Repository logo

Long distance running in Ethiopian athletes: a search for optimal altitude training.

Thumbnail Image



Journal Title

Journal ISSN

Volume Title



ABSTRACT Background The extended dominance of Ethiopian and Kenyan middle- and long- distance athletes in world athletics has resulted in researchers proposing numerous explanations to explain this success. Genetic predisposition, anthropometric, physiological, biochemical and biomechanical characteristics, environmental factors like living and training at high altitude, active lifestyles during childhood as well as nutritional practices, have all been major focus areas of past studies that involved east African endurance athletes. Of all the proposed variables, researchers have acknowledged the positive role of environmental factors in the success of these athletes. Despite the past attempts to investigate the major factors that contributed to the successes of east African athletes, to the best of the authors’ knowledge, limited studies have addressed each of the proposed physiological and environmental variables in the Ethiopian athletes, compared with the number of studies conducted on Kenyan athletes. Purpose The primary purpose of this research was to test a natural altitude training model and examine whether it enhanced the long-distance performance of junior Ethiopian athletes. The research also examined a variety of environmental factors associated with these junior athletes that included daily distance travelled to and from school, mode of transport to and from school as well as physical activity patterns after school. These factors were compared between the junior athletes who participated in the altitude training study, current and retired World and Olympic level long-distance Ethiopian athletes. The energy intake, macronutrient breakdown and energy expenditure of the junior athletes during the altitude training camp were also analysed. Methods Demographic Characteristics Study: A total of 83 endurance runners were involved in this study. The athletes were classified into three separate groups based on their current performance status and age as retired elite (n = 32), current elite (n = 31) and academy junior athletes (n = 20). The average ages of the athletes in the three groups were 38±7.6, 25±4.5, and 18±1.2 years for retired elite, current elite and junior group athletes, respectively. The study primarily employed a xvi questionnaire survey design to gather the demographic characteristics of the athletes. Along with the questionnaire, the altitudes where the athletes were born and trained, as well as the home to school distance of the athletes were measured. Data were collected from the retired athletes through both self- and interviewer-administered questionnaires and forms. Self-administered questionnaires were used to collect data from the current elite and academy junior athletes. Only 46.8% of the retired elites (n=15) filled in the questionnaire and the rest 53.2% (n=17) of the retired elite athletes responded to the questionnaire via telephone. The home to school distance of 71.8% (n = 23) and 58.1% (n = 18) of retired and current elite athletes, respectively, was measured physically using a watch with Global Positioning System (Garmin forerunner, 910X). Macronutrient Intake and Energy Balance Study: In this study, twenty (male = 16 and female = 4) junior long-distance athletes participated. The athletes were attending an eight-week training in the camp where they were living in and training in and around Tirunesh Dibaba National Athletics Training Centre (TDNATC) located at an altitude of 2500m (7°57′N latitude and 39°7′E longitude). The study used the three-day direct dietary record method. Nude body weight measurements were taken before and after the three assessment days. All food measurements were carried out when the three meals were served: breakfast (8:00 – 9:00am), lunch (12:00 am – 1:00pm), and dinner (6:00 – 7:00pm). All the measurements were taken and recorded by the principal investigator, together with the head coach of the athletes, using a digital weighing scale readable to 1 gram (Salter Housewares LTD, England) and the dietary analysis of each individual athlete, including the total energy intake, and the energy contribution and gram values for carbohydrate, fat and protein from the consumed foods was performed using the nutritional software package Nutritics (v3.7, University Edition). Training type, intensity and duration, as well as external load, including distance, time covered and speed of the training were recorded in a daily training diary over the three consecutive days. Total energy expenditure of each study participant was calculated from basal metabolic rate (BMR) using the Schofield equation (1985) and the physical activity ratio (PAR), and physical activity level (PAL). xvii Altitude Training Study: A total of 20 (male = 16 and female = 4) junior long distance athletes lived and trained in and around the Athletes Tirunesh Dibaba National Athletics Training Centre were recruited for the study. The study applied the balanced, randomised, experimental design. Before the athletes were randomly assigned to the live high - train high (LH-TH) control (n = 10 ) and live high - train high train low (LH-THTL) experimental (n = 10 ) groups, they were tested on a 5km track race at baseline (end of four pre-experimental weeks) and then assigned equally into the two groups based on their 5km performance (time) and gender. The study lasted for a continuous eight weeks where all the athletes lived in every day of the week, and trained light and moderate intensity sessions at an altitude of 2500m a.s.l. four times per week. In addition, the LH-TH and LH-THTL groups trained separately at 2500m and 1470m a.s.l. in high intensity sessions two days per week, respectively. During the study time, different haematological, autonomic, neuromuscular, subjective training monitor and five kilometre performance time