aerobic activities?" (0, 1, 2, 3, 4, 5, 6, or 7 d). Moderate-intensity physical activity was assessed by the question, "On how many of the past 7 d did you participate in physical activity for at least 30 min that did not make you sweat or breathe hard, such as fast walking, slow bicycling, skating, pushing a lawn mower, or mopping floors?" (0, 1, 2, 3, 4, 5, 6, or 7 d). Student responses to these two questions were used to create a measure of physical activity with the following categories: frequent, vigorous-intensity physical activity (engaged in vigorous-intensity physical activity for at least 20 min during 6 or more of the past 7 d); regular, vigorous-intensity physical activity (engaged in vigorous-intensity physical activity for at least 20 min during 3-5 of the past 7 d); moderate-intensity physical activity (engaged in moderate-intensity physical activity at least 30 min on 5 or more of the past 7d); insufficient physical activity (engaged in moderate-intensity physical activity at least 30 min on 1-4 of the past 7 d, or engaged in vigorous-intensity physical activity at least 20 min on 1 or 2 of the past 7 d); and inactive (engaged in 0 d of moderate- and vigorous-intensity physical activity during the past 7 d). All physical activity categories were mutually exclusive, and students were categorized into the highest level of physical activity reported. For example, if a student reported doing frequent, vigorous-intensity physical activity, he or she would not also be in the regular vigorous- or moderate-intensity physical activity groups.

Sports team participation was assessed with the question, "During the past 12 months, on how many sports teams did you play?" (0 teams, 1 team, 2 teams, or 3 or more teams). Students who answered that they played on one or more sports teams were categorized as sports team participants.

Back to Top | Article Outline

Covariate Categories

The analyses controlled for several demographic characteristics (students' age, race, and geographic region of school location). We also controlled for lifestyle behaviors that could influence our findings, including cigarette smoking, alcohol use, drug use, and unhealthy weight-control practices. Unhealthy weight-control practices have been associated with suicide (16,33), and some female athletes who participate in sports that emphasize leanness report symptoms of disordered eating, although this may reflect a dedication to excel in a given sport rather than psychopathology (27). We further controlled for select personal characteristics that may influence the physical activity/sports team participation and suicidal behavior relationships. Magnusson et al. (21) found an inverse association between BMI and suicide mortality, whereas Eaton and colleagues (14) found that weight perceptions of adolescents' may be more important than actual weight in terms of increased odds of suicidal tendencies. Unger (34) has proposed that concerns about being overweight among 9th- to 12th-grade girls may be associated with frequent physical activity and greater risk of suicide ideation, plans, and attempts. We hypothesized that discordant weight and weight perceptions (e.g., underweight body mass index and overweight perceptions) may mediate or confound the physical activity/sports team participation with suicide ideation/suicide attempt relationships, and we controlled for a combined BMI/weight perception variable. Depression or depressed mood has been found to be associated with both physical activity (13,24) and suicide (1,2,22), and, therefore, it may mediate or confound the physical activity and suicide association. We were able to control for students who reported feeling sad or hopeless. Hopelessness has been identified as an independent risk factor for completed suicide, suicide attempts, and suicide ideation (20).

