Anxiety and depression in athletes assessed using the 12-item General Health Questionnaire (GHQ-12)-a systematic scoping review

level, the full text was retrieved and screened. After merging the databases and removing duplicates, a second author screened the studies for reliability purposes. A schematic of the process is shown in

have a broad scope, covering many mental health symptoms and psychological well-being behaviours (e.g. sleep disorders, ADHD/ADD, eating disorders). [2][3][4][5][6][7] This makes it difficult to compare studies to develop interventions for depression and anxiety. In response, a scoping review of the literature focusing on studies using the GHQ-12 was performed. The aim of this review was to synthesise and compare studies using the GHQ-12 in athletes in order to inform future research by identifying trends and gaps in the literature.

Search
This review was reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. A systematic search of five electronic databases (Google Scholar, PubMed, PsychINFO, Scopus and Web of Science) was conducted on all published studies up to 1 January 2019. Key terms included: anxiety, depression, mental health, elite athletes, athletes, sport, general health questionnaires, GHQ and combinations thereof. Inclusion criteria for the review were: (1) participants were able-bodied athletes; (2) studies measured anxiety/depression using the  (3) studies were full, original articles from peerreviewed journals; (4) studies were published in English. Studies were excluded if athlete and non-athlete populations were combined as a single group. Athletes are defined in this review as individuals who train in sports aiming to improve their performance; are actively participating in sport competitions; registered in a local, regional or national sports federation; and devote several hours on most days to their sport. Studies were screened at the title and abstract level for eligibility. If a decision on the eligibility of the study was unclear at the title and abstract level, the full text was retrieved and screened. After merging the databases and removing duplicates, a second author screened the studies for reliability purposes. A schematic of the process is shown in Fig. 1.

Data extraction
The authors, year of publication, study design, purpose of the study, study population characteristics (sample size, age, active/retired, ratio of men to women, country), GHQ-12 (scoring method, cut-off point, mean score), prevalence and incidence of anxiety/depression, as well as factors associated with anxiety/depression were extracted from each study and tabulated.

Results
A total of 202 studies were identified through database searching, while five were retrieved through other sources. After duplicates were removed, 175 studies were screened by title/abstract, of which 37 full texts were assessed. Finally, 32 studies were included in the review (Table 1).

Prevalence/incidence
The prevalence of symptoms of anxiety/depression assessed by the GHQ-12 ranged from 21-48%. The incidence of symptoms of anxiety/depression ranged from 17-57%.

Discussion
The GHQ-12 is a popular tool used to screen the presence of anxiety/depression symptoms among athletes. Its popularity can be attributed to its robust psychometric properties and quick unobtrusive administration. [11][12][13] This is the first review to focus specifically on the GHQ-12 in order to compare studies and identify potential risk factors for anxiety/depression, as well as methodological considerations for future research. Not surprisingly, methodological inconsistencies between studies using the GHQ-12 were found. Sixty-eight percent (n=23) of the studies used the traditional 0-0-1-1 scoring method. Of these 23 studies, three different cut-offs were applied. For future research, we recommend the traditional scoring of 0-0-1-1 be used. Furthermore, to improve anxiety/depression prevalence and incidence comparisons, we also suggest that the recommended cut-off for the GHQ-12 for athletes be set at ≥3. In addition, the mean GHQ-12 score should be reported, as suggested by Goldberg. [15] Most of the studies used a cross-sectional or prospective cohort design with the objective of determining the prevalence and incidence of symptoms of anxiety/depression. The prevalence and incidence ranged from 21-48% and 17-57%, respectively. [9,16,17,19,23,[25][26][27][28][29][30][31][32][33][34][36][37][38][39][40]42,44,46] A sample of elite athletes from the United Kingdom (cricket, fencing, hockey, rugby union and many others) and elite Gaelic athletes presented with the highest prevalence of anxiety/depression symptoms (48%), [9,40] while the highest incidence was found in a sample of elite Dutch athletes (57%). [38] The response rates for most of these epidemiological studies were around 30%, with 40% comprising samples of European professional football players, [16,19,26,27,[29][30][31]33,34,36,41,42] thus decreasing the generalisability of the findings and highlighting a clear area for further investigation.
Several studies have identified potential risk factors for anxiety/depression symptoms. The most notable of these are recent adverse life events, [17,19,27,31,44] career dissatisfaction, [9,26,44] injuries, [19,40] surgeries, [19] social support, [21,31] osteoarthritis, pressure to perform, and career transitioning. [21,23] In one study, the GHQ score was used as a predictor for musculoskeletal injury. [42] Although no reported association was found between anxiety/depression symptoms and severe musculoskeletal injuries, the study was novel in using the GHQ score as a potential risk factor for predicting injury. Using the GHQ score as a predictor rather than an outcome suggests that the GHQ-12 could potentially be used as a monitoring tool for injury risk. Two studies attempted to reduce the GHQ-12 scores of athletes. Wilson et al. attempted to reduce GHQ-12 scores in 10 experienced jockeys using an exercise and diet intervention. [46] However, the study was limited by its small sample size and lacked a control group. The other study used a randomised control study design, where the intervention was an internetbased cognitive behavioural therapy. [45] No significant differences were found between the intervention and control group, and this was attributed to the short period of the intervention.
The GHQ-12 is proposed to measure symptoms of anxiety and depression (reported also as anxiety/depression). [2] Although often comorbid, these are two different psychological conditions. Depression is a medical condition that negatively impacts on how an individual feels, thinks and acts. [47] Symptoms include sadness, apathy, guilt, low self-esteem, trouble sleeping, decreased appetite, tiredness, poor concentration and suicidal ideation. [47] Depression can be chronic or recurrent, and can significantly affect an individual's ability to cope with daily life. [48] Anxiety is defined as the anticipation of a future concern, whilst fear is an emotional response to an immediate threat. [47] Anxiety disorders are characterised by excessive feelings of anxiety and fear. [48] Researchers and sport practitioners should be aware of the distinctions between depression and anxiety when using the GHQ-12.

Conclusion
This review compared GHQ-12 studies and identified potential risk factors for depression and anxiety, as well as methodological considerations for future research. Based on this review, we recommend the traditional scoring of 0-0-1-1 be used with the cut-off set at ≥3. Also, the mean GHQ-12 score should be reported. The prevalence and incidence of symptoms of anxiety/depression ranged from 21-48% and 17-57%, respectively. Potential risk factors for anxiety/depression include recent adverse life events, [17,19,27,31,44] career dissatisfaction, [9,26,44] injuries, [19,40] surgeries, [19] social support, [21,31] osteoarthritis, pressure to perform, and career transitioning. [21,23] Future research should broaden the spectrum of athlete populations and aim to improve response rates. Finally, researchers and sport practitioners should acknowledge that the GHQ-12 does not differentiate between symptoms of anxiety and depression.     Kenyan athletes scored lower than UK groups on the GHQ; athletes scored lower than the controls. Kenyan runners were less likely to report symptoms of anxiety/depression than UK runners.

Ivarsson et al., 2015 [41]
Observation al, prospective cohort There was a 21% prevalence of jockeys suffering from symptoms of anxiety/depression. Mean GHQ-12 score was 10.3 prior to the dietary intervention reducing to 8.9 postintervention. However, this was not statistically significant.