It is a well known fact that any sort of sports or physical activity comes with a certain risk of injury.1 With the current sports and fitness hype and the active government promotion of an active lifestyle we can consequently also expect an increased number of sports injuries.2 Fortunately, most are not life threatening, and it goes without saying that the health benefits of sports by far outweigh any of the risks involved.3 Nevertheless, we cannot ignore the fact that sometimes sports injuries do cause physical disability beyond pain or other physical discomforts, which in turn results in use of healthcare resources and possibly absenteeism from work.

It is surprising how little we know about the costs involved with sports injuries. The costs of knee injuries and knee surgery as a result of sports participation have been documented and Verhagen et al (2005) have evaluated the economic benefits of a proprioceptive balance board training programme for the prevention of ankle sprains in volleyball.4^–6 Hoy and Lindblad (1992, 1994) performed an epidemiological and socioeconomic study in badminton, European team handball and outdoor football, and there is also some information to be found on costs of occupational injuries in the USA.7^–14

Socioeconomic evaluation aims at aiding policy makers in deciding whether or not to implement or fund new healthcare measures.15 To ensure that the decision maker is supplied with all the relevant information, it is generally recommended to go with the societal perspective for any economic evaluations involved.15 This means that all relevant costs and effects need to be incorporated in the analysis, regardless of who bears the costs and who experiences the effects.15

Although injury has been recognized to be a major public health problem, data on costs are still in short supply. The lack of complete and accurate data limits the ability to focus on a variety of data-oriented aspects of injury control that potentially could reduce the incidence of sports injuries as well as healthcare costs and loss of productivity associated with them. Because of the scarcity of studies concerning the socioeconomic consequences of sports injuries and given their potential importance to determine health benefit costs of prevention strategies, our intention was to analyse all costs incurred within sports injuries occurring in Flanders, be it directly or indirectly.

METHODS

Subjects

In Flanders there are 121 sports federations.16 The term "sports federation" applies to every organisation with a private law that unites, supports, informs and represents all sports clubs within one sports discipline, this in a voluntary and generous manner.17 All sports federations that did not fit the definition of sports and were not on the list of sports disciplines as formulated by the Flemish government were excluded (n = 24). Ten of the listed Flemish sports federations were not included in the study because they were organisations that supported the sports federations without any direct relation to sports participation.17 18 Five Flemish sports federations were already part of one of the coordinating sports federations. A total of 82 sports federations were eligible for analysis.

Study design

A retrospective cohort study was set up to determine estimated injury risks and accompanying socioeconomic costs of sports injuries in Flanders.

Data collection

Injury definition

The sports federations are obligated to insure all members for sports accidents. A sports injury is defined as physical damage to the employees or sports participants due to an accident that is caused during and as a result of sports activities organized by the sports federation or sports club. Traffic accidents occurring during transport to and from an organized sports activity are covered by the insurance companies but are not included in this study. As a result this study only concerns claims to insurance companies for reimbursement of the involved costs for acute injuries. All of these injuries required medical attention; however, not all injuries requiring medical attention are claimed.

Estimated injury rate

Information on sports injuries and the rate of those injuries was obtained through retrospective epidemiological cohort designs based on insurance statistics. The actual rate of sports injury was determined by matching the total number of members against the total number of insurance claims in 2003 for all participating Flemish sports federations. The sports federations provided us with the total number of members and the injury statistics, which they in turn had obtained from the insurance companies. The estimated injury rate is expressed as a percentage and defined as the total number of injuries per 100 participants per year.

Injury characteristics

Two insurance companies cover the sports injuries of the majority of the Flemish sports federations. Sports injuries were included if they were acute in nature and if they resulted in a claim with one of the above-mentioned insurance companies. For 14 Flemish sports federations information was gathered concerning diagnosis, localisation, type, causes and severity of the injury. This was established by analysing all insurance claims in 2003 using an encoded list.

