Presenteeism in the elite sports workplace: The willingness to compete hurt among German elite handball and track and field athletes

Abstract

Playing hurt is a widespread phenomenon in elite sports that often goes along with using painkillers, disregarding medical guidelines, and hiding pain from coaches, teammates and medical staff. This paper theoretically conceptualizes the phenomenon of playing hurt as a sport-specific sickness presenteeism problem. To empirically analyse the willingness to play hurt, we refer to survey data from 723 elite German athletes, both male and female, in the sports of handball and track and field. Factor analysis, cluster analysis and binary logistic regression analysis are applied to reveal the athletes’ cognitive representation of absence legitimacy and to identify athlete groups with varying levels of willingness to compete hurt. Our results show that subtle distinctions are made between different kinds of health problems. In particular, there is a high willingness to compete despite psychosocial complaints. Cluster analysis reveals two clusters: ‘athletes conditionally willing to rest’ and ‘rest-averse and pain-trivializing athletes’. Athletes who perceive more social pressure to compete hurt, who have a higher performance level and who participate in handball, are more likely to be in the group of rest-averse and pain-trivializing athletes. The findings enhance our understanding of presenteeism and absenteeism in a highly competitive work context, and can contribute to the development of more target-group-specific health prevention programmes for athletes.

Keywords

attitudes elite athletes playing hurt presenteeism return-to-play

Competing despite having health problems is a widespread phenomenon in elite sports. The consequences of playing hurt and return-to-play decisions may affect an athlete’s career success and long-term health. Insufficient rehabilitation breaks, ignoring injuries and trivializing pain can cause irreversible physical damage, recurring traumatic injuries or chronic overuse injuries (Kujala et al., 2003; Waddington, 2000). The phenomenon of playing hurt has been amply described in sociological research on pain and injury in sport (Loland et al., 2006; Young, 2004). Case studies from different sports disciplines have shown that playing despite health problems often goes along with using painkillers, disregarding medical guidelines or hiding pain from coaches or teammates (Howe, 2001; Malcolm, 2006; Murphy and Waddington, 2007; Pinheiro et al., 2014; Roderick et al., 2000; Thiel et al., 2010). In this paper, we conceptualize the phenomenon of playing hurt as a sport-specific sickness presenteeism problem and analyse the willingness to compete hurt using a quantitative methodology. In the empirical analysis, we focus on the perspective of athletes and address the willingness to compete hurt as a cognitive representation of absence legitimacy that is affected by both contextual constraints and personal characteristics.

Studies on athletes’ injury experiences have revealed that ignoring and trivializing pain, injuries and health risks are central to practicing sport (Curry, 1993; Nixon, 1992; Young et al., 1994). Nixon (1993) speaks of a relatively coherent ‘culture of risk’ that is communicated to athletes through social networks or ‘sportsnets’, and is closely entangled with the norms and values of the sports ethic: sacrifice for the game, seeking distinction, taking risks and challenging limits (Hughes and Coakley, 1991). Both coaches and athletes support the notion that athletes must accept the risk of injury and push their bodies to the limit – even occasionally beyond the limit. However, athletes are at the same time also expected not to take excessive risks, as their bodies are the instruments they use to perform. This dilemma has been called the ‘risk-pain-injury paradox’ (Nixon, 1994a, 1996). The acceptance of high risk seems to be a general problem in competitive sports (Pike and Maguire, 2003; Pringle and Markula, 2005; Theberge, 2008) and can already be observed in youth sports (Schubring and Thiel, 2014). A strong willingness to accept physical and social risks can be found particularly among perfectionist and performance-focused elite adolescent athletes (Schnell et al., 2014). Overall, existing research has indicated that athletes in elite sports act within a more or less coherent culture of risk; this culture is assumed to promote risky practices, such as playing hurt.

Qualitative studies of team sports have indicated that injured athletes are often socially pressured to compete. For some coaches and managers, playing hurt is symbolic of the ‘right attitude’ in elite sports. Athletes who are not willing to play despite pain and injury risk being stigmatized as being soft, feminine or malingerers, and are often made fun of by teammates or ignored by managers (Malcolm and Sheard, 2002; Roderick, 2006; Roderick et al., 2000). Club physicians play an ambivalent role within this context. In general, physicians must negotiate between the culture of risk in sports and the culture of precaution in medicine (Safai, 2003). Together with physiotherapists, they often must convince athletes to take recovery breaks, sometimes in co-alliance with coaches (Thiel et al., 2010). However, studies on the role of physicians in rugby and football clubs have revealed that physicians have a relatively low status and limited authority in fitness-related decisions. Physicians are expected to provide medical treatment to keep athletes in competition mode and to get injured athletes back to training as fast as possible. The interaction between team physicians and injured athletes (as well as that between athletes, coaches and managers) is therefore often characterized by ethical dilemmas, power struggles and conflicts resulting from different perceptions about treatment strategies (Malcolm, 2006, 2009; Malcolm and Safai, 2012; Waddington, 2006). The practice of playing hurt can be understood as resulting from health-related risky decision-making processes, including complex social interactions between the athlete, their coach and medical staff. In this context, an individual athlete’s willingness to compete hurt is considered a highly relevant precondition in the decision about pain, injury or illness related lay-offs (Mayer, 2010).

The willingness to compete hurt has not been directly addressed in existing research and there has been a lack of quantitative studies of large samples of elite athletes. Thus, we do not know the extent to which athletes share a similar willingness to play hurt and which factors contribute to higher and lower levels of willingness. Moreover, the willingness to play hurt has not yet been sufficiently conceptualized from a theoretical perspective.

