Abstract
Background:
Researchers have examined associations between sport specialization and injury. However, no studies have investigated associations between sport specialization and health-related quality of life (HRQoL) after injury.
Hypothesis:
Among injured youth athletes, level of sport specialization is not associated with differences in HRQoL approximately 1 month after sport-related injury.
Study Design:
Cross-sectional study.
Level of Evidence:
Level 4.
Methods:
A multicenter, cross-sectional study was performed at 3 primary care sports medicine clinics. Eligible 8- to 18-year-old athletes who presented for musculoskeletal injury or concussion were enrolled. At the initial clinic visit, patients completed the Player Development Survey (PDS) to determine sport specialization level. Approximately 1 month after enrollment, patients completed the Patient-Reported Outcomes Measurement Information System (PROMIS) to determine HRQoL measures.
Results:
Of 720 athletes invited to participate, 336 (age, 14.2 ± 2.4 years) completed the PDS and PROMIS surveys; 22% were highly specialized, 39% were moderately specialized, and 39% were low specialized athletes. Overall, there were no differences in HRQoL scores across the 3 levels of sport specialization. In subgroup analysis based on sex, female HRQoL scores were worse than male HRQoL scores irrespective of sport specialization level in 3 categories: pain interference (mean difference between female and male scores [± SEM] = 2.3 ± 1.1; P = 0.04), depression/sadness (2.9 ± 1.1; P = 0.01), and anxiety/fear (2.8 ± 1.2; P = 0.02).
Conclusion:
Higher level of sport specialization is not associated with worse HRQoL approximately 1 month after sport-related injury. Female athletes have worse short-term HRQoL after an injury irrespective of sport specialization level. While sex differences were statistically significant, the magnitude of differences was small.
Clinical Relevance:
Sport specialization is not associated with differences in HRQoL after injury. Injured female athletes may need closer monitoring due to possibly worse short-term HRQoL, but further investigation is needed.
Participation in youth sports has clear health benefits, including decreased risk of obesity, decreased cardiovascular risk, improved bone mineral density, and improved mental health. 15 However, sport participation also includes the risk of injury. Youth sport injuries are common. Over 3.5 million children under the age of 14 years receive medical treatment for a sports injury each year. 21 Approximately 2 million high school athletes suffer a sports related injury each year. 21
One factor in youth sports that has been linked to increases in overuse injury is sport specialization.2,5,12,17,20 According to Logan et al 15 , “Sports specialization is the concept of intensely focusing on a single sport, typically year-round, while giving up other sports.” Sport specialization is becoming increasingly common among young athletes.4,11,18 Previous research in various populations of youth athletes show prevalence rates of sport specialization from 13.4% to 37.5%.2,6,12,13,17,20
Researchers have examined the association between injury and health-related quality of life (HRQoL).25,26,28 -30 Overall, HRQoL of injured athletes is similar to healthy controls in most domains and slightly worse than healthy controls in a few domains. For instance, Verma et al 30 found that injured athletes occasionally reported worse mobility (a patient’s self-perceived movement capabilities) scores than healthy controls.
Researchers have also examined the association between sport specialization and HRQoL.7,13,31 Although it is widely accepted that sport specialization may lead to an increased risk of overuse injury, the data regarding the association between sport specialization and HRQoL is limited and shows mixed results. Some studies have found a negative association between sport specialization and HRQoL, while others have found no association between sport specialization and HRQoL. Burwell et al 7 performed a study in college athletes to see whether sport specialization characteristics (age at specialization/degree of specialization) had an impact on current HRQoL. The study found moderate evidence that early sport specialization was associated with worse HRQoL when compared with late sport specialization. 7 However, a qualitative study performed by Patel and Jayanthi 19 showed different results. Patel and Jayanthi 19 interviewed young athletes and their parents about their sport participation. HRQoL scores were high in all young athlete and parent categories with no significant differences in HRQoL between specialized and multisport young athletes. 19 However, unlike our study, neither of these studies surveyed currently injured athletes.
To our knowledge, no studies have investigated the associations between sport specialization and short-term HRQoL after injury. The purpose of our multicenter study was to examine the associations between sport specialization and short-term HRQoL after injury in a clinical population of youth athletes.
