Anti-Doping research: What is left to do?

View correction statement:

Correction

Introduction

Since the formation of the World Anti-Doping Agency (WADA) in 1999, research on doping and (in particular) anti-doping has proliferated. A quick search on Scopus returns almost 9.000 scientific journal articles that are related in some way to anti-doping policy and practice.¹ The vast majority of these articles (98%) have been published in the post-WADA period. Figure 1 illustrates what we regard as the ‘birth’ and ‘development’ of anti-doping as a research field.² The field has seen an almost exponential increase in research publications over the past two decades, including numerous monographs, edited collections, research reports, and even a handbook dedicated to the topic (Møller et al., 2015). Perhaps the best indicator of the field’s maturity is the fact that there are now entire journal sections dedicated exclusively to anti-doping research, such as the recently established ‘Anti-doping Sciences & Integrity in Sport’-section of the journal Frontiers in Sport and Active Living.

Figure 1. Growth in anti-doping research from 1999 to 2022.

Given that the field has now reached ‘adulthood’, it seems timely to reflect on the question that is the theme of this special issue: ‘What is left to do?’ This question also took the center stage at the 2022 International Network of Doping Research (INDR) conference, which was held at Aarhus University in August 2022 to celebrate the 20^th anniversary of the network.³ Passionate doping scholars may find this question provocative owing to its implicit assumption that most doping-related topics are somewhat exhausted, and that there is limited basis for gaining truly novel insights in the field. However, in our opinion, this is not (yet) the case. In this editorial, we argue that there is still much to learn about the doping phenomenon and that many questions remain open (albeit some more than others). More specifically, we focus on the following three questions because they have attracted more scholarly attention than others in the domain of social science (our field), and because they are central to the anti-doping movement in a very fundamental way:

• How widespread is doping (the epidemiology question)?

• Why do athletes dope (the aetiology question)?

• How can doping be prevented (the prevention question)?

In the following, we briefly summarize and discuss what past research has taught us about the epidemiology and aetiology of doping in elite sport and beyond as well as effective prevention strategies, including which aspects of these questions are not yet fully understood. We conclude by proposing a vision for the future of anti-doping research that, if brought to fruition, will move the field forward and significantly improve our understanding of doping and the impact of efforts to curb it. We use the term ‘doping’ in this article to refer to athletes’ intentional use of substances or methods that are prohibited in sport according to the World Anti-Doping Code (WADA, 2021b).

The epidemiology of doping

Previous research and doping control testing statistics have shown that the prevalence of doping varies according to factors such as gender, age group, type of sport, and level of competition. Doping is generally more widespread among men than women, just as the frequency of doping tends to increase with age. Furthermore, as doping is – generally speaking – a way to enhance performance, we find its use especially in sports where performance depends heavily on the athletes’ physical capacity, such as athletics, cycling and weightlifting (Ntoumanis et al., 2014; WADA, 2021a). We also know that elite-level athletes engage in doping to a greater extent than amateur-level and recreational athletes. A notable exception are bodybuilders and recreational gym-goers which, at least in absolute terms, represent the largest group of anabolic steroid users (Sagoe et al., 2014).

Regarding the general prevalence of doping in elite sport, it is well known that there is a discrepancy between the proportion of positive doping controls and the proportion of elite athletes admitting doping in surveys based on self-report. The percentage of so-called ‘adverse analytical findings’ in doping samples have remained low and relatively stable since the late 1980s (0.96 − 2.45%), with less than 1% of all samples leading to a so-called ‘anti-doping rule violation’. By contrast, evidence from prevalence studies suggest that the true number of doped athletes is considerably higher, with some studies reporting prevalence estimates as high as 39%, indicating a significant number of undetected cases (de Hon et al., 2015; WADA, 2021a). However, it is equally well known that athletes tend to underreport doping due to the secrecy surrounding this behavior. Although doping scholars have attempted to circumvent this issue by employing alternative methods, such as the randomized response technique, the question of how reliable current estimates of doping are remains open, as does the question of which method(s) are best suited to measure the prevalence of doping (Gleaves et al., 2021; Petróczi et al., 2022).

