A National Survey on the Relationship of Youth Sport Specialization Behaviors to Self-Reported Anxiety and Depression in Youth Softball Players

Abstract

Background:

There are little to no data on whether any associations exist between sport specialization and mental health in youth softball athletes.

Hypothesis:

Highly specialized youth softball athletes will have worse self-reported depression and anxiety symptom scores compared with low and moderate specialized athletes.

Study Design:

Cross-sectional survey.

Level of Evidence:

Level 4.

Methods:

An online cross-sectional survey was distributed in the fall of 2021 to a national sample of female youth softball athletes between the ages of 12 and 18 years. Sport specialization status was determined using a 3-point specialization scale that classifies either low, moderate, or high. The patient health questionnaire-9 (PHQ-9) and the 7-item general anxiety disorder scale (GAD-7) were used to assess self-reported symptoms of depression and anxiety. Comparison also included sports participation and specialization behaviors between specialization groups.

Results:

A total of 1283 subjects (mean age, 15.1 ± 1.7 years) fully completed the survey. After adjusting for covariates, lower scores were reported on both the PHQ-9 and GAD-7 by highly specialized athletes compared with moderate or low specialization athletes (PHQ-9, high = 8.6 ± 0.4; moderate = 11.2 ± 0.3; low = 10.9 ± 0.5; P < 0.01; GAD-7, high = 6.5 ± 0.4; moderate = 8.6 ± 0.3; low = 8.4 ± 0.4, P < 0.01). Conversely, higher scores were reported on both scales for athletes who received private softball coaching compared with those who did not (PHQ-9, 11.5 ± 0.3 vs 9.0 ± 0.3; P < 0.01; GAD-7, 8.8 ± 0.3 vs 6.9 ± 0.3, P < 0.01). Finally, athletes who reported an arm overuse injury in the previous year reported higher PHQ-9 scores (10.8 ± 0.3 vs 9.8 ± 0.3; P < 0.01).

Conclusion:

While sport specialization, as measured by the validated 3-point scale, was not associated with increased anxiety and depression symptom scores, other aspects of specialization behavior such as private coaching or overuse injury history were associated with worse scores on these scales, indicating potential concern for anxiety and depression. However, although the differences we observed were statistically significant, they did not exceed the minimal clinically important difference values that have been established for the PHQ-9 (5 points) or GAD-7 (4 points).

Clinical Relevance:

This project is a first step toward understanding the sport specialization behaviors and their influence on the mental health of youth softball athletes. Focusing on investigating specialization behaviors further may reveal to be a better indicator of risk of developing anxiety and depression symptoms compared with utilizing the 3-point specialization scale.

Youth sport specialization is defined as year-round participation in sport at the exclusion of other sports, and recent reports have suggested that 17% to 41% of athletes are highly specialized, but this proportion is dependent on sex, age, and sport.^3,39 The American Academy of Pediatrics Clinical Report in 2016 cited that up to 70% of youth are discontinuing playing organized sports by the age of 13 years due to increasing early specialization, cost, time, parental/coach pressure and expectations, and injuries.^5,17,43 This reduction directly influences the known benefits of playing youth sports such as leadership, fun, self-esteem, teamwork, physical activity skills, and peer socialization.^5,35,36 Youth softball has not been excluded from this trend of year-round participation beyond regular middle or high school season, but now includes travel teams and private clubs. The National Federation of State High School Associations’ most recent recorded year (2018/2019) reported 362,038 female fast-pitch high school athletes, and USA Softball (the national governing body of softball) reports >120,000 teams.

Current expert opinion suggests that early specialization could predispose these athletes to burnout, poor sleep, social isolation, and increased stress.^6,19 These expert opinions suggest contrary data to the recommendation that participation in organized sport, ²³ as well as exercise outside of sport, can be beneficial to mental health when compared with adolescents who are not active in sport, as long as participation volume and behaviors do not start to influence the mental and physical well-being of the adolescent. ⁴⁷ Position statements from professional sports medicine bodies such as the American Academy of Pediatrics, American Medical Society of Sports Medicine, and American Orthopedic Society of Sports Medicine express concern about the impact of specialization on injury risk and burnout in youth athletes.^{5,9,19,22,24,33,34} In addition, early youth specialization (<12 years of age) has not been shown to correlate with higher rate of play at elite level (professional and collegiate).^4,5