trial tests were conducted. Resting haematological tests were conducted three times (baseline, week four and week eight). Sample blood was drawn from a cubital vein under standard conditions (off-training days, between 08:30 and 09:30 a.m. before breakfast and after a 10 minute rest period in a sitting position) in the haematology laboratory of the College of Health Sciences of Arsi University, Ethiopia at the specified time for complete blood count ( CBC) analysis. Like the haematological tests, three consecutive vagal related heart rate measurements (heart rate variability and one- and two-minute heart rate recovery measurements) were taken at baseline, week 4 and week 8. The heart rate variability measurements were taken early in the morning, before the athletes left for training, in their bedrooms (before leaving their beds). The one- and two-minute heart rate recovery tests were taken as soon as the three 5km time trials were completed at baseline, week 4 and week 8. Along with the CBC and vagal-related heart rate measurements, five consecutive neuromuscular fitness tests (at baseline, week 2, week 4, week 6 and week 8) using the common vertical jump tests (counter movement (CMJ) and squat jump (SJ) test) were conducted after 10 hours of light intensity training. For a total of 47 training sessions, subjective training load responses were collected using a session rating of perceived exertion (session-RPE) methods within 30 minutes after the end of the day’s workout. At baseline, week four and week eight xviii three 5km endurance performance tests were conducted on a 400m standardized synthetic track under standard conditions (at 2500m a.s.l. and between 07:00 – 08:00 a.m.). Results Demographic Characteristics Study: Although the demographic characteristics study identified significant difference between the three groups in the age at which formal training started (p < 0.001), no significant difference was identified between the groups (p > 0.05) regarding the altitudes where the athletes were born and raised. Moreover, the study reported no significant difference in the daily distance covered to and from school between the three groups during their primary education (p > 0.05) but not during their secondary education (p < 0.05). The study also revealed that there were significant regional distribution differences in the three groups (p = 0.002) where 81.3% of retired athletes and 55% of academy junior athletes were from central Ethiopia. There was also no significant difference (p = 0.05) between the three groups in the mode of transportation used to cover the daily distance to and from school. In addition to the above findings, this study also found no statistically significant difference in the types of major out-of-school activities between the three groups of athletes during their childhoods (p > 0.05). Macronutrient Intake and Energy Balance Study: There was a significant difference between the mean total energy intake (14593± ) and mean total energy expenditure (13423± 1134 , p < 0.001) during the three days’ dietary assessment. Moreover, the daily total energy intake (EI) and energy expenditure (EE) throughout the three days for all subjects were also compared in the same way as the total EI and EE. In comparison to the daily energy expenditure, on day one there was a mismatch between EI (15682 ± 1599 ) and EE (12823 ± , p = 0.000), and a positive energy balance was calculated. On day two there was no substantial difference between daily EI (14368 ± ) and EE (13688± 1618 , p = 0.146). Similarly, there was also no significant difference between the EI (13728± ) and EE (13757± , p = 0.919) on day three. This study also confirmed no significant differences in the daily energy expenditure between the three days (p = 0.091). As compared to fat and protein, it appears that CHOs were the major energy source consumed during the three days. The overall proportion of the energy derived from the foods revealed that CHO provided 65.7% (±11.7 %); protein 18.7% (±6.9 %) and fat 15.4% (±4.9 %). xix When the overall proportions of energy intake (KJ) derived from the three macronutrients were analysed on a daily basis, there were statistically significant differences in CHO, protein and fat consumption across the three days, (p < 0.001). Moreover, substantial differences were identified in the day-to-day fat (p < 0.001) and protein (p < 0.001) consumption during the three dietary assessment days. Altitude Training Study Haematological Study: No statistically significant difference in RBC count was observed between the LH - TH and LH - THTL study groups following eight weeks of endurance training (∆0.05; CL ±0.029; p = 0.741; ES = 0.12). After eight weeks of endurance training no significant difference was observed in the haemoglobin concentration (p = 0.926), but substantially declined from baseline to week eight in both groups (Experimental: ∆-0.48; CL±0.46; p = 0.040; ES = - 0.35 and control: ∆-0.51; CL± 0.46; p = 0.030; ES = -0.37). This study also identified no substantial difference in haematocrit value between the two study groups following eight weeks of endurance training (∆0.2; CL±1.9; p = 0.832; ES = 0.06). Vagal-Related Heart Rate Response: The resting HRV (RMSSD) measurements revealed no meaningful differences between the LH-TH and LH - THTL groups (∆-0.18; CL±0.43; p = 0.407; ES = -0.29) from baseline to week eight in the experimental (∆0.05; CL±0.31; p = 0.761; ES = 0.08) and control (∆0.22; CL±0.31; p = 0.145; ES = 0.37) groups; although the changes in both groups were positive. The difference between the experimental and control groups, however, was negative and small (∆-0.29). The regression analysis also revealed no significant differences, both in the one-minute (∆4.4; CL±10.8; p = 0.413; ES = 0.47) and two-minute postexercise heart rate recovery (∆3.1; CL±10.4; p = 0.550; ES = 0.31), between the experimental and control groups. Neuromuscular Fitness/Fatigue Response: The CMJ test results revealed