Back to Top | Article Outline

Covariate Measures

YRBS questions related to the age and race of survey respondents and the geographic location of their schools were used to define the demographic characteristics used in this study. In regard to lifestyle characteristics, we controlled for the frequency or number of days of cigarette smoking using the YRBS item, "During the past 30 d, on how many days did you smoke cigarettes?" We also controlled for alcohol use with and without episodic heavy drinking during the 30 d preceding the survey. We used two items from the YRBS to define alcohol use: "During the past 30 d on how many days did you have at least one drink of alcohol?" and "During the past 30 d, on how many days did you have five or more drinks of alcohol in a row, that is, within a couple hours?" Students who reported that they had five or more drinks of alcohol in a row 10 or more times during the past 30 d were defined as frequent episodic heavy drinkers. Those who drank five or more drinks in a row one to nine times during the past 30 d were defined as doing some episodic heavy drinking. Students who did not engage in episodic heavy drinking were then defined as either drinking alcohol without episodic heavy drinking (those having at least one drink of alcohol on one or more days during the past 30 d and 0 d of having five or more drinks of alcohol in a row) or as nondrinkers (those who reported zero drinks during the past 30 d). We controlled for any drug use using the following three YRBS items: 1) "During the past 30 d, how many times did you use marijuana?" 2) "During the past 30 d, how many times did you use any form of cocaine, including powder, crack, or freebase?" and 3) "During the past 30 d, how many times have you sniffed glue, breathed the contents of aerosol spray cans, or inhaled any paints or sprays to get high?" We controlled for students who reported any unhealthy weight-control practices according to the following three YRBS items: 1) "During the past 30 d, did you go without eating for 24 h or more (also called fasting) to lose weight or keep from gaining weight?" 2) "During the past 30 d, did you take any diet pills, powders, or liquids without a doctor's advice to lose weight or keep from gaining weight?" and 3) "During the past 30 d, did you vomit or take laxatives to lose weight or keep from gaining weight?"

Personal characteristics included a combined body weight status and weight perception variable that was created according to a student's BMI category and a YRBS question about self-perception of body size. Weight status was defined by a student's BMI, which was calculated from self-reported height and weight, using the formula weight (kg)/height (m2). Using CDC growth charts (19), students were categorized as 1) underweight (includes students with BMI < 15th percentile), 2) normal weight (BMI ≥ 15th percentile to < 85th percentile), or 3) overweight (includes students with BMI ≥ 85th percentile). Body weight perception was assessed by the question, "How do you describe your weight?" Students who answered very underweight or slightly underweight were categorized as perceiving themselves to be underweight. Students who responded slightly overweight or very overweight were categorized as perceiving themselves overweight. The remaining students considered themselves to be about the right weight. Student BMI category and body weight perception were combined to form a body weight status/weight perception variable coded as follows: 1) concordant BMI and weight perceptions (i.e., a BMI of underweight and an underweight perception, BMI of normal weight and about the right weight perception, and a BMI of overweight and an overweight perception), 2) discordant BMI and higher weight perceptions (i.e., a student's perceived weight was higher than his or her actual BMI weight status category), and 3) discordant BMI and lower weight perceptions (i.e., a student's perceived weight was lower than the actual BMI weight status category). Finally, we controlled for feeling sad or hopeless using the YRBS question, "During the past 12 months, did you ever feel so sad or hopeless almost every day for 2 wk or more in a row that you stopped doing some usual activities?"

Back to Top | Article Outline

Data Analysis

All hypothesis testing was performed on weighted data using SUDAAN, a software package that accounts for complex sampling (31). Adjusted odds ratios (AOR) and 95% confidence intervals (CI) were calculated to measure the association of 1) physical activity with suicide ideation, 2) physical activity with suicide attempts, 3) sports team participation with suicide ideation, and 4) sports team participation with suicide attempts. Four hierarchical regression models were examined for each of the four associations of physical activity or sports team participation with suicide ideation or suicide attempts. All four regression models were adjusted for the basic covariates of age, race/ethnicity, geographic region, number of days engaged in cigarette smoking during the 30 d preceding the survey, alcohol use during the 30 d, preceding the survey, drug use during the 30 d preceding the survey, use of unhealthy weight-control practices, and BMI/weight perception. Model 1 examined the associations of the exposure variable (physical activity or sports team participation) with the outcome variable of interest (suicide ideation or suicide attempts) by adjusting for the influence of these basic covariates. Model 2 adjusted for the basic covariates and further adjusted for sports team participation (when physical activity was the exposure variable) or for physical activity (when sports team participation was the exposure variable of interest). Model 3 was adjusted for the covariates in the basic model (Model 1) as well as for feeling sad or hopeless. The full model (Model 4) adjusted for physical activity or sports team participation as appropriate along with feeling sad or hopeless and all of the basic covariates. Results were stratified by sex.

