The SCL-90 is an instrument to estimate psychological problems with a five-point rating scale, ranging from 1 (not at all) to 5 (extremely). It has a wide coverage of various mental health disorders and contains nine subscales about mental health (i.e., somatic complaints, obsessive-compulsive symptoms, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychoticism), with a higher score suggesting a lower mental health level (Derogatis et al., 1973; Xin et al., 2019). The SCL-90 scale has a wide range of application and can accurately assess the degree of subjective disease perception and distribution characteristics (Wang, 1984). In addition, the Chinese version of the SCL-90 is the most frequently used tool for measuring Chinese athletes’ mental health levels (Liu et al., 2002; Tang et al., 2013). Thus, this study collected data from original research on the SCL-90 scores of Chinese athletes to conduct a cross-temporal meta-analysis.

The sources of literature for the current research were three academic literature databases in Chinese: CNKI, Wanfang, Chongqing VIP Information. These databases are frequently used in China and include Chinese journals of sciences and social sciences published after 1985, as well as master’s theses and doctoral dissertations. We used search terms such as “SCL-90,” “athletes,” “college athletes,” “psychological problems,” and “mental health.” Additionally, we searched five academic literature databases in English, including Elsevier, Wiley, ProQuest, PubMed, and Web of Science, using the same keywords. Only articles that provided the average SCL-90 score, sample size, and publication year were collected. In addition, the year of data collection was coded as 2 years before the publication of the previous study, unless a specific year was mentioned in the article (Twenge, 2001; Xin and Chi, 2008; Xin et al., 2019).

Studies that met the following special inclusion rules were included in the current cross-temporal meta-analysis: (1) studies using the SCL-90 as the tool; (2) studies reporting mean scores and sample sizes for the total or unselected subgroups; (3) studies in which the participants were Chinese athletes; (4) studies in which all participants were from mainland China; (5) studies published before September 2023; and (6) for studies in which the same data were published twice or more, the earliest study was selected. To illustrate the inclusion strategy and the results of the search strategy, a PRISMA diagram is depicted in Figure 1.

As shown in Figure 1 and Table 1, the final sample included 28 studies involving 4,227 athletes between 1995 and 2023 (the year of data collection). No English literature met the inclusion criteria. Detailed information for the studies included in the meta-analysis can be found in Table 1.

In the present study, publication class, sex ratio and types of athletes in every article were recorded as control variables in the cross-temporal meta-analysis because they may confound with the year. We coded the publication class was divided into three levels (Xin et al., 2019): first class (including the SSCI, SCI, CSSCI, and CSCI), second class (including publications from other sources) and third class (including master’s theses and doctoral dissertations). Publication class was controlled to avoid the effect of publication bias, which refers to the fact that studies with statistically significant results are more likely to be accepted for publication (Xin et al., 2019). And the types of athletes were divided into two levels: first class (including the professional athlete), second class (including college athletes). The sex ratio, an important variable for mental health scores, was also recorded in the database and controlled in regression analysis. Detailed information can be found in Table 2.

First, for articles that only provided research data without comprehensive study results, the weighted statistics were obtained based on the provided data using the following formula (Li et al., 2022).

Where x ¯ , S, n_i, x_i, x_i, s_i, represent the average and standard deviation after synthesis, the sample size, average and standard deviation of a substudy, and analyze the changes of each factor in each stage. In addition, since most studies of SCL-90 used a 5-point scale ranging from 1 to 5, for the sake of uniformity, we converted the 3 studies with a score ranging from 0 to 4 into a score ranging from 1 to 5 by adding the mean of each factor by 1. To investigate the trend of mental health symptoms among athletes over different eras, a scatter plot will be created to illustrate the scores of each factor in the SCL-90 against the eras. Next, Pearson correlation analyses will be conducted between the scores of each factor in the SCL-90 and the eras. Furthermore, a weighted analysis will be performed to account for sample size, using eras as the independent variable and the scores of each factor in the SCL-90 as the dependent variables in a univariate linear regression analysis. This aims to examine the relationship between eras and mental health symptoms among college students.

In order to understand the specific amount of change in the mental health of different genders and athlete types Chinese athletes, according to previous researchers (Twenge et al., 2001, 2007), which is mainly measured by the effect size d or the explanation rate r², and the formula for the two is as follows, where SD is the average standard deviation.

This is achieved by establishing a regression equation (controlling sample size) that takes the mean value of each factor of SCL-90 as the dependent variable and the age as the independent variable: y = Bx + C (where B represents the unstandardized regression coefficient, x is the year, C is the constant term, and y is the average mental health quantity). After the regression equation of each factor is established, the mean scores of 1995 and 2023 M₁₉₉₅ and M₂₀₂₃ can be predicted. Following the usual practice of Twenge et al. (2007), the mean standard deviation in this study is obtained by averaging the standard deviations across all studies, unlike the calculation of the effect size in normal meta-analyses. This approach, which uses individual-level variables, effectively avoids the ecological fallacy. The statistical software used for this study is SPSS 25.0, and the reference management software used is EndNote 20.0.