Mental health in sport has become a prominent area of discussion in recent years, both in the lay press, as well as in medicine. High profile elite athletes have shared their struggles with mental health, and how these are affected by sport.^25,30 Medical and sports organizations such as the International Olympic Committee and the American Medical Society for Sports Medicine have produced statements regarding the mental health of athletes.^8,40 In April 2019, the American Medical Society for Sports Medicine convened the Youth Early Sport Specialization Summit to synthesize research on youth sport specialization, as well as to identify areas of need for the research community.^12,20 Recommendations from this conference included research on psychosocial outcomes, as well as sport-specific outcomes related to sport specialization in youth athletes.

Research has increased regarding youth sports injuries related to early sport specialization, but there remains a paucity of data related to the potential mental health risks associated with early sport specialization. Most research in throwing sports has reported early specialization and injuries and burnout of youth baseball players.^9,36,38 We are aware of no published data specific to softball players, or highly specialized softball players that link specialization with mental health disorders; however, there have been recent high profile examples at the Division I level of softball athletes completing suicide, which has focused attention on mental health in this sport. ¹⁰

Our research is the first of its kind to evaluate active softball players for symptoms of anxiety and depression and to determine whether sport specialization and specialization behaviors are associated with these symptoms.^{11,13,32,41,42,46} Our goal is to identify sports participation and specialization behaviors that could have associations with increased mental health issues in our youth softball athletes. Our purpose of this research was to investigate the self-reported symptoms of anxiety and depression of mental health in a national sample of youth and adolescent softball players, as well as to examine any associations between sport specialization behaviors and mental health. Identifying factors that are specific to softball athletes to help develop targeted educational interventions for anxiety and depression among at-risk groups, further helping physician/providers better assess and care for an athlete.

Methods

Participants

This study was approved by the Institutional Review Board at Indiana State University. An anonymous, online survey was distributed to a national sample of female youth softball athletes between the ages of 12 and 18 years in the Fall of 2021. To be included in this study, participants must have been between the ages of 12 and 18 years and must have been an active participant on an organized fast-pitch softball team during the previous 12 months. The survey was distributed via email to the parents of youth athletes in national softball organizations (United Sports Specialty Association Fast-Pitch Softball and National Fast-Pitch Coaches Association), as well as through individual softball tournaments, and via social media. The organizations who volunteered to participate sent a recruiting email to parents of players aged 12 to 17 years, and directly to players 18 years of age. The email asked parents to offer participation to their child if the parents consented to the study, and the child (if under 18) assented to voluntary participation. Once consent and assent were determined by opting in, the child was instructed to complete the survey independently. The first 500 participants received an Amazon gift card valued at $10. All surveys were anonymous and completed with web-based software Qualtrics.com (Qualtrics), which is Health Information Portability and Accountability Act (HIPAA) compliant.

Instrumentation

The questionnaire consisted of 6 sections focusing on (1) demographics, (2) softball sport participation, (3) sport specialization status and behaviors, (4) athlete injury history, (5) depression/anxiety symptoms, and (6) perceived parental pressure and self-rated perfectionism. The demographic, softball participation, sport specialization status, and injury history sections were modified from previous questionnaires on this topic in youth baseball. ³⁸ Incomplete surveys were included in the analysis if they included all major outcome variables of interest (specialization status, upper extremity overuse injury history).

Sport Participation and Specialization Status

Softball participation information from the previous year included primary position, position(s) played, softball games played in the previous year, and months per year and hours per week of softball participation. The specialization status and behaviors section included sport specialization status, participation on a softball club team, whether the athlete had regularly traveled overnight (at least once a month) for softball competitions or showcases, and whether the athlete received private softball coaching. Year-round softball participation and year-round pitching were determined by whether the athlete played softball for >8 months and whether the athlete pitched for >8 months in the previous year.^31,40

Sport specialization status was determined using a 3-point sport specialization scale, ¹⁸ which classifies athletes as low, moderate, or highly specialized based on answers to 3 questions. In our study, the questions were modified to address softball specifically: “Do you participate in softball more than 8 months of the year?”; “Do you consider softball to be more important than the other sports you play?”; “Have you ever quit another sport to play softball?” A categorical classification system was used to assess the responses to these sport specialization questions (yes, 1; no, 0), with a total score of 3 considered high specialization, a score of 2 considered moderate specialization, and a score of 0 or 1 considered low specialization.