According to sport sociological research the action field ‘professional sports’ can be understood as a workplace (Murphy and Waddington, 2007; Roderick, 2006; Waddington, 2000; Young, 1993). This is true not only for those of paid professional status, but also for male and female athletes who pursue double careers. Firstly, being an elite athlete is a full time engagement, considering the time spent training or travelling to competitions. Secondly, elite athletes usually have contractual agreements with their federations and/or clubs. These contracts regulate the conditions and duration of the memberships – and are thus similar to working contracts. As members of clubs or squads, the athletes have to fulfil certain role expectations like being present at scheduled training sessions or being available for pre-set competitions. They have to follow general rules and regulations of the international and national sports federations, which are comparable to profession-specific ethics. Thirdly, the athletes are facing formal and informal performance-related expectations – just like every ‘normal’ worker. These expectations produce a large amount of social pressure to accomplish the tasks associated with the athlete’s role. However, the elite sports workplace is special, as good health and excellent physical fitness are fundamentally important for being able to remain a productive participant in this field.

In recent years, the question of why people attend work while being ill has been the subject of various studies. One of the most prominent approaches is the dynamic model of presenteeism and absenteeism introduced by Johns (2010). In the following, we apply the concept of presenteeism to the social context of elite sport and link it to the discussion about managing pain and injury in sports.

Theoretical background

Competing hurt as sport-specific sickness presenteeism

Sickness presenteeism is generally defined as attending work while being ill (Aronsson et al., 2000). From a sociological perspective, it refers to situations in which employees go to work despite perceiving themselves to be sufficiently ill to have legitimately called in sick (Hansen and Andersen, 2008). We define sickness presenteeism in the sports context as training or competing despite having a health problem. Presenteeism is therefore synonymous with keeping up social functioning despite having a medical condition or feeling ill. In the following, we focus on ‘competing hurt’. Because athletes often compete hurt before seeking medical advice (and therefore before being diagnosed by a physician), we do not differentiate between ‘competing hurt’ and ‘competing injured’.

Analysing presenteeism enables the ‘grey area between illness related absence and full work engagement’ to be addressed (Johns, 2010: 522). Research on presenteeism is crucial because this practice might not only exacerbate existing medical conditions but also damage the quality of working life or lead to impressions of ineffectiveness due to reduced productivity. However, presenteeism might also be viewed as personally beneficial, as it exhibits organizational citizenship and can garner praise and public recognition. From an organizational perspective, appropriately managing presenteeism can also be seen as an important source of competitive advantage (Johns, 2010).

Dynamics in competing hurt

Presenteeism can be conceptualized as a dynamic social action arising from the interplay between the social structure of an organization, personal factors and the characteristics of the current health event (Johns, 2010). It begins with the interruption of a person’s fully productive regular attendance by an acute, episodic or chronic health event. Within the frame of the given work context, organization members facing health problems must decide whether to show up for work or to be absent. People usually attend work regularly without any conscious decision-making; thus, recognizing a health event forces a person to make a decision (Nicholson, 1977). According to Nicholson (1977), health incidents fall on a continuum of avoidability. Avoidability is seen to be jointly affected by the health event and job demands. The interrelation of the individual’s health issue and their specific job demands highly preconfigures the choice between presence and absence. Thus, the particular nature of the health event to some extent dictates whether absenteeism will directly result, and the extent to which presenteeism is possible. For a soccer player, for example, a cruciate ligament rupture is likely to induce absenteeism, whereas a headache might result in presenteeism. Elite athletes often have subjective health complaints or injuries that do not necessarily require withdrawing from sports activity. Particularly in less extreme medical cases, the interplay of work context and personal factors highly influences presenteeism and absenteeism decisions (Johns, 2010).

The work-context-related factors that tend to favour the occurrence of presenteeism on a collective level can be grouped into job characteristics, organizational policies and absence/presence cultures (Johns, 2010). Job insecurity, time pressure, teamwork and low replaceability, and adjustment latitude when ill are some of the job characteristics that have been associated with presenteeism (e.g. Hansen and Andersen, 2008). Among organizational policies, the existence of a reward system for not being absent potentially fosters presenteeism (Munir et al., 2007). Organization- and profession-specific absence/presence cultures have been assumed to play a major role in the collective avoidance of the sick role (Dew et al., 2005); an absence/presence culture represents a ‘set of shared understandings about absence legitimacy and the established custom and practice of employee absence and control’ (Johns and Nicholson, 1982: 136). Studies analysing workgroup norms and supervisor attitudes concerning absenteeism have emphasized the practical relevance of such shared belief systems (e.g. Bamberger and Biron, 2007). Although personal factors associated with presenteeism are not yet well understood (Johns, 2010), attitudes towards work and health, as well as gender, age and personality characteristics, may have an effect on a person’s decisions when facing a health problem (Aronsson and Gustafsson, 2005; Hansen and Andersen, 2008).

From a temporal perspective, presence/absence decision outcomes must be regarded as distinct events occurring in sequence over time. For example, being present can negatively affect an underlying health problem and result in absenteeism some time later. On the other hand, deciding to rest can ease a health problem and finally lead to fully engaged attendance after a few days (Johns, 2010). In this regard, the aforementioned long-term consequences of presenteeism can be seen as the result of consecutive and closely entangled decisions.

Extending Johns’ (2010) dynamic model of presenteeism, we argue that the characteristics of the work context and its related social expectations influence presenteeism/absenteeism decisions in two ways. First, situation-specific personal and organizational expectations, such as a perceived deadline or specific advice from a supervisor, can directly influence how presenteeism/absenteeism decisions end up being made. Second, temporally outlasting expectations, such as those arising from the structure of the work context, indirectly affect decision-making habits by shaping individuals’ work-related attitudes (Figure 1). Within this paper, we address these temporally outlasting attitudes towards work and focus on the characteristics of athletes’ perceived absence and presence legitimacy when injured or ill. Absence and presence legitimacy is defined by the extent to which employees perceive presenteeism and absenteeism as acceptable work behaviour (Addae et al., 2013). A person’s perceived absence legitimacy is a major factor in presenteeism-related decision-making, particularly regarding minor health conditions (Harvey and Nicholson, 1999). Applying this notion to competing hurt, we argue that athletes develop a characteristic willingness to compete despite health problems and that this represents perceptions of absence and presence legitimacy in the ‘work context’ of elite sports. Thus, very mild subjective complaints would imply a low willingness to compete hurt while a high willingness would be associated with the neglect of more severe medical conditions to compete.

Figure 1.

Dynamic model of competing hurt in elite sport (based on the model introduced by Johns (2010)).

