A total of 225 subjects provided written informed consent before their participation in the study, which was approved by the University Ethics Committee and local Ethiopian authorities in Addis Ababa. The experimental procedures were in accordance with the policy statement of the American College of Sports Medicine. Subjects were comprised of 114 athletes, all members of the Ethiopian national senior and junior athletics teams, and 111 control subjects (C). Both athletes and controls were comprised of males and females. The control group, 111 students at Kotebe Teaching College, Addis Ababa, were intended to be representative of the Ethiopian general population; none were regularly training for any track or field athletic events.
The athletes were classified into one of three groups, according to their specific athletic discipline. Athletes specializing in 5,000- to 10,000-m distances (5–10 km; N = 42), marathon runners (M; N = 34), and a third group, comprised of track and field athletes, track athletes being runners up to 1500 m (TF; N = 38). Athletes in the marathon and 5- to 10-km groups were truly elite athletes, regularly successful in international distance running, and both groups included World and Olympic champions. Although Ethiopia is not successful internationally in sprinting and field events, these athletes were included in this study to investigate the possibility of a disproportionate number of athletes originating from any geographical region being the result of an increased prominence of athletics in that area.
Questionnaires translated into Amharic, the national language of Ethiopia, to ensure comprehension were administered to the subjects participating in this study. The questions included in these were designed to obtain the following information:
Place of birth.
This was classified according to the 14 provinces of Ethiopia from 1946 to 1980 (7). We wished to identify any particular regions with a disproportionately high number of elite athletes and possibly, therefore, elucidate any link between distance-running success and altitude.
Spoken language (and that of their grandparents).
This was to provide further information on ethnicity. A common language is often indicative of common origin, and a related language (i.e., a language of the same family) suggests a common origin, but one that dates farther back in time. At present, over 70 languages are in everyday usage in Ethiopia, most of which belong to the Afro-Asiatic family. The majority of these languages belong to either the Semitic or Cushitic subset of languages. To simplify the analysis in the present study, three separate language categories were used: Semitic, Cushitic, and Other (which includes the less common subsets within the Afro-Asiatic language family such as Omotic).
Mode of travel to school (walk, run, other) and distance traveled (< 5 km, 5–20 km, other).
This was to assess the validity of reports linking distance traveled to school to East-African running success.
Contingency chi-squares were performed using the Yates correction factor on all occasions given the low subject numbers in each field. For each section of the questionnaire, chi-squares were initially performed to establish the presence of significant differences between the subject groups (Place of birth:df = 12, language, distance and method of travel to school:df = 6). Individual chi-squares were then performed to identify between which groups the differences lay (Table 1). Statistical significance was declared at P < 0.05. The 14 regions appearing in the questionnaire responses were collapsed into five categories according to the most common responses. To allow statistical analysis using contingency chi-squares, regions accounting for < 7% of total subject number were collapsed into the category Other. In Figure 1, the Other category also represents certain subjects whose birthplace could not be assigned into one of the 14 regions used for classification purposes. In Figure 2, the language of both sets of grandparents was used to minimize the influence on spoken language of the recent shift toward Amharic as the national language of Ethiopia.
Place of birth.
The regional distribution of controls differed significantly from all athlete groups (Table 1). The marathon athletes differed significantly from track and field athletes, but not to those of the 5- to 10-km team. Additionally, track and field athletes did not differ significantly to the 5- to 10-km athletes. As can be seen in Figure 1, there are an excess of athletes, particularly marathon athletes (73%), from the two regions of Arsi and Shewa, which, together, account for only 15% of control subjects. The majority of controls originate from the 10 regions collapsed into the Other category (54%). All groups of athletes showed an excess of subjects from the Arsi province, the association being strongest in the marathon athletes (38%) compared to controls (3%).
The origin of language of control subjects differed significantly from all athlete groups (Table 1). The marathon athletes also differed significantly from all groups (Table 1), with a predominance of languages of Cushitic origin (75%) relative to the other groups (5–10 km: 52%, TF: 46%, C: 30%), as can be seen in Fig. 2. There was no significant difference between track and field athletes and the 5- to 10-km athletes (Table 1). Each of the athlete groups showed an excess of languages of Cushitic origin relative to controls, with the effect most pronounced in the marathon athletes, where 75% speak languages of Cushitic origin compared to 30% of controls.
Distance and method of travel to school.
In distance traveled to school, the marathon athletes differed significantly from all other groups (Table 1), with no significant differences present between the other groups. As can be seen in Fig. 3, a higher proportion of the marathon athletes traveled over 5 km to school each day (73%) than in the other groups (5–10 km: 40%, TF: 32%, C: 36%). In method of travel to school, again, the marathon athletes differed significantly from all other groups (Table 1). Controls also differed significantly from all other groups (Table 1). Figure 4 shows that a higher percentage of the marathon athletes ran to school each day, compared with other groups (M: 68%, 5–10 km: 31%, TF: 16%, C: 24%).