Back to Top | Article Outline

RESULTS

Overall, 19% (N = 1898) of students reported suicide ideation, and approximately 7% (N = 774) reported one or more suicide attempts during the past 12 months. As noted in Table 1, 23% (N = 1227) of girls reported suicide ideation, and almost 11% (N = 565) reported one or more suicide attempts. Results from Cochran-Mantel-Haenszel testing show that both suicide ideation and attempts among girls differed significantly by race/ethnicity, sports team participation, cigarette smoking, alcohol use, drug use, unhealthy weight-control practices, and feeling sad or hopeless. Girls exhibited a significant difference in suicide attempts by age and in suicide ideation by BMI/body weight perceptions.

TABLE 1

TABLE 1

Findings in Table 2 show that approximately 14% (N = 671) and 4% (N = 209) of boys reported suicide ideation and suicide attempts, respectively. Among boys, a significant difference was found for both suicide ideation and attempts by race/ethnicity, sports team participation, cigarette smoking, alcohol use, drug use, unhealthy weight-control practices, and feeling sad or hopeless. Boys also exhibited a significant difference in suicide ideation by level of physical activity.

TABLE 2

TABLE 2

Back to Top | Article Outline

Physical activity.

The odds of suicide ideation were lower among boys who engaged in frequent, vigorous-intensity physical activity than for boys who were physically inactive (Table 3) (unadjusted odds ratio (UOR) = 0.52; 95% CI = 0.33, 0.83; model 1 adjusted odds ratio (AOR) = 0.48; 95% CI = 0.29, 0.79). The association remained significant after adjusting for sports team participation (model 2), feeling sad or hopeless (model 3), and both (model 4). No significant association was observed between physical activity and suicide ideation among girls from either unadjusted or adjusted models.

TABLE 3

TABLE 3

The odds of suicide attempts were lower for boys who engaged in frequent, vigorous-intensity physical activity (Table 3) (model 1 AOR = 0.44; 95% CI = 0.21, 0.96) than for boys who were physically inactive. However, the association was diminished and became nonsignificant after further adjusting for sports participation (model 2), feeling sad or hopeless (model 3), or both (model 4). The odds of suicide attempts were lower for girls who reported that they engaged in regular (3-5 d) vigorous-intensity physical activity during the past week (model 1 AOR = 0.67; 95% CI = 0.45, 0.99). The association between regular, vigorous-intensity physical activity and suicide attempts among girls was weakened and became nonsignificant after adjusting for sports participation or feeling sad or hopeless, or both.

Back to Top | Article Outline

Sports team participation.

Boys who were sports team participants were at significantly lower odds of suicide ideation (Table 4) (UOR = 0.60; 95% CI = 0.45, 0.80; model 1 AOR = 0.65; 95% CI = 0.48, 0.86) and suicide attempts (UOR = 0.50; 95% CI = 0.33, 0.76; model 1 AOR = 0.61; 95% CI = 0.40, 0.93) than were nonparticipants. However, the associations were attenuated and became nonsignificant after further adjusting for physical activity (model 2), feeling sad or hopeless (model 3), or both (model 4).

TABLE 4

TABLE 4

Girls who were sports team participants had significantly lower unadjusted odds of suicide ideation than did nonparticipants (Table 4) (UOR = 0.76; 95% CI = 0.62, 0.94), but this association weakened and became nonsignificant after adjusting for covariates (models 1-4). Both the unadjusted (UOR = 0.66; 95% CI = 0.53, 0.82) and adjusted (model 1 AOR = 0.73; 95% CI = 0.57, 0.94) odds of suicide attempts were lower among girls who were sports team participants than among nonparticipants. This association remained significant after adjusting for the covariates in model 1 plus physical activity (model 2 AOR = 0.71; 95% CI = 0.54, 0.94), but it weakened and became nonsignificant after adjusting for feeling sad or hopeless (model 3) and feeling sad or hopeless and physical activity level (model 4).

Back to Top | Article Outline

DISCUSSION

We examined the association of physical activity and sports team participation with suicide ideation and suicide attempts among U.S. high school students. Associations were found for students who reported vigorous-intensity physical activity or sports team participation with a reduced risk of suicide ideation or suicide attempts. However, findings differed among boys and girls, and the associations were weakened by adjusting for select variables, with one exception. For boys, the relationship of frequent, vigorous-intensity physical activity with lower odds of suicide ideation persisted across all models.