Predictors of the willingness to compete hurt

Taking the dynamic model of presenteeism and the concept of presence/absence legitimacy into account, we assume that the willingness to compete hurt is primarily mediated by processes of socialization in the work context of competitive sports and is progressively shaped through social expectations within elite sports organizations. We further expected the willingness to compete hurt to be shaped by athletes’ self-expectations and commitment to pursuing sports careers, which both result from individual biographical trajectories (Figure 1).

Organizational context factors

Elite sports organizations such as league teams and sport federations’ squads have the basic goal of high performance in competition. Irrespective of the sports discipline, this is not only a central goal for these organizations as entities but also for central members – coaches, managers, doctors and athletes. As the social constraints to compete increase with a team’s performance level, we expect international top athletes to show a higher willingness to compete hurt compared with elite athletes at the top of the national level and below. At the organizational level, job security must also be considered. In elite sports, professional and amateur athletes have high levels of job insecurity due to squad membership and short-term contracts, which induce an extreme orientation towards performance. Nevertheless, being a paid employee of a club or team might have an effect on the willingness to compete hurt as well, due to specific absence regulations including statutory sick pay.

The specific structure of a sports discipline could also have an impact on presenteeism in elite sport. Athletes learn formal rules and informal expectations about how to appropriately deal with health conditions that occur within their sports discipline: they gather knowledge about legitimate causes for lay-offs, as well as knowledge about when they are socially expected to continue to play. In this regard, each sports discipline’s distinct performance requirements, load demands and attendance regulations must be considered. The regulation of attendance can be assumed to be influenced by sports organizations’ particular presence and absence cultures. In elite sport, presence and absence cultures likely induce perceived social pressure to continue to play despite having a health condition. The sport-specific culture of risk (Donnelly, 2004; Nixon, 1992) could be closely interrelated with presence and absence cultures. It can be assumed that athletes who continuously experience subtle or even direct social pressure within teams or federations develop a stronger willingness to compete hurt. However, personal factors related to athletes’ individual biographical trajectories can also affect the extent to which athletes are willing to compete hurt within the peculiar ‘work context’ of elite sport.

Personal factors

From the beginning of early adolescence, athletes are socialized into a competitive sports culture. Particularly promising athletes must successively disregard aspects of life outside of sport to fulfil athletic role expectations. Because of this process of ‘hyperinclusion’ (Riedl et al., 2010), ‘overconformity’ to the central values of the sports ethic likely ensues (Hughes and Coakley, 1991) and very high levels of ‘attendance motivation’ (Nicholson, 1977) are generated. Moreover, we must consider the fact that athletes also develop high levels of intrinsic motivation to continue sports activity by any means simply because they love competing. Thus, hyperinclusion and high levels of intrinsic attendance motivation might result in a stronger willingness to compete hurt.

In addition, an athlete’s history of ill health may shape the willingness to compete hurt. Injury-prone athletes spend more time communicating with medical staff and are more experienced with decisions to play hurt and the potential negative health consequences of such decisions. Severe health challenges can also affect an athlete’s identity and body image. In particular, long-term injured athletes describe their experiences with rehabilitation processes as painful and highly personal, with the need to individually cope with the uncertainties about their future sports career, bodily capacities and the effects of stigma (Thing, 2004, 2006). This more elaborate health-related knowledge might be reflected in the willingness to compete hurt. Moreover, career length and age must be considered. Compared with younger athletes, athletes who are older and have had longer careers are more experienced in dealing with everyday health issues. They may also have gathered more knowledge about managing typical sport injuries and the potential consequences of such injuries simply by learning from their peers. Moreover, as studies among non-athlete populations have shown, age is a general factor related to presenteeism as well (Aronsson and Gustafsson, 2005). In terms of gender-specific socialization processes, there is general evidence that females are less prone to presenteeism than men (Patton and Johns, 2007). Men and women also vary in the way they perceive illnesses as legitimate reasons for absences (Harvey and Nicholson, 1999). Qualitative research about the resumption of non-professional female handball players’ sports careers supports these findings: female amateur players tend to be less eager to return to play after a serious anterior cruciate ligament (ACL) injury than professional male players from other sports (Thing, 2006). Nevertheless, previous studies about pain- and injury-related attitudes in sport also indicated that both female and male athletes tend to accept and normalize the risks of sports in a similar way (Nixon, 1996; Schnell et al., 2014; Young and White, 1995).

The main purpose of the article is to empirically analyse the willingness to compete hurt based on a large sample of elite athletes from a typical team and a typical individual sport (handball and track and field). In particular, we address the following research questions: first, how distinguished is the athlete’s willingness to compete hurt? Second, are there different types of athletes who are more or less willing to compete hurt? Third, what are the predictors of (potentially) higher or lower levels of willingness to compete hurt?

Materials and methods

Study design

This analysis is based upon data drawn from a larger multi-method study that addressed health management in Germany’s elite handball and track and field teams. For the quantitative section of the study, a cross-sectional survey was conducted; a questionnaire that covered health-related topics including subjective health, sports injuries, sense of coherence, healthy and risky behaviours and subjective concepts of health was administered. The questionnaire used socio-demographic variables and scales that were developed by the study to address general attitudes toward risk and the willingness to compete hurt. The study was funded by the Federal Institute of Sports Science, Germany, and was approved by the Institutional Review Board of the Faculty for Economics and Social Sciences, Tübingen University. All participants were informed in writing about the study’s aims and procedures, data confidentiality, the voluntary nature of participation and the use of anonymous data for analysis. Informed consent was implied through the act of completing and returning the questionnaire. All data obtained were anonymized before being received; therefore, the data were not linked with athletes’ personal information when entered into IBM SPSS Statistics 21.

