lnternationalJournal of Sport Nutrition, 1996, 6, 307-320 O 1996 Human Kinetics Publishers, lnc.

Body Weight Changes and Voluntary Fluid Intakes During Training and Competition Sessions

in Team Sports

Elizabeth M. Broad, Louise M. Burke, Greg R. Cox, Prue Heeley, and Malcolm Riley

Fluid losses (measured by body weight changes) and voluntary fluid intakes were measured in elite basketball, netball, and soccer teams during typical summer and winter exercise sessions to determine fluid requirements and the degree of fluid replacement. Each subject was weighed in minimal clothing before and immediately after training, weights, and competition sessions; fluid intake, duration of exercise, temperature and humidity, and opportunity to drink wererecorded. Sweat rates were greatest during competition sessions and significantly lower during weights sessions for all sports. Seasonal variation in dehydration (%DH) was not as great as may have been expected, particularly in sports played indoors. Factors influencing fluid replacement during exercise included provision of an individual water bottle, proximity to water bottles during sessions, encouragement to drink, rules of the game, duration and number of breaks or substitutions, and awareness of personal sweat rates. Guidelines for optimizing fluid intakes in these three sports are provided.

Key Words: sweat loss, dehydration, seasonal variation

Dehydration negatively impacts exercise performance and impairs muscular endurance, mental functioning, thermoregulation, and gastric emptying (2,18,22). Total body dehydration increases core temperature by 0.15-0.40 "C for each 1% decrease in body weight (water) during exercise in the heat (17). High-intensity work performance decreases progressively with 1.8% dehydration (3, 21), and performance decrements of 30% occur with 5-6% body weight loss (16). In the case of team sports, impairments to mental functioning caused by dehydration (5) present a further challenge; performance in team sports requires a high degree of mental functioning for tactical changes, reading the play, anticipation, and skill delivery.

Little is currently known about fluid losses and fluid intake practices of team sport players. Most work has been done on endurance runners and cyclists, often

E.M. Broad, L.M. Burke, G.R. Cox, andP. Heeley are with the Australian Institute of Sport, PO Box 176, Belconnen, ACT 2616, Australia. M. Riley is with the Menzies Centre for Population Health Research, GPO Box 252C, Hobart Tas. 7001, Australia.

308 / Broad, Burke, Cox, et a/.

within a laboratory setting that may not accurately replicate the variable conditions under which field exercise is performed. Among team sports, soccer has been the focus of most investigations of fluid needs. Investigators have reported sweat losses of 0.85-5 L and fluid intakes of just CL1.14 L over a 90-min game, depending on the environmental conditions (8.10). These results indicate that moderate to severe . . . ,

dehydrationmay occur in soccer, whichis of concern since inmany countries soccer is played in hot and humid conditions. Indeed, at the 1988 USA Cup Youth Soccer Tournament, such conditions caused 18 competitors to collapse from heat exhaus- tion by midway through the second day of play (4).

There are a number of reasons why team sport athletes may differ from endurance athletes in fluid losses and fluid intake practices. Team sports are generally of an intermittent nature, with higher intensity efforts interrupted by periods of minimal activity. It has been estimated that 45-55% of a netball game is played at between 85 and 95% maximum heart rate (23, 24). Top-level soccer players spend over two thirds of the entire match at 85% of maximal heart rate (3) due to the frequency of sprinting and acceleration, and they may cover up to 10-1 1 km per game (19). It is still unknown whether this form of intermittent exercise affects sweat losses differently than continuous, prolonged aerobic exercise.

There are several potential barriers to team sport players achieving adequate fluid intakes. Often, the-rules of the game can limit opportunities to drink, such as in soccer where games are played in 45-min halves during which fluids are not permitted on the field. Cricket has similar limitations, with fluids only allowed every hour at specified breaks. Gore et al. (6) reported that cricket fast bowlers lost an average of 4.3% body weight after two sessions of play on a hot day despite attempts to maintain fluid intake. In contrast, basketball players have frequent opportunities to drink during time-outs, between quarters, and when players are substituted. Gastrointestinal intolerance to fluids (perceived or actual) may also limit fluid consumption, especially when a large proportion of the game is played at high intensity (16). This may be more of aproblem in intermittent activities than in continuous exercise svorts.