no significant difference between the two study groups following eight weeks of endurance training (∆0.5; CL±4.8; p = 0.829; ES = 0.06), although meaningful changes were identified both in the experimental (∆8.3; CL ±3.4; p = 0.001; ES = 0.92) and control (∆7.8; CL ±3.4; p = 0.001; ES = 0.86) groups from baseline to week eight. Significant changes in squat jump ability were xx observed, both in the experimental (∆4.8; CL±2.8; p = 0.001; ES = 0.69) and control (∆2.9; CL±2.8; p = 0.039; ES = 0.42) groups, following eight weeks of endurance training; but not between the groups (∆1.8; CL±3.9; p = 0.353; ES = 0.26). This study also confirmed no significant difference between the two study groups in the eccentric utilisation ratio following eight weeks of endurance training (∆-0.05; CL±0.16; p = 0.511; ES = -0.39). Training Load Response: The results of the analysis identified significant differences between the groups, and for all weekly training load responses of all training sessions, i.e., light, moderate and high-intensity training sessions (p = 0.019) and high intensity training sessions (p = 0.000). However, no substantial difference was identified between groups (p = 0.133) in the weekly load responses to the light and moderate intensity training sessions. Based on the results of the least significant difference, post-hoc meaningful differences were identified between the groups in their weekly load response to the total intensity training at week seven and eight; as well as at weeks one, five, seven and eight for the high intensity training sessions. Out of the 47 training sessions, light intensity sessions (< 4 RPE, less than the first ventilatory threshold) made up 87.2% of sessions in the experimental group and 68.1% in the control group, while 12.8% (Experimental group) and 31.9% (Control group) of the training sessions were completed at RPE > 4 < 7 RPE (between first and second ventilatory threshold). Five Kilometre Endurance Performance: After eight weeks of endurance training no significant difference in the 5km endurance performance was identified between the LH-TH and LH-THTL study groups (∆-12; CL ±25s; p = 0.335). Even though times for the 5km decreased significantly in the experimental group (∆-19; CL±18s; p = 0.037) from baseline to week eight, performance in the control group did not improve significantly (∆-7; CL ±18s; p = 0.440). Conclusions Demographic Characteristic Study: Significant difference was observed between the three groups in the age at which formal training started. However, no significant differences were identified between the three groups in the altitudes where the Ethiopian long-distance athletes were born and raised, the daily distance travelled to and from school, the mode of transportation and the major out-of-school activities during their childhood. Thus, the findings of this study xxi confirmed that the 20 junior athletes who were involved in the study shared common demographic characteristics with the retired and current elite Ethiopian long-distance athletes. Macronutrient Intake and Energy Balance Study: In line with the previous studies conducted on Kenyan and Ethiopian endurance athletes, the young long-distance Ethiopian athletes met the recommended daily macronutrient intake for carbohydrates and protein for endurance athletes. However, the study also identified that the athletes’ dietary fat consumption was below the recommended amount for endurance athletes. Moreover, based on the three-day dietary assessment results, the young Ethiopian endurance athletes were found to be in a state of positive energy balance one week before their first major competition of the year (albeit during the preparation phase of their yearly training plan). Altitude Study: The overall results of the current altitude study revealed that in most of the study variables (i.e., haematological, autonomic, neuromuscular, and endurance performance), except the subjective based training load response, statistically insignificant results were identified between the two study groups. However, when the results of the altitude study variables across time (baseline to week eight) were examined, athletes in the LH-THTL experimental group showed better progress in neuromuscular and lower training load responses which were accompanied with significant five kilometer endurance performance change; and lower or similar progress in haematological and autonomic regulation responses as compared with the LH-TH control group. It is noted that the ultimate purpose of any type of altitude training is enhancing the running performance while minimizing athlete’s susceptibility to injury. Taking these core concepts of athletic training and the physiological and performance changes of the current study in to consideration, the LH-THTL altitude training model was potentially the preferred optimal altitude training model to further enhance the past and existing long-distance performance of Ethiopian endurance athletes although further comprehensive studies are required to confirm the results. Future Directions In order to exhaustively investigate optimal altitude training models that better enhance the longdistance performance of athletes’ native to high altitude, more comprehensive, similar studies xxii should be designed. Moreover, to achieve stronger results, further studies should be conducted using larger sample sizes with balanced gender proportions, along with more subjective and objective training monitoring methods. Furthermore, future studies should consider additional altitude training models, be conducted over longer periods and during different phases of the yearly training plan (preparation, pre-competition, and competition). It is also recommended that future studies should design endurance performance tests at different altitudes (low and high) to enhance the local and international competition performance of Ethiopian long-distance athletes.


Doctoral Degree. University of KwaZulu-Natal, Durban.