Dietary Intake of Women Runners R. R. Pate, R. G. Sargent, C. Baldwin, M L. Burgess School of Public Health, University of South Carolina, Columbia, SC 29208
R. R. Pate, R. G. Sargent, C. Baldwin, M. L. Burgess, Dietary Intake of Women Runners. mt j Sports Med,Vol 1 l,No6,pp461—466, 1990. Accepted: February 12, 1990
One hundred and three women who were habitual distance runners and 74 age-matched physically inactive women provided three-day diet records. Intergroup dif-
ferences in intakes of energy nutrients, micronutrients, cholesterol and fiber were evaluated via analysis of variance
and, to assess qualitative differences, via analysis of covariance with total caloric intake entered as the covariate. Women runners reported consuming more carbohydrate
(192.4 vs 165.0 gd 1) and less fat (57.5 vs 66.1 gd ') than did the inactive women (p < .05). After controlling for the non-significant intergroup difference in caloric intake, these differences persisted and protein intake was lower in
the runners. Cholesterol and saturated fat intakes were lower and fiber intake was greater in the runners, with and without control for differences irs caloric intake. These data
suggest that female runners, when compared to inactive counterparts, tend to follow dietary practices that conform more closely to the current recommendations of health authorities. Key words
Physical activity, exercise, diet, energy nutrients, iron, calcium, cholesterol, fiber
The dietary habits of female runners are of interest for several reasons. First, there is a common perception
that habitual exercisers, like regular runners, exemplify the lifestyle that is currently recommended by public health officials. However, previous studies of associations among preventive health behaviors have yielded conflicting results (7), and only a few previously published investigations have been designed in a manner that would allow testing the hypothesis that the diets of women runners are qualitatively different from those of inactive women (5, 21). In the athletic domain, the eating habits of physically active women have been of interest to sports medicine specialists, exercise scientists and coaches. This interest is based on the established link between diet and endurance performance (4) and on the concern that physically active women may be at particular risk for diet-related health problems such as iron deficiency conditions (11, 16,26), osteoporosis (15, 17, 28) and eating disorders (8,9,29).
Despite the popularity of distance running and widespread concern about diet-related health problems in active women, little is known about the dietary practice of typical adult women runners. Most previous studies of dietary practices in active women have focused on highly competitive athletes and have included small groups of subjects (10, 13, 18, 27). The purpose of this study was to observe the dietary practices of a relatively large group of typical adult women runners. Comparisons were made with the dietary habits of an age-matched, physically inactive group of women.
Subjects Two groups of adult women were recruited for participation in this investigation. A group of regular runners
(N 103) was assembled by soliciting participation among
Over the past two decades a large number of women in the United States have become regular participants in endurance exercise activities. One of the most popular activities has been distance running. This is evidenced by reports that in 1985 over 50,000 women finished 10-km road races (32)
and by national surveys indicating that running is one of the most popular fitness-generating activities among women as well as men (31).
Int.J. Sports Med. 11(1990)461—466 GeorgThieme Verlag Stuttgart New York
women who were either members of runner's clubs or entrants in running events. The inclusion criteria for the runner group were reported running on a regular basis for at least the previous six months and running at least 110 mm in the seven days previous to intial collection of data. The runners included in
the study reported 2.75 2.01 ( SD) h of running in the previous seven days. Thus the typical runner in this group was running on a very regular basis but was not training or competing at a highly competitive level. A comparison group of physically inactive women (N = 74) was formed by soliciting volunteers from local women's organizations and employee groups. A subject was considered inactive if she reported participation in vigorous, aerobic exercise with a frequency of no more than once per week. Subjects provided informed consent to participate in the study in accordance with the procedures of the University of South Carolina Institutional Review Board.
Downloaded by: Queen's University. Copyrighted material.
462 mt. J. Sports Med. 11 (1990)
R. R. Pate, R. G. Sargent, C. Baldwin, M. L. Burgess
Table 1 Descriptive data ( SD) on the two groups included in the present study and for U. S. women in the 25—34 year age range as
Table 2 Physical activity habits of the two groups as determined from a seven-day activity recall; data are expressed as w±SD
observed by the National Center for Health Statistics (22) Variable
Weight (kg) Height (cm) Age (years) Triceps skinfold
61.5± 9.7 163.4± 5.7 31.4±10.4 22.5± 7.6
Type of activity
Moderately hard activity (hours/week) Hard activity (hours/week) Very hard activity (hours/week) Running + (hours/week)
U. S. Women
64.2± 15.0 163.1
3.73±2.67' 2.75 2.01 *
'Significantly different from inactive group (p < .05) + Running time is included in very hard activity time.
'Significantly different from inactive group (p < .05).
Subjects reported to the laboratory on a Friday. Height, weight and triceps skinfold thickness were measured and subjects completed a questionnaire providing information regarding demographic characteristics, health history and general dietary practices. A seven-day physical activity recall instrument was completed and was used to quan-
tify recent participation in "moderately hard", "hard", and "very hard" physical activities (6). Subjects then met with trained personnel who employed standard techniques in obtaining 24 h dietary recall information. Also, subjects were given detailed and standardized instructions to complete an additional two-day dietary record. Subjects were instructed to follow their customary eating and exercise habits during these two days. The two-day food records, which included data for one weekend day, were returned to the investigators in a selfaddressed, postage-paid envelope provided during the clinic visit. This procedure was employed so as to provide three-day
dietary records that spanned two week days (Thursday and Friday) and one weekend day (Saturday). Subjects used similar procedures and forms for the one-day and two-day records,
except that the one-day record was completed in the lab, whereas the two-day record was completed at home. If any of the information on the returned records was ambiguous, the subject was contacted via telephone to obtain clarification.
Analysis of Dietary Information All diet record data were analyzed using the
Nutritionist II software package (N-Squared Computing;Silverton, OR, USA). When a food item was not represented in the Nutritionist II food bank, that item was broken down into its components, or a similar replacement item was entered. This was necessary in only a few cases (