Another issue in fluid requirements of team sports that requires further investigation is seasonal variation. Historically, team sports have been considered as "summer" or "winter" sports. Today, many team sports are played year-round (some in a modified form of the game, such as touch football) or begin their traditionally "winter" season at the end of summer. Although it is true that sweat rates are reduced in cooler temperatures, many factors may modify this. For instance, some team sports train and play indoors in all seasons (such as basketball and netball) in a controlled environment. Brown and Bannister (I) reported higher than expected sweat rates when activity was held indoors without the benefit of adequate convective cooling. Furthermore, athletes who train outdoors tend to adjust to cooler conditions by wearing more clothes, which may increase sweat rates. In such situations, if fluid intake decreases in response to cooler outdoor temperatures, a similar net fluid deficit may occur.

An example of this phenomenon can be seen by comparing two studies undertaken in Australian Football League players (13, 14). These studies showed that over a 105-min game, as the temperature decreased (38 "C, 27 "C, and 12- 15 "C), sweat losses decreased (3.63 L, 3.19 L, and 1.57 L, respectively) but fluid intakes decreased disproportionately to the reduction in sweat losses (1.5 L, 0.74 L, and 0.19 L, respectively), leading to a similar increase in rectal temperatures over

Fluid Intake / 309

the three temperature ranges (39.9 "C, 39.3 "C, and 39.6 "C, respectively). Finally, there may be a large inter- and intraindividual variability in the rates of fluid loss during exercise in team sports, due to differences in body size of athletes, variability in the exercise load in different games, differences in maximal oxygen uptake, and different field/court positions (9). In contrast, runners and cyclists tend to be more morphogenically homogeneous as a group and undertake more defined exercise loads, which makes recommending fluid intakes for the "typical" athlete more useful.

The aim of this study was to investigate voluntary fluid intake practices and fluid losses during summer and winter conditions in soccer, basketball, and netball training and competition sessions; our intent was to provide prac- tical consideration of and recommendations for meeting fluid needs in team sports.

Methods

Subjects

Members of the men's and women's basketball, men's soccer, and women's net-ball squads living and training at the Australian Institute of Sport (AIS), Canberra, over the period 1993-1995 were studied. Summer data from the Australian Senior Women's Soccer squad were also collected during matches played in Papua, New Guinea (qualifying matches for World Cup), and dur- ing training sessions in Canberra and Sweden. Basketball players and male soccer players were aged 16-18 years, netball players 18-21 years, and female soccer players 16-28 years. Subject characteristics are presented in Table 1. The study was approved by the Ethics Committee of the Australian Sports Com- mission. Informed consent was obtained; however, to ensure usual fluid intake practices, the purpose of the study was described as a monitoring of body weight changes.

Table 1 Subject Characteristics

Number of Height (cm) Weight (kg) Skinfold (sum of 7) Sport subjects M SD M SD M SD

Netball 22 177.9 4.5 74.23 6.95 105.1 21.9 Basketball

Men 19 197.5" 7.8 92.65" 8.33 63.1 15.4 Women 12 179.8 7.1 68.16b 5.42 83.5 12.1

Soccer Men 32 181.5 7.1 76.53 6.79 54.3 11.8 Women 17 166.9" 7.1 62.83" 8.54 93.6 32.0

Note. Sum of 7 skinfolds as per Telford et al. (20). aSignificantly different from male soccer players (p < .001). bSignificantly different from female basketball players @ < .05). "Significantly different from female basketball players (p < .001) and netball players @ < .001).

3 1 0 / Broad, Burke, Cox, et a/.

Measurements

Seven skinfold sites were measured by a qualified kinanthropometrist (minimum Level I), as described by Telford et al. (20): biceps, triceps, subscapular, supra- iliac, abdomen, thigh, and calf.

Testing sessions were chosen to represent a typical program of weight training, field/court training, and competition sessions over a 1-week period. Data were collected during a minimum of two matches, four training sessions, and two weight training sessions for each team except for women's soccer, which included datacollected over two matches and three training sessions. Testing was undertaken during two periods: the Australian summer months of January to March (outside temperature was greater than 20 "C at the commencement of each testing session) and the Australian winter months of June to August (outside temperature was less than 10 "C at the commencement of each testing session). Temperature and humidity were recorded at the beginning of each session on the playing arena using a Hygrotest 6200 probe (Germany). Testing sessions for basketball and netball occurred indoors; outdoor temperature and humidity were also recorded. Soccer training and competition sessions all occurred outdoors. All testing followed at least 3 weeks of acclimatization to the outdoor temperatures.

