The Impact of Specific High-Intensity Training Sessions on Football Referees’ Fitness Levels Matthew Weston,* BSc(Hons), MSc, Werner Helsen,*† PhD, Clare MacMahon,‡ BSc, MSc, and Don Kirkendall,§ MD, PhD From the *Department of Kinesiology, Katholieke Universitat Leuven, Belgium, the ‡Department of Kinesiology, McMaster University, Hamilton, Ontario, Canada, and the §Department of Orthodaedics, University of North Carolina Background: In comparison to the amount of literature that has examined the match demands of football refereeing, there has been little attempt to assess the impact of high-intensity training. Purpose: The main goals were to get a better understanding of the long-term effect of specific intermittent training. Study Design: The authors examined the cardiovascular strain of specific high-intensity training sessions and also their impact on referees’ fitness levels. Methods: To examine the physical workload during intensive intermittent training sessions, heart rates were recorded and analyzed relative to the referees’ maximum heart rate (HRmax). To assess the referees’ fitness levels, the Yo-Yo intermittent recovery test was used. Results: Both the pitch- and track-training sessions were successful in imposing an appropriate high intensity load on the referees, at 86.4 ± 2.9% and 88.2 ± 2.4% HRmax, respectively. Following 16 months of intermittent high-intensity training, referees improved their performance on the Yo-Yo intermittent recovery test by 46.5%, to a level that is comparable with professional players. Conclusions: As match officials are subjected to a high physical load during matches, they should follow structured weekly training plans that have an emphasis on intensive, intermittent training sessions. Keywords: intermittent training; Yo-Yo test; football refereeing The physiological characteristics of football referees have received an increasing amount of focus in the scientific literature over the past decade. Studies have examined referee profiles, both physiological and anthropometric,19 as well as the movement patterns and physiological load experienced during actual match play.6,7,10–12,21 Match analysis by D’Ottavio and Castagna7 demonstrated that referees cover a mean ± SD distance, over the course of an average 95-minute match, of 11469 ± 983 m, with walking constituting 957 m, 4174 m of low-intensity running, 2585 m of medium-intensity running, and a total of 1556 m of highintensity running. Running at maximal speed accounted for 608 m, and nonorthodox directional modes—that is, sideways and backward running—accounted for 181 m and 1315 m, respectively.7 Heart rate recordings collected during matches demonstrated that the mean cardiovascular strain imposed on referees during match is approximately 85% of maximal heart rate (HRmax).9,11,21 However, as referees have been reported to spend approximately 75% of their total match time either standing, walking, or jogging,11 the amount of high-intensity activity, performed at speeds of more than 15 km/hour during a match, is a better indicator of the physical demands of matches.3,11,12 Given the high physical load imposed on top-class referees during actual match play,7,9,11 fitness levels need to be sufficient enough for the referees to be able to cope with the demands of their games through keeping up with play at all times and ensuring optimal viewing positions. This in itself is a challenge. However, when combined with the fact that referees are on average 10 to 15 years older than their playing counterparts6,7,9 and aging has a negative effect on fitness levels,20,23 referees have to work extremely hard in training to ensure that they attain, and maintain, an appropriate level of fitness. Also, in most countries referees † Address correspondence and reprint requests to Dr. Werner Helsen, Katholieke Universiteit Leuven, Tervuursevest 101, Leuven 3001, Belgium. The American Journal of Sports Medicine, Vol. 32, No. 1 Suppl. DOI: 10.1177/0363546503261421 © 2004 American Orthopaedic Society for Sports Medicine 54S Vol. 32, No. 1 Suppl., 2004 still work full-time, and their physical training sessions often have to be arranged around work commitments. Therefore, to ensure referees can attain an optimal level of match fitness, emphasis within their fitness preparation programs has to be firmly placed on quality structured training sessions that provide an appropriate training stimulus to enable the attainment of such fitness levels, especially with training time being at such a premium. To overcome this problem, the governing bodies of European and world football, the Union des Associations Européennes de Football (UEFA) and the Fédération Internationale de Football Association (FIFA), have moved to further professionalize the football referees’ environment over recent seasons. After all, the game has become more athletic and faster in recent years,22 and the referee or match official can no longer remain outside these developments. For referees, this is demonstrated by the fact that Catterall et al.6 reported a mean ± SD distance covered by the referees during matches in 1993 of 9438 ± 707 m, whereas some 8 years later it was reported to be 11376 ± 1600 m.7 Some suggest that referees should be provided with high-tech tools, for example, to detect if the ball has fully crossed the goal line or to reexamine a tricky piece of action using video replays of the ongoing game. However, others question these kinds of technical assistance, arguing their rejection in terms of the need to preserve the human input and therefore imperfect nature of the game. The point of view we take in this research study is to provide the referees with contemporary, match-specific physical preparation sessions, which replicate the physical demands of football refereeing. Despite the amount of literature that has profiled the match demands of football refereeing, to our knowledge there is no published literature in football, or any other team sports, that has examined the long-term effect of structured training programs and, more specifically, highintensity training on referees’ fitness levels. Given that the performance of referees during matches has been reported to be equal to that of midfield players and even greater than that of other outfield players,17 it is surprising that the Krustrup and Bangsbo11 study has been the only attempt to date to examine the impact of training on referees’ fitness levels. The characteristics of their training sessions were based around high-intensity running at 90% HRmax with a work-rest ratio of 2:1. The sessions were intermittent, with either long intervals (4 or 8 minutes) or short intervals (30 seconds or 1 minute). However, due to the short-term nature of the training study, the long-term effects of such high-intensity running on referees’ fitness levels remain unexamined. Also, despite reporting the intensity at which the high-intensity running was performed, the study failed to report any of the actual training heart rate data. With this in mind, the aim of this study was to present the type and exact intensity of high-intensity intermittent training sessions that are being prescribed to top-class UEFA and FIFA referees as part of their weekly training regime and how these types of sessions affect referees’ heart rates during the training sessions. Also, the longterm effectiveness of such training sessions with regard to High-Intensity Training for Football Referees 55S referees’ fitness levels, as determined by the Yo-Yo intermittent recovery test, was also examined. METHOD Participants The participants for this study consisted of 18 Belgian referees who were separated into two groups according to ability. The first group consisted of 7 Belgian referees (mean age 37.8 ± 4.12 years) who were also listed FIFA international referees. The referees had 13.7 ± 4.15 years of experience refereeing within their own country and 4.4 ± 4.16 years of experience refereeing internationally. The second group consisted of 11 Belgian elite referees (mean age 39.28 ± 4.21 years). These referees were affiliated with the Belgian football federation (KBVB) and only refereed nationally in the first and second Belgian leagues. The data were collected as part of the ongoing physiological support provided to the Belgian referees’ committee. Physical Workload During Training Sessions Match-specific weekly training plans for the referees were introduced from the beginning of the 2001/2002 Belgian football season. This involved the referees receiving a training plan by e-mail for each week of their competitive season (and also preseason), with the plan being based around four sessions per week: recovery, high intensity, speed endurance, and speed training. The focus of this study is on the specific high-intensity training sessions. However, it would be impractical to present all the highintensity sessions that the referees have been prescribed over the past two seasons since their training regimes became match specific in terms of the physical demands of football refereeing. Therefore, the sessions outlined in this study are actual examples of the high-intensity training sessions included in the referees’ weekly training plans. The referees’ heart rates during the training sessions were recorded via short-range telemetry using Polar S610 watches (Polar, Kempele, Finland) with the data being recorded every 5 seconds. Following each group training