Overall, our findings were attenuated when adjustment was made for exposure variables (i.e., when level of physical activity was adjusted for in the sports team participation analyses and sports team participation was adjusted for in the physical activity analyses). Possibly, the association between physical activity and suicide ideation and attempts, and between sports team participation and suicide ideation and attempts, were weakened after controlling for one exposure variable with the other, because the variables are at least moderately correlated. This raises the question as to which of the two variables-vigorous-intensity physical activity or sports team participation-protects against suicide ideation and attempts. Post hoc analyses among boys showed no significant interaction of sports team participation and physical activity with suicide ideation (Wald F = X; P = 0.239) or suicide attempts (Wald F = X; P = 0.355). On the other hand, post hoc analyses among girls showed a significant interaction of sports team participation and physical activity with suicide attempts (Wald F = X; P = 0.019). Analyses (controlling for all covariates, including feelings of sadness/hopelessness) stratified by sports team participation revealed that vigorous physical activity was associated with lower odds of suicide attempt among girls who did not participate on a sports team (OR = 0.53; 95% CI = 0.31, 0.91), but not among girls who did participate on a sports team (OR = 1.27; 95% CI = 0.53, 3.06). These data imply that girls who do not participate on sports teams seem to be the most likely to benefit from vigorous physical activity.

We also investigated whether an indicator of mental health (i.e., feeling sad or hopeless) confounded the relationship between physical activity/sports team participation and suicide ideation and attempts. With the exception of a lower odds of suicide ideation among boys who participated in weekly, frequent, vigorous-intensity physical activity, adjusting for feeling sad or hopeless in our models weakened the obtained associations, such that several significant findings became nonsignificant. This finding 1) suggests that distress/depressed mood or diagnosed depression may confound, or possibly mediate, the activity-suicide associations, and 2) reinforces the importance of accounting for mood disturbance or psychopathology in studies on this topic. For example, in a previous study by Brown and Blanton (6), physically active college women were found to be at higher odds of suicidal behaviors than inactive women. We did not control for a measure of mental health in this earlier research, and it would have been instructive to determine to what extent feeling sad or hopeless, depressed mood, or depression may have affected the study outcomes. However, the National College Health Risk Behavior Survey analyzed in the Brown and Blanton study (6) did not include questions related to symptoms of depression, or respondents identified as having diagnosed depression, so this was not possible.

The findings in this current study also differ from those of Simon et al. (32), who interviewed young adults aged 13-34 yr old within a few days after the interview subjects had made nonlethal suicide attempts. During the interviews, the suicide attempters were asked about their physical activity behaviors the month before their attempts and were screened for select suicide risk factors: depressive symptoms, hopelessness, alcoholism, and any serious medical condition. Physical activity behaviors and suicide risk factors were also evaluated during the month preceding the interview among case control subjects who had not attempted suicide. Suicide attempters were significantly less likely to report participation in physical activity the month preceding their suicide attempts, compared with the physical activity levels reported by control subjects the month before their interview. The association remained significant after adjusting for study covariates, including depressive symptoms. The difference between our observations and those of Simon et al. (32) may be attributable to the fact that we compared current self-reported physical activity behavior (past 7 d) with self-reported suicide attempts that occurred during the past 12 months, whereas Simon et al. compared current self-reported physical activity behavior (past month) in close proximity to actual suicide attempts. The different measures (actual vs self-reported suicide attempts) and the time frames during which physical activity behaviors and suicide attempts were assessed may account for the different study findings.

Our findings do not support a conclusion that a dose-response relationship exists between level of physical activity and suicide ideation and suicide attempts. Rather, they suggest that some physical activity threshold level may be required to reduce the risk of suicide ideation and attempts.