Costs

One insurance company was able to provide detailed information on the direct medical costs for each insurance claim for 4 of the 14 analysed sports federations. The direct medical costs consisted of all costs involved in rehabilitation, medical care, hospitalisation, medication, bandages, transport and crutches/dental prosthesis and were split into costs paid by the health insurance institution of the social security system (Rijksinstituut voor Ziekte- en Invaliditeitsverzekering (RIZIV)) and costs paid by the insurance company or the injured athlete. To determine the direct medical costs involved in sports injuries in Flanders, sports disciplines were classified according to the data available (table 1). The human capital method was applied to determine the indirect costs of sports injuries in Flanders. The absenteeism from work was expressed in days of absence and multiplied by the socioeconomic cost inherent to production loss. The most recent calculations of Securex reveal that 1-day absence from work results in a socioeconomic cost of €200.19

Calculations

Direct medical cost for sports in category A

The direct costs for the category A sports were provided directly by an insurance company. The cost per injury localisation was calculated in order to be able to determine the total direct medical cost for all other possible sports disciplines. Since there are more injuries than claims, the calculation based on the injury localisations would reveal an overestimation of the costs. This means that when we add up the costs for all the injury localisations, the sum will be higher than the total direct medical cost for all claims. To avoid this overestimation of the medical cost in other sports disciplines, a correction factor was introduced. To explain the use of the correction factor we present an example of a cycling injury. Owing to a fall a cyclist sustains a head and upper arm injury. The insurance company only provides one cost for all the injuries sustained within one insurance claim. Since the cyclist has been injured at two localisations, the cost is recorded for both the head and upper arm injury. This results in an overestimation of the total direct medical cost when only taking into account the insurance claims. To avoid this overestimation the implementation of a correction factor is necessary.

Direct medical cost for sports in category B

Table 1 shows how the different injury localisations are spread over the category B sports disciplines. For the category A sports we know the average cost per injury localisation. In order to calculate the total medical cost for the category B sports disciplines, we took the sum of the costs per injury localisation and multiplied it by the correction factor we obtained from the category A sports.

Direct medical cost for sports in category C

In cases where in category C we found a sport similar to those occurring in categories A and B, e.g. aikido, we used the average of every calculated prevalence per injury localisation for all combat sports/martial arts out of categories A and B. When this was not the case we used the average of every calculated prevalence per injury localisation for all sports out of categories A and B. The direct medical cost was consequently calculated in the same way as for the category B sports disciplines.

Indirect cost

We know the absenteeism numbers for the sports disciplines of categories A and B. For the disciplines of category C the absenteeism was estimated based on the ratio of accidents reported versus the days of absence from work reported for categories A and B.

RESULTS

Subjects

Table 2 shows the selection tree and the total number of participating Flemish sports federations (n = 72). Eleven Flemish sports federations refused to participate in this study because they were not willing to provide the total number of members or the number of sports injuries claimed to the insurance companies in 2003.

Sports injuries

In table 3 the estimated injury rates (95% confidence interval (CI)) are presented for the sports disciplines of categories A and B. The highest rate of injury is found in European team handball (8.96%; 95% CI 8.95–8.97), basketball (7.45%; 95% CI 7.45–7.45) and outdoor football (7.42%; 95% CI 7.42–7.42). Swimming sports (0.62%; 95% CI 0.62–0.62), tennis (0.91%; 95% CI 0.91–0.91) and karate (1.36%; 95% CI 1.36–1.36) show the lowest injury rates. For the remaining 58 sports disciplines, rugby (13.1%; 95% CI 13.1–13.1), korfball (8.94%; 95% CI 8.93–8.95) and baseball/softball (8.75%; 95% CI 8.72–8.78) showed the highest injury rates (data not shown).

Direct medical costs

The total medical cost of sports injuries as claimed to insurance companies is €4 228 970 on 23 432 sports injuries. The social security system takes care of the bulk of this cost, being 54.0%. €1 143 891 (27.1%) is covered by the insurance companies and €801 812 (18.9%) is paid by the injured athletes themselves. From previous research we know that the total number of sports participants in Flanders is 3 620 190 with an associated injury risk of 2.3% of insurance-claimed injuries. This means that in 2003 the total medical cost of all insurance-claimed injuries occurring in Flanders was €15 027 423. This reveals that an average of €180 direct medical costs are involved per acute insurance-claimed injury, of which €97 is paid by the social security system, €49 by the insurance company and €34 by the injured athlete. The total direct medical and indirect costs are shown in table 4.