Protocol

The following standard protocol was designed to minimize disruptions to normal routine. Body weight was determined prior to the session with either a Precision Health Scale UC300 (A&D Company, Limited, Tokyo, Japan) or Wedderburn Scales DS-260 (Tanita Corporation, Japan), both accurate to 50 g; subjects wore minimal clothing during weighing and did not wear taping, ankle guards, orjewelry. Weights of full water bottles or other drink containers were taken on portable food scales (Salter Microtronic Electronic Kitchen Scale, Model 2001, Salter Housewares Limited, Kent, England, accuracy to 2 g). Throughout the session, water bottles were weighed, refilled, and reweighed when empty to monitor fluid intake. Athletes were supervised to ensure they drank only from their own water bottle (or other measured fluid) and to prevent spillage and use of water from bottles for purposes other than drinking. Fluid intake opportunity was determined by the observer and classified on one of three levels:

Level 1 : Fluid was readily available, and drinking was encouraged through regular breaks in sessions or players being handed a water bottle whenever they were not in play.

Level 2: Fluid was available but was difficult to obtain on a regular basis.. Level 3: Fluid generally was unavailable (no water bottles and/or no drink breaks).

If athletes needed to use the bathroom, body weight changes were determined. At the completion of the session, water bottles were removed and weighed, and athletes were reweighed after they toweled down and removed taping, prior to the consumption of any further fluid. The length of time spent by each athlete in training or competition was recorded. During competition, body weight was measured prior to warm-up, with athletes being allowed to drink during warm-up. Sweat losses and fluid intakes were expressed per hour of actual exercise time, including warm-up. Hence, fluid intakes per hour represent an overestimate of actual intake during

Fluid Intake / 31 1

exercise, since fluid was consumed during breaks in playltraining that were not included in the calculation of exercise time. Reported sweat loss may represent an underestimate of that occurring during actual competition activity, since it includes a warm-up that involved some lower intensity exercise.

Sweat losses were estimated using the following equation:

Sweat loss (g) = [change in body weight (g) + fluid intake (g) - (urine + fecal output) (g)].

No correction was made for respiratory water loss or metabolic fluid changes. In this article, dehydration (%DH) is defined as the body weight loss achieved during the exercise session and was calculated using the following equation:

(Body weight change - urine output) 1 initial body weight X 100.

Statistical Analysis

Three factors proposed to be important for fluid balance and for which data were collected were type of sport, type of activity, and season. Individual subjects on each team did not contribute data to every observed session. The means and standard deviations were calculated for the weight, height, and skinfolds of all members in each team who provided any measurement for this study. The means were compared between teams using independent sample Student's t tests.

Means and standard deviations for sweat loss, fluid intake, and percentage dehydration were calculated on the crude data stratified by sport, activity, and season. The differences between means for summer and winter for each activity and each sport were assessed using an ANOVA model with subject (the sportsperson) treated as a random effect. Subjects were therefore regarded as being randomly selected from a population of elite sportspersons within each sport. To assess differences in means between sports for each activity, an ANOVA model was used with sport (netball, men's basketball, women's basketball, men's soccer, and women's soccer) treated as a fixed effect. To control for season, season was once again entered as a fixed effect. Each subject played one sport only. Post hoc testing of means was by t tests with no adjustment for multiple comparisons.

Statistical analysis was carried out using the SAS statistical software package (SAS statistical software, SAS Institute Inc., 1989, SASISTAT User's Guide, Ver. 6,4th ed., SAS Institute Inc., Cary, NC, PROC GLM). All significance tests were two-sided, using a critical value of p = .05.

Results

Physical characteristics of subjects are presented in Table 1. These data indicate that, as a group, male basketball players were taller and heavier and had greaterbody fat levels than male soccer players 0, < .001). The mean height of female basketball players was not significantly different than that of netball players; however, mean body mass was significantly greater in netball players 0, c: .05). Female soccer players were on average shorter and lighter than the players from each of the other teams 0, < .001).

