Article Text

Download PDFPDF

Economic burden of physical activity–related injuries in Dutch children aged 10–12
  1. Dorine C M Collard1,
  2. Evert A L M Verhagen1,
  3. Willem van Mechelen1,
  4. Martijn W Heymans2,
  5. Mai J M Chinapaw1
  1. 1Department of Public & Occupational Health, EMGO Institute for Health and Care Research, VU University Medical Center, Amsterdam, The Netherlands
  2. 2Department of Epidemiology and Biostatistics, VU University Medical Centre, Amsterdam, The Netherlands
  1. Correspondence to Evert Verhagen, Department of Public and Occupational Health, EMGO Institute for Health and Care Research, VU University Medical Center, Van der Boechorststraat 7, 1081 BT Amsterdam, The Netherlands; e.verhagen{at}vumc.nl

Abstract

Background Injuries in children occur most often in physical activity–related activities. A lot of these injuries result in direct and indirect costs. A detailed overview of the economic burden of those injuries in children is lacking.

Method A prospective study was conducted with 996 children in Dutch primary schools to describe the economic burden of injuries that occur during organised sports, leisure time and physical education (PE) class activities. Injuries were continuously monitored by PE teachers during the school year 2006–2007. An injury was recorded if it occurred during PE class, leisure time or organised sports activity and caused the child to at least stop the current activity. If an injury was recorded, parents received a cost diary to report the direct and indirect costs of the child's injury. Costs were collected from a societal perspective.

Results During one school year, a total of 119 injuries were reported by 104 children. The mean total costs as a result of an injury were €188±317. The mean direct costs as a result of an injury were much higher than the mean indirect costs (€131±213 and €57±159, respectively). The highest costs were found for upper extremity and leisure time injuries.

Conclusion Physical activity–related injuries are common in children and result in medical costs. Injuries that lead to the highest costs are those that occur during leisure time activities and upper extremity injuries. Intervention programmes for children to prevent upper extremity injuries and leisure time activity injuries may reduce direct (ie, healthcare) and indirect costs.

Trial registration: ISRCTN78846684

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

Paediatric injury-related visits to emergency departments are mainly due to physical activities.1 In children aged 6–12 years, home was the most common place where injury occurred leading to an emergency department visit (28% of injuries), but sport/recreation areas (12%), streets and highways (11%) and schools (10%) were the other important sites.1 In The Netherlands, sports injuries among 5–14-year-old children can be considered as a substantial public health problem. Belechri and colleagues2 found in 2001 that 42 000 sports and physical activity–related child injuries are seen in Dutch hospitals every year. Of these, 13 000 injuries occurred in a school setting, 17 000 during organised sports and 12 000 during leisure time activities.

Besides the short- and long-term consequences of physical activity–related injuries on physical and mental well-being, these injuries also lead to high direct and indirect costs for society. For instance, in the USA, over one-third of school-aged children sustain an injury severe enough to be treated by a doctor or nurse.3 In 2003, the associated annual medical costs were estimated to be about US$1.8 billion.3

The costs of injuries can be a useful measure for injury severity by identifying specific injuries that are ‘expensive’ for society,4 thereby guiding injury prevention efforts towards the reduction of particularly high-cost injuries. Moreover, descriptive cost data may act as a facilitator for political agenda setting and policy change towards prevention.

Most studies that report on costs of injuries include only sport-related injuries. However, during childhood, children participate not only in organised sport activities but also in leisure time activities. For Dutch children, the mean hours of leisure time activities per week are twice as high as those of organised sports activities.5 The annual direct and indirect healthcare costs as a result of physical activity–related injuries (organised sport, leisure time and physical education (PE) class activities) in children are unknown.

In this article, we report on the economic burden of injuries that occurred during organised sports, leisure time and PE class activities in children. The analyses include (1) the volume of healthcare use and associated costs and (2) the mean total direct and indirect costs in general and by gender (boys/girls), type of physical activity (organised sport activity, leisure time activity and PE class) and location of injury (upper or lower extremities).

Methods

Population

This study was part of the iPlay study, an injury prevention study in 10–12-year-old children, that has been carried out in primary schools located throughout The Netherlands.6 From the 7000 primary schools throughout The Netherlands, 520 primary schools were randomly selected and invited to take part in the iPlay study. A total of 40 primary schools (consisting of 2208 children of 10–12 years) were eligible for participation and were randomised to an intervention group or a control group. All 20 schools assigned to the control group formed the prospective cohort described in the present study, resulting in a cohort of 1091 children aged 10–12 years. The study was approved by the Medical Ethics Committee of the VU University Medical Center, Amsterdam, The Netherlands. Informed consent was given by each child's parent or guardian by means of a passive informed consent.

Data collection

At the start of the school year (September 2006), all children completed a questionnaire that collected information on demographic variables including age, gender, participation in organised sport and leisure time activities.

