Heat-related injuries resulting in hospitalisation in Australian sport

Heat-related injuries resulting in hospitalisation in Australian sport

Journal of Science and Medicine in Sport (2008) 11, 40—47 ORIGINAL PAPER Heat-related injuries resulting in hospitalisation in Australian sport Timo...

117KB Sizes 1 Downloads 26 Views

Journal of Science and Medicine in Sport (2008) 11, 40—47

ORIGINAL PAPER

Heat-related injuries resulting in hospitalisation in Australian sport Timothy Robert Driscoll a,∗, Raymond Cripps b, John R. Brotherhood c a

School of Public Health, University of Sydney, NSW 2006, Australia Research Centre for Injury Studies, Flinders University, Australia c School of Exercise and Sport Science, The University of Sydney, Australia b

Received 26 October 2006 ; received in revised form 13 April 2007; accepted 13 April 2007 KEYWORDS Heat; Sport; Triathlons; Lawn bowls; Cricket; Golf; Running

Summary The aim of this study was to summarise the extent and characteristics of cases of illness due to environmental heat, significant enough to result in hospitalisation, and arising during sporting activity in Australia. Cases were identified from the hospital separations database compiled by the Australian Institute of Health and Welfare, using the allocated external cause and diagnosis codes and the activity code ‘‘While engaged in sports’’. Hospital separations for the 2 years 2002—2003 and 2003—2004 were used. One hundred and forty eight cases were identified (68% male). Cases were fairly evenly distributed across 10-year age groups starting from age 15 years, apart from fewer cases between 55 and 64 years. Nearly two thirds of the cases occurred in the summer months (December to February inclusive). The most commonly involved individual sports were lawn bowls, cricket, softball, golf, marathon running and walking, and the rate was highest for triathlons, lawn bowls, cricket, and running. Rates for persons aged 65 years or older were more than twice the rates at younger ages. Heat-related disorders are an uncommon cause of significant morbidity in Australians participating in sporting activity. However, particular sports have a relatively high rate of occurrence and these sports would provide an appropriate focus for prevention activity. The availability of a specific code in the International Classification of Diseases and Injuries to cover excessive endogenous production of heat would assist future analyses of the role of thermoregulatory disturbance in leading to morbidity in persons participating in sporting activity. © 2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

Introduction There are estimated to be nearly 13 million Australians aged 15 years or more involved in ∗

Corresponding author. E-mail address: [email protected] (T.R. Driscoll).

‘‘exercise, physical activity or sport’’.1 This is about 80% of the population in this age group. It is not possible to accurately separate activities undertaken for sport from those undertaken for leisure, although for many individual ‘‘sports’’ the context of the activity should be reasonably clear, and the type of activity involved should not differ

1440-2440/$ — see front matter © 2007 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

doi:10.1016/j.jsams.2007.04.003

Heat-related injuries resulting in hospitalisation in Australian sport importantly regardless of the primary motivation for the activity. Heat-related illness associated with sport has been an area of concern for many years. This is particularly the case in hot countries such as Australia, where people can undertake very vigorous physical activity in conditions of high ambient temperature and high humidity. Several disorders considered to be associated with environmental heat stress and exertion have been described: heat stroke, heat exhaustion, heat syncope, heat cramps and heat rash.2 Different authors have sometimes used slightly different terms for the conditions, and the patho-physiology underlying the conditions is still debated, especially for heat stroke.3 With regard to exercise and sport, the heat-associated conditions that are likely to be admitted to hospital are heat stroke, collapse with hyperthermia, and severe dehydration. Heat stroke results from gross thermoregulatory failure with deep body temperature >41 ◦ C, accompanied by central nervous system disturbance. Heat stroke is potentially lethal and may lead to serious injuries in survivors.4 Exertioninduced heat exhaustion has been described as collapse during or after exertion with deep body temperature >38 ◦ C.5 It usually involves circulatory instability and in cases with hyperthermia there may also be central nervous system disturbance.6 Some authors have suggested that there may be a continuum between exertion-induced heat exhaustion and heat stroke.5 Heat-related illness has classically been associated with dehydration related to extreme exercise in high-temperature environments. However, several authors have noted that most significant heat-related illness is primarily related to the rate of physical exertion relative to the fitness of the individual, rather than to the environmental conditions. This explains why heat-related conditions in sport are more common in events of shorter duration, often in athletes of limited experience, and can occur in temperate environments.6,7 In these cases, dehydration is not seen as the primary pathophysiological problem, and drinking too much may actually result in significant pathology.8 Hyponatraemia may well be due to over-consumption of hypotonic solutions rather than excessive loss of sodium as a result of sweating.9 This study was designed to focus on injurious effects of environmental heat on sports participants. Although occasional cases of sports-related heat stroke have been reported in Australia4,10,11 and cases of collapse with hyperthermia are a regular occurrence in community running events such as the Sydney City to Surf,7 there is not a lot of information available on the incidence of heat-