In this paper, by establishing the scatterplot of all factors and ages in SCL-90 scale, we can intuitively judge the change of athletes’ mental health in the process of age change. The scores of the 9 factors all improved to varying degrees with the change of years. See Figure 2 for the change of each factor with the change of years.

In order to explore whether the trend of change in the above scatter plot is significant, the correlation analysis of the 9 weighted factors in the SCL-90 scale is carried out. In this paper, the regression equation is established with age as the independent variable and scores of each factor as the dependent variable. Results As shown in Table 3 depression, hostility, phobic anxiety and paranoid ideation were significantly positively correlated with age (p < 0.05). The other factors were positively correlated with age, but none of them were significant. It can be seen that the age change can explain 25.6% of the variation of depression, 27.2% of the variation of hostility, 23.8% of the variation of phobic anxiety, and 18.8% of the variation of paranoia ideation.

The results in Table 3 indicate that the three factor scores of depression, hostility, and terror have increased over the years. The factor weighting values of 2023 and 1995 are brought into the publicity to calculate the interpretation rate and effect size, as shown in Table 4.

From Table 4, it can be seen that from 1995 to 2023, the three factors in the SCL-90 scale of Chinese athletes all show an upward trend, with depression increasing by 0.4 points, hostility increasing by 0.28 points, and phobic anxiety increasing by 0.42 points. According to the judgment of the effect size, the effect size between 0.2 and 0.5 is small effect. The medium effect is greater than 0.5–0.8, and the hostile is “medium effect.” Greater than 0.8 is a large effect, and depression factor and phobic anxiety are “large effects.” All of the above indicate that from 1995 to 2023, the mental health of Chinese athletes gradually deteriorates in depression, hostility and phobic anxiety, and the change speed is relatively fast, showing a large degree.

In this part, the correlation between the mean value of various factors of mental health of athletes of different genders and the age was statistically analyzed. It can be seen that the somatic complaints, obsessive compulsive, hostility are positively correlated, but they are not statistically significant. Among female athletes, interpersonal sensitivity and paranoid ideation paranoia were negatively correlated with age, while the other factors were positively correlated. Among the depression factors, age could explain 67.2% of the variation, and among the psychoticism, age could explain 48.9% of the variation, indicating that the scores of depression and psychoticism psychosis of female athletes were gradually increasing with the increase of age. It can also be seen that the change of depression in female athletes is the main cause of the change of depression factors in the mental health of the whole athlete group (Table 5).

This part makes a statistical analysis of the correlation between the mean value and age of each factor of mental health of different types of athletes. Each article reports the types of athletes, which can be divided into two types: professional athletes and college athletes. The relationship between the scores of each factor of the SCL-90 scale and the age of different types of athletes is shown in Table 6.

Overall, different types of mental health status over the years have obvious differences. The scores of mental health factors of professional athletes and college athletes increased gradually with the change of years, and the hostility of professional athletes showed an upward trend with the change of years, the increase rate was 29.4%; The factors of depression, anxiety, hostility, phobic anxiety, paranoid ideation, psychoticism in college athletes were on the rise with the change of years, and the increase rate was between 29.4 and 49.4%. Depression, phobic anxiety, paranoid ideation and hostility change of college athletes and hostile change of professional athletes are the main causes of the change of depression, hostility, fear and paranoia of athletes.

In order to explore the change trend of the overall mental health status of Chinese athletes in recent 28 years, this study used cross-sectional historical meta-analysis to explore the relationship between SCL-90 scale score and age. The results showed that the scores of the 9 psychological problems in the SCL-90 scale increased gradually with the age, among which depression, hostility, phobic anxiety and paranoid ideation were the most significant.

This is consistent with previous studies showing an increased risk of mental health problems among athletes (Sundgot-Borgen et al., 2004; Baum, 2005). A national survey of elite athletes in Australia found that almost half of the respondents had a mental health problem, with a prevalence similar to that of the community group (Gulliver et al., 2015).Recent studies have shown that retired elite athletes have a particularly high risk of mental illness (Gouttebarge et al., 2016).Foreign countries pay close attention to the mental health of athletes, especially elite athletes, but the overall research is relatively small at present. However, the view that elite athletes lack mental health has been paid more and more attention by sports medicine experts (Wolanin et al., 2015). Although much has been done to promote and disseminate the improvement of athletes’ mental health, as well as the prevention, identification and early treatment of mental health problems, it is hoped that the relevant sports bodies will continue their efforts to minimize the extent of mental illness in athletes (Reardon et al., 2010). If athletes do not have timely access to psycho-mental health care, or do not see this as important themselves, then this will have an impact on them to have psychological problems (Hamer et al., 2009). Almost all studies have indicated that physical exercise has a positive impact on mental health (Daley, 2008), but some studies have found that intense physical exercise by elite athletes is detrimental to mental health and will excessively increase anxiety, depression symptoms, injuries and mental exhaustion (Peluso et al., 2005). Many domestic research results show that the mental health of Chinese athletes is not optimistic, and the scores of many indicators of mental health level are higher than the national norm (Niu et al., 2009; Tan et al., 2014).The emergence of this situation is influenced by many factors, especially the social macro environment. Since China successfully held the Beijing Olympic Games in 2008, competitive sports have been booming, and with it comes social expectations for athletes, which makes athletes pay more attention to performance. The special status of athletes also leaves them with less time and experience to devote to their studies, and when it comes to college, they have to pay more than non-athletes to get into college. The common problem faced by professional athletes is the problem after retirement. The retirement of athletes has always been one of the main problems plaguing the development of competitive sports in China. In addition, there are other studies that show that the national team athletes have a serious lack of learning of cultural lessons, and 48% of the surveyed people have not taken cultural lessons in the latest year (Xu et al. 2011; Zhou et al., 2012). The management system and mode within the sports system also directly affect the future development of retired athletes. Due to the lag of system reform, the mobility, development and re-employment of athletes are objectively affected (Zhou et al., 2012).