Physical activity–related injuries were continuously monitored by PE teachers during one school year. They were instructed to question children explicitly every week about whether they have been injured as a result of physical activities (including leisure time activities) in the past week. The 1-week recall period was specifically chosen to avoid recall bias and to prevent loss of information from failure to remember. In case of an injury, an injury registration form was completed. This registration form provided information on the injury location, injury type, injury diagnosis, direct cause of injury and activity performed at the time of injury. To enhance compliance on reporting injuries, all teachers were contacted each month to remind them of the injury registration.

An injury was recorded when it occurred during PE classes, leisure time or sports activity and if it caused the child to (1) stop his/her current activity; (2) not be able to (fully) participate in the next planned physical activity; (3) not be able to go to school the next day; and/or (4) seek medical attention.7

Costs

In case of a reported injury, teachers handed a cost diary to the child. The parents of the injured child reported all injury-related costs in a cost diary. This cost diary was previously successfully used in other economic evaluations of sports injuries.8 9 All direct and indirect costs from the moment of injury onwards, until full recovery, were registered in the diary. After completion of the diary, the parents were asked to forward the cost diary to the investigators. If a cost diary was not returned within 5 weeks of injury onset, parents were contacted via mail, telephone or school.

Costs were collected from a societal perspective. Table 1 provides an overview of the costs.10,,12 Direct healthcare costs included costs as a result of an appointment with a general practitioner; visits to a sport physician, physical therapist, sports masseur or dentist; visits to emergency department, outpatient clinic, day care admission or hospital admission; and diagnostic interventions such as x-ray. Costs of medication and medical devices (eg, crutches) were also included. Indirect costs included costs as a result of absence of parents from paid work, the presence of a carer and transportation to and from daily activities in a different way from usual as a result of the injury.

Table 1

Prices used in the economic evaluation

Dutch guideline prices were used to value resource use.10 The direct costs of hospital treatment (eg, x-ray) were estimated on the basis of prices from the Health Care Insurance Board.11 Costs of drugs and medical devices were based on prices of the Royal Dutch Society for Pharmacy12 or the Health Care Insurance Board.11

The costs of absence from paid work were calculated from the average age and wage in The Netherlands.10 All prices were standardised to the year 2007 and adjusted for inflation.13

Statistical analyses

Total costs were estimated for each child by multiplying resource data by cost prices. Total, direct and indirect costs were calculated by adding costs per category of utilisation of healthcare resources. Missing cost data were completed by multiple imputation using the Multiple Imputation by Chained Equations (MICE) procedure.14 Five multiple imputed datasets were generated, whereupon we used the average data of those five imputed datasets.

The mean total, direct and indirect costs and the associated SDs were calculated for the complete cases, that is, by first excluding cases with missing cost information, and for the multiply imputed data. In the latter case, cost analysis was first applied in each multiply imputed dataset and subsequently pooled according to Rubin's rules.15

Furthermore, differences in costs for boys, girls, organised sport activity, leisure time activity and PE class, and upper and lower extremities were calculated for the multiple imputed dataset. As costs were not normally distributed, mean differences in costs and associated 95% CIs were obtained by bias corrected and accelerated bootstrapping (2000 replications) applied in each imputed dataset.16 Subsequently, results were pooled according to Rubin's rules.15 All analyses were performed with SPSS and R statistical software.17

Results

Of the 20 schools that were analysed for the study, all completed the entire follow-up period. Subject characteristics are shown in table 2.

Table 2

Subject characteristics

Injuries

During one school year in the sample of 996 children, a total of 119 injuries were reported by 104 children. The injury incidence density (IID) for this population was reported in a previous study by Verhagen and colleagues.5 The overall IID was 0.48 per 1000 h of exposure (95% CI 0.38 to 0.57) The injury IID was lowest for leisure time activities (0.39, 95% CI 0.28 to 0.50), followed by PE (0.50, 95% CI 0.29 to 0.71) and organised sports activities (0.66, 95% CI 0.46 to 0.87). More injuries were reported in girls than in boys (58% vs 42%). The lower extremities (68%) were the most injured body parts. Bruises were most commonly reported (43%), mainly occurring during leisure time physical activity. The second most frequent occurring injury was a ligament sprain, most commonly resulting from participation in sports. In addition, fractures were commonly reported. Falling or stumbling as well as missteps were most common, each accounting for 28% of all injuries. Fractures were significantly more often reported in upper extremity injuries (n=11) than in lower extremity injuries (n=2). For the upper extremity injuries, 9 of the 11 fractures were to the hand/wrist and forearm. It appeared that 8 of the 11 fractures were reported during leisure time activities.