41

related disorders associated with sport. Research in this area is complicated by difficulties with definitions and the overlap between ‘heat-related’ illness to which the environment has made a significant contribution and heat-related illness that is primarily due to endogenous factors. Furthermore, some heat-related cases of collapse may recover spontaneously or with medical care at the sports ground or in the emergency department. Currently these cases are not formally recorded on a widespread basis and are thus lost to enumeration. Therefore, the aim of this study was to summarise the extent and characteristics of cases of illness due to environmental heat, significant enough to result in hospitalisation, and arising during sporting activity in Australia.

Methods Cases were identified from the hospital separations database compiled by the Australian Institute of Health and Welfare (AIHW).12 The database records all hospital separations. A separation is ‘‘the process by which an episode of care for an admitted patient ceases’’ and an admission is defined as ‘‘Admission is the process whereby the hospital accepts responsibility for the patient care and/or treatment’’.13 This means that a discharge from an emergency department after a prolonged stay may or may not be counted as a separation for the purposes of the database, depending on the administrative approach of the hospital. Each visit can be categorised in four different ways: external cause, diagnosis, activity, and place. All cases in which an injury contributed to the admission or the stay in hospital are allocated at least one external cause code (E-code). Up to 31 such E-codes can be allocated. Similarly, every case is allocated one or more Diagnosis codes, with up to 31 such codes possible. External cause codes are coded using the Australian Modification of the 10th revision of the of the International Classification of Diseases and Injuries (ICD-10AM).14 Diagnosis codes also come from ICD-10AM. In addition, each case should be allocated an activity code, which identifies the activity at the time of injury for external cause cases. The activity coding frame includes a code for ‘‘While engaged in sports’’ and a separate code to cover leisure activity (‘‘While engaged in leisure’’). Since July 2002, cases with an activity code of ‘‘While engaged in sports’’ can also be allocated a separate code identifying the exact sport involved. These codes can be grouped to combine similar types of sports (e.g. ‘‘team ball sports’’; ‘‘individual athletic activities’’). There is

42

T.R. Driscoll et al.

some overlap between sport and leisure activities, and instructions to coders using ICD-10AM are to use the relevant sport code when both a sport and a leisure code applies.15 All external cause cases should be allocated a place code, which identifies the place of the injurious incident for external cause codes. Using this information, the cases of interest were identified using a two-step process. Firstly, all heat-related hospitalisation cases were identified using the external cause codes and diagnosis codes. Heat-related hospitalisation cases met the following definition: - External cause code equal to X30 (environmental sources of heat) in any external cause field; or - External cause code equal to W92 (man-made source of heat) in any external cause field; or - Diagnosis code equal to T67 (effects of heat and light) in any diagnosis field. Sport-related cases were identified within this subset by choosing all people with an activity of ‘‘While engaged in sports’’. The justification for this approach was that admissions for heat-related illness should be allocated an E-code of either X30 (environmental sources of heat) or W92 (man-made source of heat). The diagnosis code for heat-related cases is likely to be T67 (effects of heat and light), which has 10 subcategories, but might occasionally be expected to be one of several different diagnosis codes, such as those related to dehydration (E86), hypernatraemia (E87.0) or hyponatraemia (E87.1). Similarly, some heat-related cases might have a diagnosis of T67 but not have either X30 or W92 as an external cause code, although this should not happen according to the ICD-10AM coding rules. Therefore, all heatrelated cases should theoretically be able to be