This study found that depression and psychoticism scores of female athletes gradually increased with the growth of years, and such changes were the main reason for the change of depression factors in the mental health of the entire athlete group, while there was no significant change in male athletes. This is consistent with the results of some studies, which have found that vulnerability is related to depressive symptoms (Yamaguchi et al., 2018), and the vulnerability score of women is significantly higher than that of men (Yamaguchi et al., 2019). Female athletes have more symptoms of depression and suspicion than male athletes (Zhou et al., 2005).

From an individual perspective, female athletes face a greater variety of sources of pressure compared to male athletes (Pascoe et al., 2022). These sources include pressures related to family planning, motherhood, caregiving responsibilities (Moesch et al., 2012), media scrutiny (Wensing et al. 2003; Biscomb et al. 2019), and the prevalence of violence against women in the sports industry (Donegan, 2012). Additionally, inequalities in areas such as compensation, employment opportunities, and societal roles contribute to the stress experienced by female athletes (Pascoe et al., 2022). If these pressures are not actively addressed, they can lead to mental health issues. Furthermore, considering traditional biomedical differences (Zhou et al., 2005), women exhibit greater variations in hormone systems throughout their lifecycles compared to men (Seaman, 1997), potentially increasing the risk of factors like depression and mental illness among female athletes.

From the perspective of cultural causes, the decline of women’s mental health is due to the double pressure from life and work, especially the conflict of multiple social roles and the squeeze of gender culture (Ye, 2010). Campbell et al. (2020) believes that in countries with greater gender equality, women face a double burden of balancing economic income and political participation while also adhering to traditional female responsibilities and norms. For example, the pressure on women to succeed in highly competitive environments, while also having to navigate traditional gender roles such as childbirth, may exacerbate stress levels and increase the risk of mental health issues among female athletes.

Although domestic sports institutions provide mental health education for athletes with different levels of needs by inviting psychological experts to carry out training under the team and set up basic psychological education courses, at present, professionals with clinical psychological counseling experience and familiar with elite sports are very scarce, which greatly restricts the development of mental health of athletes in China. Therefore, in order to prevent the spread of female athletes’ psychological problems and improve their mental health as soon as possible, it is also necessary to integrate various social forces and provide a more equal and harmonious social, cultural and institutional environment.

This study found that the scores of depression, anxiety, hostility, phobic anxiety, paranoid ideation, psychoticism of college athletes were significantly increased year by year, and the scores of hostility factors of professional athletes were significantly increased year by year. It has become an indisputable fact that there are certain problems in the mental health of college athletes. Studies have found that the rate of college athletes and club members seeking treatment from college mental health organizations is relatively low, and 46.4% of college athletes show symptoms of depression, eating disorders and anxiety disorders (Gulliver et al., 2015). Alcohol abuse is more common among college athletes than non-athletes (Ford, 2009). Studies have shown that athletes with high education have more personal, career, academic and family problems, and their mental health status is relatively poor (Chen, 2001).

On the one hand, such a result may be related to the dual pressure of college athletes who need to balance academic work and performance. College athletes have not only academic tasks but also competition tasks, and long-term conflicts in academic training will inevitably lead to certain pressure and negative emotions of college athletes. On the other hand, it may be related to the stage of college athletes. Just entering college is a key period of self-awareness cultivation, but it may be negligent management, resulting in college athletes not establishing a correct and positive self-awareness.

The present study has the following limitations. First, there are other instruments to test mental health besides the SCL-90, such as the PHI (Zhou et al., 2009; Ma et al., 2015). Moreover, this study is the reliance on the SCL-90 tool for assessing mental health factors among athletes. While the SCL-90 is a widely used instrument, its effectiveness in capturing the full spectrum of mental health issues in athletes may be subject to interpretation and measurement bias. Future research could further explore the underlying mechanisms of athletes’ mental health status by comparing the results of cross-temporal meta-analyses using other questionnaires with the current results. Second, although the current study examined the trend of mental health level of Chinese athletes and provided possible reasons for this trend, it did not further reveal the specific influencing mechanisms. Therefore, future studies should adopt a longitudinal design and corresponding statistical analyses to test the relevant influencing factors. At Last, the longitudinal meta-analysis approach employed in this study, spanning different time periods and locations, introduces potential biases related to variations in data collection methods, cultural factors, and evolving societal norms. These factors could impact the generalizability and robustness of the findings.