Cost diaries

Complete cost diaries were returned by 73 of the 104 parents (70%). An independent sample t test showed no differences with respect to sex, ethnicity, socioeconomic status and body mass index between children whose cost diary was completed, missing or incomplete. Furthermore, injuries without a cost diary were not significantly different from injuries with a cost diary in terms of variables such as severity, anatomical location and cause.

Healthcare use

Table 3 shows the volume of healthcare use and the associated costs for the complete cases (n=73). In addition, volume and associated costs are given for absence of parents from paid work, the presence of a carer and transportation to and from daily activities.

Table 3

Frequency and costs (€) per category of direct, indirect and total costs for the complete cases in the control group (n=73)

Of the 73 parents who completed the cost diary, 43 (58%) reported costs as a result of their child's injury. Of these, 31 parents reported a total of 38 visits to a general practitioner and 7 parents reported a total of 18 visits to a physical therapist. A total of 43 visits to an emergency department or outpatient clinic was reported by 31 parents. During these visits, 28 x-rays were taken. In total, 17 children had one x-ray and four children had more than one x-ray. Furthermore, 18 plaster bandages were applied and six children needed a second plaster bandage.

The highest accumulated costs were found for emergency department and outpatient visits (€3234 and €1244, respectively). Costs as a result of x-ray and plaster bandages were also high (€1220 and €1141, respectively). Visits to a general practitioner and therapists were associated with lower costs. Remarkably high indirect cost for absence of parents from their work as a result of the injury of their child was recorded (€5329). Eleven parents reported in total 18 days of absence from paid work. The accumulated total direct and indirect costs for the 73 complete cases were €13 716 (see table 3).

Mean direct, indirect and total costs as a result of injury

Table 4 shows the mean direct, indirect and total injury costs (±SD) for the complete cost data (n=73) and the imputed data set (n=104). The results described below are based on the imputed data set.

Table 4

Mean direct, indirect and total costs (€, 2007) (SD) for total, boys, girls, sport club, leisure time, PE class, upper and lower extremity injuries

Total costs

Mean total costs per injury were €188±317. Mean total costs were calculated from mean direct and indirect costs. Mean direct costs as a result of injury were twice as high as the mean indirect costs (€131±213 and €57±159, respectively).

Gender

Table 4 also shows the mean direct, indirect and total costs for injuries reported by boys and girls. The mean total injury costs were higher in girls (€226±350) than in boys (€185±304), mainly as a result of higher direct costs. No significant difference in mean total costs between girls and boys (cost difference = €41, 95% CI −126 to 198) was found.

Type of physical activity

The mean total costs for injuries that occurred during leisure time activities (€271±416) were higher than the costs for injuries that occurred during PE classes (€182±269). However, the difference was not significant. Organised sport activities had the lowest injury costs (€122±240) (table 4).

Borderline significant higher total costs were found for leisure time injuries than for organised sport injuries (cost difference = €149, 95% CI −5 to 308). The difference in costs for leisure time injuries and organised sport injuries was mainly caused by the higher indirect costs (cost difference = €93, 95% CI 30 to 174). The costs for leisure time injuries were also higher than the costs for PE class injuries, but this difference was not significant (cost difference = €89, 95% CI −75 to 265).

Location of the injury

The mean total costs for upper extremity injuries were €316±401 compared with €75±232 for lower extremity injuries (table 4). The difference in costs was significant (−€241, 95% CI −364 to −92).

Discussion

Main findings

The mean total costs as a result of a physical activity–related injury in children aged 10–12 years in The Netherlands were €188±317. The mean direct costs were much higher than the mean indirect costs, €131±213 and €57±159, respectively. The highest costs were found for upper extremity injuries and for injuries that occurred during leisure time activities.

The higher costs of upper extremity injuries were mainly due to fractures to the hand/wrist and forearm. Upper extremity injuries were significantly more often reported during leisure time activities than during organised sport or PE class activities. Most of the fractures were reported during leisure time activities. The higher costs for injuries incurred during leisure time activities were caused by a higher incidence of fractures to upper extremities.

It must be mentioned that although physical activity–related injuries do have associated costs, there is also a cost to society of physical inactivity.18 19 It is important to stimulate children to participate in physical activities to enhance health. Because a lot of physical activity–related injuries are preventable, costs as a result of those injuries can be reduced.

Strengths and limitations

This study is the first to describe not only the medical and non-medical costs as a result of sports injuries but also the medical and non-medical costs as a result of leisure time and PE class activities in children.

Another strength of this study is that costs were measured prospectively by means of cost diaries. Most studies only report costs provided from medical records (ie, emergency department records). Such data only take into account injuries severe enough to result in a visit to hospital. In our study not only injuries severe enough to result in a visit to the hospital but all injuries were taken into account. It is known that minor injuries without medical need for hospitalisation account for a substantial part of healthcare costs.20 Moreover, our study also included costs of general practitioners, sport physicians, physical therapists, sports masseurs and dentists and indirect costs such as the costs of parental absence from paid work and the presence of a carer. In addition, parents were contacted by mail, telephone or school if a cost diary was not returned within 5 weeks of injury onset. Parents could contact the researchers if they had difficulties with completing the cost diary.