Table 5 shows the direct medical cost for each injury localisation presented. The highest total direct medical cost is seen for ACL injuries (€1358 per injury) followed by knee (€527 per injury), tooth (€465 per injury), upper arm (€364 per injury) and finger injuries (€315 per injury). Ankle sprains stand for a total medical cost of €197 per injury. Lowest direct medical costs are observed for foot, toe, abdominal, wrist and thorax injuries with a total cost of €52, €61, €69, €72 and €77 per injury, respectively.

Indirect costs

The indirect costs are presented in table 4. For the sports federations analysed the total indirect cost is €31 355 698. This means an indirect cost of €1338 per injured athlete. Taking into account the sports participation level in Flanders, the estimated indirect cost consequently amounts to €111 420 813.

DISCUSSION

Estimates of the incidence of and costs associated with sports injuries provide an objective basis for assessing the need for health and safety measures in the sports environment. Until now no one has attempted to estimate and publish the total socioeconomic cost of sports injuries in a certain region or country. This is the reason why this current investigation aims to study healthcare costs, both direct medical and indirect costs, for injuries related to sports participation.

In this paper we combined a variety of data sources and from that derived an estimation of the total cost involved in sports injuries. The numbers of members and of insurance-claimed injuries were provided by the Flemish sports federations, which enabled us to calculate the estimated injury rates. Data concerning the epidemiology were derived from injury statistics of 2003 put at our disposal by two insurance companies. From one insurance company we also obtained a detailed overview of the direct medical costs involved in each injury. The sports participation level of the Steunpunt Re-Creatief Vlaanderen and the number of Flemish inhabitants were used to determine the total number of sports participants in Flanders.20 21 The most recent calculation of Securex revealed that the cost involved in production loss due to absenteeism is €200 per day.19

The total direct medical cost for injuries in Flanders taking into account all the sports participants is €15 037 423. €8 113 468 is covered by the social security system, €4 064 757 by the insurance company and €2 849 198 by the injured athletes themselves. The total expenditure by the social security system for healthcare in Belgium is €18 billion on a yearly basis.22 From this amount approximately 60% goes to Flanders, which results in an expenditure of €10 to 11 billion.22 This means that 0.07−0.08% of the healthcare budget is spent on sports injuries, severe enough to claim to insurance companies. Before interpreting this result, one has to take into account the restrictions of this study, as mentioned below. In The Netherlands the total cost for healthcare associated with sports injuries is €140 million compared to €744 million spent on health problems caused by an inactive lifestyle.23 This reveals that there is a net positive amount of €604 million, which is a direct result of sports participation or physical activity.23 The total amount of medical costs is lower in Flanders due to the lower number of inhabitants compared to The Netherlands and the inclusion of overuse injuries in the study of Proper et al. (2004).23 In this latter study the average cost per injury observed is €100 compared to €180 in Flanders.23 The higher cost seen in this investigation is probably the result of the origin of the injury data and the definition as formulated by the insurance companies. In Belgium, only one study estimated costs for other health disorders.24 The prevalence of brain disorders in Belgium in 2004 was 2.9 million and amounted to a total cost of €10.6 billion.24 The total direct medical cost for brain disorders amounted to €3.1 billion and the direct non-medical costs to €2.1 billion, which resulted in a direct medical cost of €1069 per patient.24

The direct medical cost of an ankle sprain calculated by Verhagen et al. (2005) was €56.77.6 The fact that we only take those ankle sprains severe enough to result in an insurance claim into consideration most likely accounts for the higher cost in our study as they are more severe than those reported by Verhagen et al. (2005).6 De Loës (2000)4 determined the direct medical costs involved in knee injuries for which medical treatment was provided. The cost for ACL/PCL ruptures was US$ 2773.50 or €2147, while the cost for all knee injuries was US$1114.00 or €862.4 These results are comparable to the cost of ACL injuries, which is €2062 (only those involved in costs), and the cost of all knee injuries together, which is higher in this study (€1098; only those involved in costs).

Hoy and Lindblad (1992, 1994) 7^–9 reported the monetary loss, which is the loss of income/wages and medical expenses for the injured individuals. Separate data on the direct medical costs were not mentioned, which makes comparison with our study unfeasible.