Methods
Study Design
A multicenter, cross-sectional study was performed from 2018 to 2020 at 3 academic primary care sports medicine clinics in the United States. Each site received the appropriate Institutional Review Board (IRB) approval. Eligible patients were recruited at the time of their clinic visit with consent and/or assent obtained via research electronic data capture (REDCap, Vanderbilt University). If a patient was aged 8 to 17 years, consent to participate was provided by a parent or legal guardian of patient. If patient was between age 12 and 17 years, assent was also obtained from the patient. If the patient was 18 years old, only the patient provided consent. If the patient was <12 years, the parent or legal guardian completed the surveys in the presence of the patient. If the patient was between age 12 and 17 years, the parent or legal guardian was asked to supervise the patient while completing the surveys. Recruitment ended on March 17, 2020 as all clinical research activities ceased due to the onset of the COVID pandemic.
This particular study was a secondary analysis focusing on sport specialization. The primary analysis was focused on the relationship between HRQoL and injury type for young athletes. The primary analysis can be found in the paper authored by Verma et al. 30
Inclusion and Exclusion Criteria
We invited 8- to 18-year-old patients who presented to a sports medicine clinic for an acute/overuse musculoskeletal injury or concussion. Exclusion criteria included patients who were not in organized sport, patients who were unable to read or write in English, and patients who were already enrolled in another research study.
Patient Characteristics
The following data were extracted from the medical record: age, sex, clinic note, and the International Classification of Disease-10th revision (ICD-10) code for the patient’s primary musculoskeletal diagnosis at time of enrollment. Patients were assigned an injury cohort based on ICD-10 code and clinic note in accordance with a previously published classification system. 12
Outcome Measures
At the initial clinic visit, patients completed the Player Development Survey (PDS) adapted from Jayanthi et al. 11 This survey classified patients as having low, moderate, or high level of specialization. Of note, the PDS is a commonly used survey; however, it has not been formally validated. Degree of specialization was determined by the number of positive responses to the following sport specialization criteria: (1) year-round training (>8 months per year), (2) chooses a single main sport, and (3) quit other sports to focus on 1 main sport. Highly specialized athlete was defined as 3 positive responses; moderately specialized athlete was defined as 2 positive responses; low specialized athlete was defined as 0 or 1 positive responses.
Approximately 1 month after enrollment, patients were emailed a survey entitled Patient-Reported Outcomes Measurement Information System (PROMIS) Pediatric-25 Profile Version 2.0. 8 The 1-month follow-up was modeled after a previous study that examined HRQoL after injury. 24 The PROMIS survey measures HRQoL via 6 domains: depression/sadness, anxiety/fear, mobility, pain interference, fatigue, and peer relationships. Up to 3 weekly reminders were sent to patients via email or text message if the survey had not been completed. Patients were assigned to acute injury, overuse injury, serious overuse injury, or concussion cohort based on a previously published classification system. 12 Acute injuries were defined as occurring during a single traumatic event, and overuse injuries were defined as occurring gradually. Overuse injuries were considered serious if the physician typically recommended ≥1 month of rest from sport participation. Examples of serious overuse injuries included spondylolysis, stress fractures to spine or extremity, stress injury of the growth plate, and osteochondritis dissecans. Concussions were diagnosed as per Berlin Guidelines. 16
Statistical Analysis
The sample size was fixed, based on youth athletes who were enrolled in the primary study. For each PROMIS HRQoL domain, T-scores have a mean of 50 and standard deviation of 10 based on a reference population of patients aged 8 to 17 years. 10 Patient T-scores were calculated using the HealthMeasures Scoring System (HealthMeasures, Northwestern University). Of note, HealthMeasures recommends using the pediatric measures for all respondents if the population sample is predominately ages 8 to 17 years with a small number of very young adults (e.g., 18-20 years of age). 23 This recommendation was consistent with our sample; therefore, we used the pediatric measures. For reference, the main difference between the pediatric survey and the adult survey is the language used in the questions. The language is designed specifically to be age-appropriate for the specific target population. The domains are similar between the adult survey and the pediatric survey.