session, the data were downloaded onto a computer for further analysis. Analysis was performed using the Polar Precision Performance software, version 3.0 (Polar, Kempele, Finland). There are two types of sessions presented, depending on if they were performed on a football pitch (exercises 1 and 2) (Fig. 1) or on a 400-m running track (exercises 3 and 4). For the purpose of this study, only the heart rate data recorded during the high-intensity and active-recovery running within sets are reported. The heart rate data during the recovery in between sets were not included in the analysis. For exercise 1, all referees perform the exercise in pairs. On a signal from the coach, one referee (A) from each pair starts running at 90% HRmax clockwise around the diagonal carrying a vest. The partner referee (B) from each pair walks on the outside part of the diagonal running area. 56S Weston et al. The American Journal of Sports Medicine B A A B A B cone 4. They then perform the same exercise but in reverse order so that set 1 consisted of 8 high-intensity runs, of various durations, at 90% HRmax. A 5-minute active-recovery break was taken after set 1 was completed, and then the whole proceedure was repeated 3 times giving a total of 24 runs of varying distances at 90% HRmax over the whole training session. Along with the two pitch-based high-intensity training sessions, two specific interval training sessions at 90% HRmax are also presented. These sessions were performed around a running track, and the structure of these sessions was as follows: EXERCISE 1 Exercise 3 30 seconds at 90% HRmax followed by 30 seconds of active recovery C walking 60 seconds at 90% HRmax followed by 60 seconds of active recovery 20" B 1 jogging 10" A 2 3 90 seconds at 90% HRmax followed by 90 seconds of active recovery 4 120 seconds at 90% HRmax followed by 120 seconds of active recovery 75 m 100 m 135 m 170 m 34" 15" 20" 27" EXERCISE 2 xx Start Figure 1. Illustration of two different types of high-intensity pitch-training sessions performed by the referees in study 2. Referee A has to cover 3 clockwise high-intensity runs around the diagonal course. Once this has been completed, the referees exchange roles by handing over the vest. Referee B then completes 3 clockwise runs of the diagonal, while referee A walks (or jogs) on the outside part of the diagonal running area. Once this has been completed, again the referees switch roles by exchanging the vest, and referee A now performs 2 clockwise runs of the diagonal. The referees then repeat this sequence down to 1 clockwise diagonal run and then reverse the protocol to 1, 2, and 3 diagonal runs, respectively (resulting in 12 diagonal runs of 150 m each, or approximately 1800 m in total). After a 2- to 3-minute recovery break, the same diagonal run is now performed but counterclockwise. Each referee now has to cover 1 round at the first run, then 2, 3 and again 3, 2, 1 (for 12 diagonal runs of 150 m each or approximately 1800 m in total). The running was performed at 90% HRmax, as the main training objective was high intensity (85% to 95% HRmax). Exercise 2 consists of three sets of the following highintensity runs. The referees start jogging in pairs from the start position (A) to cone B, then walk from cone B to cone C. From cone C, they then run at 90% HRmax around cone 1 back to the start position. For lap 2, the referees repeat the sequence from cone A to cone C, but this time they turn around cone 2, on lap 3 around cone 3, and on lap 4 around 90 seconds at 90% HRmax followed by 90 seconds of active recovery 60 seconds at 90% HRmax followed by 60 seconds of active recovery 30 seconds at 90% HRmax followed by 30 seconds of active recovery Two sets, with 5-minute active recovery in between sets Exercise 4 30 seconds at 90% HRmax followed by 30 seconds of active recovery 45 seconds at 90% HRmax followed by 45 seconds of active recovery 60 seconds at 90% HRmax followed by 60 seconds of active recovery 75 seconds at 90% HRmax followed by 75 seconds of active recovery Repeated twice, followed by 5-minute active recovery before set 2: 75 seconds at 90% HRmax followed by 75 seconds of active recovery 60 seconds at 90% HRmax followed by 60 seconds of active recovery 45 seconds at 90% HRmax followed by 45 seconds of active recovery 30 seconds at 90% HRmax followed by 30 seconds of active recovery This sequence was also repeated twice. Vol. 32, No. 1 Suppl., 2004 Training Monitoring As previously mentioned, from the beginning of the 2001/2002 season, the referees received a training plan, sent via e-mail, at the start of every week. Obviously, the structure in terms of training intensity, frequency, and duration varied with the stage of the season and