Share this article on:

Physical Activity and Body Mass Index of Children in an Old Order Amish Community


Medicine & Science in Sports & Exercise: March 2007 - Volume 39 - Issue 3 - p 410-415
doi: 10.1249/mss.0b013e31802d3aa7
BASIC SCIENCES: Epidemiology

The prevalence of childhood obesity is rapidly increasing in many countries throughout the world. Advances in modern technology may have contributed to this problem by reducing children's physical activity levels. Thus, it is interesting to study children belonging to groups that have refrained from adopting modern technology.

Purpose: This study examined the physical activity levels and body mass index (BMI) of Old Order Amish children and adolescents living in a nontechnological farming community.

Methods: One hundred thirty-nine Amish youth (6-18 yr of age) were recruited for the study. BMI was computed from measured height and weight. The schoolchildren were instructed to wear sealed step counters for 7 d. Each school day, research assistants opened the step counters, recorded their steps, reset and resealed them, and placed them back on the children. Older children recorded their steps on a step-counter log sheet.

Results: The number of steps per day (mean ± SD) was 17,525 ± 4443 (measured for four weekdays), 10,661 ± 4208 (measured over Friday, Saturday, and Sunday), and 15,563 ± 3702 (measured for 7 d). All of the step variables (P < 0.001) showed gender effects, but there were no significant age effects. Only 7.2% of the youth were overweight, and only 1.4% were obese.

Conclusion: These Amish youth have high levels of physical activity, and obesity is rare. Comparing our results with those of other studies, it is evident that these Amish youth are more physically active and have a lower rate of obesity than children living in modern, industrialized societies.

1College of Education, Health, and Human Sciences, The University of Tennessee, Knoxville, TN; 2College of Kinesiology, University of Saskatchewan, Saskatoon, Saskatchewan, CANADA; 3Statistics Canada, Ottawa, Ontario, CANADA; 4Department of Kinesiology, University of Lethbridge, Lethbridge, Alberta, CANADA; and 5Department of Anthropology, University of Michigan, Ann Arbor, MI

Address for correspondence: David R. Bassett, Jr., Ph.D., Department of Exercise, Sport, and Leisure Studies, The University of Tennessee, Knoxville, TN 37996; E-mail:

Submitted for publication July 2006.

Accepted for publication October 2006.

A decline in physical activity is one of several factors being implicated in the tremendous increase in rates of obesity, diabetes, and other health problems in many parts of the world. Recent U.S. statistics indicate that 16.5% of children and adolescents (ages 6-19 yr old) are overweight or at risk of overweight (6), roughly four times as many as 40 yr ago (12). Similar trends have been observed in Canada (18,22). In addition, diseases formerly seen mainly in adults (such as type 2 diabetes) are now appearing in youth (7,13).

Descriptive studies are emerging that suggest that North American children in contemporary society are less physically active than those living in other industrialized nations. For instance, elementary school children in the American Southwest take fewer steps per day than those in Sweden and Australia (25). In that study, the American children were found to be heavier than the Swedish and Australian children. Other studies report that children living in Belgium (2) and Cyprus (11) also have higher daily step counts than American children.

Few studies have measured physical activity and physical fitness in children living in modern, nonindustrialized societies. One exception is a recent study that measured time spent in activity, aerobic fitness, handgrip strength, body composition, muscular fitness, and flexibility in Old Order Mennonite children living in Canada (21). The study showed that the Mennonite children were fitter and more physically active than rural and urban Canadian children living in modern societies.

The Amish, like the Mennonites, are a religious group that traces its origin back to the Swiss Anabaptist movement of the 16th century. They immigrated to America in the 1700s and 1800s, and today no Amish are left in Europe. The Amish have a distinctive style of dress, and they live in rural communities that adhere to a lifestyle similar to that of rural Americans in the 1800s. Unlike most societies, they purposefully refrain from adopting modern technologies such as highline electricity and automobile ownership. Amish children walk to school and friends' homes rather than being driven. They perform farm and household chores and engage in active play. Amish children avoid many of the sedentary leisure time pursuits common among children living in contemporary society (e.g., television watching, video games, listening to music, and computer use).