Participants and procedure

Elite German male and female athletes were included in the study. Participants had to be (a) engaged in handball or track and field, (b) a member of one of the highest-level national squads (i.e. the A, B, C or D/C squads) and/or (c) a competitor in the first or second division of Germany’s professional handball league. The study was supported by the National Handball Federation, the National Handball League and the National Track and Field Federation. Due to sport-specific organizational differences, the questionnaire was distributed as follows. (1) An existing mailing list that included all of the squad members was used to contact track and field athletes by mail. In total, 328 of the 559 athletes completed and returned the questionnaire (58.50%). (2) For the handball players, a complete mailing list did not exist; thus, contact was made via their home teams: packages containing sealable questionnaires were sent to all 30 premier league teams and to the best 26 teams in the second division (either to the headquarters of the home teams or directly to team physicians). In total, 395 players from 43 different teams responded. We approximated a response rate of 50.38% based on an estimated 14 players on each of the 56 contacted teams. In comparison with common response rates in organizational research (Baruch, 1999; Baruch and Holtom, 2008), our response rates are significantly above average. The final study sample comprised 723 elite athletes (395 handball players and 328 track and field athletes) with an average age of 22.83 years (SD = 5.38). There were slightly more male athletes (59.6%) in our sample; 38.6% of respondents considered themselves professional athletes – that is, athletes who made a living from sport. The average career length in elite sports was 6.94 years (SD = 4.66).

Measures

Willingness to compete hurt scale

To assess the willingness to compete hurt we created a battery of questions that inquired regarding legitimate reasons for not competing. Based on theory, we developed 25 items that detailed typical health-related situations that could potentially result in the athlete having a lay-off. Our items encompassed different definitions of health to include the physical, psychological and social dimensions of health – for example, some questions were premised on health as the absence of disease, whereas others were premised on health as well-being. The items were worded as ‘I’-statements, for example: ‘I abstain from competing if… I have a feverish cold, or… if I am not feeling well’. The athletes were asked to assess the extent of their agreement with the statements on a five-point Likert scale ranging from ‘totally agree’ to ‘totally disagree’. The wording and intelligibility of the scale was pre-tested with several athletes and adjusted accordingly.

Variables with potential impact on the willingness to compete hurt

Different types of variables representing organizational context and personal factors were incorporated to identify predictors for the willingness to compete hurt; the statistical method used is described in more detail in the Data analysis section below. In addition to age and gender, we incorporated variables such as type of sport engaged in, performance level, professional status and career duration in elite sports. In addition, we constructed variables to assess the perceived pressure to compete hurt (direct social pressure, indirect social pressure and intrinsic pressure) and the subjective degree of inclusion in elite sports. Additionally, we asked athletes whether they had received medical treatment in the last season due to injury or illness. Furthermore, we assessed the number of necessary training and competition lay-offs lasting longer than one week that athletes had had in their careers to date due to injury.

Data analysis

The statistical analyses were computed using IBM SPSS 21 for Windows. Descriptive statistics, including means and standard deviations, were performed for the 25 items pertaining to the willingness to compete hurt. To identify athlete groups with a varying willingness to compete hurt, cluster analysis was applied on the basis of a preliminary exploratory factor analysis.

Factor analysis

Principal component analysis was performed using the varimax rotation method with Kaiser normalization (Bühl, 2012). The reliability of the willingness to compete hurt scale was high (Cronbach’s alpha = 0.909); thus, eliminating items would not have led to increased reliability. To minimise information loss, the explained variance was set to a minimum of 65%. Seven interpretable factors with eigenvalues of >0.9 were extracted based on the Kaiser criterion and the scree plot. Six items with factor loadings of <0.5 and/or loadings on two or more factors were eliminated before a second analysis was performed. This procedure finally produced seven interpretable factors that accounted for 71.8% of the variance. Unweighted factor mean values were then calculated to enable the content to be interpreted more effectively.

Cluster analysis

To identify different types of athletes, we conducted a cluster analysis that included these factors. Cluster analysis is a type of multivariate analysis that is used to find homogeneous subgroups within heterogeneous samples (Backhaus et al., 2006). In order to identify an appropriate number of clusters, we applied the hierarchical clustering method with Ward’s linkage and referred to the resulting dendogram (Milligan and Sokol, 1980). We then specified the identified clusters using partition-clustering with k-means (Bühl, 2012). The cluster analyses revealed two clusters that could be very well interpreted. The level of significance of the difference between the final clusters was assessed based on one-way analysis of variance (ANOVA) (p < 0.001).

Binary logistic regression analysis

To estimate which members’ specific attributes were influential in regard to the two clusters identified on the basis of the willingness to compete hurt, binary logistic regression analysis was carried out. Binary logistic regression analysis simultaneously identifies and quantifies the effects of predictor variables (either non-continuous or continuous) on a dichotomous dependent variable. Using this analysis, the probability of an event’s occurrence can be estimated (Menard, 2002). Possible correlations among predictor variables were examined before conducting the analysis. Because of the exploratory character of our study, we used forward stepwise selection for the covariates (Forward: LR).

Results

Willingness to compete hurt

Descriptive analysis of the items showed that very subtle distinctions are made between different kinds of health issues when it comes to withdrawing from competition (see Figure 2). Athletes were most likely to consider it legitimate to rest when their coaches order them to take a break or if they had a fever. By contrast, having aching muscles or having a cold without a fever appears to be absolutely illegitimate as reasons for a lay-off.

Figure 2.

Legitimate reasons for withdrawing from competition.

Factor analysis identified seven factors with an explained variance of 71.7%. Table 1 shows the identified factors and the factor mean values. It is obvious that athletes are especially willing to rest when their coaches, doctors or physiotherapists had issued orders to do so (m = 1.66; SD = 0.81). We also identify a low willingness to compete with the occurrence of feverish colds (m = 2.04; SD = 1.10) and traumatic muscle injuries (m = 2.24; SD = 1.05). A higher willingness to compete exists when there is the risk of worsening an injury or illness (m = 2.65; SD = 1.22) and when painkillers and antibiotics must be used (m = 2.78; SD = 1.17). The occurrence of severe pain (m = 3.62, SD = 1.16) and poor well-being (m = 4.13; SD = 0.81) are broadly rejected as reasons to withdraw from competition. In sum, a high willingness for presenteeism was observed for factors that are linked to poor subjective well-being and un-associated with specific medical diagnoses.

Table 1.