The inclusion of indirect costs of disease or injury in economic appraisal is subject to considerable debate and scepticism.25 The disagreement on the relevance of indirect costs resulting from reduced productivity can be illustrated by considering recent guidelines for economic evaluations in healthcare that have been issued by Australian and Canadian (Ontario) health authorities in the context of decisions on reimbursement of pharmaceutical products.25 In our study we used the human capital approach to estimate the value of potentially lost production as a consequence of sports injuries. Many authors have suggested that the real production losses for society may be much smaller because in cases of short-term absenteeism, work may be taken over by others and non-urgent work may be cancelled or made up for by the injured employee upon his/her return to work.25 For long-term absences, work can be taken over by someone drawn from the rank or by reallocating employees over jobs.25 This is why Koopmanschap et al.25 introduced the friction cost method. The essence of this method is that absent workers will be replaced after an adaptation period (the friction period) and in this way further production losses may subsequently be prevented.15 In this study, however, we have discussed productivity losses with reference to the number of days absent from work. The estimated indirect cost in Flanders due to sports injuries is €111 420 813. An indirect cost of €6.6 billion is estimated to be a direct result of absenteeism in Belgium.19 If we assume that half of that cost is produced in Flanders, the indirect cost is €3.3 billion. The report of Securex revealed that the cost that is down to absenteeism is 20% lower in Flanders than in the Walloon provinces in Belgium.19 A percentage of 3.4% of the cost caused by production loss in Flanders is attributed to sports injuries. The estimated indirect cost for brain disorders in Belgium was €5.3 billion or €1827 per patient.24

The injury rate and cost data for sports injuries in Flanders presented in this paper are conservative estimates based on available data and methodologies. The results here are subject to some restrictions. This paper only shows the costs involved in sports participation and does not mention the benefits. When we extrapolate the cost for all Flemish sports participants we assume that all sports participants are susceptible to the same injury risk, which was calculated based on all the members and all the insurance claims together. In this extrapolation a sports participant has been considered to participate in only one sport. We know that in real life situations most participants are active in more than one sport, revealing an injury rate that is the sum of the rates of all sports he or she participates in. The data on the epidemiology and costs only takes into account sports injuries severe enough to result in claims to insurance companies and, consequently, excludes costs arising from overuse injuries as well as from less severe acute injuries. From previous investigations we know that these insurance-claimed injuries make up to 34% of the total number of injuries and 71% of the total number of acute injuries resulting in a sports inactivity of more than 1 week.26 Of the total number of injuries, 52% are overuse injuries.26 Owing to the fact that not all sports injuries are claimed to insurance companies the calculation presented here is an underestimation of the total cost. In the literature, it was never reported what the average medical costs are involved in overuse injuries and we question if the remaining acute injuries even cause a medical cost since they were not claimed to the insurance companies for reimbursement. As a result an exact estimation of the direct medical and indirect costs could not be provided for all sports injuries occurring in Flanders on a yearly basis. The costs of sports disciplines of category A were exactly calculated. Costs involved in sports disciplines of category B were established by determining the cost per injury localisation and taking into account a correction factor derived from category A. For sports disciplines of category C, the injury risk was known and the costs were estimated using the average percentage for each injury localisation after which the estimation was refined by incorporating a correction factor. The indirect costs were obtained using the human capital method. The study of Hutubessy et al.27 showed that the indirect costs as a result of absenteeism in the Netherlands caused by back pain was US$3104 million using the human capital method compared to US$1504 million using the friction cost method. In general, as the friction cost method assumes production loss to be limited to the short term, indirect cost for injuries that cause disability are much lower if based on the friction cost method than if they are based on the human capital method.27 For injuries entailing short-term absence, the difference is much smaller.25

We are aware of the fact that this calculation is associated with restrictions and that it is an approximate estimation. However, since only very few of the studies published have attempted to make this estimation on a particular cost (eg, knee injuries, etc), we feel justified in using conventional methods.

CONCLUSION

Costs have become critical statistics in medical care debates. From this study it can be concluded that the direct medical costs and the indirect costs involved in acute insurance-claimed injuries in Flanders are €15 027 423 and €111 420 813, respectively. Injuries and their inevitable costs are the negative side effects of sports participation. Although we know that the costs involved in sports injuries exist, the benefits outweigh them.23 However, sports injury prevention can have a considerable impact on those socioeconomic consequences. In order to reduce short- and long-term consequences, there is a need for sports injury prevention. The obtained data concerning the cost of each injury localisation will enable us to determine the cost−benefit of preventive measures in the future.