Back to Top | Article Outline

Limitations

This study is subject to several limitations. The major one is its cross-sectional design, which means that the direction of causality related to our findings cannot be determined. Perhaps, physical activity or sports team participation leads to reduced risk of suicide ideation and attempts. On the other hand, students who are not predisposed in some way to suicide ideation or attempts may be more likely than students who are at risk for suicidal thoughts or behaviors to participate in physical activity and sports. For example, select personality traits have been found to be associated with both suicide (29) and successful sports participation (26). Conversely, mental disorders increase the risk of suicide (2), and psychopathology impairs sports performance (26). This may mean that many persons with certain personality traits or mental disorders will tend to self-select out of physical activity or sport participation.

Second, the data are self-reported and may have associated biases. For example, a tendency may exist for some students to answer in a socially desirable manner (i.e., they may report being more physically active than they actually are).

Third, students may not accurately recall activities they did in the past. However, because the timeline for the YRBS physical activity items is the past 7 d, students may more easily recall their physical activities in this recent time period. In addition, a test-retest reliability study has shown that students generally report consistent information about their risk behavior participation (3).

Fourth, we hypothesized that the association of physical activity with suicide ideation and attempts would differ according to the amount of physical activity the students reported. The students were categorized according to their highest levels of self-reported activity. Because the groups were mutually exclusive, some students who were in the vigorous-intensity groups were also moderately active, and some students who were categorized in the moderate-intensity physical activity group may have also been doing some (1 or 2 d of) vigorous-intensity physical activity. Although this introduces some measurement error in our analyses, it also reflects the reality that daily physical activity is a composite of varying intensity activities. Using questionnaire items, we are unable to completely and precisely disentangle these physical activity measurement issues. When possible, research questions related to physical activity and suicide should use more precise monitoring methods to assess physical activity or sports participation. These measures may include the use of one or more methods, such as accelerometers, physical activity logs, or direct observation.

Fifth, different types of physical activities and sports teams may protect against suicide ideation and attempts in different ways. We were unable to determine, from the YRBS, the specific physical activities students were engaged in, or on which types of sports teams they played. Therefore, we were unable to evaluate whether certain types of physical activities (e.g., aerobic activities vs resistance training; running vs Tai Chi) or sports (individual vs team sports; sports focusing on leanness or body image vs sports that do not; school- vs community-based sports teams) were more protective against suicide ideation and attempts than others, and whether such differences varied by sex. Future research is needed to examine the extent to which specific types of physical activities and sports are protective against suicide ideation and behaviors.

Sixth, the ability to identify a significant association may have been constrained by excluding a large portion (30%) of the YRBS sample because of missing data. Despite missing data on one of the physical activity or demographic variables, 33% of the students in the exclusion group reported suicide ideation, and 13% reported suicide attempts. This compares with 19% of the study population who reported suicide ideation and 7% who reported suicide attempts.

Seventh, although we controlled for numerous variables that can covary with the physical activity/sports team participation and suicide ideation/attempts associations, we did not control for all variables that may have confounded our findings (e.g., we did not assess personality traits that may be related to both sports participation and suicide ideation and attempts). Also, the current study did not include data from students who ultimately died by suicide. Quite possibly, the associations we observed may be stronger among other high-risk populations than among samples included in the current study.

Back to Top | Article Outline

Strengths

Strengths of our study include, first, that the YRBS is a unique surveillance survey that assesses health and risk factor behaviors of a nationally representative sample of U.S. high school students. Second, the YRBS allows for evaluating associations among health outcomes and risk factor behaviors for both boys and girls. Third, the YRBS provided us with data on the exposure and outcome variables related to our research question, and it enabled us to control for numerous variables that could confound our findings.