The main purpose of this study was to describe physical activity levels in youth residing in an Old Order Amish community located in Ontario, Canada. A secondary purpose was to examine body mass index (BMI) and the prevalence of obesity in this population. On the basis of our previous studies of Amish adults (1) and Mennonite children (21), it was hypothesized that these Amish children would be leaner and more physically active than children in modern, industrialized societies who have been studied by other investigators.

Back to Top | Article Outline


A written informed consent form (approved by the University of Tennessee institutional review board and University of Saskatchewan research ethics board) was sent home with each of the children before the start of the study. Children from all of the one-room Amish schoolhouses in the community were examined, and 98% of the school children participated.

Amish schoolteachers distributed the parental informed consent forms to students in class before our arrival. Because Amish schools end after the eighth grade, teachers and students (in many cases younger siblings) spread the word that older children (14-18 yr of age) were invited to participate. Approximately 70% of the 14- to 18-yr-olds living in the community chose to participate. The nature of the study was explained to the children, and they were asked to sign or initial a simple assent form if they wished to participate. Parental consent and child assent were obtained from all children before their involvement in the study. The data were collected between April 18 and April 27, 2005.

Back to Top | Article Outline

Anthropometric characteristics

Physical characteristics were assessed in addition to age, grade, and gender. Body mass (kg) and height (cm) were measured in one layer of light clothing, without shoes, using a physician's scale and stadiometer. BMI was computed from the formula:



Overweight and obesity were determined using the international definition for child obesity developed by Cole et al. (3). That study used six nationally representative growth studies and constructed BMI growth curves, so that at 18 yr of age, the curves passed through the BMI cutoff points of 25 and 30 kg·m−2 for adults. The resulting curves were then averaged to arrive at age- and gender-specific cutoff points for overweight and obesity.

Back to Top | Article Outline

Electronic step counter

Participants in the study were asked to wear a Yamax SW-200 step counter (Yamasa, Japan) for a 7-d period. This device is widely used in research studies and has been previously validated for counting steps (4,8,10,16,17,26). The participants were asked to position the step counter on a black nylon belt worn at the waist, directly above the right kneecap. Because younger children might have had difficulty recording their daily steps on a log sheet, the pedometers were sealed shut with a cable tie before they were placed on the children. At school each morning, research assistants collected the pedometers, removed the cable ties, recorded steps taken, reset the step counters to zero, resealed them, and placed them back on the children. Older children (14-18 yr of age), who had completed their schooling, recorded their own steps using a 7-d step-counter log that was identical to the one we used in our previous study of Amish adults (1).

We wanted to determine the number of steps per day the Amish children accumulated in their everyday lives, so we instructed them not to change their activity habits during the study. In addition, the following instructions were given:

  • The step counter should be worn at all times for exactly 7 d, except when bathing or sleeping.
  • We are interested to know the total number of steps you take each day.
  • As soon as you wake up each morning, put the step counter on, and wear it all day.
  • Just before you go to bed each night, please remove the step counter.
  • Repeat this the next day, until you have worn it for 7 d.

On the pedometer log sheet, children also were instructed to write down several activities that they had engaged in that day.

To obtain a record of steps taken over the weekend, the school children wore the pedometers for 3 d (Friday, Saturday, and Sunday), and the total number of steps was recorded. The data reported here are part of a larger study funded by the Canadian Population Health Initiative that included accelerometer measurements of physical activity and measurements of physical fitness.

Back to Top | Article Outline

Data analysis

The Amish youth were grouped according to age (6-8, 9-11, 12-14, and 15-18 yr old) and gender. These age groups were chosen to facilitate comparisons with the age or grade groups used in other studies (9,23,25). Means and standard deviations on each of the anthropometric variables were computed for each age- and gender-specific group. Step values were computed as the mean of four weekdays (i.e., Monday to Thursday), Friday to Sunday, and 7 d. Two-way ANOVA (age group × gender) were used to test for between-group differences in anthropometric variables and daily step counts. When age-group main effects were significant, we ran Tukey post hoc tests for multiple comparisons to determine which age groups differed from the others. When gender main effects were significant, we examined the means to see which one (male vs female) was higher. There were no significant age × gender interactions for any variables; hence, no follow-up tests were used on the interactions. For all comparisons, the alpha level was set at 0.05.