Table 1 Age 0—14 15—24 25—34 35—44 45—54 55—64 >65 Total

identified using the external cause codes X30 and W92, and the diagnosis code T67. It is possible that some cases of ill health related to heat will be missed using this approach, if the connection to heat was not recognised by the treating physician and/or the coders. However, it was considered not appropriate to include all cases with codes related to related to dehydration (E86), hypernatraemia (E87.0) or hyponatraemia (E87.1) because there are many other causes of these conditions that are not related to heat (nor to exercise). Hospital separations for the 2 years 2002—2003 and 2003—2004 were used for the analysis. Cases were identified in a special file provided routinely by the AIHW to the National Injury Surveillance Unit (NISU). This file has all hospital separations which have one or more E-codes and/or one or more diagnoses in the S or T range. Since the database records all separations, it includes as separate cases transfers from one hospital to another for the same injury. This means the number of identified cases will be slightly higher than the true number of incident cases. The data item ‘Source of referral’ was used to omit hospital separations records shown to be inward transfers from one hospital to another and to estimate incident admissions. Data were extracted from the file using SPSS, and analyses performed using SPSS and Excel. Rates of heat-related hospitalisation cases were calculated using participation data obtained in a population survey conducted in 2003. The sample was selected using a random telephone survey covering the whole of Australia and was conducted in February, May, August and November of 2003. There was a 45% response rate, with results available for 13,644 persons.1 The survey excluded persons aged less than 15 years. So, rates presented here on the basis of sporting participants are based

Age distribution and rate of sport-related heat cases by age Numbera 15 24 24 23 17 9 36e 148

%

Participantsb

Ratec

95% CId

10.1 16.2 16.2 15.5 11.5 6.1 24.3

— 2480.3 2554.8 2449.8 2164.3 1534.6 1666.9

— 4.8 4.7 4.7 1.8 2.9 10.8

— 3.1—7.2 3.0—7.0 3.0—7.0 2.3—6.3 1.3—5.6 7.6—14.9

100.0

12850.7

5.2

4.3—6.1

Hospital separations. Australia, 2002/2003—2003/2004. a Cases over 2 years. b 000s (participation data were not available for persons under the age of 15 years). c Cases per 1,000,000 per year: excludes persons aged less than 15 years. d 95% confidence interval. e Twenty persons were aged between 65 and 74 years inclusive; 16 persons were aged 75 years or more.

Heat-related injuries resulting in hospitalisation in Australian sport only on cases aged 15 years or more. Rates are expressed as cases per 1,000,000 participants per year. Confidence intervals for rates were estimated from the observed number of cases, based on the Poisson distribution. The rate of injury based on the general population (not just the sporting participants) used all cases, regardless of age, and denominator data based on the average of the resident population aged 5 or over in June 2002 and June 2003.16,17

Results There were 1,648,264 separations for incident cases from July 2002 to June 2004. One thousand three hundred and fifteen cases (0.08%) met the study definition of heat-relatedness. Of these 1315 cases, 148 (11%) had an activity code of ‘‘While engaged in sports’’. These 148 cases were the focus of the results presented here. One hundred and one (68%) of the cases were male. Cases were fairly evenly distributed across 10-year age groups starting from age 15 years, apart from fewer cases between 55 and 64 years (Table 1). Eighty one percent (120) of cases had a principal diagnosis of T67 (effects of heat and light) and another 20 cases had T67 in another diagnosis field. Eighty-six percent (128) had X30 (environmental sources of heat) as the left-most external cause code, and another four had X30 as another external cause. Eighty-four percent of cases had both T67 as a diagnosis code and X30 as an external cause code. Heat-related hospitalisation cases with a T67 principal diagnosis were allocated to 1 of 4 of the 10 sub-categories—–T67.0 (heatstroke and sunstroke): 36%; T67.1 (heat syncope): 15%; T67.3 (heat exhaustion, anhydrotic): 7%; T67.5 (heat exhaustion, unspecified): 43%. The most commonly involved sports were team bat and stick sports (primarily cricket and softball); target and precision sports (golf and lawn bowls); and individual endurance sports (marathon running, jogging, triathlon and cycling). The most commonly involved individual sports were lawn bowls, cricket, softball, golf, marathon running and walking (Table 2). The overall rate of hospitalisation per sporting participant was 5.2 cases per million. This rate varied considerably between sports, being highest for triathlons, softball, lawn bowls, cricket and running (Table 3). The rate also varied considerably with age, with rates for persons aged 65 years or older being more than twice as high as the rate at younger ages (Table 1).