Table 2 represents the mean sweat rates, fluid intakes, and dehydration recorded for players of team sports in summer and winter conditions during weight training, training, and competition sessions. After we controlled for person and

31 2 / Broad, Burke, Cox, et a/ .

season, both mean sweat rate and mean fluid intake differed by type of session, being greatest in competition and least in weight training sessions ( p < .05), except for women's soccer (no significant difference in sweat rates between training and competition sessions). For all sports, mean %DH was significantly lower after weight training than after either training or competition sessions. For men's soccer, mean %DH was significantly greater after competition sessions than after training sessions, but in the other sports there was no significant difference after either competition or training sessions.

After we controlled for person and type of session, mean fluid intake rates were significantly greater in summer than winter for basketball players but were not different for netball players or male soccer players. Mean sweat rate was signifi- cantly greater for male basketball and male soccer players in summer, and mean %DH was significantly greater in summer for netball players and male soccer players. Mean %DH was significantly lower in summer for male basketball players, and there was no significant difference between summer and winter for female basketball players. Male and female soccer players had consistently lower mean fluid intake rates during training and competition and higher mean %DH than other sports; however, these differences were not significant. Male basketball players had significantly greater fluid intake rates than players for each other team sport, across all sessions. The durations and conditions of exercise sessions are summarized in Table 3. The variations in mean %DH between sports are presented in Figures 1-4.

Discussion

The results of this study provide descriptive data on sweat losses and fluid intakes in elite women's and men's basketball, women's and men's soccer, and women's netball teams during different training sessions. Overall, competition sessions elicited the greatest sweat rates and fluid intakes. Sweat rates, fluid intakes, and mean %DH varied between sports with few consistencies except for the highest %DH andlowest fluid intakes being reported by soccerplayers and the lowest %DH being maintained by netball players.

The type of exercise session undertaken by team sport athletes is an important determinant of sweat loss rates. Greater sweat rates were found during competition, with sweat rates during weight training significantly lower than competition or courdfield training. These differences are most likely due to varying intensities of exercise rather than environmental conditions. This assumption is supported by evidence from Woolford (23, 24), who found that time spent at high intensity in netball matches was greater than during training sessions.

An interesting finding, which was suggested in the introduction, was that dehydration in summer was not consistently greater than in winter for indoor team sports (i.e., netball and basketball). For instance, the degree of dehydration seen in male basketball players, after we controlled for the type of session and for subject, was greater in winter than in summer. This was due to a significantly greater fluid intake in summer despite little difference in sweat rates (viz., mean fluid intake during summer, 1,079 ml - hrl , represented 67% of sweat loss, while mean winter fluid intake, 917 ml . hrl , represented 58% of sweat loss). For netball players, there was no significant difference in mean fluid consumption by season, and the mean degree of dehydration was significantly lower in winter. After we controlled fortype of session and person, the mean sweat rates and the lack of difference in fluid intake

Table 2 Sweat Rates, Fluid Intakes, and Dehydration of Team Sport Players in Summer Versus Winter Exercise Sessions

Weights Training Competition

Sweat rate Fluid intake Sweat rate Fluid intake Sweat rate Fluid intake (mlihr) ( m a ) %DH (d lhr ) (mllhr) %DH ( d / h r ) (a) %DH

Sport M SD M SD M SD M SD M SD M SD M SD M SD M SD

Netball W S

Soccer Men-W Men-S Women-S

Basketball Men-W Men-S Women-W Women-S

aSignificantly different ( p <

different rate between seasons (p < .05) after controlling for person. bWithin each sport, sweat rate and fluid intake were significantly Z!

.05) from each other session type except where otherwise indicated. 'Not significantly different from fluid intake rate for competition. 2. 1l-

dNot significantly different from sweat rate for competition. "% Dehydration significantly different from each of the other session types after - 3

controlling for person. B B

Table 3 Exercise Conditions, Duration, and Number of Measurements (M f SD)

Netball Soccer

W S Men-W Men-S Women-S

Weight training No. measures Temp ("C) inside Temp (OC) outside Humidity (%) inside Humidity (%) outside FA

Coudfield training No. measures Duration (min) Temp (OC) inside Temp ("C) outside Humidity (%) inside Humidity (%) outside FA

Competition No. measures Duration (min) Temp PC) inside Temp ("C) outside Humidity (%) inside Humidity (%) outside FA

W --L

P

Basketball -. m

Men-W Men-S Women-W Women-S 2 .Q

m %

2 1 18 7 23 .m

20.1 f 0.0 22.5 f 0.0 20.9 21.4 f 0 . 4 g 11.4 f 2.5 27.1 f 3.9 15.2 15.8 f 0 . 3 37.0 k 0.2 52.1 f 6.4 65.9 48.6f 3.5 63.7 f 23.4 33.8 k 14.9 85.8 62.9 It 7.0 3 3 1 1

*FA = fluid availability rating.