also with the game demands of the referees. However, in general the plans consisted of four training sessions per week. The first training session was a general recovery from the weekend match, sessions 2 and 3 were high-intensity intermittent sessions, with session 3 periodically being speed endurance. The final training session of the week, training session 4, was always a match-preparation session where the focus was on short sprints with sufficient recovery in between. Heart rates were used as the tool to prescribe and control the training load of these sessions, as this method has been demonstrated to be a reliable index of cardiovascular strain.1,18 For the international referees, adherence to the training plan was mandatory, and performance during weekly training sessions was analyzed by means of a monthly training dairy filled in by the referees in which the mean and maximum heart rates during the training sessions, and also the number and distribution of their weekly training sessions, were recorded. Following analysis of the individual heart rate data, the referees were informed as to whether the aims of their training sessions had been met in terms of time spent working within the appropriate activity category. Every Tuesday during the preseason and competitive season, the international referees attended a group training session at the Katholieke Universiteit Leuven. On these occasions, the referees performed a highintensity intermittent session, such as those presented previously, and these sessions were designed and strictly administered by the second author of this study. The elite referees were instructed to follow the weekly training plans and also to keep a training diary. However, this group did not receive any feedback with regard to their training diaries and heart rate data obtained during their training sessions. The group training sessions for the elite referees took place every 6 weeks at the Katholieke Universiteit Leuven, where again the referees performed a high-intensity intermittent session designed and strictly administered by the second author of this study. High-Intensity Training for Football Referees 57S below 40 b.min–1, the data were corrected using the error correction algorithm within the Polar Precision Performance software and checked by visual inspection of the data. If there were still periods of more than 30 seconds that remained uncorrected after this procedure, then the data were excluded from analysis. The referees’ HRmax was calculated using the highest 5second peak value observed during matches, high-intensity/ speed endurance training sessions, or the Yo-Yo intermittent recovery test, which has been demonstrated to produce a maximal physiological response.13 Using the highest 5-second peak obtained during either practice, tests, or competition enables a more accurate determination of HRmax than a single maximal fitness test, as Palmer et al.15 and Helsen and Bultynck9 reported consistently higher HRmax values during competition when compared to maximal laboratory tests. For the determination of the various heart rate zones relative to the referees’ HRmax, a distinction was made between five different activity categories that were based on findings within the football literature4–5 and were also in line with previous studies on the physiological demands of football refereeing.7,9 These categories were the following: 1) maximal effort (ME, >95% HRmax), 2) high intensity (HI, 86% to 95% HRmax), 3) low intensity (LI, 76% to 85% HRmax), 4) active recovery (AR, 65% to 75% HRmax), and 5) passive recovery (PR, <65% HRmax). Data Analysis Differences in the referees’ mean and peak heart rates (expressed as a percentage of HRmax), and also exercise duration, between the high-intensity pitch- and tracktraining sessions were determined by a paired t-test. To further explore impact differences between two types of training sessions, a two-way analysis of variance (ANOVA) was performed using a 2 (training) (pitch sessions, track sessions) by 5 (activity category) (ME, HI, LI, AR, and PR) design with repeated measures. For the analysis of the distance covered on the Yo-Yo test, a two-way ANOVA was performed using a 2 (group) (international referees, elite referees) by 4 (testing period) (four separate testing occasions) repeated measures design. When a significant interaction was detected, the data were subsequently analyzed using a Scheffe test. Significance was set at P < 0.05. Yo-Yo Intermittent Recovery Test RESULTS To assess the long-term efficacy of the referees’ training programs, the referees were assessed using the Yo-Yo intermittent recovery test. Specific details of this are given in Krustrup et al.13 The referees were