Back to Top | Article Outline

CONCLUSIONS AND RECOMMENDATIONS FOR FUTURE RESEARCH

Our findings indicate that boys who participated in frequent, vigorous-intensity physical activity or on a sports team were less likely to engage in suicide ideation and suicide attempts than were their inactive or sports team nonparticipant peers, respectively. Regular, vigorous-intensity physical activity and sports team participation among girls were associated with a lower likelihood of attempting suicide. From the outset, we wanted to evaluate the independent effects of physical activity and sports team participation with suicide ideation and attempts. Our findings suggest that regular vigorous-intensity physical activity for girls, frequent vigorous-intensity physical activity for boys, and sports team participation among boys and girls contribute independently to lowered odds of suicide ideation and/or attempts. Yet, physical activity and sports team participation are significantly associated to such a degree that adjusting for one behavior attenuates the relationship with the other behavior. This finding indicates that a measure of total energy expenditure or a combined physical activity and sports team participation measure may be a more precise indicator of the activity-suicide association than assessment of either behavior alone. Importantly, our findings also suggest that a third variable-feeling sad or hopeless-may confound or, perhaps, mediate, the associations of physical activity and sports team participation with suicide ideation and suicide attempts among U.S. high school students. Further research on this topic is needed to identify the temporal relationships between physical activity and sports team participation with suicide ideation and suicide attempts. Research is also needed to identify the biological, psychological, and/or sociological mechanisms that mediate the associations between physical activity or sports participation and suicide ideation and attempts, and to determine whether these mechanisms differ between boys and girls.

The findings and conclusions in this paper are those of the authors and do not necessarily represent the views of the U.S. Centers for Disease Control and Prevention.