The aetiology of doping

In their attempt to explain why some athletes dope while others stay ‘clean’, researchers have made use of various methods, both quantitative and qualitative. Two common approaches include the application of existing theories of behavior to explain and predict doping such as Social Cognitive Theory or the Theory of Planned Behavior, and the development of domain-specific models such as the Sport Drug Control Model and the Life-cycle Model of Performance Enhancement (Backhouse et al., 2016). The predictive validity of these models - that is, how well they perform in terms of explaining behavior - is still debated (Blank et al., 2016). Taken together, doping (in most cases) does not seem to be a linear, rational decision, but is an ‘interplay of sociocultural, socioeconomic, and personality factors’ (Ntoumanis et al., 2014). Especially the above-mentioned models are mainly based on certain risk factors and predictors which were revealed to explain doping intentions, susceptibility and behavior. But, when explaining doping we have to take into account not only personal factors (e.g., beliefs, attitudes, and norms) but also environmental factors that influence athletes’ decision-making in relation to doping (e.g., situational temptation, coach climate) (Blank et al., 2016; Ntoumanis et al., 2014). When asking athletes who are currently doping or have previously done so, they often justify their behavior by pointing at external pressure (e.g., peer pressure) (Engelberg et al., 2015). Summarizing research on this topic, scholars have indeed been able to identify many correlates, but this research is limited by the cross-sectional nature of many studies. This means that factors which are commonly regarded as determinants of doping may turn out to be consequences, or both. Therefore, research cannot give a universal answer to the aetiology question yet. Decisions about doping are still individual, complex choices with irregular patterns that are not easily explicable by one single model. In more recent years doping scholars have started to approach the topic from a new perspective, moving on from the examination of risk factors to predict doping towards exploring what makes an athlete be or stay clean (Englar-Carlson et al., 2016; Overbye et al., 2015).

The prevention of doping

The key aim of the international anti-doping program, aside from detecting and sanctioning doped athletes, is to prevent the onset of doping through deterrence as well as education of athletes and their support personnel, such as coaches and parents. This two-pronged approach has been WADA’s core prevention strategy since the first World Anti-Doping Code entered into force in 2004, although the importance ascribed to education in the Code relative to deterrence has increased over the years (WADA, 2003, 2021b). However, despite more than three decades of research into the doping prevention question, very little is known about the effectiveness of this strategy, if we by ‘effective’ mean a documented ability to reduce the incidence of doping (Bates et al., 2019). As regards deterrence, only two studies have investigated the effects of drug testing on athletes’ use of performance enhancing drugs, and the findings from these studies were mixed (Goldberg et al., 2003; Goldberg et al., 2007). Thus, although in theory doping controls may be effective in deterring athletes from taking prohibited substances, there is little experimental evidence to support this strategy (Donovan et al., 2002; Strelan & Boeckmann, 2003). In relation to this, research has shown that although the risk of getting caught and punished do influence athletes’ willingness to dope, many athletes regard factors such as the health risks of doping and the anticipated condemnation from others as greater deterrents than the sanction itself (Dodge & Jaccard, 2007; Huybers & Mazanov, 2012; Overbye, 2017; Overbye et al., 2015; Strelan & Boeckmann, 2006).

The evidence base underpinning education as a prevention strategy is considerably stronger, and many doping scholars seem to put great faith in education as a key ingredient of effective anti-doping (see e.g., Backhouse et al., 2012; Blank et al., 2022; Hanson, 2009; Hauw, 2017; Hoberman, 2013; Woolf, 2020). In a recent systematic review of interventions to prevent the use of anabolic steroids and other doping substances, Bates et al. (2019) identified 14 distinct interventions, of which 12 were categorized as educational. Most of these interventions are drug education programs delivered to adolescent athletes in school settings. Common to these programs is that they are based on the idea that adolescence is a critical period where attitudes and values are formed, and that early education can serve as a form of ‘psychological inoculation’ that will protect athletes when they are later exposed to ‘dopogenic’ environments (Backhouse et al., 2012). However, as the situation stands, it is not possible to confirm or disconfirm this assumption. With a few notable exceptions such as the Athletes Training and Learning to Avoid Steroids (ATLAS) program, none of the reviewed interventions had any impact on doping (most interventions did not even include a behavioral outcome). Positive intervention effects have been reported on behavioral influences such as attitudes, norms, and intentions, but these effects are generally small, and it is unclear whether they persist due to lack of long-term follow-up and, if they do, whether they translate into sustained abstinence from doping (Bates et al., 2019). Several intervention studies have been published after this review, yet this has not improved our understanding in any significant way of what works in relation to doping prevention (Galli et al., 2021; Kavussanu et al., 2022; Kavussanu et al., 2021; Nicholls et al., 2020; Pöppel, 2021; Sagoe et al., 2021; Yager et al., 2023).

Until a strong evidence base has been developed, national anti-doping agencies and others seeking to prevent doping through education are best served by implementing the strategies applied in exemplar interventions (e.g., ATLAS), and by relying on best practice recommendations for educational programs within the field of traditional substance use prevention (Backhouse et al., 2009; Elliot et al., 2008; Goldberg et al., 1996). Recent research has shown that there is a significant gap between the strategies that anti-doping authorities employ in their educational programs and the strategies that are known to be most effective in preventing traditional substance use and, presumably, doping. This indicates that there is much room for improvement (Gatterer et al., 2020).

So what’s left to do?