Table 2

43

Type of sport for sport-related heat cases

Sport Team ball sports Football Team bat and stick sports Cricket Softball

Number

%

5 4

3.4

32 16 15

21.6

Boating sports

..

Individual water sports Fishing Swimming

10 4 6

6.8

Individual athletic activities Walking Track and field

13 10 ..

8.8

Individual endurance sports Jogging and running Marathon running Triathlon Cycling

28 7 10 6 5

18.9

5 4

3.4

34 20 14

23.0

Racquet sports Tennis Target and precision sports Lawn bowls Golf Equestrian activities Other sports Canoeing Unspecified boating sport Rock climbing Motorcycling Go-carting Athletic activities involving fitness equipment, nec Wood chopping Other specified sport and exercise activitya Unspecified Total

.. 15

10.2

.. .. .. .. .. .. .. 8 .. 148

100.0

Hospital separations. Australia, 2002/2003—2003/2004. Note: ‘..’ means there was at least one case, but less than four. a This category covers sporting and exercise activities that were specified in the source documents but for which separate categories do not appear in the ICD-10-AM coding frame.

Nearly two thirds of the cases occurred in the Summer months (December to February inclusive), with approximately 15% in each of Spring and Autumn months and 6% in the Winter. The predominance of cases in summer months appeared to be present for all the major sports except for the individual endurance sports, for which there was almost equal number of cases in autumn and summer (see Table 4).

44 Table 3

T.R. Driscoll et al. Rate of sport-related heat cases by sport (for selected sports) Numbera

Sport Triathlon Softball Lawn bowls Cricket Running Golf Fishing Cycling Tennis Football Walking Swimming Other Total

Participantsb

5 10 20 14 16 14 .. 5 4 4 9 .. 16 133

Ratec

95% CId

30.4 75.6 364.7 696.7 1181.1 1282.6 402.6 1471.8 1407.3 1670 6808.1 2383.1 —

82.2 66.1 27.4 10.0 6.8 5.5 3.7 1.7 1.4 1.2 0.7 0.6 —

26.3—192.4 31.7—121.7 16.7—42.4 5.5—16.9 3.9—11.0 3.0—9.2 0.7—10.9 0.5—4.0 0.4—3.6 0.3—3.1 0.3—1.3 0.1—1.8

12850.7

5.2

4.3—6.1

Hospital separations. Australia, 2002/2003—2003/2004. Persons 15 years and older. Note: ‘..’ means there was at least one case, but less than four. a Cases over 2 years. b 000s (participation data were not available for persons under the age of 15 years). c Cases per 1,000,000 per year: excludes persons aged less than 15 years. d 95% confidence interval.

Table 4

Seasonal distribution of sport-related heat cases

Sport

Season Summer

Team ball sports Team bat and stick sports Individual water sports Individual athletic activities Individual endurance sports Racquet sports Target and precision sports Other sports Unspecified

.. 27 4 10 11 .. 25 11 ..

Autumn .. .. .. 0 9 .. .. .. 0

Winter 0 0 .. .. 4 0 0 .. ..

Spring

Total

0 4 .. .. 4 0 6 .. ..

5 32 10 13 28 5 34 18 ..

Total

95

21

9

23

148

Percent (of total cases)

64.2

14.2

6.1

15.5

100.0

By sport. Hospital separations. Number and percent. Australia, 2002/2003—2003/2004. Note: ‘..’ means there was at least one case, but less than four.

The age distribution of cases varied considerably between different sports. Persons playing team bat and stick sports and individual endurance sports were nearly all aged less than 54 years. In contrast, 70% of persons who were playing golf or lawn bowling were aged 65 years or older (Table 5).