Back to Top | Article Outline

REFERENCES

1. Angst, J., F. Angst, and H. H. Stassen. Suicide risk in patients with major depressive disorder. J. Clin. Psychiatry 60(Suppl. 2):7-62, 1999.
2. Blumenthal, S. J. Suicide: a guide to risk factors, assessment, and treatment of suicidal patients. Med. Clinics N. Am. 72:937-971, 1988.
3. Brener, N. D., L. Kann, T. McManus, S. A. Kinchen, E. C. Sunberg, and J. G. Ross. Reliability of the 1999 Youth Risk Behavior Survey Questionnaire. J. Adolesc. Health 31:336-342, 2002.
4. Breslau, N., L. R. Schultz, E. O. Johnson, E. L. Peterson, and G. C. Davis. Smoking and the risk of suicidal behavior: a prospective study of a community sample. Arch. Gen. Psychiatry 62:328-334, 2005.
5. Brown, D. R. Physical activity, sports participation, and suicidal behaviors: a literature review. Int. J. Sport Exerc. Psychol. 3:484-500, 2005.
6. Brown, D. R., and C. J. Blanton. Physical activity, sports participation, and suicidal behavior among college students. Med. Sci. Sports Exerc. 34:1087-1096, 2002.
7. Brown, D. W., D. R. Brown, G. W. Heath, et al. Association between physical activity dose and health related quality of life. Med. Sci. Sports Exerc. 36:890-896, 2004.
8. Centers for Disease Control and Prevention. Methodology of the youth risk behavior surveillance system. Morb. Mortal. Wkly. Rep. 53:1-13, 2004.
9. Centers for Disease Control and Prevention. Methods of suicide among persons aged 10-19 years-United States, 1992-2001. Morb. Mortal. Wkly. Rep. 53:471-498, 2004.
10. Cerel, J., T. A. Roberts, and W. J. Nilsen. Peer suicidal behavior and adolescent risk behavior. J. Nerv. Ment. Dis. 193:237-243, 2005.
11. Chaouloff, F. The serotonin hypothesis. In: Physical Activity and Mental Health, P. Morgan (Ed.). Washington, DC: Taylor and Francis, pp. 179-198, 1997.
12. Colcombe, S., and A. F. Kramer. Fitness effects on the cognitive function of older adults: a meta-analytic study. Psychol Sci. 14:125-130, 2003.
13. Dishman, R. K., R. A. Washburn, and G. W. Heath. Physical Activity Epidemiology. Champaign, IL: Human Kinetics, 2004.
14. Eaton, D. K., R. Lowry, N. D. Brener, D. A. Galuska, and A. E. Crosby. Associations of body mass index and perceived weight with suicide ideation and suicide attempts among U.S. high school students. Arch. Pediatr. Adolesc. Med. 159:513-519, 2005.
15. Etnier, J. L., P. M. Nowell, D. M. Landers, and B. A. Sibley. A meta-regression to examine the relationship between aerobic fitness and cognitive performance. Brain Res. Rev. 52:119-130, 2006.
16. French, S. A., M. Story, B. Downes, M. D. Resnick, and R. W. Blum. Frequent dieting among adolescents: psychosocial and health behavior correlates. Am. J. Public Health 85:695-701, 1995.
17. Hallfors, D. D., M. W. Waller, C. A. Ford, C. T. Halpern, P. H. Brodish, and B. Iritani. Adolescent depression and suicide risk:association with sex and drug behavior. Am. J. Prev. Med. 27:224-231, 2004.
18. King, R. A., M. Schwab-Stone, A. J. Flisher, et al. Psychosocial and risk behavior correlates of youth suicide attempts and suicidal ideation. Am. Acad. Child Adolesc. Psychiatry 40:837-846, 2001.
19. Kuczmarski, R. J., C. L. Ogden, S. S. Guo, et al. 2000 CDC growth charts for the United States: methods and development. National Center for Health Statistics. Vital Health Stat. 11:1-190, 2002.
20. Kuo, W. H., J. J. Gallo, and W.W. Eaton. Hopelessness, depressions, substance disorder, and suicidality-a 13 year community-based study. Soc. Psychiatry Psychiatr. Epidemiol. 39:497-501, 2004.
21. Magnusson, P. K. E., F. Rasmussen, D. A. Lawlor, P. Tynelius, and D. Gunnell. Association of body mass index with suicide mortality: a prospective cohort study of more than one million men. Am. J. Epidemiol. 163:1-8, 2006.
22. Mann, J. J. A current perspective of suicide and attempted suicide. Ann. Intern. Med. 136:302-311, 2002.
23. Mann, J. J., D. A. Brent, and V. Arango. The neurobiology and genetics of suicide and attempted suicide: a focus on the serotonergic system. Neuropsychopharmacology 24:467-477, 2001.
24. Martinsen, E. W. Exercise and depression. Int. J. Sport Exerc. Psychol. 3:469-483, 2005.
25. Morgan, W. P. Physical Activity and Mental Health. Washington, DC: Taylor and Francis, 1997.
26. Morgan, W. P. Selected psychological factors limiting performance: a mental health model. In: Limits of Human Performance, H. Clarke and M. Eckert (Eds.). Champaign, IL: Human Kinetics Publishers, pp. 70-80, 1997.
27. O'Conner, P. J., and J. C. Smith. Physical activity and eating disorders. In: Lifestyle Medicine, J. M. Rippe (Ed.). Malden, MA: Blackwell Science, Inc., pp. 1005-1015, 1999.
28. Oler, M. J., A. G. Mainous III, C. A. Martin, et al. Depression, suicidal ideation, and substance use among adolescents: are athletes at less risk? Arch. Fam. Med. 3:781-785, 1994.
29. Paffenbarger, R. S. Jr., I.-.M. Lee, and R. Leung. Physical activity and personal characteristics associated with depression and suicide in American college men. Acta. Psychiatr. Scand. Suppl. 377:16-22, 1994.
30. Raglin, J. S. Anxiolytic effects of physical activity. In: Physical Activity and Mental Health, W. P. Morgan (Ed.). Washington, DC: Taylor and Francis, pp. 107-126, 1997.
31. Research Triangle Institute. SUDAAN Language Manual, Release 9.0. Research Triangle Park, NC: Research Triangle Institute, 2004.
32. Simon, T. R, K. E. Powell, and A. C. Swann. Involvement in physical activity and risk for nearly lethal suicide attempts. Am. J. Prev. Med. 27:310-315, 2004.
33. Thatcher, W. G., B. M. Reininger, and J. W. Drane. Using path analysis to examine adolescent suicide attempts, life satisfaction, and health risk behavior. J. Sch. Health 72:71-77, 2002.
34. Unger, J. Physical activity, participation in team sports, and riskof suicidal behavior in adolescents. Am. J. Health Promot. 12:90-93, 1997.
35. U.S. Department of Health and Human Services. Healthy People 2010. With Understanding and Improving Health and Objectives for Improving Health. 2 vols, 2nd ed, Washington, DC: U.S. Government Printing Office, 2000.
Keywords:

EXERCISE; ATHLETICISM; SUICIDE IDEATION; SUICIDE ATTEMPT; ADOLESCENTS; YOUTH

©2007The American College of Sports Medicine