assessed on four separate occasions: November 2001, January 2002, October 2002, and March 2003. Data Reduction During data analysis, when data appeared to be recorded outside the normal physiological range expected for the match officials, that is, more than 210 b.min–1 (beats per minute) or There were no significant differences between the mean percentage HRmax recorded by the referees during the pitchand track-training sessions (86.4 ± 2.9 versus 88.2 ± 2.4 %HRmax, P > 0.05) despite the pitch sessions being shorter in duration than the track sessions (19.9 ± 0.5 versus 27.0 ± 5.19 minutes, P < 0.05). However, there were differences between the peak heart rates recorded during the sessions, with the peaks being higher during the track sessions (94.8 ± 2.6 versus 97.4 ± 2.1 %HRmax, P < 0.05). The analysis of the time spent working within the five different heart rate zones showed a significant effect only 58S Weston et al. The American Journal of Sports Medicine TABLE 1 Mean Values and Standard Deviations of the Distance Covered on the Yo-Yo Intermittent Recovery Test for Both Belgian International (n = 7) and Elite Referees (n = 12) (in meters) International referees Mean SD Elite referees Mean SD January 2002 October 2002 March 2003 1720a 276 1735a 318 2167a 234 2330a 268 1290 407 1431 409 1777 393 1985 279 80 70 60 50 40 30 20 10 0 % Practice time November 2001 PITCH TRACK Low Passive intensity recovery Active High recovery intensity Maximal effort a Significant difference between international and elite referees (P < 0.05). Figure 2. Means and standard errors of the percent practice time spent in the different activity categories during highintensity training sessions. 2200 2100 2000 Yo-Yo intermittent recovery test (in m) for activity category (F4,104 = 144.52, P < 0.0001). In line with the training objectives, the referees spent significantly more time training within the high-intensity heart rate zone than in the other four heart rate zones (Fig. 2). Table 1 provides an overview of the means ± SD of the distances covered by the referees during the Yo-Yo intermittent recovery test. The results of the statistical analysis showed a significant effect only for group (F1,16 = 6.152, P < 0.0246) and testing period (F1,3 = 35.330, P < 0.0001). Specifically, international referees outperformed the elite referees on each testing occasion. From Figure 3, it becomes clear that the mean distance covered on the test increased significantly from test 1 to test 3 (P < 0.01), from test 2 to test 3 (P < 0.01), test 1 to test 4 (P < 0.01), and test 2 to test 4 (P < 0.01). However, mean distance covered did not increase between test 1 and test 2 (P > 0.05) and between test 3 and test 4 (P = 0.0579), although this value did approach conventional levels of statistical significance. Figure 4 presents a comparison of a referee’s heart rate response from test 1 and test 4. It is clear that the improvement in performance by this particular referee was, as expected, concomitant with a decrease in submaximal heart rates throughout the duration of the test along with an increase in the amount of work performed while working at HRmax. 1900 1800 1700 1600 1500 1400 1300 II I III IV Test periods Figure 3. Mean values and standard errors of the distance covered during the Yo-Yo intermittent recovery test for the total group of match officials. 200 Nov-01 190 Mar-03 180 -1 Heart rate (b.min ) 170 150 140 130 120 110 21:05 20:05 19:05 18:05 17:05 16:05 15:05 14:05 13:05 12:05 11:05 10:05 09:05 08:05 07:05 06:05 05:05 04:05 03:05 02:05 100 01:05 In this study, we examined the long-term effect of structured training, and more specifically, high-intensity training sessions, on football referees’ fitness levels. We used heart rate analysis to determine the physical load imposed on the referees during training sessions as this method has been demonstrated to be an accurate indicator of overall cardiovascular strain10,18 and is the most commonly used method of assessing exercise intensity used in the field.1 To determine the impact of such training sessions, we used the Yo-Yo intermittent recovery test as this test has been 160 00:05 DISCUSSION Test duration (min:sec) Figure 4. An example of a referee’s heart rate response during two separate testing occasions on the Yo-Yo test. Vol. 32, No. 1 Suppl., 2004 reported to be highly correlated to referees’ match performance.11 To our knowledge, this is the first study to report actual heart rate responses during specific intensive intermittent training sessions and to also examine the long-term impact of such training sessions on referees’ fitness levels. The results obtained in the present study illustrate that both the pitch- and track-training