Factor loadings and means for legitimate reasons to withdraw from competition.

	Factor loadings (>0.5)							M (SD)
Factor 1: Not feeling well								4.13 (0.81)
Lack of drive	.812
Feeling mentally not resilient	.786
Feeling prone to getting a cold	.783
Lack of well-being	.774
Feeling injury-prone	.755
Feeling physically not resilient	.618
Factor 2: Orders from coach and medical staff								1.66 (0.75)
Coach’s order		.775
Doctor’s prescription		.772
Physiotherapist’s prescription		.755
Factor 3: Risk of deterioration of an injury or illness								2.65 (1.22)
Risk of deterioration of an injury			.833
Risk of deterioration of an illness			.832
Factor 4: Severe pain								3.62 (1.16)
Severe backache				.806
Severe headache				.803
Factor 5: Common cold								2.04 (1.10)
Bronchitis					.796
Cold with fever					.766
Factor 6: Traumatic muscle injury								2.24 (1.05)
Muscle fibre rupture						.841
Muscle strain						.779
Factor 7: Use of pain relief and antibiotic drugs								2.78 (1.17)
Use of antibiotics							.784
Use of painkillers							.697

Eliminated items: lack of energy (double loading); joint pain while moving (double loading); joint pain without moving (<0.5); aching muscles (double loading and <0.5); gastro-intestinal problems (double loading and <0.5); cold without fever (<0.5)

Differences in athletes’ willingness to compete hurt

To differentiate among athletes in regard to the willingness to compete hurt, we ran a cluster analysis on the basis of the described factors. Cluster analysis revealed two clusters of approximately the same size (see Table 2).

Table 2.

Willingness to compete hurt clusters.

	Cluster 1:	Cluster 2:
	Athletes conditionally willing to rest	Rest-averse and pain-trivializing athletes
Factors (sorted by factor means of Cluster 1)	M (SD)	M (SD)
F5_Common cold	1.40 (.59)	2.68 (1.1)
F2_Order of coach and medical staff	1.49 (.61)	1.83 (.83)
F6_Traumatic muscle injury	1.68 (.76)	2.79 (1.0)
F3_Risk of deterioration of injury or illness	1.86 (.81)	3.45 (1.0)
F7_Use of pain relief and antibiotic drugs	2.10 (.89)	3.48 (.99)
F4_Severe pain	2.93 (1.1)	4.31 (.77)
F1_Not feeling well	3.72 (.81)	4.53 (.57)

All factor means for cluster 1 and cluster 2 differ significantly (p < .001).

Cluster 1: Athletes conditionally willing to rest

The athletes within this first cluster (50.2% of all athletes) categorically refuse to compete with feverish colds. They indicate that they strictly follow the recommendations of their coaches, physicians and physiotherapists. For these athletes, traumatic muscle injuries and the risk of worsening an existing injury or illness are both important reasons not to compete. To some degree, this is also true in regard to needing to take painkillers and antibiotics and the occurrence of severe pain. However, among this group of athletes, withdrawing from a competition on account of not feeling well is not considered appropriate.

Cluster 2: Rest-averse and pain-trivializing athletes

Athletes within the second cluster (49.8% of all athletes) indicate that they would completely refuse to withdraw from competition because of severe pain or not feeling well. For them, withdrawing from a competition is absolutely illegitimate when a health complaint can be dealt with by taking painkillers and antibiotics. Such athletes also believe that it is illegitimate to withdraw from competition to avoid the risk of exacerbating an injury or illness. Moreover, these athletes also indicate that they would compete with a feverish cold or traumatic muscle injuries. Among these athletes, the only acceptable reason not to compete is if a coach, doctor or physiotherapist orders rest.

Determinants of athletes’ willingness to compete hurt

Finally, we determined which contextual and personal factors influence membership in one of the two identified clusters. Table 3 summarizes the results from the binary logistic regression analysis with cluster membership as the dependent variable. The percentage of correctly classified cases was 78.8% and the model’s explanatory power was determined to be good (Nagelkerke R² = 0.467). The Lemeshow GoF-test did not yield significant results, indicating good fit. Correlation analyses showed that the independent variables included in the model were broadly independent.

Table 3.

Estimated parameters in binary logistic regression analysis for willingness to compete hurt clusters (n = 637).

		Wald	OR	95% CI	p
Type of sport		124.194	10.335	6.9–15.6	<.001**
Indirectly perceived social pressure to play hurt		13.273	1.328	1.1–1.5	<.001**
Directly perceived social pressure to play hurt		10.644	1.312	1.1–1.5	.001**
Performance level		2.165	.491	.3–.9	.023*
Intrinsic pressure to play hurt		5.223	1.215	1.0–1.4	.022*
Sex					NS
Age					NS
Professional athlete					NS
Career length in elite sports					NS
Degree of inclusion in elite sports					NS
Medical treatment in last season due to injuries					NS
Medical treatment in last season due to illnesses					NS
Training and competition lay-offs >one week due to traumatic and overuse injuries in the athlete’s career					NS
Nagelkerke R²	.467
Hosmer and Lemeshow GoF-test	NS
Percentage correctly classified (%)	78.8

n = 637, NS = not significant, * significant at p < 0.05, ** significant at p < 0.01; OR = odds ratio, CI = confidence interval 95%.

The regression analysis demonstrates that the type of sport, perceiving indirect or direct social pressure to play hurt, performance level and intrinsic pressure to play hurt were relatively strong predictors for membership in one of the two clusters. Thus, handball players are 10 times more likely than track and field athletes in the rest-averse cluster (Wald 124.194, OR 10.335; p < 0.001). Furthermore, athletes who perceive direct (Wald 10.644, OR 1.312; p = 0.001) and indirect social pressure (Wald 13.273, OR 1.328; p < 0.001) to play hurt are more likely in the rest-averse cluster. Athletes who feel a high intrinsic pressure to compete despite health problems are also more likely to be found in the rest-averse cluster (Wald 5.223, OR 1.215; p < 0.05). By contrast, not competing at the top international level increases the probability of an athlete to belong to the willing-to-rest cluster (Wald 2.165, OR 0.491; p < 0.05). Gender and age have no influence on cluster membership. Moreover, career length, professional status, subjective degree of inclusion in elite sports, and injury and illness experiences are also no significant predictors of cluster membership.