Fluid Intake / 31 5

b Mens Soccer

c Womens Soccer

Mens Basketball

% Dehydration

Figure 1 - Distribution of %DH during summer training sessions.

b Mens Soccer

c Womens Soccer

d Mens Basketball

% Dehydration

Figure 2 - Distribution of %DH for summer competition sessions.

31 6 / Broad, Burke, Cox, et a/ .

% Dehydration

Figure 3 - Distribution of %DH during winter competition sessions.

a 17 Netball

b 17 Mens Soccer

c 17 Mens Basketball

d 17 Wornens Basketball

7

n 70 C 0 .- w crr 60

5 50

8

0

<I % 1 TO 1.9% 2 TO 2.9%

% Dehydration

a I

e - d

Figure 4 - Distribution of %DH during winter training sessions.

b -

b

a -

-

a Netball

b Mens Soccer

c Mens Basketball

d Womens Basketball

a -

d

a

-

d

b

-

-

I

-

e - d - d - e

b -

b

a

-

Fluid Intake / 31 7

between seasons led to a greater degree of dehydration for male soccer players in summer. These results indicate that athletes do not respond predictably or sensi- tively to environmental conditions to meet their fluid needs during exercise; fluid intake should be considered a learned behavior requiring encouragement and ready availability of fluid, rather than a spontaneous action.

Fluid intake is a major controllable variable in preventing dehydration during exercise. Factors influencing fluid intakes in this study included provision of an individual water bottle, proximity to water bottles during traininglcompetition, encouragement from coaches to drink, the rules of the game, duration and number of breaks or substitutions, and awareness of personal sweat rates. All of these areas need to be addressed if fluid intake practices are to improve, particularly for sports in which many of these factors impact heavily (such as soccer).

In this paper, changes in body weight during exercise were used to represent dehydration. This does not represent true hydration status since it fails to account for unabsorbed fluid in the gastrointestinal tract and urine unemptied from the bladder. However, the results suggest that the level of dehydration during exercise is influenced by individual differences and rates of fluid intake rather than rates of sweat loss. During weight training, dehydration is minimal across all sports due to low intensity and short duration of exercise; however, during training and matches, changing rates of fluid intake greatly affect dehydration. Greenleaf described this phenomenon as involuntary dehydration that begins at sweat rates greater than 280 g hr ' (7). Both basketball and netball players replaced 5675% of their sweat losses

during training and competition, whereas female soccer players achieved 49-54% replacement and male soccer players 3544%. Although these responses are comparable to those of Greenleaf's troops, who replaced only 50% of sweat losses when sweat rates were 1,200 g - hr', individual differences within and between sports remain an important variable.

The mean level of dehydration observed for each team for a session may not be sufficient to substantially affect performance. However, the standard deviation indicates that wide variance occurred within a sport. Within each sport, a proportion of players achieved high levels of dehydration (see Figures 1 4 ) , which suggests they could be specifically targeted for education to improve fluid intake, particu- larly in soccer and men's basketball.

Upper limits for fluid replacement set by maximal gastric emptying rates are 1-1.5 L - hr' for the average adult male but are decreased during high-intensity exercise (>75% VO,max), dehydration, and heat stress (17). In general, voluntary fluid intake by endurance athletes appears to be in the order of 400-600 ml . hr' (1 1). It has been suggested that the upper figure in this range might represent the maximum level of fluid intake, dictated by opportunity or comfort during endurance exercise. The results from this study indicate that fluid intakes of up to 1,800 ml - hr' can be tolerated by some individuals, particularly large males. Achieving larger fluid intakes requires practice, so players should be encouraged to drink early and take advantage of volume effects that facilitate gastric emptying (12) (i.e., drink the largest comfortable amount just prior to exercise and continually top up during exercise). The advice to practice these strategies in training remains sound, both to improve hydration and performance in training sessions and to learn good strategies for matches.