sessions were equally successful in imposing a high-intensity training load on the referees as the mean heart rates for both types of training session fell within the high-intensity category highlighted by previous authors.4,5,9 Therefore, for those referees who do not have the opportunity to train together in groups or have access to pitches (geographical, work constraints, and so forth), an appropriate high-intensity training session can be achieved through specific track running. From a coaching perspective, it was also encouraging to see the lack of variability in the heart rate recordings during the training sessions from referee to referee. There was a relatively small difference between the percentage of time spent working within the maximal effort and high-intensity activity categories during the training sessions and the data reported during the UEFA 2000 Championships9 (71.3% versus 60.7% of total time, respectively) and also within the scientific literature.7,11,12,21 Therefore, despite the fact that the heart rates during matches may represent an exaggeration in terms of the physical load placed on the referees, due to emotional stress, rising core temperature, and/or dehydration,11 the similarity between these data suggests that the training sessions have a high level of cardiovascular specificity with regard to the match demands of refereeing. For the referees to achieve optimal levels of physical fitness, it is crucial that they train at and above the intensity at which they referee as training adaptations are highly specific to the type of activity and to the volume and intensity of the exercise performed.23 The recommended guidelines for maintaining and improving aerobic fitness have been reported to be an intensity of training between 55%/65% and 90% of HRmax, with the lower intensity values being most applicable to individuals who are quite unfit, at a frequency of 3 to 5 times per week.2 However, Krustrup and Bangsbo11 and Helgerud et al.8 have since reported that specific highintensity aerobic training, involving a session total of 16minute high-intensity running performed 2 to 3 times per week, improved fitness levels in football referees and players, respectively. Specifically, for referees, Krustrup and Bangsbo11 reported that following a 12-week period of intense, intermittent training (90% HRmax) performed 3 times per week, the distance covered by referees on the YoYo test improved by 31%. For players, Helgerud et al.8 reported that 8 weeks of high-intensity training sessions, performed twice per week at 90% to 95% HRmax, brought about a significant increase in VO2max, blood lactate threshold, and running economy in football players. With this in mind, we have demonstrated that the highintensity pitch- and track-training sessions presented in this study are of a suitable intensity (86.4 ± 2.9 and 88.2 ± 2.4 % HRmax, respectively) and duration (19.9 ± 0.5 and 27.0 High-Intensity Training for Football Referees 59S ± 5.19 minutes, respectively) to bring about significant training adaptations in referees. However, it is important to highlight that when the training load becomes too intense, the body’s ability to recover is impaired, which may lead to underperformance.23 Therefore, in line with Krustrup and Bangsbo11 and Helgerud et al.,8 we recommend that the referees do not perform more than two to three high-intensity training sessions within 1 week. Krustrup et al.12 reported that games are characterized by long, low-intensity activity periods interspersed with frequent brief intense exercise periods. Therefore, speed endurance and good recovery in between intense periods of exercise are vital for referees and assistant referees.12 Consequently, the training programs that referees follow on a weekly basis should have a blend of high- and lowintensity aerobic sessions, complemented with training sessions dedicated toward the improvement of speed and the ability to perform repeated bouts of intense exercise with incomplete recovery (speed endurance). Previous work11 has examined only the short-term (12 weeks) effect of intense, intermittent training in relation to referees’ fitness levels. However, our results clearly demonstrate that over a period of almost two football seasons, following a structured training regime that replicated match demands through intermittent high-intensity training, referees significantly improved their aerobic and anaerobic fitness levels as both energy systems have been reported to be highly taxed during the Yo-Yo intermittent recovery test.13 What is even more encouraging was that both groups of referees continued to progressively improve their fitness levels with each test. The international referees had