Back to Top | Article Outline


The physical characteristics of the Amish youth are shown in Table 1. Data are reported as mean ± SD for height, body mass, and BMI. For height, we found significant effects for age group [F(3, 130) = 190.95, P < 0.001] and gender [F(1, 130) = 11.98, P < 0.001], but there was no significant age × gender interaction [F(3,130) = 2.35, P = 0.076]. As expected, older children were taller than younger children (all age groups differed from each other on height), and boys were taller than girls.



For body mass, we found significant main effects for age group [F(3, 130) = 113.69, P < 0.001] but not for gender [F(1, 130) = 0.51, P = 0.476] or for age × gender interaction [F(3, 130) = 0.61, P = 0.605]. Older children were heavier than younger children (all age groups differed from each other on body mass). Boys and girls did not show significant differences in body mass.

The average BMI for the Amish boys was 18.2 kg·m−2, and for the Amish girls, it was 19.1 kg·m−2. There were significant main effects for age group [F(3, 130) = 33.96, P < 0.001] and gender [(1,130) = 4.35, P = 0.039], but there was no significant interaction [F(3, 130) = 1.989, P = 0.119]. Older children had higher BMI values than younger children, and all age groups differed from each other on BMI (with the exception of 13- to 15-yr-olds vs 16- to 18-yr-olds, P = 0.523). The prevalence of overweight among these Amish youth was 7.2%, and the prevalence of obesity was 1.4%, according to the definitions proposed by Cole et al. (3).

The step counts of the Amish youth are shown in Table 2. Daily step counts (mean ± SD) were 17,525 ± 4443 (measured for four weekdays), 10,661 ± 4208 (measured over Friday, Saturday, and Sunday), and 15,563 ± 3702 (measured for 7 d). For the four-weekday average, we observed a significant gender effect [F(1, 131) = 39.44, P < 0.001] but no significant age effect [F(3, 131) = 0.57, P = 0.636] or interaction [F(3, 131) = 2.215, P = 0.089]. For the 3-d average (Friday to Sunday), we observed a significant gender effect [F(1, 131) = 14.17, P < 0.001] but no significant age effect [F(3, 131) = 0.99, P = 0.401] or interaction [F(3, 131) = 0.91, P = 0.439]. For the 7-d average, we observed a significant gender effect [F(1, 131) = 42.94, P < 0.001] but no significant age effect [F(3, 131) = 0.83, P = 0.482] or interaction [F(3, 131) = 1.79, P = 0.153]. In summary, all three of the step variables showed that boys took more steps than girls, and there were no significant differences between age groups. Although the age × gender interactions were not significant, girls showed a trend toward decreased step counts with increasing age, whereas boys showed a trend toward increased step counts with increasing age (Fig. 1). Including height as a covariate in the analysis did not change the results with respect to age.





Back to Top | Article Outline


The main findings of this study were that Amish children had high levels of ambulatory activity (as determined by a step counter) and a low rate of obesity. The Amish boys averaged 17,174 steps per day, and Amish girls averaged 13,620 steps per day, for 1 wk. These values are similar to those previously reported for Amish adults residing in the same traditional farming community. Amish men averaged 18,425 steps per day, and Amish women averaged 14,196 steps per day, for 1 wk (1)-nearly three times the average amounts for American adults (24,27).

Previous studies have examined step counter-determined activity levels of children (6-12 yr of age) living in contemporary societies in other parts of the world. Vincent et al. (25) have reported that for 6- to 12-yr-old boys, mean step counts were 16,166 steps per day for Sweden, 14,337 for Australia, and 13,231 for America. For 6- to 12-yr-old girls, mean step counts were 13,564 for Sweden, 11,710 for Australia, and 10,992 for America. It should be noted that their data represent the average of four weekdays; by comparison, Amish boys of the same age averaged 19,068 steps per day, and Amish girls averaged 16,071 steps, averaged for four weekdays. Thus, the younger children in this Amish community accumulated about 45% more steps than American elementary school children and 20% more than European elementary school children. The high step counts may partly explain why these Amish children (ages 6-12 yr) have lower rates of overweight and obesity than European children, who, in turn, have lower rates than American and Canadian children.