The fluid intakes and sweat rates of male soccer players in this study are similar to those of other studies (8, 10). Thus, this study confirms reports of poor hydration

31 8 / Broad, Burke, Cox, et a/.

practices of male soccer players. However, since this finding was apparent in training as well as matches, it is evident that factors other than the rules of the game limit fluid intake. In view of the number of factors influencing fluid intake in all sports, we have identified a number of strategies in each of the sports to better match fluid intake to fluid needs during exercise; these are summarized below.

Recommendations

The results of this study, and other considerations, give rise to several recommen- dations for each team sport. The following recommendations are applicable to all team sports:

1. Players should always begin a competition or training session fully hydrated. This will require complete rehydration from previous exercise sessions as well as attention to fluid intake throughout the day and prior to the next session.

2. Each player should be aware of his or her own fluid needs by regularly weighing before and after exercise sessions and several times over the year in different environmental conditions, and should work on training him- or herself to drink more.

3. A simple method of improving fluid intake and minimizing dehydration is to ensure that all players have a water bottle at each exercise session and that they are provided with opportunities and encouragement to use it. This not only improves access to fluid but also provides a means of monitoring how much has been consumed. Following are other ways of optimizing fluid intake opportunities:

Coaches should ensure that a supply of palatable, uncontaminated fluid is available. This may be a challenge in some locations (e.g., unsafe water supply in some countries, isolated playing arenas). Coaches should schedule specific drink breaks every half hour during training. Coaches and trainers should remind players to drink regularly, specifically targeting players who have particularly high fluid losses andlor poor intake practices. Players should drink during the warm-up period. Players should drink a small amount at each break in play (such as time-outs or breaks between quarterslhalves, during shooting practice, or at drill changes during training). Players should maintain a regular fluid intake when on the bench during a game.

4. A sports drink is a suitable fluid; carbohydrate supplementation may be of benefit during prolonged training sessions or competition situations such as tournaments where optimal preparation of body carbohydrate stores has not been possible between matches. Furthermore, in team sports where matches last 90 min or more (including warm-up), carbohydrate supplementation has been shown to improve performance.

The specific recommendations outlined below have been shown to be realistic fluid intake goals that are within the gastrointestinal comfort levels of most players. Importantly, the same goal applies for all outdoortemperatures, even when training or matches occur indoors.

Fluid Intake / 31 9

A general fluid intake goal for male basketball players is 800-1,400 ml - hrl and for female basketball players 600-1,000 ml . hr'. The amount of fluid that trainers or coaches should have available on court for a 2-hr training session or match is 30 L for men and 20 L for women, for squads of 10-12 players.

Netball

A general fluid intake goal for netball players is 600-1,000 ml - hrl. The amount of fluid that trainers or coaches should have available on court for a 2-hr training session or match is 20 L for a squad of 10-12 players.

Soccer

General fluid intake goals during training and competition are 800-1,200 ml . hrl for men and 600-1,000 rnl . hr' for women. This may not be realistic during matches due to the currentrules limiting fluidintake on the field; indeed, current fluid intakes are well below this level. However, since this level of intake is within gastric emptying rates at rest, theoretically it should be achievable during training sessions. It is important to pay attention to fluid intake in winter, as present evidence suggests this is not well attended to.

Realistic goals for a match include adequate hydration on the day of the match, with a preload of 500 rnl of fluid to be consumed in the 30 min before the match. At halftime, a minimum of 400 ml should be ingested, and complete rehydration should occur at the completion of the match. Opportunities to drink during matches should be taken, particularly in hot conditions; this may necessitate rule changes andlor strategies to allow intake of fluid from bottles placed near the field. Keeper's bottles should be placed behind the goal.

We recommend revision of current rules for soccer matches, particularly those played in summer conditions, to allow players more regular fluid intake. Such revisions may also be required in other team sports, such as cricket and football.

The amount of fluid required by men during a training session for 18 players is 40 L, and 35 L for a match. Women require 30 L during training and 25 L during a match.

References

1. Brown, S.L., and E.W. Bannister. Thermoregulation during prolonged actual and laboratory-simulated bicycling. Eur. J. Appl. Physiol. 54:125-130, 1985.

2. Buskirk, E.R., and S. Puhl. Nutritional beverages: Exercise and sport. In Nutrition in Exercise and Sports, J.F. Hickson and I. Wolinsky (Eds.). Boca Raton, FL: CRC Press, 1989, pp. 201-231.