significantly higher levels of fitness at the start of the study when compared to the elite referees. This finding is consistent with Rontoyannis et al.19 who also reported that fitness levels were lower among those referees from a lower standard of competition. It has been demonstrated that the initial level of fitness largely governs the training response with relative improvements being greater for less fit individuals.23 Our results are consistent with this finding as the elite group of referees, whose initial level of fitness was significantly lower than the international referees, demonstrated an average 53.9% improvement from test 1 to test 4. However, as a result of comprehensive feedback of their training diaries and training data, along with weekly group-training sessions, the international referees still improved their performance on the Yo-Yo intermittent recovery test by an average 35.5% from test 1 to test 4. Despite demonstrating a higher initial fitness level when compared to the elite referees, the international group of referees also managed to improve their performance on the Yo-Yo test with every subsequent testing session. Also, the distances attained on the Yo-Yo intermittent recovery test during tests 3 and 4 were highly comparable with professional football players as Krustrup et al.13 reported a mean ± SD distance covered during the Yo-Yo intermittent recovery test of 2075 ± 156 m for 17 players, with a mean age of 28 years (range, 25 to 36). However, when training referees it is important not to lose sight of the main goal, which is improving match per- 60S Weston et al. formance. It is no good improving fitness levels if it does not improve match performance. It has previously been demonstrated that intense intermittent training does not only improve fitness levels in football referees, but this improvement in fitness is concomitant with an improved match performance.11 Specifically, the amount of highintensity running performed by the referees increased by 23 ± 8%, mostly because of a marked increase in highintensity running during the second half.11 Also, the number of high-intensity running bouts was higher after than before the intermittent training, and the highest recorded values of distances from match incidents were less after the intermittent training.11 With this in mind, we are confident that improvements in the referees’ fitness in the present study will have had a positive impact on the field of play as the Yo-Yo intermittent recovery test is highly correlated with a referees’ match performance.11 This improvement in match performance will have been achieved through a marked increase in the referees’ ability to perform repeated intense work, thus enabling the referee to keep up with play more efficiently and obtain optimal viewing positions, especially during the second half of matches when literature has consistently demonstrated a decreased work rate by the referee during the second half of matches owing to fatigue.6,7,11,12,16 The training sessions and results presented within this study may also have implications for referees involved in other field sports such as hockey and rugby where inadequate fitness levels may impair the referees’ ability to keep up with play and subsequently make the correct decisions. For example, Martin et al.14 reported that rugby referees are required to have well-developed anaerobic energy systems to enable them to respond to fast passages of play and to be “on the spot” to make decisions. Therefore, the intensive intermittent training sessions performed by the football referees in the present study may help to develop appropriate levels of aerobic and anaerobic fitness levels in referees involved in other team sports. The absence of any significant improvements between test 1 and test 2 can be attributed to the timing of the testing sessions; only 2 months separated the two tests. The fact that the testing sessions were not performed at evenly spaced intervals over the duration of the study period was a limitation associated with the study design. This was due to the problems associated with getting all the referees to the testing venue at the same time as the referees within this study have full-time professions outside that of football, and very often work and match commitments interfere with training and testing schedules. Also, another limitation was the lack of a control group against which to gauge the improvements in fitness. However, as fitness has been reported to decrease with age,20,23 the fact that the referees improved their performance on the Yo-Yo test from November 2001 to March 2003 by an average 46.5% clearly demonstrates the effectiveness of such training session regimes for