Other studies support the view that European children have higher step counts than those in the United States and Canada. Cardon and DeBourdeaudhuij (2) measured physical activity in elementary school children (6-12 yr of age) in Belgium. They averaged 15,308 steps per day for 7 d, with boys taking significantly more steps than girls (16,628 vs 13,008 steps per day, respectively). Loucaides et al. (11) examined physical activity levels in urban and rural elementary school children (11-12 yr of age) in Cyprus and examined seasonal effects. The urban children took an average of 14,507 steps per day, whereas the rural children took an average of 14,443 steps per day (averaged for four weekdays). The urban children were significantly more active in winter, whereas the rural children were significantly more active in summer (P < 0.001).

In the present study, teenage Amish boys averaged 20,292 steps per day for four weekdays, considerably more than teenage boys in modernized societies. LeMasurier et al. (9) have found that American boys in grades 7-9 and 10-12 accumulated 11,082 and 10,828 steps per day, respectively. Teenage Amish girls averaged 13,558 steps per day for four weekdays), which was more than that of teenage girls in modern society. LeMasurier et al. (9) have found that American girls in grades 7-9 and 10-12 accumulated 10,080 and 9706 steps per day, respectively. In another study, Trost et al. (23) placed accelerometers on 375 American school children in grades 1-12. Similar to the findings of Vincent et al. (25) and LeMasurier et al. (9), they observed that the number of minutes spent in moderate to vigorous physical activity was inversely related to age for children in grades 1-12 living in modern society.

In this Amish community, the number of steps per day was remarkably similar in children and adults. In contrast, in modern America, pedometer scores are approximately 12,000 steps per day for U.S. elementary school children (25) and only 5000-6000 for U.S. adults (24,27). This observation is consistent with the hypothesis that modern, technology-based lifestyles may promote an age-related decline in physical activity. However, it is interesting to note that in western Europe, both children and adults are more active than in America. This is largely attributable to their transportation system, which relies heavily on walking, cycling, and use of public transit (14,15). In some western European nations, these forms of active transport account for more than 50% of trips taken, whereas in the United States, they account for fewer than 10% of trips (20). Pucher and Dijkstra (15) suggest that this is one reason that obesity rates in North America are higher than in Europe.

Obesity was rare among the Amish youth we studied. Only 1.4% of them were obese according to BMI, computed from measured height and body mass. In contrast, 6.5-9.5% of American boys ages 6-19 are obese, and 6.6-11.7% of American girls the same age are obese according to similar methods of measurement (5). Obesity rates for Canadian children are 7 and 9% for 2- to 17-yr-old girls and boys, respectively (18). Only 7.2% of the Amish youth we studied in this age group were overweight, compared with roughly 25% of American and Canadian children ages 6-19 yr (5,18). The additional calories expended in physical activity are one factor contributing to the low rates of childhood overweight and obesity in this Amish community.

Amish children reported performing a variety of physical activities. Morning chores included carrying firewood, feeding livestock, collecting eggs, and milking. For the most part, the Amish children in this community follow traditional gender roles. Boys are more likely to help with outdoor farm chores, and girls are more likely to assist with gardening and with household activities such as cooking, child care, laundry, and quilting. (Child care and laundry may require few steps but considerable activity.) These household activities result in fewer steps than farm work, which helps to explain why boys accumulated more steps than girls. In addition, active play is likely to be more vigorous in boys than in girls. Most Amish children walked to school, even in inclement weather, unless they lived more than a couple of kilometers away (in which case they were transported by horse and buggy). At school, despite the lack of formal physical education classes, there was plenty of opportunity for sports and recreation. A 15-min recess in the morning, a 10-min recess in the afternoon, plus most of a 60-min lunch break were spent out of doors, and children engaged in games such as softball, volleyball, bombardment, King's base, and freeze tag with their teachers.