Depression and Anxiety

The validated short-form 9-question patient health questionnaire (PHQ-9) was used to evaluate symptoms of depression and is a validated short-form 9-question screening test for depression that is appropriate for pediatric patients. ⁴¹ The PHQ-9 generates a score between 0 and 27, with higher scores indicating higher depression symptoms (0-4, minimal to no depression; 5-9, mild depression; 10-14, moderate depression; 15-19, moderately severe depression; 20-27, severe depression). ²¹ In adolescent populations, a cut-point ≥11 has been identified as indicating the need for further evaluation for depression. ⁴¹ The minimal clinically important difference (MCID) for the PHQ-9 has been calculated in adults as 5 points on the 27-point scale. ²⁶

The validated 7-question generalized anxiety disorder scale (GAD-7), which is appropriate for adolescent patients, was used to evaluate symptoms of anxiety. ³² The GAD-7 results in a score between 0 and 21, with higher scores indicating increasing concern for generalized anxiety disorder (0-4, minimal anxiety; 5-9, mild anxiety; 10-14, moderate anxiety; 15-21, severe anxiety). ³² Similar to the PHQ-9, research in adolescent populations has identified a cut-point ≥ 11 or higher to classify pediatric populations as having at least “moderate” anxiety. ³² The MCID for the GAD-7 has been calculated as 4 points on the 21-point scale. ⁴⁵

Statistical Analysis

Data were summarized using means with SDs, medians with interquartile ranges (IQR), and frequencies with percentages. Continuous variables were assessed for normality using skewness/kurtosis values and via visual inspection of histograms.

Bivariate parametric analyses (independent t tests, 1-way analysis of variance (ANOVA), Pearson’s correlations) examined differences and associations between variables of interest and the PHQ-9 or GAD-7 scale total scores. The type of bivariate analysis depended on the variable type (continuous or categorical) and the normality of the specified variables. Independent t tests were used to assess differences in PHQ-9/GAD-7 total scores for playing pitcher, playing softball >8 months in the previous year, participation on a club softball team, regular travel for softball competitions or showcases, and whether the athlete had sustained an overuse injury to their throwing arm in the past 12 months. One-way ANOVAs were used to assess differences in PHQ-9/GAD-7 total scores based on specialization status (low, moderate, high). Pearson’s correlations were calculated to examine associations of PHQ-9/GAD-7 total scores with age, months per year of softball participation, and hours per week of softball participation. Bonferroni adjustments were made to account for the large number of bivariate analyses, with the adjusted alpha-value thresholds listed in footnotes to the relevant tables.

Separate multivariable linear regression models were created to examine the association of variables of interest with either the PHQ-9 or GAD-7 total scores. Only variables that were identified as statistically significant in the bivariate analyses were included in the final multivariable regression models. This bivariate selection approach was used due to the lack of previous research on specialization and mental health in youth sports that could have otherwise been used to guide a priori selection of variables for the multivariable model. Least-square means (LS-Mean) were calculated with SEs for each scale’s total score. The multivariable linear regression model was assessed to determine whether it met the assumptions of linear regression using the Global Validation of Linear Models Assumptions package in R statistical software (R Foundation for Statistical Computing) and via visual inspection of quantile-quantile and residual plots. Statistical significance was set a priori at a 2-sided P value < 0.05, and all analyses were performed in R statistical software.

Results

A total of 1489 youth softball players accessed the survey, 1309 players finished the survey, and 1283 youth softball players (mean age, 15.1 ± 1.7 years) fully completed all questions in the survey related to sport specialization and the PHQ-9 and GAD-7 and were included in data analysis (86.2% completion rate). Demographics and sport participation characteristics are provided in Table 1. Approximately one-fifth (19.9%, N = 255) of all participants were classified as highly specialized, with 70.1% (N = 899) classified as moderately specialized, and 10.0% (N = 129) classified in the low specialization category. One-fourth (25.7%, N = 330) of participants reported playing softball >8 months in the previous year. The majority of participants reported participating on a club softball team (86.4%, N = 1109), traveling overnight regularly for softball (86.0%, N = 1103), or receiving private softball coaching (80.5%, N = 1033). For the overall sample, the mean PHQ-9 total score was 11.5 ± 5.1, and the mean GAD-7 total score was 8.9 ± 4.1. For the PHQ-9, 11.1% (N = 142) of participants’ scores were in the minimal-to-no depression symptom category, 18.4% (N = 236) in the mild depression symptoms category, 41.9% (N = 538) in the moderate depression symptom category, 25.9% (N = 332) in the moderately severe depression symptom category, and 2.7% (N = 35) in the severe depression symptom category. For GAD-7, 15.9% (N = 204) of participants’ scores were in the minimal anxiety symptom category, 34.4% (N = 442) in the mild anxiety symptom category, 43.5% (N = 558) in the moderate anxiety symptom category, and 6.2% (N = 79) in the severe anxiety symptom category.