Discussion

The main purpose of the study was to analyse the willingness to compete hurt among elite athletes based on a large sample of elite German handball and track and field athletes. Compared with previous studies, our sample of more than 700 German elite athletes from a typical team and a typical individual sport provides more representative results within a research field dominated by qualitative methodology. Applying the concept of presenteeism as a framework to analyse the social practice of playing while being hurt added a new perspective to sociological research about pain and injury in sports.

According to our descriptive analyses, very subtle distinctions are made between different kinds of health problems when it comes to withdrawing from competition. This observation is in line with research that has been conducted regarding the legitimacy of various minor health conditions in normal work contexts (Harvey and Nicholson, 1999). The findings support our theoretical assumption that sport-specific environmental expectations play a central role regarding the presence and absence legitimacy in the sports workplace. Breaks are only legitimate when injuries or illnesses can lead to reduced functionality or strong long-term health risks. Furthermore, our descriptive findings support previous observations in professional rugby that the process of professionalization reduces athletes’ acceptance of competing with severe injuries but does not reduce the willingness to compete in pain (Howe, 2001; Malcolm and Sheard, 2002). We also observe a relatively high willingness to compete while medicated, such as with painkillers. This finding is also in line with results from qualitative case studies (Murphy and Waddington, 2007; Thiel et al., 2010). In practice, one very important finding is that the recommendations of coaches and medical staff play a relevant role in the decision to compete hurt or to take a break. However, as we know from qualitative analysis, coaches and medical staff often tend to ascribe responsibility for the final decision (and for the negative consequences of that decision) solely to the athlete (Mayer, 2010). Letting the athlete decide, in turn, increases the likelihood of presenteeism.

Although all athletes are fundamentally willing to compete hurt, our cluster analysis revealed that in regard to the degree of risk acceptance, there are two types of athletes. The ‘rest-averse and pain-trivialising’ and ‘conditionally willing to rest’ clusters differ primarily in regard to the risk of developing severe health problems through competing hurt. Compared to athletes in the other cluster, the rest-averse and pain-trivialising athletes were far more willing to accept potential health risks in favour of competing. Binary regression analysis showed that the type of sport, performance level and perceived climate of pressure within the team are the dominant factors explaining cluster membership. The sports discipline, particularly its formal rules, informal regulations and performance requirements, has a central relevance for the perceived absence/presence legitimacy. Due to the fact that elite handball teams generally have more than one top player per position, coaches can easily sanction players who are not committed enough to compete despite being in pain. Thus, it can be argued that team sport coaches have more structural power than individual sports coaches because they are dependent on individual athletes to a lesser degree. Furthermore, sports disciplines have distinctive physical performance requirements that allow competing despite relatively severe health problems to varying degrees. A handball player who has a feverish cold can still contribute to team performance if he or she scores one important goal as a substitute. By contrast, for most track and field athletes, having a feverish cold directly limits the ability to perform.

Our findings correspond with previous observations regarding differences between team and individual sports in college sports (Nixon, 1994b). A recent study of elite adolescent athletes (Schnell et al., 2014) revealed that the general culture of risk in elite sport is differently translated into specific risk cultures depending on the sports discipline. Against this background, our findings suggest the existence of discipline-specific presence and absence cultures. These seem to be both entangled with prevailing risk cultures in the different kinds of sports and with the general culture of normalizing and accepting health risks in sport. However, our empirical findings also reveal that the sports’ discipline-specific culture is not the only factor that explains the extent of the athlete’s willingness to compete hurt.

Moreover, the finding that social pressure has a strong influence on the willingness to compete hurt supports both Nixon’s (1992) network-theoretical assumptions and qualitative findings obtained from professional soccer and rugby clubs (Howe, 2004; Roderick, 2006). As social pressure is to a relevant degree generated by the expectations expressed by coaches or other team members, it has to be assumed that communication structures in sports networks play a crucial role in shaping the athletes’ attitudes towards presenteeism-related decision-making. The relevance of strong intrinsic pressure highlights the role of the experience of the joy of competing in and practicing sports. This is in line with qualitative research addressing injury experiences of competitive athletes (Thing, 2006). Nevertheless, it cannot be ruled out that the intrinsic pressure to compete hurt or ill results from general psychic barriers to rest, such as that caused by exercise addiction (Freimuth et al., 2011).

The observation that the most elite top-level athletes are more likely in the rest-averse and pain-trivialising cluster could be explained by the higher risk of losing money or public recognition as a result of missing a competition. In particular, in regard to important international competitions, the socially shared acceptance of severe negative health consequences can be expected to be much higher than usual.

Contrasting with health risk and presenteeism research in general, this study found that gender has no significant influence on the degree of willingness to compete hurt. However, this finding is in line with findings in previous studies that have dealt with pain- and injury-related attitudes in sport. Although many of these attitudes have been interpreted as stemming from stereotypically male perceptions of risk, pain and injury (Messner, 1990), female athletes seem to accept and normalize the risks of sports in a similar way to male athletes (Nixon, 1996; Schnell et al., 2014), even in recreational sports (Weinberg et al., 2013). This phenomenon can be explained by the similar manner in which people of both genders experience sport-specific socialization (Young and White, 1995; Young et al., 1994). All athletes engage in the sports system under the same general principles and basic structure; thus, we surmise that general performance-oriented expectations in competitive sports over-influence potential gender-specific attitudes.

Cluster membership was also not affected by age, career length or the degree of professionalization. We assume that socialization into the culture of risk in elite sports has already been completed when athletes reach the top level. This could also be the reason for the fact that among these athletes, the degree of subjective inclusion in elite sports had no influence on the willingness to compete hurt. Athletes at an elite level already have had to invest so much into their sport that it necessarily predominates in their personal lives. Differences among the athletes in the subjective relevance of elite sport were therefore attitude-related differences of no practical import. By contrast, among adolescent elite athletes, the degree of inclusion in the elite sports system correlates positively with risk acceptance (Schnell et al., 2014).