As briefly touched upon in the introduction, the scientific community still can’t give a clear answer to the question of how widespread doping is. In fact, the question is whether this will ever become possible since the nature of doping and the secrecy surrounding this behavior makes it difficult to investigate. Today, there are some promising, but far from flawless methodological alternatives to the traditional, self-report survey for measuring prevalence. These methodological problems impede our ability to provide good answers to the prevalence question, which further impacts our ability to answer the other questions properly (in particular the prevention question). There is evidence to suggest that athletes’ decision to dope is influenced in part by their perception of the prevalence of doping within their sport, the so-called ‘false consensus effect’ (Petróczi et al., 2008). Against this background, it will be important to know what the actual prevalence is in order to indirectly influence the incidence of doping. A satisfactory answer to the prevalence question would not only influence research on risk and protective factors, but also support the monitoring and evaluation of anti-doping policies and prevention programs.

We also need more research on the aetiology of doping and especially on what makes athletes stay clean. We see the current trend to put more focus on drivers for clean sport behavior, i.e. protective factors, instead of risk factors, as a potentially promising way for future research to better address this question. By stressing the benefits of competing clean and empowering athletes to withstand the enticement of doping, those aiming to prevent doping might be able to counter the perceived ‘dopogenic’ environment and strengthen the clean sport attitude in the next generations of athletes.

A major problem from our perspective is the preponderance of cross-sectional studies compared with longitudinal ones. In a comprehensive review of the literature on doping, Backhouse et al. (2016) identified 153 studies that examined factors that influence doping among athletes, yet only six of these studies adopted a longitudinal design. As cross-sectional designs are not suitable for drawing causal inferences, a lesson we teach our students but tend to forget in our own research, our understanding of protective and risk factors is limited to correlates which may not be the determinants we are actually looking for. With seven years having passed now, the tendency to prefer cross-sectional designs seems to prevail, although scholars in the field agree about the importance of using more appropriate research designs (Boardley et al., 2021). As with any long-term study, the problem lies mainly in the financing of such projects, which are costly and can only deliver results after several years of study. However, such durations would be necessary in order to be able to examine changes in variables over time.

The suggestion to rethink our methodological approach also applies to the field of prevention and intervention studies. We see a problematic discrepancy here between recommendations on how to design and implement interventions for prevention and how interventions are actually designed (Bates et al., 2019; Gottfredson et al., 2015). One of the most profound gaps we see is the lack of evaluation of anti-doping programs, especially long-term follow-ups. Before inventing new programs, it seems reasonable to evaluate existing ones and revise them if necessary. This could save scarce resources both in research and practice as systematic intervention development can be cost-intensive. Another problem that has been increasingly addressed in recent years is the lack of collaboration between academic institutions and anti-doping stakeholders such as regional and national anti-doping organizations, sports federations as well as athletes. The latter in particular should be actively involved in decision-making and development processes in order to establish the basis of any intervention – a thorough understanding of the target population. Not only should we expand collaboration in the developmental phase, but also afterwards by not keeping outcomes to oneself but sharing existing knowledge with other organizations and countries. Especially countries with less developed anti-doping structures would benefit from such a transfer of knowledge since the unequal distribution of resources contributes to the problem that international competitions cannot be held on a level playing field, since the same conditions do not prevail in every country.

In summary, we believe that it is impossible to eliminate doping in sport, as anti-doping authorities will always be one step behind the doping industry. Nevertheless, if in the future we address and answer the above questions against the background of the listed, especially methodological difficulties, we are convinced that efforts to prevent doping will be better fit for purpose.

Closing statement

Finally, we would like to stress that the points we have made are not meant to belittle the work of the researchers who came before us. It is only through this work that we are where we are today. As we have tried to show in this editorial, the fact that anti-doping has now become a well-established research field with a large and growing body of publications does not mean that there is nothing left for doping scholars to do in the future. We encourage our colleagues to challenge the prevailing ‘publish or perish’ – mentality in academia which often leads to prioritizing research designs that most effectively boost publication records, but which may not provide a satisfactory answer to the questions we ask.

Disclosure statement

The authors report there are no competing interests to declare.

Correction Statement

This article was originally published with errors, which have now been corrected in the online version. Please see Correction (http://dx.doi.org/10.1080/09687637.2024.2310876)

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.

Notes

1 We searched for articles published between 1999 and 2022 that included ‘anti-doping’ or ‘antidoping’ in all fields of the article. We deliberately refer solely to ‘anti-doping’ and omitted ‘doping’ somewhat from our discussion. Our aim is to focus on science as it relates to the global anti-doping movement and the prevention of doping, rather than to delve into the study of doping substances and methods per se.

2 We recognize that a significant body of research into the use of performance enhancing drugs in sport existed before 1999, which includes studies that are still regarded as exemplars (see e.g., Goldberg et al., 1996), but the formation of WADA appears to have stimulated an exceptional scholarly interest in the doping issue.

3 https://ph.au.dk/indr/