Table 1.

Participant demographics and sport participation characteristics

Variable	N (%), mean (SD), or median [IQR]
Participants	1283 (100%)
Age	15.1 (1.7)
Primary position
Pitcher	160 (12.5%)
Catcher	162 (12.6%)
First base	190 (14.8%)
Second base	171 (13.3%)
Shortstop	125 (9.7%)
Third base	145 (11.3%)
Left field	111 (8.7%)
Center field	110 (8.6%)
Right field	109 (8.5%)
Pitched for team in previous year, yes	343 (26.7%)
Months per year of softball participation	4 [4-5]
Hours per week of softball participation	14 [7-21]
Specialization scale
Low	129 (10.0%)
Moderate	899 (70.1%)
High	255 (19.9%)
Play softball >8 months per year, yes	330 (25.7%)
Participate on club team - yes	1109 (86.4%)
Travel regularly for softball - yes	1103 (86.0%)
Receive private softball coaching - yes	1033 (80.5%)
Arm overuse injury in past year - yes	567 (44.8%)
PHQ-9 total score	11.5 (5.1)
GAD-7 total score	8.9 (4.1)

GAD-7, validated 7-question generalized anxiety disorder scale; IQR, interquartile range; PHQ-9, short-form 9-question patient health questionnaire.

PHQ-9 and GAD-7 scores based on sport participation characteristics are presented in Tables 2 and 3, respectively. PHQ-9 scores were significantly lower among highly specialized athletes and athletes who played softball year-round. Conversely, PHQ-9 scores were significantly higher among athletes who reported participating on a club softball team, traveling regularly for competitions, receiving private softball coaching, and those who reported an arm overuse injury in the previous 12 months. These significant differences were identical for GAD-7, with lower scores among highly specialized athletes and athletes who played softball-year round, and higher GAD-7 scores among athletes who played on a club team, traveled regularly, received private coaching, or reported an arm overuse injury in the previous 12 months. Table 4 presents correlations between continuous variables of interest and PHQ-9 or GAD-7 scores. For both the PHQ-9 and GAD-7, months per year of softball were correlated negatively with total scores, while total scores were correlated positively with hours per week of softball participation. All variables from Tables 2-4 that were statistically significant were included in the final multivariable regression models.

Table 2.

Bivariate differences in PHQ-9 scores based on sport participation characteristicsa

	PHQ-9 Total Score	P
Specialization		<0.01
Low	11.7 (4.8)
Moderate	12.5 (4.2)
High	8.1 (6.4)
Pitcher as one of positions		0.22
Yes	11.2 (5.3)
No	11.6 (5.0)
Softball >8 months		<0.01
Yes	8.8 (6.4)
No	12.5 (4.1)
Club team participation		<0.01
Yes	11.9 (5.0)
No	9.4 (4.8)
Regular travel for competitions/showcases		<0.01
Yes	11.8 (5.1)
No	9.8 (4.2)
Private softball coaching		<0.01
Yes	12.4 (4.6)
No	7.9 (5.2)
Arm overuse injury in the past 12 months		<0.01
Yes	12.5 (4.5)
No	10.8 (5.3)

PHQ-9, short-form 9-question patient health questionnaire.

^aData reported as mean (SD). Alpha value after Bonferroni adjustment for 7 analyses in group was 0.007.

Table 3.