The cross-sectional analysis of each HRQoL PROMIS T-score domain included both univariable and adjusted analyses. For the univariable analysis, a separate 1-way analysis of variance (ANOVA) was performed for each PROMIS T-score domain to assess the relationship with the categorical variable sport specialization (low, moderate, high). The normal distribution of residuals and homogeneity of variances were checked. Results were summarized with the mean and 95% CI for each level of sport specialization.
For the adjusted analysis, a standard linear fixed effects model was used for each PROMIS T-score domain outcome. The model included sport specialization, sex, the interaction between sex and sport specialization, and study site. Due to missing covariate data, the analysis was implemented with the SAS MIXED Procedure (SAS/STAT, Version 15.1). Missing data were assumed to be missing at random, meaning the probability of data being missing depended only on observed data, not on the unobserved missing data itself. Statistical significance was defined as a 2-sided P value of <0.05.
Results
Patient Demographics
Of the 720 patients enrolled in the study, 55% were female. Out of the 720 patients, 357 (49.6%) completed the PROMIS survey. Out of the 357 patients that completed the survey, 336 were athletes. Only the 336 athletes were included in this study. Out of the 336 athletes, 64% were female. Of note, there was a higher response rate among the female patients. Average age of athletes was 14.2 years. The pool of respondents had 111 acute injuries (33%), 132 overuse injuries (39%), 51 serious overuse injuries (15%), and 41 concussions (12%). A total of 19 (6%) of the injuries were managed surgically. Of the 336 participants, 75 (22%) were low specialized, 130 (39%) were moderately specialized, and 131 (39%) were highly specialized. Time of follow-up survey completion was approximately 1 month (mean, 38 days; median, 31 days) (Table 1).
Patient demographics
Data given as mean ± SD or n (%).
Sport Specialization and Short-Term HRQoL After Injury
There were no statistically significant differences in short-term pain interference, peer relationships, depression/sadness, fatigue, anxiety/fear, or mobility scores after injury among various sport specialization levels. For full details, see Table 2.
Univariable analysis of HRQoL after injury by sport specialization level
HRQoL, health-related quality of life.
Sport Specialization and Short-Term HRQoL After Injury Based on Sex
Female HRQoL scores were worse than male HRQoL scores irrespective of sport specialization level in 3 categories: pain interference (mean difference between female and male score [± SEM] = 2.3 ± 1.1; P = 0.04), depression/sadness (2.9 ± 1.1; P = 0.01), and anxiety/fear (2.8 ± 1.2; P = 0.02) (Table 3).
Adjusted analysis of HRQoL by sport specialization and sex
Interaction between sport specialization and sex.
Discussion
To our knowledge, this is the first study to evaluate the association between sport specialization and short-term HRQoL after injury in a clinical population of youth athletes. Overall, this study provides evidence that sport specialization is not associated with short-term differences in HRQoL after injury. Another significant finding was that pain, depression/sadness, and anxiety/fear scores were worse in female athletes compared with male athletes irrespective of sport specialization level. The magnitude of the differences in HRQoL scores between girls and boys was small. Further research is needed to determine the clinical importance of these findings.
When the injuries were examined overall (including nonsurgical and surgical management), this investigation showed no statistically significant differences in pain interference, peer relationships, depression/sadness, fatigue, anxiety/fear, or mobility scores among different sport specialization levels after injury. This novel finding suggests that sport specialization level was not associated with differences in how youth athletes rate their quality of life after injury in this clinical population. For the HRQoL domains pertaining to anatomical complaints, this result is intuitive. The type of injury sustained would logically be the strongest predictor of pain and mobility after injury. However, for the HRQoL domains pertaining to relational and emotional complaints (peer relationships, depression/sadness, fatigue, and anxiety/fear), this result is important for youth athletes who choose to specialize in 1 sport. One explanation could be that both highly specialized and low specialized athletes may have much of their identity associated with being an athlete. When athletes become injured and can no longer participate in sport, it may affect their identity as an athlete, whether they are highly specialized or not. This is important because research shows that a strong and stable sense of identity is associated with improved mental health. 22
Another theory that could explain the lack of difference in relational and emotional related HRQoL scores is that injury could affect the interpersonal dynamics between people involved in the athlete’s sport for both highly specialized and low specialized athletes. Research studies have shown that participation in youth sports leads to the social benefits of relationships with coaches and friends, decreases in social isolation, improvements in social acceptance, and decreases in suicidal behavior.1,3,9,27 Due to injury, athletes often miss practices, games, travel, and team bonding activities. These factors can affect the athlete’s ability to engage in the beneficial social aspects of sport whether they are highly specialized or low specialized.