The finding that the professional status of an athlete did not affect cluster membership can be explained by the characteristics of professional sports in Germany. Although payment regulations differ between the two disciplines,¹ all (full or semi) professional athletes get paid during shorter health-related absences, and are covered by their mandatory health insurance. Surprisingly, athletes’ individual histories of ill health do not influence cluster membership either. This suggests that even athletes who are less illness- and injury-prone have the ‘right’ attitudes towards pain and injury (Roderick, 2006) and have an understanding of ‘legitimate’ reasons for lay-offs.

Conclusions

Analysing presenteeism and absenteeism among elite athletes contributes to our understanding of how health is handled in a highly competitive social context. In conclusion, the results of our analysis highlight the fact that, in elite sports, a generally high willingness to compete hurt is a given among both male and female athletes. However, the fact that some athletes are highly willing to risk their long-term health is a challenge for medical staff and coaches. In particular, excessive social pressure within the sports network can be targeted by establishing athlete-centred communication strategies. In this regard, coaches can play a crucial role in preventing rest-averse and pain-trivialising athletes from competing despite severe injuries.

Our study has several limitations. First, our results may not be representative for the entire elite sports system. The sample only consists of elite athletes drawn from two sports disciplines, although it includes a high percentage of all German elite athletes in this field. Second, the study has the typical limitations of cross-sectional survey studies that rely on self-reported variables. Third, it is not clear at which point the increased risk for developing long-term health problems begins. Moreover, a certain amount of willingness for presenteeism may be – even from the perspective of the long term – highly instrumental in maintaining a successful sports career. For example, if competitions are decisive for an athlete’s career, contract negotiations or decisions about squad memberships lie ahead, being too careful could lead to a career slump or even to exclusion from the elite sports system altogether.

For future research, it is therefore recommended that longitudinal studies be conducted in order to analyse the willingness to compete hurt in relation to long-term health development. To better understand the influence of factors pertaining to social structure, such longitudinal studies should include athletes of different cultures, sports disciplines and performance levels. Furthermore, future studies should thoroughly address the impact of adjustment latitude, absence regulations and discipline-specific absence cultures on playing hurt and return-to-play-decisions – using both qualitative and quantitative approaches. In this regard, the introduced dynamic model about presenteeism in elite sports offers a general framework to guide and integrate future research in this field.

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The study was funded by the Federal Institute of Sports Science (BISp) as part of the German Ministry of the Interior. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Notes

References

Addae

Johns

Boies

(2013) The legitimacy of absenteeism from work: A nine nation exploratory study. Cross Cultural Management: An International Journal 20: 402–428.

Aronsson

Gustafsson

(2005) Attendance presenteeism: Prevalence, attendance-pressure factors, and an outline of a model for research. Journal of Occupational and Environmental Medicine 47: 958–966.

Aronsson

Gustafsson

Dallner

(2000) Sick but yet at work: An empirical study of sickness presenteeism. Journal of Epidemiology and Community Health 54: 502–509.

Backhaus

Erichson

Plinke

et al . (2006) Multivariate Analysemethoden [Multivariate Methods of Analysis]. Berlin: Springer.

Bamberger

Biron

(2007) Group norms and excessive absenteeism: The role of peer referent others. Organizational Behavior and Human Decision Processes 103: 179–196.

Baruch

(1999) Response rate in academic studies – A comparative analysis. Human Relations 52: 421–438.

Baruch

Holtom

(2008) Survey response rate levels and trends in organizational research. Human Relations 61: 1139–1160.

Bühl

(2012) SPSS 20: Einführung in die moderne Datenanalyse [SPSS 20: Introduction to Modern Data Analysis]. München: Pearson.

Curry

(1993) “A little pain never hurt anyone”: Athletic career socialization and the normalization of sports injury. Symbolic Interaction 16: 237–290.

10.

Dew

Keefe

Small

(2005) ‘Choosing’ to work when sick: Workplace presenteeism. Social Science & Medicine 60: 2273–2282.

11.

Donnelly

(2004) Sport and risk culture. In: Young

(ed.) Sporting Bodies, Damaged Selves: Sociological Studies of Sports-Related Injury. Oxford: Elsevier Ltd, pp. 29–58.

12.

Freimuth

Moniz

Kim

(2011) Clarifying exercise addiction: Differential diagnosis, co-occurring disorders, and phases of addiction. International Journal of Environmental Research and Public Health 8: 4069–4081.

13.

Hansen

Andersen

(2008) Going ill to work – What personal circumstances, attitudes and work-related factors are associated with sickness presenteeism? Social Science & Medicine 67: 956–964.

14.

Harvey

Nicholson

(1999) Minor illness as a legitimate reason for absence. Journal of Organizational Behavior 20: 979–993.

15.

Howe

(2001) An ethnography of pain and injury in professional rugby union. International Review for the Sociology of Sport 36: 289–303.

16.

Howe

(2004) Sport, Professionalism and Pain: Ethnographies of Injury and Risk. London; New York: Routledge.

17.

Hughes

Coakley

(1991) Positive deviance among athletes – The implications of overconformity to the sport ethic. Sociology of Sport Journal 8: 307–325.

18.

Johns

(2010) Presenteeism in the workplace: A review and research agenda. Journal of Organizational Behavior 31: 519–542.

19.

Johns

Nicholson

(1982) The meaning of absence: New strategies for theory and research. Research in Organizational Behavior 4: 127–172.

20.

Kujala

Orava

Parkkari

et al . (2003) Sports career-related musculoskeletal injuries: Long-term health effects on former athletes. Sports Medicine 33: 869–875.

21.

Loland

Skirstad

Waddington

(2006) Pain and Injury in Sport: Social and Ethical Analysis. London; New York: Routledge.

22.

Malcolm

(2006) Unprofessional practice? The status and power of sport physicians. Sociology of Sport Journal 23: 376–395.

23.

Malcolm

(2009) Medical uncertainty and clinician–athlete relations: The management of concussion injuries in rugby union. Sociology of Sport Journal 26: 191–210.