The data obtained from this study are relevant to the ongoing discussion of how much physical activity is needed for good health. Current physical activity recommendations call for every child to perform a minimum of 60 min of moderate physical activity per day (19). The U.S. President's Council on Physical Fitness and Sports encourages the attainment of step goals for children ages 6-18 yr. To qualify for the Presidential Active Lifestyle Award (PALA), 11,000 steps per day for girls, or 13,000 steps per day for boys, are recommended. In fact, 96% of the Amish boys we studied, and 92% of the Amish girls, met the PALA requirement. These high activity levels were achieved despite inclement weather during the testing period. The Amish youth averaged 18,422 steps per day on clear days versus 15,770 on rain/snow days (P < 0.001). Thus, it seems that precipitation impacts physical activity behavior in youth in this farming community.

This study has both strengths and limitations. The strengths of the study are that we obtained direct, objective measurements of physical activity, height, and body mass, thereby avoiding inaccuracies inherent in self-report. In addition, nearly all of the children in this community were surveyed, so it is unlikely that selection bias affected the step counts and obesity data. A limitation of the study is that we only assessed ambulatory activity, and thus other types of activities such as arm movements, lifting and carrying objects, walking uphill, and stair climbing went undetected. We did not measure seasonal differences in physical activity; instead, we chose to assess the Amish youth during the springtime, when activity levels are probably between the winter and summer values. In addition, we did not measure energy intake, another factor that contributes to obesity. Finally, it should be noted that we only studied the children in one technologically conservative Amish community, where nearly 80% of households are involved in farming. In other Amish communities where a smaller percentage farm, and where use of modern technology is more prevalent, childhood obesity rates may be higher; hence, our findings should not be generalized to other Amish groups.

In summary, this study has described physical activity and BMI of Amish youth residing in a nontechnological farming community. The Amish children have higher levels of physical activity and a lower rate of obesity than children living in industrialized societies. This study of a nontechnological farming community can help us understand the impact of modern technology on children's physical activity and fitness. By studying a community that refrains from automobile dependency, use of modern labor-saving devices, and electronic entertainment, it is possible to estimate how these advances have impacted our lives. Assuming that the physical activity levels of these Amish youth resemble those of rural North American children 150 yr ago, one can roughly estimate the effects of modernization on this important behavioral trait. Although there are acknowledged limitations to our research design, the results suggest that physical activity levels of children have declined during the past century and a half, although not as markedly as in adults.

The authors express their appreciation to children in the Amish community for participating in the study, to David Luthy for sharing the resources in the Heritage Historical Library, and to Cary Springer (University of Tennessee Statistical Consulting Services) for conducting the data analysis. This research was supported by the Canadian Population Health Initiative of the Canadian Institute for Health Information and by the Charlie and Mai Coffey endowment in Exercise Science at the University of Tennessee.