The findings of the present study are that elite Ethiopian athletes, particularly the marathon athletes, are of a distinct background, both ethnically and environmentally. The marathon athletes display a distinct environmental background relative to the other groups in all categories. Seventy-three percent of the marathon athletes are from the two regions of Arsi and Shewa, compared with controls, of whom only 15% are from these two regions. Similar patterns are found in the language results with the marathon athletes exhibiting an excess of languages of Cushitic origin in comparison with control subjects (M: 75%, 5–10 km: 52%, TF: 46%, C: 30%). A higher percentage of marathon athletes traveled farther to school than the other groups (M: 73%, 5–10 km: 40%, TF: 32%, C: 36%) and had a higher prevalence of running to school (M: 68%, 5–10 km: 31%, TF: 16%, C: 24%). These results indicate that a number of factors other than genetic endowment may be influencing the athletic potential of elite Ethiopian athletes.
Although intended to be representative of the general population, the control group is comprised of students at an Addis Ababa teaching college, which may account for the excess of control subjects from Addis Ababa relative to early census data, which shows only 4% of subjects from Addis Ababa (15) compared with 25% in our control group. The significant differences between controls and all groups of athletes suggest some link between place of birth and athletic ability. As can be seen in Figure 1, the marathon athletes have a distinct appearance relative to all other groups. Thirty-eight percent of marathon athletes are from the region of Arsi, which accounts for less than 5% of the Ethiopian population and comprises only 3% of control subjects, 18% of track and field athletes, and 24% of the 5- to 10-km athletes. It is interesting to note that of the marathon athletes, 24% are from any of the other 12 regions with a total of 73% from the regions of Arsi and Shewa. A similar phenomenon of regional overrepresentation has been documented in elite Kenyan athletes. Up until 1989, 45% of Kenyan world elite results were made by runners belonging to the Nandi tribe, who account for less than 3% of the Kenyan population (13). Reasons for the geographical imbalance of elite Kenyan athletes have been primarily attributed to cultural factors within the Kalenjin tribe. Kipchoge Keino, the first famous Kenyan runner was part of the Kalenjin people, and his success is thought to have spurred a tradition of distance running, sustained by talent recruitment to high school with formal training and competition (13).
There have been suggestions that the success of Kenyan runners is in some way linked to their proximity to the Rift Valley, as the Nandi people inhabit an area adjacent to it (4). It seems, from Figure 5, that parallels may be drawn to Ethiopia, as the Rift Valley intersects the regions of Arsi and Shewa along their common border. Although much of Ethiopia is altitudinous, the regions of Arsi and Shewa are both in the central highlands of Ethiopia and reach altitudes of over 4000 m. Up to 80% of the population in these regions live around and above 2000 m. The finding that the marathon athletes seem to be clustered in particularly altitudinous regions may support the suggestion of a link between altitude and endurance success. However, an alternative hypothesis is presented when it is considered that Arsi is also considerably overrepresented in track and field athletes (18%). There is no perceived benefit to throwers or jumpers from altitude inhabitation for sea level performance, yet it is found that the track and field athletes also exhibit an excess from Arsi, which may suggest that altitude is not the main influence on the prevalence of elite Ethiopian athletes from Arsi. One of the senior Ethiopian athletic coaches informed the investigators that most of the marathon athletes would be found to be from Arsi. If those in charge of athletic development believe this, it may cause a self-fulfilling prophecy through talent scouts focusing more attention to this region or through increased regional development of athletics.
Figure 3 shows that a higher proportion of the marathon athletes traveled greater distances to school each day than those of the other three groups. This is comparable to the method of travel to school, where it was found that almost 70% of them ran to school each day compared with only 25–40% of the other groups. A higher percentage of the 5- to 10-km athletes traveled farther to school, and a significantly higher proportion of them traveled by running, relative to controls. When considered alongside the findings that Kenyan boys who walked and ran long distances to school each day but did not formally train for athletics had V̇O2max values some 30% higher than those who did not (13), the results implicate childhood endurance activity as a key selection pressure in the determination of Ethiopian endurance success. With the prevalence of childhood obesity in the United States and Great Britain at an all time high (3), and physical activity levels among such populations in stark contrast to the daily aerobic activity of Ethiopian children, these factors may offer an explanation for the success of East-African athletes on the international stage.
The marathon athletes exhibit an excess of people speaking languages of Cushitic origin (Fig. 2), most of these speaking Oromigna, which is the language of Oromo people. The official language of Ethiopia is now Amharic, which is of Semitic origin, so the language distribution of the marathon athletes appears to be strongly influenced by some other factor. Studies into the frequency of certain gene polymorphisms in Ethiopia have, in some cases, shown significant differences in allelic frequencies and haplotypes between the two main ethnic groups of Ethiopia: Amhara and Oromo (5). Given the identification and association of particular gene polymorphisms with endurance performance (8,10), the excess of elite Ethiopian endurance athletes originating from one ethnic group may reflect a high frequency of potential “performance genes” within this particular group. However, it is perhaps more likely that the distinctive ethnic origin of the marathon athletes is a reflection of their geographical distribution, as primarily Oromo people populate Arsi.