24.

Malcolm

Safai

(2012) The social science of sports medicine. In: Malcolm

Safai

(eds) The Social Organization of Sports Medicine. New York: Routledge, pp. 1–22.

25.

Malcolm

Sheard

(2002) “Pain in the assets”: The effects of commercialization and professionalization on the management of injury in English rugby union. Sociology of Sport Journal 19: 149–169.

26.

Mayer

(2010) Verletzungsmanagement im Spitzensport [Injury Management in Elite Sports]. Hamburg: Czwalina.

27.

Menard

(2002) Applied Logistic Regression Analysis. Thousand Oaks, CA: Sage.

28.

Messner

(1990) When bodies are weapons: Masculinity and violence in sport. International Review for the Sociology of Sport 25: 203–220.

29.

Milligan

Sokol

(1980) A two-stage clustering algorithm with robust recovery characteristics. Educational and Psychological Measurement 40: 755–759.

30.

Munir

Yarker

Haslam

et al . (2007) Work factors related to psychological and health-related distress among employees with chronic illnesses. Journal of Occupational Rehabilitation 17: 259–277.

31.

Murphy

Waddington

(2007) Are elite athletes exploited? Sport in Society 10: 239–255.

32.

Nicholson

(1977) Absence behaviour and attendance motivation: A conceptual synthesis. Journal of Management Studies 14: 231–252.

33.

Nixon

(1992) A social network analysis of influences on athletes to play with pain and injury. Journal of Sport & Social Issues 16: 127–135.

34.

Nixon

(1993) Accepting the risks of pain and injury in sport: Mediated cultural influences on playing hurt. Sociology of Sport Journal 10: 183–196.

35.

Nixon

(1994a) Coaches’ views of risk, pain, and injury in sport, with special reference to gender differences. Sociology of Sport Journal 11: 79–87.

36.

Nixon

(1994b) Social pressure, social support and help seeking for pain and injuries in college sport networks. Journal of Sport & Social Issues 18: 340–355.

37.

Nixon

(1996) Explaining pain and injury attitudes and experiences in sports in terms of gender, race, and sports status factors. Journal of Sport & Social Issues 20: 33–44.

38.

Patton

Johns

(2007) Women’s absenteeism in the popular press: Evidence for a gender-specific absence culture. Human Relations 60: 1579–1612.

39.

Pike

ECJ

Maguire

(2003) Injury in women’s sport: Classifying key elements of ‘risk encounters’. Sociology of Sport Journal 20: 232–251.

40.

Pinheiro

Pimenta

Resende

et al . (2014) Gymnastics and child abuse: An analysis of former international Portuguese female artistic gymnasts. Sport, Education and Society 19: 435–450.

41.

Pringle

Markula

(2005) “No pain is sane after all”: A Foucauldian analysis of masculinities and men’s experiences in rugby. Sociology of Sport Journal 22: 472–497.

42.

Riedl

Borggrefe

Cachay

(2010) Elite sports versus university studies? Organizational changes and networking as structural solutions to the inclusion problem facing elite student athletes in Germany. In: Wagner

Storm

Hoberman

(eds) Observing Sport: Modern System Theoretical Approaches. Schorndorf: Hofmann, pp. 163–196.

43.

Roderick

(2006) Adding insult to injury: Workplace injury in English professional football. Sociology of Health & Illness 28: 76–97.

44.

Roderick

Waddington

Parker

(2000) “Playing hurt”: Managing injuries in English professional football. International Review for the Sociology of Sport 35: 165–180.

45.

Safai

(2003) Healing the body in the “culture of risk”: Examining the negotiation of treatment between sport medicine clinicians and injured athletes in Canadian intercollegiate sport. Sociology of Sport Journal 20: 127–146.

46.

Schnell

Mayer

Diehl

et al . (2014) Giving everything for athletic success! – Sports-specific risk acceptance of elite adolescent athletes. Psychology of Sport and Exercise 15: 165–172.

47.

Schubring

Thiel

(2014) Coping with growth in adolescent elite sport. Sociology of Sport Journal 31: 304–326.

48.

Theberge

(2008) “Just a normal bad part of what I do”: Elite athletes’ accounts of the relationship between health and sport. Sociology of Sport Journal 25: 206–222.

49.

Thiel

Mayer

Digel

(2010) Gesundheit im Spitzensport. Ein sozialwissenschaftliche Analyse [Health in Elite Sports. An Analysis from a Social Sciences Perspective]. Schorndorf: Hofmann.

50.

Thing

(2004) Scars on the body: The risk management and self-care of injured female handball players in Denmark. In: Young

(ed.) Sporting Bodies, Damaged Selves: Sociological Studies of Sports-Related Injury. Oxford: Elsevier, pp. 195–210.

51.

Thing

(2006) “Voices of the broken body”. The resumption of non-professional female players’ sports careers after anterior cruciate ligament injury. The female player’s dilemma: Is she willing to run the risk? Scandinavian Journal of Medicine & Science in Sports 16: 364–375.

52.

Waddington

(2000) Sport, Health and Drugs: A Critical Sociological Perspective. London; New York: E & FN Spon.

53.

Waddington

(2006) Ethical problems in the medical management of sports injuries: A case study of English professional football. In: Loland

Skirstad

Waddington

(eds) Pain and Injury in Sport: Social and Ethical Analysis. London; New York: Routledge, pp. 182–199.

54.

Weinberg

Vernau

Horn

(2013) Playing through pain and injury: Psychosocial considerations. Journal of Clinical Sport Psychology 7: 41–59.

55.

Young

(1993) Violence, risk and liability in male sports culture. Sociology of Sport Journal 10: 373–396.

56.

Young

(2004) Sporting Bodies, Damaged Selves: Sociological Studies of Sports-Related Injury. Oxford: Elsevier.

57.

Young

White

(1995) Sport, physical danger and injury: The experiences of elite women athletes. Journal of Sport & Social Issues 19: 45–61.

58.

Young

McTeer

White

(1994) Body talk: Male athletes reflect on sport, injury and pain. Sociology of Sport Journal 11: 175–194.