Back to Top | Article Outline


1. Bassett, D. R., P. L. Schneider, and G. E. Huntington. Physical activity in an Old Order Amish community. Med. Sci. Sports Exerc. 36:79-85, 2004.
2. Cardon, G. M., and I. M. M. DeBourdeaudhuij. A pilot study comparing pedometer counts with physical activity minutes in elementary school children (Abstract). Med. Sci. Sports Exerc. 36:S31, 2004.
3. Cole, T. J., M. C. Bellizzi, K. M. Flegal, and W. H. Dietz. Establishing a standard definition for child overweight and obesity worldwide: international survey. BMJ 320:1-6, 2000.
4. Crouter, S. E., P. L. Schneider, M. Karabulut, and D. R. Bassett. Validity of 10 electronic pedometers for measuring steps, distance, and energy cost. Med. Sci. Sports Exerc. 35:1455-1460, 2003.
5. Flegal, K. M., C. L. Ogden, R. Wei, R. L. Kuczmarski, and C. L. Johnson. Prevalence of overweight in US children: comparison of US growth charts from the Centers for Disease Control and Prevention with other reference values for body mass index. Am. J. Clin. Nutr. 73:1086-1093, 2001.
6. Hedley, A. A., C. L. Ogden, C. L. Johnson, M. D. Carroll, L. R. Curtin, and K. M. Flegal. Prevalence of overweight and obesity among US children, adolescents, and adults, 1999-2002. JAMA 291:2847-2850, 2004.
7. Kaufman, F. R. Type 2 diabetes mellitus in children and youth: a new epidemic. J. Pediatr. Endocrinol. Metab. 15(Suppl 2):737-744, 2002.
8. Leenders, N., F. A. Ramierz-Marrero, B. A. Smith, C. Munford, T. E. Kirby, and W. M. Sherman. Evaluation of a step counter in African American children (abstract). Med. Sci. Sports Exerc. 29:S72, 1997.
9. LeMasurier, G. C., A. Bieghle, C. B. Corbin, et al. Pedometer-determined physical activity levels of youth. J. Phys. Act. Health 2:159-168, 2005.
10. LeMasurier, G. C., and C. Tudor-Locke. Comparison of pedometer and accelerometer accuracy under controlled conditions. Med. Sci. Sports Exerc. 35:867-871, 2003.
11. Loucaides, C. A., S. M. Chedzoy, and N. Bennett. Differences in physical activity levels between urban and rural school children in Cyprus. Health Educ. Res. 19:138-147, 2004.
12. Ogden, C. L., K. M. Flegal, M. D. Carroll, and C. L. Johnson. Prevalence and trends in overweight among US children and adolescents, 1999-2000. JAMA 288:1728-1732, 2002.
13. Pinhas-Hamiel, O., and P. Zeitler. The global spread of type 2 diabetes mellitus in children and adolescents. J. Pediatr. 146:693-700, 2005.
14. Pucher, J. Transportation trends, problems, and policies: an international perspective. Transportation Res. A Policy Pract. 33:494-503, 1999.
15. Pucher, J., and L. Dijkstra. Promoting safe walking and cycling to improve public health: lessons from the Netherlands and Germany. Am. J. Public Health 93:1509-1516, 2003.
16. Schneider, P. L., S. E. Crouter, O. Lukajic, and D. R. Bassett. Accuracy and reliability of 10 pedometers for measuring steps over a 400-m walk. Med. Sci. Sports Exerc. 35:1770-1784, 2003.
17. Schneider, P. L., S. E. Crouter, O. Lukajic, and D. R. Bassett. Pedometer measures of free-living physical activity: comparison of 13 models. Med. Sci. Sports Exerc. 36:331-335, 2003.
18. Shields, M. Overweight and obesity among children and youth. Health Rep. 17:27-42, 2006.
19. Strong, W. B., R. M. Malina, C. J. R. Blimke, et al. Evidence based physical activity for school-age youth. J. Pediatr. 146:732-737, 2005.
20. Transportation Research Board. Making Transit Work: Insight from Western Europe, Canada, and the United States. Washington, DC: National Academy of Sciences Press, pp. 1-170, 2001.
21. Tremblay, M. S., J. D. Barnes, J. L. Copeland, and D. W. Esliger. Conquering childhood inactivity: is the answer in the past? Med. Sci. Sports Exerc. 37:1187-1194, 2005.
22. Tremblay, M. S., P. T. Katzmarzyk, and J. D. Willms. Temporal trends in overweight and obesity in Canada, 1981-1996. Int. J. Obes. 26:538-543, 2002.
23. Trost, S. G., R. R. Pate, J. F. Sallis, et al. Age and gender differences in objectively measured physical activity in youth. Med. Sci. Sports Exerc. 34:350-355, 2002.
24. Tudor-Locke, C., S. A. Ham, C. A. Macera, et al. Descriptive epidemiology of pedometer-determined physical activity. Med. Sci. Sports Exerc. 36:1567-1573, 2004.
25. Vincent, S. D., R. P. Pangrazi, A. Raustorp, L. M. Tomson, and T. F. Cuddihy. Activity levels and body mass index of children in the United States, Sweden, and Australia. Med. Sci. Sports Exerc. 35:1367-1373, 2003.
26. Welk, G. J., J. A. Differding, R. W. Thompson, S. N. Blair, J.Dziura, and P. Hart. The utility of the Digi-Walker step counter to assess daily physical activity patterns. Med. Sci. Sports Exerc. 32:S481-S488, 2000.
27. Wyatt, H. R., J. C. Peters, G. W. Reed, M. Barry, and J.O.Hill. A Colorado statewide survey of walking and its relation to excessive weight gain. Med. Sci. Sports Exerc. 37:724-730, 2005.


©2007The American College of Sports Medicine