Bivariate differences in GAD-7 scores based on sport participation characteristicsa

	GAD-7 Total Score	P
Specialization		<0.01
Low	9.0 (4.3)
Moderate	9.6 (3.5)
High	6.4 (4.9)
Pitcher as one of positions		0.13
Yes	8.6 (4.2)
No	9.0 (4.0)
Softball >8 months		<0.01
Yes	6.9 (5.0)
No	9.6 (3.5)
Club team participation		<0.01
Yes	9.2 (4.0)
No	7.2 (4.1)
Regular travel for competitions/showcases		<0.01
Yes	9.1 (4.1)
No	7.5 (3.9)
Private softball coaching		<0.01
Yes	9.6 (3.8)
No	6.2 (4.3)
Arm overuse injury in the past 12 months		<0.01
Yes	9.5 (3.7)
No	8.5 (4.3)

GAD-7, validated 7-question generalized anxiety disorder scale.

^aData reported as mean (SD). Alpha value after Bonferroni adjustment for 7 analyses in group was 0.007.

Table 4.

Correlations of continuous softball participation characteristics with PHQ-9 and GAD-7 scores

Variable	PHQ-9 Total Score Pearson’s r (95% CI)	P
Age	−0.01 (−0.07 to 0.04)	0.61
Months per year of softball	−0.40 (−0.44 to −0.34)	<0.001
Hours per week of softball	0.17 (0.14 to 0.24)	<0.001
Variable	GAD-7 total score Pearson’s r (95% CI)	P
Age	0.05 (−0.004 to 0.11)	0.07
Months per year of softball	−0.34 (−0.38 to −0.29)	<0.01
Hours per week of softball	0.19 (0.14 to 0.24)	<0.01

GAD-7, validated 7-question generalized anxiety disorder scale; PHQ-9, short-form 9-question patient health questionnaire.

The results of the multivariable linear regression models for both PHQ-9 and GAD-7 total scores are presented in Table 5. After adjusting for covariates, PHQ-9 scores were lower among highly specialized athletes compared with both moderate (LS-Mean [SE]. 8.6 [0.4] vs 11.2 [0.3], P < 0.01) and low specialization athletes (LS-Mean [SE]. 8.6 [0.4] vs 10.9 [0.5], P < 0.01). In addition, PHQ-9 scores were higher among athletes who reported receiving private softball coaching compared with athletes who did not have private coaching (LS-Mean [SE]. 11.5 [0.3] vs 9.0 [0.3], P < 0.01) and were higher among athletes who reported an arm overuse injury in the previous 12 months compared with athletes with no history of injury (LS-Mean [SE]. 10.8 [0.3] vs 9.8 [0.3], P < 0.01). For the GAD-7, total scores were lower among highly specialized athletes compared with both moderate (LS-Mean [SE]. 6.5 [0.4] vs 8.6 [0.3], P < 0.01) and low specialization athletes (LS-Mean [SE]. 6.5 [0.4] vs 8.4 [0.4], P < 0.01). Finally, GAD-7 total scores were higher among athletes who reported receiving private softball coaching compared with athletes who did not have private coaching (LS-Mean [SE], 8.8 (0.3) vs 6.9 (0.3), P < 0.01).

Table 5.

Least-square mean estimates for PHQ-9 or GAD-7 scores based on multivariable linear regressiona

	PHQ-9 Least-Squares Mean Estimate (SE)	P
Sport specialization scale
Low	10.9 (0.5)	0.86
Moderate	11.2 (0.3)	<0.01 b
High	8.6 (0.4)	<0.01 c
Softball >8 months		0.72
Yes	10.2 (0.3)
No	10.3 (0.4)
Club team participation		0.67
Yes	10.4 (0.3)
No	10.1 (0.4)
Regular travel for competitions/showcases		0.70
Yes	10.3 (0.3)
No	10.1 (0.4)
Private softball coaching		<0.01
Yes	11.5 (0.3)
No	9.0 (0.3)
Arm overuse injury in the past 12 months		<0.01
Yes	10.8 (0.3)
No	9.8 (0.3)
	GAD-7 Least-Squares Mean Estimate (SE)	P
Sport specialization scale
Low	8.4 (0.4)	0.98
Moderate	8.6 (0.3)	<0.01 b
High	6.5 (0.4)	<0.01 c
Softball >8 months		0.97
Yes	7.8 (0.3)
No	7.8 (0.3)
Club team participation		0.72
Yes	7.9 (0.2)
No	7.8 (0.3)
Regular travel for competitions/showcases		0.38
Yes	8.0 (0.3)
No	7.6 (0.3)
Private softball coaching		<0.01
Yes	8.8 (0.3)
No	6.9 (0.3)
Arm overuse injury in the past 12 months		0.09
Yes	8.0 (0.2)
No	7.7 (0.2)

GAD-7, validated 7-question generalized anxiety disorder scale; PHQ-9, short-form 9-question patient health questionnaire.

