The killing of a sacred veneer : depressive symptoms in athletes

Within the past decade, there has been a growing research interest on mental health issues in athletes. Within this increasing area of research, research on depressive symptoms in athletes has been central. However, the overall depression-related evidence-base is still fragmented and several important areas of research remain under explored. The overall aim of this PhD was to map topics that have received little scholarly attention in the past, and to empirically explore novel research questions that could contribute to future research and applied work to improve support and prevention initiatives in athletes. Study 1 The aim was to describe methodological characteristics of the research that has assessed depressive symptoms in athletes, and to map the variables that have been tested concerning these symptoms. A review framework proposed by Arksey and O’Malley (2005) was utilised, and of 6983 records screened, 157 studies were included. Most studies were cross-sectional, with samples including current male and female athletes from multiple sports and levels. Non-athlete comparison groups frequently consisted of student samples. Twenty-eight different depression scales were utilised, of which CES-D, BDI, BDI-II, and the PHQ-9 were most common. The most frequently tested variables in relation to depressive symptoms were identified as proximal contextual and interpersonal factors (67.9%), including sport-specific (e.g., type of sport) (36.4%), and generic (e.g. social support) (31.5%) factors. Within-individual factors (e.g. cognitive vulnerability) accounted for 17.2% of all observed topics/variables tested in relation to depressive symptoms, and 9.3% tested depressive symptoms in relation to comorbid disorders. Macro-level variables (e.g. ethnicity) accounted only for 5% of all observations. Considering that current knowledge about depressive symptoms in athletes is largely based on cross-sectional data, and few studies have explored potential underlying mechanisms (e.g. cognitive vulnerability), more longitudinal research is needed to identify underlying vulnerabilities that predict individual differences in depressive symptoms over time. This would further improve future applied work to develop evidence-based intervention and prevention to target relevant mechanisms that increase athletes’ likelihood of experiencing elevated depressive symptoms. The type of measures utilized across the reviewed studies were also highly variable, and different cut-off scores were used to identify athletes with clinically significant depressive symptom severity. Considering the methodological heterogeneity across studies, future studies could benefit from conducting more fine-grained analyses to explore the type of symptoms athletes may be experiencing rather than merely reporting prevalence rates based on a single cut-off score. This could improve our current understanding of the type of issues that may be especially relevant in the athlete populations. Study 2 As identified in study 1 interpreting depressive symptom prevalence across previous studies is complicated considering the range of different measures and cut-off scores that have been utilized in previous studies. Furthermore, we know little about the type of symptoms that may be especially relevant in athletes. Hence, the aim of study 2 was to explore the prevalence of specific symptoms of depression in athletes, and to test differences in athletes’ likelihood of exhibiting these symptoms depending on their age, sex, the type of sport, and the level of competition in which they engage. A sample of Icelandic male and female team sport athletes competing in football, handball, and basketball (N=894, 18-42 years) were included in the study. The football sample represented 20.3% of the Icelandic adult football population (N=2170 across 105 teams) with a total of 441 participants included (age range 18-41 years, male 70.1%). For basketball, the sample represented 36.1% of the Icelandic adult basketball population (N=659 across 56 teams) with a total of 238 participants (age range 18-41 years, male 62.6%). For, handball, sample represented 26.5% of the Icelandic handball population (N=812 across 20 teams) with a total of 215 participants (age range 18-42 years, male 51.2%). Of the athletes exhibiting clinically significant depressive symptoms on the Patient Health Questionnaire (PHQ-9), 37.5% did not exhibit core symptoms of depression (i.e., depressed mood, a lack of interest). Compared to males, females were significantly more likely to exhibit depressed mood, feelings of worthlessness/guilt, problems with sleep, fatigue, appetite, and concentration. Within males, differences were mostly related to neurovegetative aspects of depression (sleep and appetite), whereas in females, differences were related to cognitive/emotional aspects (e.g. depressed mood, guilt/worthlessness). The findings underline the importance of exploring specific symptoms of depression to provide a richer understanding of depressive symptomology in athletes – consequently allowing future research to identify and target risk factors that may be linked to these specific symptoms. Study 3 As identified in study 1, understanding individual differences in vulnerability to depression are still under explored in athletes. Therefore, the main aim of this study was to fill this gap by testing the influence of depressive rumination (repetitive thought processes in response to depressed mood) on the likelihood of experiencing clinically significant depressive symptoms in athletes. Depressive rumination, as defined in the Response Styles Theory, is a well-supported cognitive vulnerability factor to depression within the general and clinical populations – but previous research in athletes has not explored this relationship. In this study, athletes’ profiles on the two underlying factors of depressive rumination, brooding (maladaptive) and reflective rumination (adaptive) were tested on athletes’ likelihood of exhibiting clinically relevant depressive symptoms. A total of 286 competitive athletes from 54 different types of sports were included in the study (62.0% male, age range 18-69 years). The majority of athletes were UK citizens (88.0 %) of white/Caucasian ethnic background (92.2 %). More than half of the athletes (53.5 %) had been selected to represent their country at some point during their athletic careers, and 30.5 % were currently competing at international/top tier professional level. The Patient Health Questionnaire 9 (PHQ – 9) was utilized to measure depressive symptoms, and the Ruminative Responses Scale (RRS-short form) was used to measure the two underlying dimensions of depressive rumination, brooding and reflective rumination. Compared to athletes with a low brooding/reflection profile, athletes with a high brooding/reflection profile had significantly higher odds of experiencing clinically relevant depressive symptoms (OR=15.24, 95% CI=4.37–53.24). A low brooding/ high reflection profile was not, however, related to increased odds. The findings validate findings in the general and clinical populations in the current athlete sample, suggesting that brooding rumination may be an important vulnerability factor explaining individual differences in depressive symptoms in athletes. This implicates that applied work may benefit from targeting brooding tendencies to help vulnerable athletes to develop more beneficial responses to negative mood. However, future research should validate the theoretical model of the response styles theory in longitudinal designs. Furthermore, as implicated by the theoretical model, depressive rumination should be related to increased depressive symptoms when individuals are experiencing stressful life situations. Hence, future studies should test whether athletes with a higher tendency to engage in depressive rumination (especially brooding) compared to those with a low tendency, are more likely to exhibit increased levels of depressive symptoms when athletes experience increased levels of stress in their lives. This would provide stronger empirical support for depressive rumination as a potential underlying mechanism, consequently supporting the utility of targeting depressive rumination in treatment and prevention of depressive symptoms in athletes. Study 4 In-line with suggestions for future research proposed by study 3, the aim was to validate the vulnerability-stress account of the response styles theory using a longitudinal research design. That is, study four tested whether between-athlete differences in the tendency to brood and/or reflect in response to negative mood, measured at the beginning of the study (i.e., baseline), would predict increases in depressive symptoms when levels of stress increased over the one-year study period. A total of 79 Icelandic elite and national team athletes were included in the study (M= 23.5, SD=4.8, age range 18-37), with the majority being female athletes (n=60, 75.9%). Athletes competed in handball (n=22, 27.8%), football (n=14, 17.7%), basketball (n=26, 32.9%), Icelandic equitation (n=8, 10.1 %), and mixed martial arts and/or Brazilian jiu-jitsu (n=9, 11.4%). While depressive rumination (brooding and reflection) were measured at baseline, stress and depressive symptoms were assessed at baseline, at 6-months and 12-months post-baseline. The results showed that higher perceived stress and brooding, but not reflective rumination, independently predicted higher depressive symptom scores over the study period.