The Influence of Brooding and Reflective Rumination on Depressive Symptoms in Athletes in the United Kingdom

4 Individual differences in vulnerability to depression are still underexplored in athletes. We 5 tested the influence of different brooding and reflective rumination profiles (i.e. repetitive 6 thought processes in response to low/depressed mood) on the odds of experiencing clinically 7 relevant depressive symptoms in competitive athletes (N=286). The Patient Health 8 Questionnaire 9 (PHQ – 9) and the Ruminative Responses Scale (RRS-short form) were 9 utilized to measure depression and rumination, respectively. Compared to athletes with a low 10 brooding/reflection profile, athletes with a high brooding/reflection profile had significantly 11 higher odds of experiencing clinical levels of depressive symptoms (OR=13.40, 95% 12 CI=3.81– 47.11). A high reflection/low brooding profile was not, however, related to 13 increased odds of depressive symptoms. Future research could extend our findings by 14 exploring determinants of ruminative tendencies, especially brooding, in athletes. 15 Furthermore, psychological interventions targeting rumination could be examined as a 16 potential prevention and treatment approach to tackling depressive symptoms in athletes. 17

causes and implications of ones' negative feelings and problems. Consequently, engaging in 119 repetitive, abstract, and evaluative thought cycles in response to depressed mood is likely to 120 exacerbate this state and to disrupt effective problem-solving. Depressive rumination can, 121 therefore, be understood from a process perspective (e.g., attention or memory), rather than by  testing the relationship between these factors, they found that brooding was related to higher 152 levels of depression concurrently and longitudinally. Although reflection was also related to 153 more depression concurrently, it was associated with less depression over time. This finding 154 led the authors to conclude that a tendency to engage in brooding may be more maladaptive 155 than engaging in reflection. This assumption was supported by Lo, Ho, and Hollon (2008) 156 who demonstrated that brooding, but not reflection, mediated the effects of negative 157 attributional/cognitive style on depression. Despite these findings, the role of reflection as an 158 adaptive trait is still uncertain and is likely to be dependent on individuals' tendency to brood 159 as well as on their current levels of depression (Joormann et al., 2006).

161
It has been suggested that a ruminative response style can develop as a coping 162 mechanism to highly controlling and critical parenting style, or when individuals are 163 socialized through environments where expression of thoughts and opinions may be restricted 164 (Watkins, 2016). Considering that some sport contexts may expose athletes to similar 165 environments where coach mediated control behaviours may involve manipulation or verbal 166 abuse, excessive personal control or dismissal of athlete individuality and autonomy 167 (Bartholomew, Ntoumanis, & Thogersen-Ntoumani, 2009), examining the link between rumination and depression in athletes may be highly relevant. In addition, maladaptive 169 rumination may also involve a repetitive and evaluative approach to thinking about oneself in 170 relation to others and the discrepancy between current and desired state (Watkins, 2016   The Current Study 189 Our study aims to extend mental health research in athletes by seeking to clarify the 190 relationship between depressive rumination and depressive symptoms, therefore responding to depressive symptoms in the current athlete sample, (2) explore potential differences in 194 depressive symptom and depressive rumination scores (brooding and reflection) across athlete on previous studies, we hypothesized that female gender, lower level of competition, 198 engaging in individual sports, and being currently injured would relate to higher depressive 199 symptom scores (aim 2). In terms of brooding and reflection, we did not set any specific 200 hypotheses regarding differences across sport-related variables, however, we did expect to see 201 that brooding and reflective rumination scores would be higher in female athletes than in male 202 athletes (aim 2). Finally, we hypothesized that athletes with a high brooding/reflection profile  items, such as "Analyse recent events to try to understand why you are depressed", and five 237 brooding items, such as "Think why do I have problems other people don't have?".

238
Respondents rate each of the 10 items in the questionnaire from 1 (almost never) to 4 (almost 239 always) with higher scores representing higher tendency to engage in reflective and brooding 240 rumination when feeling low, sad or depressed. The internal consistency of the scales in the 241 current sample was α= .82 for brooding and α= .79 for reflection.