^aAll models were adjusted for all other variables listed, in addition to months per year and hours per week of softball participation.

^bComparison between moderate and high specialization categories.

^cComparison between low and high specialization categories.

Discussion

We believe this study to be the largest attempt to evaluate the associations of sport specialization and specialization behaviors on symptoms of anxiety and depression in adolescent softball players. Our sample indicates a broad level of specialization and indicates a high level of symptomatology in this group. Overall, only 11.1% of respondents scored in the minimal-to-no depressive symptoms range on PHQ-9, while 15.9% scored in the minimal anxiety symptom range on GAD-7.

After adjusting for covariates, lower scores were reported on both PHQ-9 and GAD-7 by highly specialized athletes compared with moderate or low specialization athletes, while higher scores were reported on both measures of anxiety and depression symptoms by athletes who reported receiving private coaching compared with those who did not. Although these differences were statistically significant, they did not meet or exceed the MCID values that have been established for either PHQ-9 (5 points) or GAD-7 (4 points), calling into question any clinical significance between the groups. Conversely, there were significantly higher scores in both PHQ-9 and GAD-7 across various specialization behaviors such as club team participation, traveling a least once a month for competition and showcase, and private coaching - behaviors that have been recognized to indicate specialization in recent publications.^6,19 Explanations for increased distress in one’s mental health related to increasingly intense sports participation include possible isolation from peer groups outside of single sport, poor sleep, and coach/parent pressure, which all potentially lead to burnout. The difference in the results with specialized athletes when identified by the 3-point scale with lower scores, but with specific behaviors that may identify athletes as more specialized having higher scores, again calls into question whether an improved instrument may help better identify “specialization.”^6,19,31 Although there is no linear association between participation in club sports/travel ball (which are all designed to have a youth athlete play year round) and sports specialization, very few youth athletes are able to successfully compete in a club sport during only certain parts of the year. As such, using the definition of “>8 months in a year” to define specialization, thereby equates club play with sports specialization. Previous publications have also indicated a role for injury affecting one’s identity as an athlete in increasing psychological distress in athletes.^9,14,16

The suggestion that participating in sport, even specialized sport, does not appear to be supported by the prevalence of published literature focused on sport, exercise, and mental health, with the preponderance of evidence suggesting that simple physical activity,^2,7 and probably sport participation, are good for one’s mental health.^35,37,44 It may be true that, in sport, there are differences associated with level of specialization, but it seems true that being physically active, and participating in sport, is good for us. It also may be true that a greater degree of specialization in sport perhaps results in an athlete receiving more parenting support, time, access to mental health resources, and financial support at baseline.

One unexpected finding from our analysis was the significantly lower scores on both the PHQ-9 and GAD-7 in highly specialized athletes compared with low/moderate specialized athletes when measured by the 3-point scale. With adolescent female athletes in particular, Watson et al ⁴⁸ report lower levels of well-being in more specialized soccer athletes, with no statistical difference in reported stress in their sample of 52 American female soccer players; Watson et al ⁴⁹ report on 1482 youth volleyball players, indicating lower quality of life in highly specialized youth volleyball players compared with athletes categorized as moderate/low on the same scale used in our study.