Discussion The results provide a firmly-based national estimate of the extent of heat-related illness, significant enough to result in hospitalisation, during sporting activity in the community. They show

that heat-related illness in Australian sport is neither a trivial issue nor a major problem, with approximately 75 cases requiring hospitalisation in each of the 2 years covered by the study. Appropriate preventive action is not informed by the information, but the results do identify priority targets for such intervention. However, these statistics probably underestimate the true occurrence of significant sport-related heat illness. For example, 969 heat casualties occurred between 1971 and 2005 (about 30 cases, <0.1% of starters, per event) in the Sydney City to Surf. Most of these were successfully treated on site for exertion-induced heat exhaustion, and would not be recorded in in-patient

Heat-related injuries resulting in hospitalisation in Australian sport Table 5 Age

45

Age distribution of sport-related heat cases Team bat and stick

Individual endurance

Target and precision

Number

Number

Number

0—14 15—24 25—34 35—44 45—54 55—64 65—74 >75

7 10 6 6 .. 0 0 0

Total

32

% 21.9 31.3 18.8 18.8 0.0 0.0 0.0 100.0

%

.. 4 6 11 .. .. .. 0 28

14.3 21.4 39.3

0.0 100.0

0 0 .. 0 5 4 14 10 34

% 0.0 0.0 0 14.7 11.8 41.2 29.4 100.0

By selected sport. Number and percent. Hospital separations. Australia, 2002/2003—2003/2004. Note: ‘..’ means there was at least one case, but less than four.

statistics. About 100 were referred to hospital, usually for persisting central nervous system disturbance (DAB Richards personal communication). Many of these persons would have been discharged from the emergency department and so also may not have been recorded in in-patient statistics. Similar cases are known to occur in other community running events. Which of these cases are properly considered ‘‘serious heat-related illness’’ cannot be determined from the available information, which is the main reason hospital admission was used as the criteria for inclusion in this study. Given the high metabolic heat load likely to arise from sustained high-level exercise, and the challenges this poses in terms of required heat loss when the environmental temperature is high, the high rate of cases in triathletes is not surprising. Runners face a similar problem in terms of metabolic heat load (note that rates could not be separately determined for marathon runners because appropriate denominator data were not available). In contrast, the high number and rate of cases in persons playing lawn bowls and golf would not be expected in terms of internal metabolic heat load. However, nearly three-quarters of the cases were persons aged 65 years or older, and thermoregulatory control is known to function less well in older persons.18 In addition, there are a number of medications that can interfere with the physiological mechanisms involved in thermoregulatory control,19 and most of the medications are more commonly taken by older persons than younger persons. Finally, both lawn bowls and golf commonly involve the participants standing in the sun for several hours, increasing the participants’ exposure to radiant heat. Cricket also involves many hours spent standing in the sun. The high number and rate of cases in softball (see Table 5) is unexpected, but appears to be

almost entirely due to a single incident in February of 1 year when a junior softball tournament was held in very hot weather, with air temperature reaching about 40 ◦ C, and many of the players were admitted to a local hospital. Information from the hospital reveals that 10 persons aged between 12 and 16 years inclusive were admitted to the hospital with heat-related illness on the same day (unpublished information from the hospital medical records department). The activity code was not available for these persons, but it seems very likely that all, or nearly all, were the players known to have been admitted to the hospital on that day. Although the focus of this paper was heat-related hospitalisations in persons undertaking sporting activity, it was not possible to exclude activities that involved exercise but which might be better considered as leisure rather than as sport. For example, walking might be undertaken primarily as a social activity rather than for exercise, but it is not possible to distinguish the two. Indeed, distinguishing between ‘sport’ and ‘leisure’ is extremely difficult in many circumstances, and the activity codes used in conjunction with ICD10 AM 4th edition (U50-U73) do not attempt to make such a distinction because ‘“sport’ overlaps with ‘leisure’ and sufficiently specific and generally accepted definitions for these activities, suitable for use in clinical coding, are not available’’.15 For most sports, this should not be a significant issue, because most activities included as sports are easily recognised as such, and they involve similar basic requirements regardless of the reason the activity is undertaken. As mentioned earlier, ‘heat-related illness’ arises from an imbalance between the heat load acting on the body arising from metabolic heat production and the environment, and the ability of the body to lose heat, usually as a result of