Furthermore, the study population – children from different primary schools in urban and suburban areas throughout The Netherlands – was a good representation of the Dutch population, and consequently our data can be generalised to injury-related costs in 10–12-year-old children in The Netherlands.

Unfortunately, our study also has some limitations. First of all, due to practical reasons, it was not possible to validate the questionnaire responses of all the 2200 children. Second, the costs were calculated from cost diaries completed by parents. These data are self-reported and have limitations such as recall bias. In addition, only cost diaries were used to assess cost data. A combination of medical records, injury reports and cost diaries, for example, would have led to accurate data on costs as a result of injuries. Unfortunately, this was not possible due to logistic reasons.

Furthermore, during the study period it appeared that not all cost diaries were returned, some 30% being missing or incomplete. Although this is a common issue in cost of injury studies in which costs and injuries are collected prospectively,21 22 this should be taken into account when interpreting the results. The MICE procedure14 is a reliable method to impute missing data, but still the overall results are based upon only two thirds of available data.

A third limitation of this study is that the sample size, although larger than in previously published studies, is still small and this may lead to a less precise estimation of the mean costs. That this study reports on the cost of injuries that occur in a population that is healthy at baseline is a problem. Hence, although the baseline sample is large, the actual number of events that incur costs is relatively small. This is not problematic for the overall results, but does not allow one to calculate costs per sport activity, cause of injury (eg, falls) or injury diagnosis because the number of injuries per subgroup was too small.

Comparison with the literature

Very few studies have attempted to quantify the economic burden of physical activity–related injuries in children. Most studies focus only on sports injuries in the total population. This is, to our knowledge, the first study that describes medical and non-medical costs of injuries as a result of organised sports, leisure time and PE class activities in children.

de Loës21 showed that the medical care and economic costs of medically treated sports injuries in the total population of a municipality in Sweden was US$209±348. Cumps and colleagues22 assessed costs of sports injuries in the Belgium population from insurance company data. They found that direct medical costs per acute insurance-claimed injury and indirect costs were €180 and €1338, respectively. To our knowledge, Knowles and colleagues have conducted the only study on the costs of injuries specifically in children. They described the costs of medically treated sport injuries in a population of North Carolina high school athletes.23 Adjusted mean medical costs per injury were US$709 (95% CI 542 to 917). This study focused only on severe injuries occurring during organised sports activities. Our study investigated the entire scope of physical activity modalities in which children participate, that is, organised sports activities as well as leisure time and PE class activities. Obviously, the mean costs per injury in our study were relatively low as we also included minor (less severe) injuries with few or no costs. Furthermore, comparison of costs as a result of injuries between studies is hampered by differences in injury definition, study design, population, methods and healthcare systems.

A remarkable result in our study was the high cost of upper extremity injuries in children. This finding is consistent with the findings of Meerding and colleagues. They found that upper extremity injuries in terms of healthcare costs are relatively high during childhood. From the age of 15 onwards injuries to the lower extremities (eg, knee and lower leg fractures, hip fractures) increasingly dominate healthcare costs.20

Conclusions

This study has identified the economic burden of physical activity–related injuries in Dutch children aged 10–12. This information could usefully inform policy for injury prevention programmes. For example, this study has identified leisure time activity injuries as a major cause of healthcare costs. Similarly, the costs of upper extremity injuries (often incurred during leisure time activities) are relatively high. Current injury prevention mainly focuses on the prevention of sports injuries. Effective intervention programmes to prevent leisure time activity injuries, including upper extremity injuries, may reduce the average costs of injuries in children considerably.

What is already known on this topic

Paediatric injury-related visits to emergency departments are mainly the result of physical activities. Besides the short- and long-term consequences of physical activity–related injuries on physical and mental well-being, these injuries also lead to high direct and indirect costs for society. For instance, in the USA the associated annual medical costs have been estimated to be about US$1.8 billion.

What this study adds

This study is the first to investigate in a prospective cohort the costs of injuries that occur in the entire scope of physical activity modalities in which children participate, ie, organised sports activities as well as leisure time and physical education class activities. Therefore, results of this study may guide injury prevention efforts towards the reduction of particularly high-cost injuries and may facilitate policy change towards prevention of physical activity–related injuries in children.

References

Footnotes

  • Funding This study was supported by a grant from The Netherlands organisation for Health Research and Development (ZONMW), grant number 622000333.

  • Competing interests None.

  • Ethics approval This study was conducted with the approval of the VU University Medical Center, Amsterdam, The Netherlands.

  • Provenance and peer review Not commissioned; externally peer reviewed.