3. Ekblom, B. Applied physiology of soccer. Sports Med. 350-60, 1986. 4. Elias, S.R., W.O. Roberts, and D.C. Thorson. Team sports in hot weather. Phys.

Sportsmed. 19(5):67-78, 1991. 5. Gopinathan, P.M., G. Pichan, and V.M. Sharma. Role of dehydration in heat stress-

induced variations in mental performance. Arch. Er~viron. Health 43(1): 15- 17, 1988. 6. Gore, C.J., P.C. Bourdon, S.M. Woolford, and D.G. Pederson. Involuntary dehydration

during cricket. Int. J. Sports Med. 14(7):387-395, 1993.

320 / Broad, Burke, Cox, eta/.

7. Greenleaf, J.E. Problem: Thirst, drinking behaviour and involuntary dehydration. Med. Sci Sports Exerc. 24(6):645-656, 1992.

8. Hawley, J.A., S.C. Dennis, and T.D. Noakes. Carbohydrate, fluid and electrolyte requirements of the soccer player: A review. Int. J. Sports Nutr. 4(3):221-236,1994.

9. Maughan, R.J. Fluid balance and exercise. Int. J. Sports Med. 13(Suppl. 1):S132-S 135, 1992.

10. Maughan, R.J., and J.B. Leiper. Fluid replacement requirements in soccer. J. Sports Sci. 12:S29-S34, 1994.

1 1. Noakes, T.D., B.A. Adams, K.H. Myburgh, C. Greff, T. Lotz, andM. Nathan. The danger of inadequate water intake during prolonged exercise. Eur. J. Appl. Physiol. Occup. Physiol. 57(2):210-219, 1988.

12. Noakes, T.D., N.J. Rehrer, and R.J. Maughan. The importance of volume in regulating gastric emptying. Med. Sci. Sports Exerc. 23:307-313, 1991.

13. Pohl, A.P., M.W. O'Halloran, andP.R. Pannall. Biochemical and physiological changes in football players. Med. J. Aust. 1:467-470, 1981.

14. Pyke, F.S. and A.G. Hahn. Body temperature regulation in summer football. Sports Coach 4:41-43, 1980.

15. Rehrer, N.J. The maintenance of fluid balance during exercise. Int. J. Sports Med. 15(3):122-125, 1994.

16. Saltin, B., and D.L. Costill. Fluid and electrolyte balance during prolonged exercise. In Exercise, Nutrition and Metabolism, E.S. Horton and R.L. Terjung (Eds.). New York: Macmillan, 1988, pp. 150-158.

17. Sawka, M.N. Physiological consequences of hypohydration: Exercise performance and thermoregulation. Med. Sci. Sports Exerc. 24(6):657-670, 1992.

18. Sawka, M.N., and K.B. Pandolf. Effects of water loss on physiological function and exercise performance. In Perspectives on Exercise Science and Sports Medicine: Vol. 3. Fluid Homeostasis During Exercise, C.V. Gisolfi and D.R. Lamb (Eds.). Carmel, CA: Benchmark Press, 1990, pp. 1-38.

19. Shepherd, R.J., and P. Leatt. Carbohydrate and fluid needs of the soccer player. Sports Med. 4: 164-176, 1997.

20. Telford, R.D., W.J. Egerton, A.G. Hahn, and P.M. Pang. Skinfold measures and weight controls in elite athletes. Excel 5(2):21-26, 1988.

21. Walsh, R.M., T.D. Noakes, J.A. Hawley, and S.C. Dennis. Impairedhigh-intensity cycling performance time at low levels of dehydration. Int. J. Sports Med. 15(7):392-398,1994.

22. Williams, M.H. Nutritional Aspects of Human Physical and Athletic Performance (2nd ed.). Springfield, IL: Charles C Thomas, 1985.

23. Woolford, S., and M. Angove. A comparison of training techniques and game intensities for national level netball players. Sports Coach October-December, pp. 19-21, 1991.

24. Woolford, S., and M. Angove. Game intensities in elite level netball: Position specific trends. Sports Coach April-June, pp. 28-32,1992.

Acknowledgments

We acknowledge the assistance of Mareeta Grundy for initial development of the testing protocol. This research was funded by Benivale Orchards Pty Ltd. (Isosport sports drink).

Manuscript received: July 21, 1995 Accepted for publication: April 30, 1996