46 restriction of sweat evaporation by environmental factors, including clothing, or inadequate physiological sweat production. The focus of this paper was the development of heat-related illness serious enough to result in hospitalisation during sporting activity in hot environmental conditions. Cases could only be identified using relevant ICD-codes and the application of those codes is likely to be subject to error. For example, if a marathon runner collapses and is admitted to hospital with a raised core body temperature in winter, would this be coded as being due to environmental heat? Should it be coded as being due to environmental heat? Presumably the air temperature and humidity made some contribution to the condition developing, but the major contributing factor is likely to be the endogenous heat load, and the case could be expected to be coded on some occasions as resulting from environmental conditions and on other occasions as resulting from over exertion rather than external heat. Davies showed thermoregulatory control can be lost, and core temperature can rise progressively (to exceed 40 ◦ C) when running for 60 min at competitive intensity even when the environmental conditions are mild (air temperatures of 21 ◦ C and relative humidity less than 50%).20 If the same condition developed in the summer, the major contributing factor might again be the endogenous heat load, but the condition is very likely to be coded as being due to environmental heat. On other occasions, environmental heat might indeed be the over-riding factor. Clinicians would have difficulty separating the endogenous and environmental contribution in many cases, and it must be even more difficult for persons coding the admission information at a later time. This issue means that separately identifying cases to which high environmental temperatures have clearly made a significant contribution can be expected to be difficult. In fact, there may be more benefit in the future in focusing on the spectrum of pathological conditions that fall under the umbrella of so-called heat-related cases, regardless of the environmental temperature, than trying to identify the sub-group to which environmental heat made a major, or significant, contribution. Diagnosis codes such as R55 (syncope and collapse), T733 (exhaustion due to excessive exertion), E86 (volume depletion), E87.0 (hypo-osmolality) and E87.1 (hyper-osmolality), and the external code X50 (overexertion and strenuous or repetitive movements), might be useful when studying heat-related disorders, and when studying the broader question of thermoregulatory stress and resulting circulatory disturbance regardless of the environmental temperature. However, it should be noted that

T.R. Driscoll et al. identifying all relevant cases, regardless of environmental temperature, on the basis of ICD-10 codes is difficult, because these codes could also be validly used to describe circumstances not related to thermoregulatory stress, although this would presumably be much less likely in the context of sporting activity. To this end, it would be beneficial if a specific code under the ‘‘Overexertion and strenuous activity’’ category (X50) could be allocated to cover excessive endogenous production of heat. Regardless, this broader analysis was not performed in this case because the stimulus for the paper was the contribution of environmental heat. Nine cases occurred in the winter months, which seems unusual at face value. These cases might reflect diagnosis of heat illness in fact arising principally from endogenous heat production. Exercise-related heat casualties do occur in the winter months as evidenced by the cases in the Sydney City to Surf 14 km fun run held in August when conditions are usually cool to mild.7 In the northern parts of Australia, the ambient temperature is certainly warm enough in winter to contribute to the development of a heat-related disorder, especially in endurance events. The analysis relied on the applied External Cause and Diagnosis codes, and there was no way for these to be independently verified. In theory it would have been possible to check the weather conditions on the day and in the place where the incident occurred, but this was not possible with the data available for the current analysis. This is a limitation of the present analysis, but not likely to a significant source of error, because only nine cases occurred in the winter months and four of these involved marathon runners, who might well have sustained a thermoregulatory illness due in part to environmental heat. The diagnosis code T67 was used in 96% of the heat-related hospitalisation cases, with 4 of the 10 sub-categories specified. Given the overlap between diagnostic criteria for the terms used for the different sub-categories, it is not clear whether the specific sub-categories are of use from a prevention perspective. X30 and T67 were used in 84% of cases. It would be expected that all cases would have both codes, given the criteria for these two categories. It was not apparent from the data why some cases had only one of these two codes. The number of cases identified in this study is probably an underestimate of the total number of eligible cases because of problems with the coding of activity and of heat-related illness, but this is not likely to have been an important source of error. A sizeable proportion of hospital separations records have a missing or indefinite code for activity. The

Heat-related injuries resulting in hospitalisation in Australian sport activity code is likely to be missing when the relevant information is not recorded anywhere in the medical record. Information on activity is likely to be recorded in the medical record of nearly all cases related to heat and sport, because this is relevant to the diagnosis. This means that the vast majority of these cases are likely to have a specific activity code and so are likely to have been identified. A separate factor that might tend to overestimate the number of cases is that it was not possible to completely eliminate the chance of double counting due to multiple admissions for the same condition. Use of the ‘‘source of referral’’ data item should largely eliminate this, but it does not cover situations where a person is re-admitted from the community for further treatment or rehabilitation. Such re-admission is not likely for the type of conditions considered in this paper, so this source of error is unlikely to be important in the presented analysis.