The results of this study suggest that elite Ethiopian distance runners are of a distinct ethnic and environmental background, relative to the general Ethiopian population. The marathon athletes are found to have a distinct profile for all variables investigated in comparison with the general Ethiopian population, in that they tended to speak languages of Cushitic origin and ran a long way to school each day in Arsi or Shewa. This profile is associated with endurance success, and it is likely that these environmental conditions along with the cultural and motivational factors inherent in Ethiopia are contributors to East-African distance-running success. Although not excluding any genetic influence, the results of the present study highlight the importance of environment in the determination of endurance athletic success.
The authors acknowledge the invaluable assistance of the Ethiopian Olympic Committee, Ethiopian Athletics Federation, and Kotebe College. The Royal Society, The Carnegie Trust for the Universities of Scotland, and The Wellcome Trust are gratefully acknowledged for part-funding this research. The cooperation of all subjects is greatly appreciated.
1. Bosch, A. N., B. R. Goslin, T. D. Noakes, and S. C. Dennis. Physiological differences between black and white runners during a treadmill marathon. Eur. J. Appl. Physiol. Occup. Physiol. 61: 68–72, 1990.
2. Coetzer, P., T. D. Noakes, B. Sanders, et al. Superior fatigue resistance of elite black South African distance runners. J. Appl. Physiol. 75: 1822–1827, 1993.
3. Ebbeling, C. B., D. B. Pawlak, and D. S. Ludwig. Childhood obesity: public-health crisis, common sense cure. Lancet 360: 473–482, 2002.
4. Entine, J. Race and sport: the race to the swift—if the swift have the right ancestry. Peak Performance Special Issue: Race and Genetics. 158: 1–6, 2001.
5. Gennarelli, M., M. Pavoni, F. Cruciani, G. De Stefano, B. Dallapiccola, and G. Novelli. CTG repeats distribution and Alu insertion polymorphism at myotonic dystrophy (DM) gene in Amhara and Oromo populations of Ethiopia. Hum. Genet. 105: 165–167, 1999.
6. Hamilton, B. East African running dominance: what is behind it? Br. J. Sports Med. 34: 391–394, 2000.
7. H enze, P. B. Layers of Time: A History of Ethiopia.
London: Hurst & Co., 2000, p. xxiii.
8. Jones, A., H. E. Montgomery, and D. R. Woods. Human performance: a role for the ACE genotype? Exerc. Sport Sci. Rev. 30: 184–190, 2002.
9. Passarino, G., O. Semino, L. Quintana-Murci, L. Excoffier, M. Hammer, and A. S. Santachiara-Benerecetti. Different genetic components in the Ethiopian population, identified by mtDNA and Y-chromosome polymorphisms. Am. J. Hum. Genet. 62: 420–434, 1998.
10. Rankinen, T., L. Perusse, R. Rauramaa, M. A. Rivera, B. Wolfarth, and C. Bouchard. The human gene map for performance and health-related fitness phenotypes: the 2001 update. Med. Sci. Sports Exerc. 34: 1219–1233, 2002.
11. Saltin, B. Exercise and the environment: focus on altitude. Res. Q. Exerc. Sport 67: S1–S10, 1996.
12. Saltin, B., C. K. Kim, N. Terrados, H. Larsen, J. Svedenhag, and C. J. Rolf. Morphology, enzyme activities and buffer capacity in leg muscles of Kenyan and Scandinavian runners. Scand. J. Med. Sci. Sports 5: 222–230, 1995.
13. Saltin, B., H. Larsen, N. Terrados, et al. Aerobic exercise capacity at sea level and at altitude in Kenyan boys, junior and senior runners compared with Scandinavian runners. Scand. J. Med. Sci. Sports 5: 209–221, 1995.
14. Schmidt, W., K. Heinicke, J. Rojas, et al. Blood volume and hemoglobin mass in endurance athletes from moderate altitude. Med. Sci. Sports Exerc. 34: 1934–1940, 2002.
15. Tesfiaghiorghis, H. The growth of urbanization in Ethiopia, 1966–1984. East. Afr. Econ. Rev. 2: 157–167, 1986.
16. Weston, A. R., O. Karamizrak, A. Smith, T. D. Noakes, and K. H. Myburgh. African runners exhibit greater fatigue resistance, lower lactate accumulation, and higher oxidative enzyme activity. J. Appl. Physiol. 86: 915–923, 1999.
17. Weston, A. R., Z. Mbambo, and K. H. Myburgh. Running economy of African and Caucasian distance runners. Med. Sci. Sports Exerc. 32: 1130–1134, 2000.
18. Yu, N., F. C. Chen, S. Ota, et al. Larger genetic differences within Africans than between Africans and Eurasians. Genetics 161: 269–274, 2002.