In previously published studies evaluating mental health in youth athletes, Gerber et al ¹⁴ reported depression levels of 9% in a study of 257 elite young athletes in Switzerland. Perhaps more surprising, however, was the volume of responding athletes scoring outside of the normal range on both GAD-7 (84.1%) and PHQ-9 (88.9%). It is notable that our survey did not assess existing diagnoses of mood disorder or anxiety. Similarly, another study of the mental health of adolescents during the COVID-19 pandemic by McGuine et al^27-29 found that the majority of athletes scored outside normal ranges on both GAD-7 (68.9%) and PHQ-9 (82.5%). This survey was conducted in the fall of 2021, while the COVID-19 pandemic was ongoing, but the survey did not address the pandemic, or shutdowns as an influence on our results. The total participant mean scores for both the PHQ-9 (11.5 ± 5.1) and GAD-7 (8.9 ± 4.1) warrant highlighting on their own. These mean scores place the overall sample of adolescent athletes in the “moderate depression” symptom category for PHQ-9, and the “mild anxiety” symptom category for GAD-7. We did not ask participants any questions related to how COVID-19 had impacted their sport participation or mental health, both of which have been shown to be impacted significantly by the pandemic. For example, previous research has demonstrated that athletes who were not able to participate in sports due to the pandemic had higher depression and anxiety symptom scores, and we similarly found that athletes who participated in softball for <8 months in the past year had elevated PHQ-9 and GAD-7 scores.^1,28,29 The overall mean PHQ-9 and GAD-7 scores were high in our study, but similar to other studies of adolescent female athletes performed during the COVID-19 pandemic. ²⁷ In a sample of nearly 7000 adolescent female athletes surveyed in summer 2020, McGuine et al ²⁷ reported a mean GAD-7 score of 8.5 and PHQ-9 score of 9.7 (compared with 8.9 and 11.5, respectively, in our study). It is almost certain that the COVID-19 pandemic influenced the results of our study, and contributed to the high number of athletes scoring in the abnormal ranges of anxiety and depression, as well as the overall high average scores.

We believe our investigation calls for researching more behaviors and factors into what qualifies as a “specialized” athlete, and how we identify these athletes. The 3-point specialization scale used in this study has not been validated as a measurement or categorization of specialization but is used routinely to identify and qualify athletes at different levels of specialization in the published literature. ⁵ Our project used multiple questions to attempt to further quantify athletes’ level of specialization; the results of some of these data conflict with the findings of the specialization scale. Further studies could identify specific specialization categories, behaviors, and factors influencing depression and anxiety symptoms in our athletes.

Limitations

There are several limitations to note with our study. We are unable to report response rate due to the opt-in nature of the survey; as a survey asking questions of the previous year, it likely suffers from recall bias; there is also the possibility that those who responded to the survey did so because of current issues with mental health, leading to an overestimate of the volume of softball athletes with symptoms of anxiety and depression. We were not able to verify that only the athletes filled out the survey; as an anonymous survey sent to the parents’ email address, it is possible that someone other than the intended subject completed or influenced the answers to the survey. Although we included a variety of previously validated instruments in our questionnaire, we did not pilot test the overall questionnaire in a sample of adolescent female softball athletes before the study.

Another limitation was the use of a stepwise approach, in this case bivariate selection, to guide the development of the multivariable regression model. This approach has known limitations, including falsely low P values and the potential to incorrectly reject variables for inclusion in the multivariable analysis due to uncontrolled confounding variables. To minimize the risks of this approach, we used a Bonferroni adjustment during the bivariate selection stage to limit the inclusion of variables due to Type I errors.

Finally, as mentioned previously, the differences in PHQ-9 and GAD-7 scores that we observed in this study were statistically significant but did not meet or exceed previously established MCID values for those scales, raising the question of whether the results we observed are truly clinically meaningful. However, the interpretation of meaningful change should also take into consideration the difference between clinical meaningfulness for an individual patient versus the potential impact of a change in illness status on the health of a larger population. ¹⁵ While the small differences we observed (typically between 1 and 3 points on the PHQ-9 or GAD-7 based on various sport behaviors) may not be clinically meaningful on an individual level, they could represent a sizable difference in the overall mental health of the total population of adolescent female softball players.

We believe the overall strength of this study to be the volume of respondents, as well as the diverse level of specialization included in the sample. This study is a direct result of a conference identifying research gaps in youth athletes, ²⁰ and contributing to the call for investigation in this area.

Conclusion

This study, the first of its kind, exhibits the results from a diverse survey of >1300 adolescent softball players, and will help inform players, parents, coaches, and other providers of the mental health impact of early specialization in softball. The most notable finding was that >80% of responding athletes reported symptoms of anxiety or depression. We found that, based on the 3-point specialization scale, highly specialized athletes scored better on both PHQ-9 and GAD-7. However, with more specific specialization behaviors, such as receiving private softball coaching or sustaining an overuse arm injury, these athletes exhibited worse scores. Our secondary findings suggest that further research on whether the 3-point scale evaluates the full scope of sport-specific specialization is needed.