referees. Also, the involvement of age-matched sedentary subjects as a control group would not have been representative of a football refereeing population and has not been used in previous work.11 The American Journal of Sports Medicine CONCLUSION The data of the present study demonstrate that a structured training program with intensive intermittent training sessions can significantly improve the fitness levels of football referees, as determined by the Yo-Yo intermittent recovery test. The improvement in fitness levels will have helped improve the referees’ match performance through an increased ability to perform high-intensity running, thus enabling them to better keep up with play and to obtain optimal viewing positions in relation to critical match incidents. ACKNOWLEDGMENT We would like to thank FIFA's F-MARC (Medical Assessment and Research Centre) for their financial support in this project. We also sincerely thank all the members of the FIFA and the Belgian referees' committee for their collaboration in this project. The cooperation of the Belgian match officials was of invaluable importance. We are also very grateful to Roel De Clerck for assistance with the data reduction. Finally, we want to acknowledge Dr. Jiri Dvorak and Dr. Astrid Junge for providing us with valuable comments on earlier drafts of the paper. REFERENCES 1. Achten J, Jeukendrup A: Heart rate monitoring: Applications and limitations. Sports Med 33(7): 517–538, 2003 2. American College of Sports Medicine: The recommended quantity and quality of exercise for developing and maintaining cardiorespiratory and muscular fitness, and flexibility in healthy adults: Position stand. Med Sci Sports Exerc 30(6): 1998 3. Bangsbo J, Nørregaard L, Thorsøe F: Activity profile of competition soccer. Int J Sports Med 13: 125–132, 1991 4. Bangsbo J: The physiology of soccer—with special reference to intense intermittent exercise. Acta Physiol Scand 151(suppl): 619, 1994 5. Bangsbo J: Energy demands in competitive soccer. J Sports Sci 12: S5–S12, 1994 6. Catterall C, Reilly T, Atkinson G, et al: Analysis of the work rates and heart rates of association football referees. Br J Sports Med 27: 193–196, 1993 7. D’Ottavio S, Castagna C: Analysis of match activities in elite soccer referees during actual match play. J Sports Med Phys Fitness 41: 27–32, 2001 8. Helgerud J, Engen LC, Wisloff U, et al: Aerobic endurance training improves soccer performance. Med Sci Sports Exerc 33(11): 1925–1931, 2001 9. Helsen WF, Bultynck JB: Physical and perceptual-cognitive demands of top-class refereeing in association football. J Sports Sci: In press 10. Johnston L, McNaughton L: The physiological requirements of soccer refereeing. Aust J Sci Med Sport 26(3/4): 67–72, 1994 11. Krustrup P, Bangsbo J: Physiological demands of top-class soccer refereeing in relation to physical capacity: Effect of intense intermittent exercise training. J Sports Sci 19(11): 881–891, 2001 12. Krustrup P, Mohr M, Bangsbo J: Activity profile and physiological demands of top-class soccer assistant refereeing in relation to training status. J Sports Sci 20: 861–871, 2002 13. Krustrup P, Mohr M, Amstrup T, et al: The Yo-Yo intermittent recovery test: Physiological response, reliability and validity. Med Sci Sports Exerc 35(4): 697–705, 2003 Vol. 32, No. 1 Suppl., 2004 14. Martin J, Smith N, Tolfrey K, et al: Activity analysis of English premiership rugby football union refereeing. Ergonomics 44(12): 1069–1075, 2001 15. Palmer G, Hawley JA, Dennis SC, et al: Heart rate responses during a 4-d cycle stage race. Med Sci Sports Exerc 26(10): 1278–1283, 1994 16. Rebelo N, Krustrup P, Soares J, et al: Reduction in intense intermittent exercise performance during a soccer match. J Sports Sci 16: 482–483, 1998 17. Reilly T, Thomas V: A motion analysis of work-rate in different positional roles in professional football match play. J Hum Mov Stud 2: 87–97, 1976 18. Reilly T: Energetics of high intensity exercise (soccer) with particular reference to fatigue. J Sports Sci 15: 257–263, 1997 High-Intensity Training for Football Referees 61S 19. Rontoyannis GP, Stalikas A, Sarros G, et al: Medical, morphological and functional aspects of Greek football referees. J Sports Med Phys Fitness 38(3): 208–214, 1998 20. Stamford BA: Exercise and the elderly. Exerc Sport Sci Rev 16: 341–379, 1988 21. Weston M, Brewer J: An investigation into the physiological demands of soccer refereeing. J Sports Sci 20: 59–60, 2002 22. Williams AM, Lee D, Reilly T: A Quantitative Analysis of Matches Played in the 1991–92 and 1997–98 Seasons. London, Football Association, 1999 23. Wilmore JH, Costill DL: Physiology of Sport and Exercise. Second edition. Champaign, IL, Human Kinetics, 1998