The previously published primary analysis study compared HRQoL scores after injury based on sex. 30 The primary analysis study showed worse pain, depression/sadness, and anxiety/fear scores in female athletes compared with male athletes after injury. General studies of HRQoL surveys have shown that female athletes are at risk for worse HRQoL scores. 14 This secondary analysis study showed that female athletes have worse HRQoL scores after injury irrespective of sport specialization level. This is an important finding as injured female athletes may need closer monitoring due to possibly worse short-term HRQoL. However, the magnitude of these statistically significant differences in HRQoL scores between female and male athletes was small. Further studies are needed to determine the clinical importance of these results.
Limitations
First, there are elements of selection bias. All our study participants presented to academic primary care sports medicine specialists in large cities, some of our participants were college athletes (more likely to be highly specialized), and there was approximately a 50% survey response rate for completing the PROMIS survey. These elements of selection bias may limit our ability to generalize the results to all populations. Second, quality of life surveys were not collected before injury. Therefore, we cannot claim a causal relationship between injury and changes in HRQoL. We can only comment on HRQoL scores approximately 1-month after injury. Third, our study was a short-term 1-month follow-up. Fourth, the time from injury to clinic presentation was not recorded. Although the patient’s symptoms most likely started or peaked around the time of the clinic visit, the variability in time from injury to follow-up survey could have affected the results. Fifth, we did not collect data on return to play during the 1-month postinjury survey period. Early return to play could have impacted HRQoL scores at the time of survey. Sixth, we stopped recruitment at the onset of the COVID pandemic in 2020. Thus, near the end of our recruitment, the HRQoL scores could have been affected by circumstance unique to the pandemic. Seventh, the HRQoL scores could have been affected by circumstances in the patient’s life that were unrelated to the injury (e.g., school stress, family stressors, financial burden, etc). Eighth, our population was predominately female (64%); therefore, our overall results may have been slightly skewed toward the female responses. This was likely due to a higher response rate among the female patients. Finally, diagnoses were extracted retrospectively using ICD-10 codes and clinic note as opposed to prospectively at the time of the clinic visit. A prospective study would allow for the most accurate diagnoses.
Conclusion
Overall, this study provides evidence that sport specialization level is not associated with short-term differences in HRQoL after injury. However, this secondary analysis does identify 1 group of athletes that may need further monitoring after injury due to worse short-term HRQoL scores: injured female athletes. Female athletes have worse short-term HRQoL after an injury irrespective of specialization level. The magnitude of the differences in HRQoL scores between female and male athletes was small. Further studies are needed to determine the clinical importance of these results. Future studies should assess possible interventions to improve the short-term HRQoL for injured female athletes.
Footnotes
Acknowledgements
The authors thank their research staff for making this study possible, especially Jamie Burgess, Jacob Wild, Danielle Hunt, Tiffany Dumas, Marina Gearhart, Gina Johnson, Beau McGinley, Becky Parmenter, Mario Ramirez, Vignesh Sundaram, and Lavanya Veda. This study was supported by funding through the American Medical Society for Sports Medicine Foundation Grant and the Emory Orthopaedics Intramural Seed Grant.
The following authors declared potential conflicts of interest: N.J. is the Chief Sports Medicine Advisor at Pickup Sports and Director of the Pickup Sports Foundation.
Ethics Statement
Pickup Sports is a mission-based company that was started to encourage kids to play multiple sports. Each author certifies that his or her institution approved the human protocol for this investigation and that all investigations were conducted in conformity with ethical principles of research.