243
The relevant University ethics committee granted ethical approval for this study. The 244 online survey consisted of an information page that briefly described the study objectives. A 245 link to a detailed information letter was included on the information page, which the 246 participants were encouraged to read before consenting to participation. In addition to study 247 details, the information letter also included contact details for various mental health   First, we conducted a median split to categorize athletes based on their responses to RRS-265 scale: high/low brooding and high/low reflection categories (coded as low=0, high=1, 266 respectively). After this, we computed a "vulnerability to depressive symptoms" variable with four categories: "low = low brooding/low reflection", "moderate = low brooding/high 268 reflection", high = high brooding/low reflection, and "very high = high brooding/high     We also found that brooding was higher in off-season athletes when compared to in-365 season athletes. We are unable to infer any causal directions due to our cross-sectional design, 366 however, it is possible that levels of brooding vary across the athletic season. Although 367 brooding rumination is considered a stable trait, there is evidence suggesting that this stability

373
Exploring reflective rumination across sample characteristics, the only difference was 374 found between injured and uninjured athletes. Specifically, injured athletes showed 375 significantly higher tendency to engage in reflective rumination than uninjured athletes.

376
Similar to findings on higher brooding in off-season than in-season athletes, it is possible that 377 higher reflection scores in injured than uninjured athletes reflected increased coping efforts in 378 response to stressors associated with athletic injury. As discussed by Roy et al. (2016), higher 379 reflective rumination has also been linked to a lower ability to shift attention between tasks.

380
Perhaps athletes with a higher tendency to engage in reflective rumination in our sample were 381 more likely to be injured due to a decreased ability to switch focus, and consequently being 382 less likely to react to or avoid situations that may lead to injury.  Whitmer & Gotlib, 2011). Considering that we found that reflective rumination was higher in 411 injured than uninjured athletes, it is possible that reflection is, in fact, a maladaptive trait in an adaptive function over time and/or through other outcomes, such as increased meaning in 414 life and/or sport. 415 It is likely that the stressors that trigger depression in athletes may often be linked to 416 sport-related issues such as athletic failure, injury, or career termination (Appaneal et al.,  Targeting rumination may also be effective, not only in treating levels of depression, 450 but also co-morbid issues (e.g., anxiety) and residual symptoms of depression (e.g., sleep) that 451 may significantly impair daily functioning (Watkins et al., 2007). From this perspective, 452 theoretical approaches that have a more generic conceptualization of rumination (e.g., the S- youth sport environments (e.g. outcome-oriented, early initiation or specialization) and/or 465 youth coaching environments (e.g. criticism, excess control). If some contexts are more likely 466 to contribute to an increased tendency to ruminate, early prevention strategies could be 467 designed to attenuate these trends.

468
Our study findings should be interpreted with an understanding of its' main 469 limitations. Firstly, due to the cross-sectional study design, causal or temporal inferences 470 cannot be concluded. Also, due to the convenience sampling methodology, self-selection bias 471 may have influenced our findings. We also had a broad inclusion of athletes in terms of the 472 type and the level of sports. As we did not have sufficiently large sample sizes across different 473 sports, we were not able to conduct specific analyses to disentangle potential sport-specific 474 effects. It is also important to note, that our findings were limited to athletes from the United

475
Kingdom and hence, interpretations should be made with an understanding of this cultural 476 specificity.

477
In spite of these limitations, we believe that our study is an important addition to the 478 mental health literature in sport psychology. It provides a theory-driven glance into the 479 relationship between depressive rumination and depression symptomology in an athlete 480 sample, which to our knowledge is first of its kind in the field. Furthermore, the study 481 responds to a call voiced by other scholars to explore and validate cognitive vulnerability differences. Although informative of differences, these types of analyses are not sensitive to 487 the clinical relevance of the observed differences. Therefore, in addition to mean differences,