Conclusions Heat-related disorders are an uncommon cause of significant morbidity in Australians participating in sporting activity. Nevertheless, there are approximately 75 cases resulting in hospitalisation each year. Particular sports have a relatively high rate of occurrence and these sports would provide an appropriate focus for prevention activity. The availability of a specific code under the ‘‘Overexertion and strenuous activity’’ category (X50) to cover excessive endogenous production of heat would assist future analyses of the role of thermoregulatory disturbance in leading to morbidity in persons participating in sporting activity.

Practical implications • Heat-related disorders are an uncommon reason for the hospitalisation of Australians playing sport. • Heat-related disorders are more common in particular sports (e.g. triathlon, lawn bowls and cricket) and in older participants. • Improvements in ICD disease coding would improve future analyses of heat disorders in sport.

47

References 1. Standing Committee on Recreation and Sport. Participation in exercise, recreation and sport survey 2003 annual report. Canberra: Australian Sports Commission; 2004. 2. Leithead C, Lind A. Heat stress and heat disorders. London: Cassell; 1964. 3. Noakes T. Fluid and electrolyte disturbances in heat illness. Int J Sports Med 1998;19(Suppl. 2):S146—9. 4. Lee R, Bishop F, Ashton C. Severe heat stroke in an experienced athlete. Med J Aust 1990;153:100—4. 5. Hales J, Richards D. Principles for the prevention of death from heat stress. In: Hales J, Richards D, editors. Heat stress: physical exertion and environment. Amsterdam: Elsevier Scientific Publishers; 1987. p. 7—13. 6. Sutton J. Physiological and clinical consequences of exercise in heat and humidity. In: Harries M, et al., editors. Oxford textbook of sports medicine. Oxford: Oxford University Press; 1994. p. 231—7. 7. Richards C, Richards D. Physiological factors predisposing to physical exhaustion. In: Hales J, Richards D, editors. Heat stress: physical exertion and environment. Amsterdam: Elsevier Scientific Publishers; 1987. p. 419—26. 8. Noakes T. Fluid replacement during marathon running. Clin J Sports Med 2003;13(5):309—18. 9. Noakes T. Hyponatraemia in distance runners: fluid and sodium balance during exercise. Curr Sports Med Rep 2002;1(4):197—207. 10. Forgan-Smith J. Exertion-induced heat stroke. Med J Aust 1987;146:154—5. 11. Savdie E, Prevedoros H, Irish A, et al. Heat stroke following Rugby League football. Med J Aust 1991;155:636—9. 12. Australian Institute of Health and Welfare. National hospital morbidity data collection. Canberra: Australian Institute of Health and Welfare; 2006. Accessed in September, 2006 http://www.aihw.gov.au/hospitals/ nhm database.cfm. 13. Australian Institute of Health and Welfare. National health data dictionary, version 12. Canberra: Australian Institute of Health and Welfare; 2003. Accessed in December, 2006 http: // www. aihw. gov.au/ publications/ index.cfm/ title/ 8964. 14. National Centre for Classification in Health. The international statistical classification of diseases and related health problems, Tenth revision, Australian modification (ICD-10-AM). 3rd ed. Sydney: NCCH; 2002. 15. National Centre for Classification in Health. Activity codes. Coding Matters 2004;11(2):4. 16. Australian Bureau of Statistics. Australian demographic statistics—–December 2002. Canberra: ABS; 2003. 17. Australian Bureau of Statistics. Australian demographic statistics—–December 2003. Canberra: ABS; 2004. 18. Kenney W, Munce T. Invited review: aging and human temperature regulation. J Appl Physiol 2003;95:2598—603. 19. Lomax P, Schonbaum E. The effects of drugs on thermoregulation during exposure to hot environments. Prog Brain Res 1998;115:193—204. 20. Davies C. Influence of skin temperature on sweating and aerobic performance during severe work. J Appl Physiol 1979;47(4):770—7.

Available online at www.sciencedirect.com